(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Transactions of the Philological Society"

Google 



This is a digital copy of a book that was preserved for generations on library shelves before it was carefully scanned by Google as part of a project 

to make the world's books discoverable online. 

It has survived long enough for the copyright to expire and the book to enter the public domain. A public domain book is one that was never subject 

to copyright or whose legal copyright term has expired. Whether a book is in the public domain may vary country to country. Public domain books 

are our gateways to the past, representing a wealth of history, culture and knowledge that's often difficult to discover. 

Marks, notations and other maiginalia present in the original volume will appear in this file - a reminder of this book's long journey from the 

publisher to a library and finally to you. 

Usage guidelines 

Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the 
public and we are merely their custodians. Nevertheless, this work is expensive, so in order to keep providing tliis resource, we liave taken steps to 
prevent abuse by commercial parties, including placing technical restrictions on automated querying. 
We also ask that you: 

+ Make non-commercial use of the files We designed Google Book Search for use by individuals, and we request that you use these files for 
personal, non-commercial purposes. 

+ Refrain fivm automated querying Do not send automated queries of any sort to Google's system: If you are conducting research on machine 
translation, optical character recognition or other areas where access to a large amount of text is helpful, please contact us. We encourage the 
use of public domain materials for these purposes and may be able to help. 

+ Maintain attributionTht GoogXt "watermark" you see on each file is essential for in forming people about this project and helping them find 
additional materials through Google Book Search. Please do not remove it. 

+ Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just 
because we believe a book is in the public domain for users in the United States, that the work is also in the public domain for users in other 
countries. Whether a book is still in copyright varies from country to country, and we can't offer guidance on whether any specific use of 
any specific book is allowed. Please do not assume that a book's appearance in Google Book Search means it can be used in any manner 
anywhere in the world. Copyright infringement liabili^ can be quite severe. 

About Google Book Search 

Google's mission is to organize the world's information and to make it universally accessible and useful. Google Book Search helps readers 
discover the world's books while helping authors and publishers reach new audiences. You can search through the full text of this book on the web 

at |http: //books .google .com/I 





LIBRARY 



OP THt 



UNIVERSITY OF CALIFORNIA. 



T{eceived 



tAccession No. 




, '89 t 





. Class No. 



i 
I 




-■^ 



^ :^ 



M .^ 



t*3?<» 



PHILOLOGICAL SOCIETY, 



1891-4 




HERTFORD : 

rRIHTBD BT ITBPHBN AVITIN AND SOMB. 







CONTENTS. 



PAOB 



I. — On the Metrical Glossaries of the Mediseval Irish. 

By Whitley Stokes, D.C.L 1 

II. — The Celts and the other Aryans of the F and Q 

Groups. By Prof. John Rhts 104 

I II. — Notes on English Etymology. By the Rev. Professor 

Skeat, Litt.D 132 

IV. — On the Bodleian Fragment of Cormac's Glossary. 

By Whitley Stoeies, D.C.L 149 

V. — Note on the Pronunciation of the English Vowels 
in the Seventeenth Century. By Etjssell 
Martineau, M.A. 207 

YI. — The Greek Indirect Negative. By E. R. Whaeton, 

M.A 211 

VII. — The Compensatory Lengthening of Vowels in Irish. 

By Prof. J. Strachan, M.A 217 

Appendix. — Reports on the Progress of the Society's I^ew 
Engli$h Dictionary — 

1. By Henry Bradley, M.A., president 261 

2. By J. A. H. Murray, M. A., LL.D., vice-president 268 

3. Etymologies of some Co- words by Dr. Murray 279 

VIII.— Etymologies by Prof. J. Stbachan, M.A 289 

IX. — On the Assimilation of Pretonic iVin Celtic Suffixes. 

By Whitley Stokes, D.C.L 297 

X. — Old-Irish Glosses on the Bucolics. By Whitley 

Stokes, D.C.L 308 

XI. — Some Greek Etymologies. By E. R. Whakton, M.A. 329 



IV CONTENTS. 

FAOB 

XII.— Bare Words in Middle English. By the Eev. Prof. 

Skeat, Litt.D 359 

XIII.— The Old English Alliterative Line. By Prof. H. 

FKAJiK Heath, Ph.D 375 

XIV. — On Gaelic Phonetics. By J. H. Staples . . . . 396 

XV. — The Accentual Element in Early Latin Verse, with 
a New Theory of Satumian Metre. By W. M. 
liDfDSAr, M.A 405 

XVI. — Contrihutions to the History of the Deponent Verb 

in Irish. By Prof. J. Steachan, M.A 444 

^ Appendix II. — Notes on the Orthography of the Ormulum. 

By Arthur S. Napier, M. A., Ph.D l*-4* 

Index 569 

Treasurer's Cash Accounts : 1891 in Part I. 

„ „ „ 1892 in Part 11. 

„ „ „ 1893 in Part III. 

List of Meubebs, corrected to January, 1893 . . (see Part I.) 
List of Membebs, corrected to November, 1893 (see Part II.) 
List of Mekbebs, corrected to November, 1894 (see Part III.) 



TRAXSACTIOXS 

OP THS 



PHILOLOGICAL SOCIETY, 

189I-2-a 



L— OX THE METRICAL GLOSSARIES OF THE 
MEDIAEVAL IRISH. By Whitlby Stokes, D.C.L. 

With the exception of the modern Irish Celts, all races 
posseased of an ancient literature desire to understand it, and 
many races, including the Greeks,' the Norsemen and the 
Iriahy haye, at some stage of their civilization, taken a strange 
delight in Terse of which archaisms and wilful obscurities 
are the chief characteristics. It is, therefore, remarkable 
that the obyious aid of metre has not been oftener used to 
help the memories of the hearers, readers, or makers of the 
compositions aboTe referred to. But metrical vocabularies 
of rare or obsolete words belonging to the glossarist's mother- 
tongue ' are found, so far as I know, only among the Hindus,^ 
the Norsemen * and the Irish. The last-named people pos- 
sessed such glossaries as early, at least, as the beginning of 
the twelfth century. Thus in the vellum called Lebor na 
hUidre (the Book of the Dun), the scribe of which was 
murdered in the year 1105, we find, fo. 7^ 11. 19-22 : 

' Mur ' immed tall isind recht, ' coph ' buaid, is briathar Idnchert, 
' du ' bale, ' du ' duthaig lat, ' col ' ^ comet, is ' cul ' carpat. 

And further on, in the same column, 11. 36-38 : 

' To * ainm do maith is do mfad, ' H ' ainm d'ulc ocu» d'anHad/ 
' an ' fir, is n£ foriM fand, ' fath' mind, oetu ' fath ' ferand. 

^ See, for example, the Cassandra of Lycophron, which Soidas caUed o'lrortiy^y 
vodf^io. and which the Byzantines admired so greatly ; and consider the remarks 
of Prof. Sayce, in Mahaffy's History of Classical Greek Literature, on the 
artificial language of the Homeric poems. 

' The metri(^ glossaries composed in England and printed hy Wright are 
intended to teach the English Latin or French, not Anglo-Saxon. 

' I am informed hy Prof. Biihler that, with the exception of the so-called 
Vedic Nigha^tns, all the Sanskrit glossaries known to us are in verse. One would 
expect to find metrical glossaries among the Arabs. But I learn from Prof. D. 
H. Miiller of Vienna that they have none. **The most ancient Arabic 
gloasariee are arranged according to the last letters of the words. They were 
compiled for the use of poets, and the arrangement is intended to facilitate the 
composition of rhymed verses.** 

* See the Thulor printed in the Corpus Poeticum Boreale^ ii. 423 et seq. 

* This is eail in Rawl. B. 502, fo. 66» 2, and cal in H. 3. 18, p. 610^ and 
H. 2. 16, col. 686. 

* nooeus d*etriad. This is the reading of Rawl. B. 502, fo. 66* 2. 

FMl. trans. 1891-2-3. 1 



2 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

The same quatrains are found in the Liber Hymnorum 
ff. 34* 2, 34^ 1, and in Rawl. B. 502, fo. 56% which vellums 
also belong to the first half of the twelfth century. Further- 
more, in Rawl. B. 502, fo. 56» 2, and in H. 3. 18, p. 610% we 
find the following quatrain : 

' Dia' inchian, is coem in mod, is cfan uad o senfocul, 
ocus * derc * suil fri suairc son, oetu ' derc' in scathobor.^ 

The following stave occurs twice in Kawl. B. 502, S. 66^ 2, 
59» 2 : 

* Mos ' ar b^s robsB co cian, * buich ' ar brissiud, ni bsethchial, 
' sab ' ar tren tacrait doine, ocus * ong ' ar ecdine. 

This quatrain is found in a corrupt form in the Lebar 
Brecc, p. 92, marg. inf., and is quoted in modernized spelling 
by O'Clery. See Revue Celtique, v. 25. 

So in Cormac's Glossary s.v. bachall (LB. 264' 9-12) the 
following quatrain is quoted : 

' lath ' ainm do chlug consi f sbid, noco chelad in glanghaith, 
' bach ' buain ina dhortM tall, ' bricht ' ocus < bacc ' is bachall. 

The same quatrain is in H. 3. 18, p. 611% the last two 
lines being 

' bacht ' buain isin dorus tall, ' bricht ' minn ocus ' bach ' bachall. 

The Lebar Brecc, p. 188, marg. sup., has this stave : 

Is ' indless ' cech maithius mor, is * suan odar ' cech n-imrol, 
' folaid ' [ ] cech n-amrai, ni ba ri nach rodamnai. 

Again, in H. 3. 18, p. 611% we find the following six 
quatrains : 

' £ ' truagh in sc61 rodus-sgrus, uile bith ba h6 a duchus, 
is 6 roglic ar ' each n-eing, an t-' 6 ' breac dobf a mBoaing. 

' Ai' caingen, is ' ae" dan dron, donf each sai co solum, 
inti ba segdha ar in son bu sai can denma ar doman. 

1 The scribe adds : .i. in menb bis triasin Ags ngr^e. This quatrain is also 
in H. 2. 16, col 686. 
* no fris. 
' no ai. 



FRAGMENTS OF IRISH METRICAL GLOSSARIES. 3 

* Doctns ' foircthi each rechta, isna leabraibh lancerta, 

pcus * dinn ' [cain] each cnuasaigh, * dinn ' ainm each fir arduasail.* 

* Dibadh ' ecus * bath ' na menn, * ba ' oeus ' temel ' nar timeheall, 
ag sloinned ega gan f eall, as lor a med ron medhrann.' 

* Sneid ' each suaill snimach sund, 'seim' each seang, each sircumang, 
' sab ' each soabb each forba, oeus ' sab ' each comorba. 

* Tir dombaisi ' each sliab slan, * airbri' each n-iumat n-iumlan, 

* iath ' ferann, *iath' cloth ein coll,* iat[h] ' cloe oeus * iath* eochall. 

In the same MS., p. 612* : 

* Mos ' each sob6s sidhamail, * mos ' tuile, oeus * mo8 ' aithbe, 

* moB ' each eeol ciuin cichamail, na erinann cumsear caithme. 

* Robustus ' each sonairt elan, * reos ' each bidbadh, buan a bdgh, 
' seis ' each sofis sloindit sain, is ' castus ' each gnim genmnaid.^ 

Is ' cai ' eonair cacha huird, ' ae * each ceird na eumgaid buirb, 
cain rofitir * eonruiter,' * reeht ' dfreeh* na dianbruiter. 

' Caissi ' miscais, ' caisi ' sere, mar innisit liubair MnCert,' 
doscail nert na tromslaaig de, dias d[i]ana[d] comdhual eaisi. 

' Glinne ' na laegha rit M, ' glinne ' na bu bleaehtmara, 
0CU8 * glinne ' in luaide, * glinne ' sut[h] • eo sir-buaine. 

Synonyms, also, were treated metrically by the mediaeval 
Irish. We have already had a stave on four words, for 
' death.' So in the Lebar firecc, p. 92, marg. inf., we find 
the following quatrain on four words meaning ' good ' : 

'Dag' oeu$ 'fo/ clu cen brath, ' so' is *mo/ een cob gnathach, 
anmanna sin do maith mas, derb Hum ni saeb in senchas. 

The following specimen of a glossary of synonymous 
words of different genders is taken from Bawl. B. 502, 

> is 'dinn' ainm each ardhuasaily RawL B. 502, fo. 57* I. 

^ Here the scribe adds : Dibad 7 bath 7 baa 7 b(i 7 eel 7 bas 7 macht 7 ort 
7 teme de [leg. ix P] nomina mortis. 

> MS. geanmnaigh. 

* MS. reacht direach. 

* For mar innisit liubair read with H. 2. 16, col. 692, indisith libur. 

' .i. lacht. The above quatrains from H 3. IS, pp. 611, 612, are taken, not 
from the MS., but from Curry's transcript, an anastatic copy of which is in 
the Bodleian. 



4 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

fo. 58* 2. It occurs also in Cormac's Glossary, s.v. adba 

otIiDoe : 

Ibs ^ in ' lia/ lith rolass, iar srethaib suad hi senchas, 

[ijs ed * onn ' iar n-aicniud ail, is i * cloch ' iar sairdataid. 

Whence we learn that lia * stone ' is he, i.e. masculine, 
cloch * stone ' is she, i.e. feminine, and onn * stone ' is * it ' 
i.e. neuter. Verses dealing with the synonyms for water, 
fire, sea, wolf, horse, shield, spear, head, eye, truth, etc., will 
be found infra. 

Homonyms, too, were not forgotten. Thus the Yellow 
Book of Lecan (H. 2. 16), a MS. in the library of Trinity 
College, Dublin, contains in col. 121 the following quatrain 
on the four different meanings of saeglonn ('judge,' 'old 
man,' * king,' * column ') : 

' Saoglonn ' brithemh, buan a blad, ' saoglonn ' senoir sir-saoglach, 

* saoglonn ' each ri fora du, ocus * saoglonn' columnu. 

The same MS., col. 693, has the following stave on the 
three meanings oiferb (' cow,* * blotch,' * word ') : 

' Fearb ' as ainm do boin iar fir, oem do boilg can imshnim, 
ocus cen uair n-eithtyA n-airc, da breithir can cbantobairt.^ 

So we have had, pp. 2, 3, two quatrains dealing with the 
different meanings of ^ and mo8y and we shall find below 
verses dealing with the different meanings of triath, tuirighin, 
eo, etc. 

Besides metrical glossaries of obsolete Gaelic words, of 
Gaelic synonyms, and of Gaelic words with different mean- 
ings, the mediaeval Irish had glossaries in verse of Latin 
vocables. The following specimens will suffice : 
' Keus'^ each bidba cona blaid, ocus * castus ' cmH genmnaid,' 

* rectum ' csLch ndiriug,* dal cirt, ocus * robustus ' sonirt. 

Rawl. B. 502, fo. 57» 2. 

* Condio ' saillim suarccote,* ocus * doctus ' each forcthi, 
' custos ' cometaid rom-char, octu * oboedens ' humal. 

Ihid.io. 56^ 1. 

' This qaatrain is cited in O'Clery's Glossary, s.v. Fearb, £ev, Celt. iv. 416. 
> MS. Keos. 

* MS. ngenmnaid, but geiim[ii]aid, H. 2. 16, col. 691. 
« MS. duiud, but in H. 2. 16, col. 69, direch. 

* suarcaide, H. 3. 18, p. 613*. 



DESCEIPTION OF THE GLOSSARIES NOW PUBLISHED. O 

* Fero ' taircim dar each leth, oeus * ligo ' cech cumrech, 

* finem ' in crich coir is lib, ocus * nouit ' rofitir. 

Ibid. fo. 58» 2. 

* Reatus ' bibdanos baig, ocus ' demitus ' digbail, 

' mulgeo ' * bligim cen baegul, is * imber ' cslcH ard-broenud. 

Ibid, fo. 58^ 2. 

The metrical glossaries which I have now the honour to 
lay before the Philological Society are three in number. 

I. The first is commonly called Farm Focal, 'knowledge of 
vocables/ from the words with which it commences. This 
glossary contains 75 quatrains, and defines about 350 words. 
The first ten quatrains are here given in two recensions, one 
from p. 395 of the Book of Leinster, a vellum in the library 
of Trinity College, Dublin, the other from fo. 95 of the Stowe 
MS. No. III., now in the library of the Royal Irish Academy. 
Page 395 of the Book of Leinster was written in the sixteenth 
century. The glossary beginning on fo. 95 of the Stowe 
MS. is thus described by Dr. O'Conor, Bibl, M8. Stowensis, 
1818, vol. i. p. 52 : 

"Fol. 95. — O'Duvegan^s Metrical Dictionary of Ancient 
Irish Words, beginning Forus Focal luaiter libh. O'Duvegan 
died old in 1372. It may be doubted whether any Dictionary 
of the Northern languages is older than this. The verses are 
240.' The transcript is by Cormac og o Corrain in the 
beginning of the last century. But it is written with the 
greatest care, by one of the best Irish scholars of his 
age. 

Upon which it is to be remarked that the colophon proves 
the scribe's name to have been Corman og 6 Cormin : that 
he wrote in December, 1734,' and that although native 
Irish scholars agree in attributing the Form Focal to John 
O'Duvegan, there is not, so far as I know, any trustworthy 
evidence for such attribution. The Stowe MS. omits quatrains 
56-66 both inclusive. 



^ MS. mulceo. 

• The quatrams are 60. — ^W.S. 

s Goiman og 6 Corrnin ro scribh sia a Mi na nodhlac aiifto 1734. 



b IRISH METRICAL GLOSSARIES. — MR. STOKES. 

Two other MSS. have been used in forming the text, viz. 
a paper MS. in the library of Trinity College, Dublin, marked 
H. 2. 12, No. 6, and written in 1698,^ and a small quarto 
paper MS. in the library of the Royal Irish Academy, marked 
23. L. 21. Of these the former is here denoted by H., the 
latter by L. H. is written in an Irish hand, by Eoghan mac 
Gilleoin for Master Lachlan Campbell at Campbeltown, in 
Argyle. This scribe was grossly ignorant, and his work 
needs much emendation. The bulk of L. was written about 
1787 ; but it contains four pages in the handwriting of 
Malachi, a brother of the late Eugene O'Curry. 

Thirty quatrains of the Forua Focal have already been 
published, to wit : 

a. Twenty- three by O'Reilly in his Irish-English Dictionary 
Dublin, 1821, s.w. aidhbeis (15), Aodh (39), Art (34), caise 
(76), cobhra (37), colg (36), crom (40), dreimne (17), duirbh 
(26), eigh (20), fiodhrach (43), gabhar (19), gaodh (23), 
garbh (77), osar (60), pattric (19), reacaire (62), ren (42), 
risidhe (12), robhar (47), scuird (35), searrdha (49), snuadh 
(45), tinfeadh (64), troghan (14), tuaithcheall (11). The 
quatrain cited s.v. gabhar is repeated s.v. pattric, 

b. Seven by O'Donovan in his Supplement to O'Reilly's 
Dictionary s.w. breas (6), dreann (9), eo (5), fead (8), fo (4), 
ruiceadh (7), triath (2). 

With these exceptions, no part of the Forua Focal has 
been published. 

II. Our second glossary takes its name from its first line 
Deirbhiiur don eagna inn Sigsi, * Poetry is sister to Wisdom.' 
The copy now printed contains 62 quatrains, and defines 
about 193 words. It is taken from a photograph of the copy 
of this glossary in the library of the Royal Irish Academy, 
marked 23. L. 21. This copy contains corrections by 
Peter O'Connell, and was written about 1787. The various 
readings have been derived from the above-mentioned MS. in 
the library of Trinity College, Dublin, marked H. 2. 12, No. 6. 

* For a loan of this MS. I am indebted to the kindness of the Board of Trinity 
College. 



DESCRIPTION OF THE GLOSSARIES NOW PUBLISHED. 7 

The Derbhiiur glossary must have existed before 1643, for 
Michael O'Clery used it in compiling his Focldir^ published at 
Louvain in that year.^ 

Three other copies are known : 

(1) A vellum in the library of the Royal Irish Academy, 

23 
marked ,^— - , of which I do not know the date. This copy, 

Jr. o 
says Mr. Gilbert, the learned librarian of that institution, is 
" written in very bad ink — almost faded — wants portion at 
end, and is in a poor style of penmanship." The glossary 
begins on fo. 19. 

(2) A paper copy, transcribed at Limerick in 1768 by 

Andrew MacMahon, with corrections by P. O'Connell, in 

23 
the library of the Royal Irish Academy, marked ^ — — r. The 

glossary (62 quatrains) begins at p. 52. 

(3) Another paper copy in a large quarto MS. in the same 

23 . 
library, marked ^^oo* ^^ *^® handwriting of Michael 6g 

O'Longan, between the years 1805 and 1832. The glossary 
begins at p. 179 and contains 65 quatrains. 

I cannot find that any part of this glossary has ever been 
published. 

III. Our third glossary is unfortunately only a fragment and 
often corrupt. It is taken from fo. 1 7* of a vellum in the British 
Museum, marked Egerton 90, contains 29 quatrains, and de- 
fines about 154 words. O'Curry, in his MS. Catalogue of the 
Irish MSS. in the Museum, says that this glossary '^ appears 
to be part of a metrical glossary called Forus Focal, generally 
ascribed to Shane Mor O'Dugan,^ chief OUave of Hy Maine, 
who died in the year 1372. This and the three following 
folios are in his handwriting, and are part of the ' Book of 
Hy Maine,' now in the possession of Lord Ashburnham." 

But though the metre in which the Egerton glossary 
is written is the same as that of the Form Focal, the 

1 See Bevue Celtigue, iv. 364. 

' O* Conor's ' 0*Dayegan.' The Irish spelling is Dnbhagldn. 



> 



<• <. 



K 



.-?N 



/• ".•'" o:' r.rr, ^ ^. N 






8 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

words explained are different, and O'Curry's suggestion is 
therefore groundless. The writer of a prose wordroU in the 
Book of Lecan, a vellum in the library of the Royal Irish 
Academy, appears to have used a. good copy of this Egerton 
glossary. If so, it must have existed before 1416, when, accord- 
ing to O'Curry, the Book of Lecan was compiled. But the 
occurrence in it of firat^si, a loan from the Middle -English breche, 
shows that it cannot have been written much before 1300. 

No part of this glossary has been published. 

The three metrical glossaries will now be faithfully set 
forth. The third is supplemented by the corresponding 
portion of the prose vocabulary in the Book of Lecan. 
Then follows an alphabetical index to the glossed words, 
including (a) references to the extant Irish prose glossaries, 
(b) belegatellen from the Irish literature,^ and [c) such 
etymologies and comparisons as seem fairly probable. 

I. FoRus Focal. 

[0 Dubhagdn .i. Sean, c^cmt^.*] 

Book of Leinster, p. 395. Stowe MS. No. III. fo. 95*. 
(Cited as LL.) (Cited as S.) 

1. Forw* focul luaidht^r libh, 1. Foras focal luaiter libh, 
a eolcha in dana dlighidh !' a eolcha dhdna in dlighidb ! 

gach n{ anois ara bhuil ainm ga(;h ni a nois is ainm 

caidhe agaibh a thsenainm ? caidhe aguibh a senainm ? 

2. 'Triath' righ go. rogha 2. 'Triath' rfgh go rogha 

ndealbha, ndealbhna, 

'trfath* ainm do each tigh- Hriath' ainmdoga<?htig<fma, 

ema,* 

[is*] *trfath'tond co n-ilwr 'trfath' tonn go n-iolar a 

ndath, dath, 

*triath' tore trom, is Hriath' *triath' turc [trom] is 

tula(;h. * triath ' tulach. 

^ For eighty-two of these I ara indebted to Dr. Euno Meyer. He also 
pointed out to me the quatrain in LB. 188, cited supra p. 2. 
' «»<?, L. 

» dUghti, LL. dhirigh, L. 
^ tiagherma, LL. 
• *u?, L. 



FORUS FOCAL. 



8. 'Tuirighin' ri^ ruamna* gal, 
' tuirighin' brtf^Aeamh blath- 

* tuirighin ' tuir fuilnges * 

tech, 

* tuir[i]ghiii ' teangadh tuir- 

meach. 

4. * F6 ' ainm do mhaith is do 

mhiadh, 
*fi' ainmd'ulc ocus* d'aimh- 

riar, 
' an ' fir, is ni fonis fann, 

* aoth ' minn, ocus * iath ' 

f^rann. 

5. * £o ' dealg oeut *eo' iubar, 
' eo * eigne nach eisidhan,® 

* aedh * octts * tnu' ^ teine tra, 

* bolg' ® bema, ocu$ * ladhg ' 

sne&elitsL, 

6. 'Bres' is» *oU' gachnl as" 

mor, 
octu * breas ' " gach greadhan- 
glor. 

* irchaill ' *' ursa buidhne 

hechtf 
ocus * toicheall ' " gach n- 
imthechL 

7. * Ruiced,* " ni haiom gan 

monur, 
do thogbhail is d'ardugud, 

* ««, L. righ, LL. 



3. * Tuirighfn ' rf ruamnt^^ gal, 
'tuirighfn' breithemh bladh- 

mhar, 

* tuirighfn ' tuir fuilnges 

teach, 
is * tuirighin ' tenga thuir- 
mhech. 

4. * Fo ' ainm do mhaith is do 

miadh, 

* fi ' ainm d*ulc ocus * d'aim- 

riar, 

* in ' f irfios, nf foras fann, 

* aoth ' minn, ocus * iath ' 

feronn.* 

5. * Eo ' dealg, 7 * 6o ' iubhar, 

* eo ' eigne nac[h] eisiodhan, 
*aodh' ocus *tnuth' teine thra, 

* bolg ' bemna, is * ladhg ' 

Bnechta, 

6. * Bres ' ocus * oU ' gach n£ is 

mor, 
ocus 'breas' gach greadhan- 
ghlor, 

* iorchuill ' ursa buidhne 

hechtf 
ocus 'toicheal' gach n-im- 
ihecht, 

7. *Ruichet,' nf hainm gan 

mhonar, 
do thogbhail is d'airdughutHi, 



ngr 
' maomus, L. 

' fuilnges, L. 

* ifl, LL. S. 

* Identical with the quatrain from LU. 7^ cited supra p. 1. 

* eisighan, LL. eisiodhan, L. 
' tnuth, L. 

* bladh, L. 

' aicy L. ocus, LL. 

*" #«?, L. is, LL. 

** «<?, L. bras, LL. 

*' earcuil, L. 

'* toichim, L. and 0*Don. Supp. s.v. breas, 

** micheat, L. 



10 



IRISH METRICAL GLOSSARIES. — MR. STOKES. 



is ainm ' fuirmheadh ' os ^ 

gach mud 
do tumadh ' is d'fsliugui. 

8. * Fet ' ainm d'innisin iar- 

s[o]m, 
*dil'> farraid/ 'foacht'*^ 
fiarfoigh,* 

* riadh ' ^ rith, is * riadh ' ' 

8(maclit gan on), 
*drucht'* 6irglii, is 'drecht'* 
airchetol." 

9. 'Drenn' deabhaidh," is 

* dreann' doilghis, 
'grind*" daingen, is *grinn*" 

cuibh[dh]ea8/* 
'xnont/r' gach nfdonf neach," 

* gle ' glan, octM * gle ' 

gleithech." 

10. * Coindealg ' comairle nach 
cam, 

* f uidhair ' briathar baile 

bann," 

* rose ' tuigsin gach neich 

ma le," 
'lothar'^' gach ciall incb^he. 



is ainm ' fuirmh^^h ' os go^h 

modh 
do thurnoi^ is d'fsliught/ih. 

8. ' E6t ' ainm d'innisin iarsin, 
'ail* iaruigh,fo(?A^ ffafraighe, 

* Ha ' rith, is * Ha ' smacht 

gan on, 

* dracht ' 6irghe, is ' drucht ' 

airdtf^hadh. 

9. Oreann deabaufh, is greann 

doilghes, 
*grinne' daingen, is 'grinn' 

coibhdhes, 
*monar' gach ni do n{ neach, 
' gle ' glan^ is * gle * gl6thech. 

10, 'Coinnealg* comharle naeh. 
cam, 

* f uidher ' brfathar builidh 

bann, 

* rosg ' tuigsin gacA neith 

male, 
*lathar' gach. cfall n{ cleithe. 



^ siCf L. OS, LL. 

' thurnamh, L. 

' aiCf L. ail, LL. 

^ i&iTuidh, L. iarraigli, LL. 

* fothocht, L. f6cht, LL. 

^ fiarfaighe, LL. fiarfruigli, L. 
"* siCf L. ria, LL. 

* dmchty L. dracht, LL. 
' dr^hty L. drucht, LL. 

1" airiugtM?, LL. The quatrain is thus quoted in O^DonoTan*8 Supplement 
s.y. Fead: Fead ainm d'inisin iar soin | Ail iarraidh, fothacht fiarfoidh | Kiadh, 
rith, is riadh smacht ran 6n | Drucht ergeadh, is dreacht airchedal. | For. Focal, 

>^ deabhuidh, L. deabhaigh, LL. 

w grinn, L. grinde, LL. 

*' aie. L. ffrinne, LL. 

1^ cuimhdecht, L. 

** HCf L. nocA, LL. 

1^ aiCy L. ^leithach, LL. 

^^ In L. this couplet runs thus : * Coindealg * comhairle iar sin, is ' fuighioll ' 
briathor bhuilidh. 

^^ HCf L. maleith, LL. 

i» l&thor, L. 



FORUS FOCAL. 11 



Stowe MS. No. III. fo. 95^ {continued). 

11. ^Droeht* gach dubh, *drocht^ gach dorchay ^edroeht' gach 

glan gach sorca, 
'tuaichiol' ba hainm do ghliocus, is 'edtuaicheal'^ aimhghli- 

OCtM. 

12. 'Reisi" ainm [each] sgeil gan chaire, *reisidhe'' ainm [do] 

sgeluidhe,* 

* aidhbhsi * ' ceol, * sceo* tuicsi ad chlos, * rosal ' breath,' * basal * 

diomtM. 

13. • De' is ' deichen ' gach daP dle&cht, 'fabhaU' br6g, is ' dolbh ' 

dra6idhe(7A^, 
' gart ' eineach, is ' neoid ' gach gann, ocus ' seoid ' gach crodh 
coitchenn. 

14. *Bolg' bo, briathar *ferbh' anainm, *buich'® brisiodh is 

* tethra ' » ar bhdidhbh, 
don *® bhrad^n is comhainm * fach,' octM ' trogan ' ar 
bhrain[f]iach. 

15. ' Bior' ocus *aii' is 'dobhar,' tri hanmann uisge in domhuin," 

• lear,' 'aibheb/ *bochna' bladha," anmanna gach" drdmhara. 

16. ' Faol,' *cuib,* is" *luan' ar chonuibh, a ttri senainm iar 

sodhoin, 
dha senainm ar bhuaibh bladha, 'ferbh' is 'laithri' lionn- 
mhara." 



^ ^ttoaichiol, L. 

' lis, L. and 0*£. s.v. Risidhe. 

' risidhe, O'R. risighe, L. 

* ainm do s^alaighe, L. 
^ »u^f L. aibhsi, S. 

* roisiol br&th, O'R. rasal breith, L. 
7 aoi ifl caingen gach d&il, L. 

® nCf L. buith, S. 

* teta, S. tethra, L. teatra, H. 
»« *«r, L. do, 8. 

^^ tri hanmanna duisge ar domhan, L. 

1* blagha, S. bladha, L. and H. 

^ 8ie,L. tri hanmann, S., but anman do gach, H. 

»* #M?, L. Om, S. 

lA lionmhara. 



12 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

17. DhaV senainm ar mhuic mhiadhuigh, 'feis'* is ' mada ' 

moirdhiamhuin,' 

* ceathnaid * * caora, * dreimne ' • gal, ' seghach 'is * * cadhla ' 

gabhar. 

18. *Peat" ainm d'oirfidedh gan bhrath, *c^m' buaidh, ocus® 

Hlas ' » oenach,*® 
' neas ' ainm ratha, li nach lag, sen-ainm na slighed ^^ ' rdmhad.' 

19. * Gabhar* oeus *mairc'" is *peall* ar na heacha»3h cantur 

ann, 

* paitric * cennsridn, carmad ^' * cab,* " * ulaid ' " sratbar, * cul ' 

20. *Grith* grian ocus 'eig'" esga,** *colt' bfadh mar agras^* 

eigsi, 
'lothar' edach, li** n-amhra, do eineach'*^ ba bainm * easla- 
bhra.' 

21. *!N'eid' ainm catha, *flann' [ainm^] d*fuil, *lear' iomad, 

ainm d*aon * uathadh,' ^ 

* grib * ** luas, is michert" in modh, * ridnacht ' ^* ainm do 

thiodhlacadh. 

* Da, L. 

* seis, L. 

' m6rdhiainbair, L. 

* cethnaid, L. 

* dreimlme, L. 

* 8ead gach slighe, L. 
' Peit, L. 

8 is eige, S. is 6ag, L. 

* tlacht, L. 

^^ eineach, S., but aonach L., and cf. 0*C1. s.v. tlas. 
1* sleighe, S. sUgheadh, L. 
^' marc, L. 

" carma, H. is cearma, L. leg. is earr, as in 0*R. s.v. pattric. Ajid so 
P. O^ConneU in marg. of L. (is carr cap). 
** ceap, L. 
** nc, L. ulad, S. 
*• carbat, L. 
*' is eige, S. 
'* is 6ag ^ga, L. 

'» adhrwa, S. agrus, H. adhras, L. agras, O'R. s.v. eigh. 
«> lith, L. 

" inneach, S. doineach, L. 
** siCf L. 

*^ aon ainm uathaidh, L. 
** cirb, L. 
•* m6ircbeart,L. 
** righnecA^, S. riodhnacbt, L. 



FORUS FOCAL. 13 

[fo. 96».] 22. *Calb'* cruas [is] 'naoinneall'' gaisg^dh, ^eallamh'^ 
ingiiadh* no^h aisder,^ 
'galann' gath namha' go nert, ocus 'barann' gach b6im- 
neart. 

23. *N(a' trenfear/ is *mal' mflidh, *lulghach.' gach® Idoch go 

bhfiorbhladh, 
' cusal ' gach coimhneart gan choir, 'gdodh' ocus *ge6nadh* 
gerghuin.' 

24. ' Sgal ' ^® 0CU8 * arg,* luait^ leat, ^m (?) anmann " na bhfer go 

heehty 

* ainner ' is ' frac,' radh gan cheilg,^ ama mnaibh 'san tsengaoi- 

dhUg." 

25. 'Bath* octu *ort' marhadh ier^ 'cearr' ocus *ciochladh'" 
ciorrbaih, 

* eanglonn ' " gabhadh " nocha go, ocus fuirech *fumaidheo.' ^"^ 

26. * Ana ' saidhbhrios ^ iarmotha, ocus * una ' " ainm do ghorta, 

* duirb ' " gach galar immalle,'^ * easdoth,® ba hainm do 

shlainte.^ 



1 cakdh, L. 

* is naoinneal, L. 
' tUj L.ealla, S. 

* iongnadh, L. eangiuidh, S. 

* tistear, L. 

* namhaid, L. 

' nCf L. ti^n, S. 
^ tiCf L. Ivlgachy S. 

* gaodh is geoghnadh gach gearghoin, L. 

11 anmaima, L. 

" ainder is fiacbra gan cheilg, L. 

>' sengfaaidheilg, L. 

" cachladh, L. ceachladh, O'Cl. 

^ easglaniiy L. 

" jir, L. gabha, S. 

** Mr, L. nirniiigbeo, 8. 

1^ aaidhbhreas, L. 

» fina, L. 

» duirbh, L. 

'^ iomalldth, 6. ima 1^, L. 

** eassaoth, L. 

*> This quatrain, omitted by S., is taken from H. and L. It is quoted by 0*R. 
f.T. duirbh. 



14 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

27. ' Cuanna ' ' cnoc is * coice ' * slfabh, ' ail ' cloch, ' tec " 

cnaimh, is ' conn ' ^ ciall, 

* thicht ' talamh comhadhbhal cain,' ' tabhartha ' ainm * 

tuarasduil.^ 

28. Sen-ainm na trfucha' 'fonn' tra, sen-ainm na tuaithe *forba," 
' dun ' ainm [do] baile, is bladh binn, is 'fath' sean-ainm ^^ gach 

f^uinn." 

29. 'Aincis'^^ is 'miscaith' namd, dhd senainm na^ malla^A^a, 

* ordit ' " heimsLcht ima le,'* * gesca ' " hi hainm do soillsi. 

30. * Cobh ' buaidh oeus " brfathar borr, ainm d'feoil * cdma' ecus ^* 

* s^coll,' 

* dfu ' cfan, is ' dwe * su[i]l abhus, * cul ' ^' coimh6d, is ' an ' *^ 

aoibhnios. 

31. * Faosamh ' is ' cuime,' '^ nf ch61, dhd ainm cumairce ** gan 16n, 

* oibid ' is * umhla ' go be^A^, dha senainm do[n] ^isgidhecht.** 

32. * Annoid ' eagluis in gach tan, * sgal ' ^ Idoch is * axal ' " uasal, 
'ailcne' ail** ar lW*uib soin,*' ocas'* *anno'*' ainm do 

bliadhuin. 

siCf L. cuadhna, S. 

nCf L. coidhce, S. 

sec, L. teach, S. 

siCf L. cunn, S. 

can, L. 

For tabartha ainm, H. has othar ba haimn. 

' othar * ainm do thuarastal, L. 

do thri(ichad, L. 

na ttuath * forbadha,' L. 

neas ainm diath, S 

* D(in * ainm do bhaile, is biadh * mann.' senainm * iath ' do gach fearann, L. 
Acais, L. 
do, L. 
oirbidh, S. 

* 6raoid ' ar bhemuu;A^ ma le, L. 
ggsca, L. 

Cobh buaidh agal, L. Cuibh 7 bCiaidh, S. 
is, S. agas, L. 
cdl, L. 

is an, H. agas kn, L. rian, 9. 
coimhcheasa, L. 
comairce, L. 

^ ^isdidhecA^, S. n. eiisgnidhee;^^, H. ^scaidheacht, L. 

»* sc&l, L. 

** acsal, L. 

'• * airtne ' oil, L. 

" ailcne 6il ar chloich mar sin, H. 

•« is, S. and L. 

" ne, L. ando, S. 



FORUS FOCAL. 15 

83. • Ara ' ainm giolla gan chol, is ' airrdhe ' ainm do leasnghadli,^ 
'aidhmirt'* ar gheis, is garbh gal, ocua *ainf§in' gaeh n-iongnadh. 

[fo. 96^] 34. Cniaid 'art' a senainm go ftr, is 'anart' ainm 
do mhaoithmbfn, 
' mioin ' ^ balbh, ' tdoi ' bodhar nacb dis, ' main ' gr^h [is] 
' anmain ' mioscais.^ 

35. *Fuan' brat, is 'rocan'* ionar, is * stiall'^ fuatbrog ro fionnadh, 
'scuird/ 'caimsi,'* leine gan on, *obbrat" hi hainm do 

cbenbbor.*° 

36. 'Gen'^' octu 'colg,' toluibh^' gal, dbd senainm cloidhimh 

curadh, 
'cealtair' doig^,^' 4uiblme' dbe, dha senainm gach dirdsleigbe. 

37. Ceitbre banmann** in sc6itb gan feall 'fraic,'^* 'cobbra,'" 

'failte,' 'finneall,' 
* dinfatb ' is " ' troniatb,' " go ttairm, don cbathbbarr is da 
senainm. 

38. Anmann^' in cbinn, is 6ol damb, tre sengbaoidbilg na bfil^^, 
'trull,' *coll,' 'itropa' trom[d]a, noeba coll a cbombfodbla.** 

39. *Aodb'" oeui * d^c ' octu *cais,' tri banmann in ruisg 

rionngblais, 
'bra' oeu8 <laba'^ na^b gnatb gairm, don mbala»^b is da 
senainm. 

' is airidin leasoghadh, H. is airidin lesaghadh, L. is airidin leasaghodh, 
0*R. 8.V. Garbh. 

* aimiirt, L. leg. airmit. 

' ainbfeiii, H. ainmh^id, L. ainffein, 0*£. s.v. garbh. ainbeidh, S. 

* maon, L. 

* maoin gr&dh is anmbaoia mioscais, L. 

* rotan, S. roc&n, L. 

"* giall, S. diall, H. stiall, L. and 0*R. s.v. scuird. 

* tie, L. cuimsi, S. 

* 6bnit, H. obhrat, L. 

^ cheannbharr, L. o ba hainm do cheannbhodar, O'R. s.v. scuird. 

" Tol, 8. ; but H., L. and O'R. s.v. colg have Gean. 

» tola, L. 

*' is, H. agas, L. 

1* hanmanna, L. 

» fraiF, L. 

^ eaochbhrat, S. cobra, H. cobhra, 0*R. s.v. eohhra, 

" 7 8. 

u dimaih is tr6ithiathy L. 

^ anmanna, L. 

^ Tromcholl, is troi>a tromdha, nach ar choll a chomhaifoirala, L. 
« odh, 8.=a6dh, H. 
^ bi^ is Inbha, L. 



16 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

40. * (Jreann ' ainm d'ulchain, Hth nach. lochtf * feac ' fiacuil 

treabha[r] * teiohhnocht, 
*com[m]ar' sron, *6'' cliias ganlen, *cui[n]8eal" aighedh* 
nach aimhreidh.^ 

41. * Coidhche ' ascal* farmothd, < eochair ' thenga thagartha,^ 
*luc,'® bni,' 0CU8 *guaire' folt fionn, *gulba/ is *bile'" bel 

bldithbhinn. 

42. *Glaim'" guala is 'doid' Idmh gan 16n, 'scibh'" glac, ocus*' 

* luibhne ' m6r, 
*lua'" cos, is *treathan'^* troigh, [*ren*] reisi, is *nena' ainm 
ordloigh.*' 

43. * Meallach ' is ' mdoin ' ocus " ' muaidh/ ionann sin is maith no 

miadh,** 
* grib * toirmeisc, [is] * c^bhach ' " creach, * fiodbradh ' nos,** 
'fireach'" biseach. 



[fo. 97*.] 44. * Tul ' ga<?b gnuis is fodhan aim, * caor ' cainneal,*^ 
is * cuar ' cam, 
* abrann ' olc is dirsan ann, * blach ' ^ saill is * salar * " salann.** 



1 tubhra, L. treabair, H. 

» 6o, S. 

' cuinnsi, H. 

^ aidhigh, S. 

^ cuinnse agbaidh nach kimbgheur, L. 

* caoiche is coll, L. 

7 teanga thagartha, L. 

^ luch, H. and L. 

« br6, L. 
10 Mic, L. bil, S. 
" glang, L. 
** 8cib, L. 

»* iCiagh, S. Inath, H. 

^^ treaghan, S. trethan, L. 

^^ orluidh, S. L. has reon reisi . nean ainm ordlaigh. 

" is, S. 

1^ Meallach agas maoin is rauadh, ionann sin is maith r^a iomlnadh, L. 
Melleadb is maoin agus muadh, Inann sin is maith, re imluadh, O'K. s.v. 
fiodhrach. 

^' toirmeasc, \b cearrbhach L. 

*> siCf L. no, S. 

'^ fioghradh n6s, is fiodhrach bisech, H. fiodhrach no fiodhradh biseach, O'R. 
fiodhradh n68, fiodhrach oiseach, L. 

'2 cuinneal, S. 

» leg. bluch ? 

»* ejOar, S. And so O'Br. salor, O'Cl. 

^ H. and L. omit this quatrain. 



FORUS FOCAL. 17 

45. ' Broth '» ainm f6ola, 'flann'' [ainm] d'fuil, 'grith' 6olt«, 

* fraic " folt farsin. 
'conn' ainm c6ille iarmotM, 'turbhuidh' ainm gac[b] urbhadha. 

46. 'Meamur' congna ferrdha^ fir, is 'caisi' ainm [do] gach 

caingin,^ 
' ain '* ainm doibhinn s^mli farsin,^ ocim 'aoin' ainm do luacbair. 

47. *Ribhar'* cHatbar, foram ngle, 'rumro*' radbarc go ngeire," 
' coig ' " ainm runa, nf baidb bbeag, otms ' lu ' " ainm do 

laigbeadb. 

48. ' Manaois' sen- ainm sleighe rigb, is ' goitbne ' foga" go fir, 
agiM *luigbne,'** go ffr dbe, ainm na" sleighe dfubbruic[th]e.*' 

49. * Searrdha ' faobhar go " ngaile, oeus * serrdha' tesgaidhe," 
*[d]inn' ainm gfl£?h dniimne d6ra," *fuince' ingne** airmgh6ra." 

50. * Mur ' ** iomad thall isin re^A^, * cob ' ^ buaidh, is brfatbar 

[14n]cert,** 
*du'** baile, *du ' dutbatyh lat, 'cul' coimb^d, is *cul ' carbad. 

51. 'Ob' ainm gacb airdf{adb[a] amacb, 'fuincbe' is** 'criombtan' 

ar sionnacb, 
' patdn ' mfol muigbe ma seacb," ' earc ' mil, ocm ' earc ' 
firbbeacb. 



^ Brat, S. Broth, H., L. and 0*£. s.t. tnuadh, 
' snuaidh, H. snuadh, L. and 0*R. s.v. tnuadh. 

* frag, L. 

^ Meamar congnadh feardha, L. 

* agas caise ainm do chaingin, L. 
^ am, 8. 

^ ain ainm cicbe go s&imh sin, H. &in ainm siodha go suim soin, L. 

* Ribhadh, S. Kobhar, L. and 0*R. s.v. robhar. Robha, H. 

* ramhra, H. romhra, L. rumna, O'R. s.v. robhar. 
M ngl^, L. 

^^ Coic, L. 
" «^, H. l&oi, S. 
'' tie, H. fada, S. 
'* Inibhne, H. and L. 

S«h, L. and H. 
ubhraicthe, L. 
" med, L. 
^'* teaflcaidhe, L. 
1' dionn gach dmimne adera, L. 
* fuinche, 8. and L. 

'^ aimghera, H. kimgh^ara, L. and O'R. s.v. searrdha. 
^ MOr, L. 

" tie, H. cobh, L. cod, 8. 
M l&incheart, L. 
» dfi, L. 
» oeut, 8. 
» f& seach, L. 

Phil. Ttam. 1891-2-8. 2 



18 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

52. Dh£ senainm do Idoch^ gan on, 'b^ ' otms 'c6/ nf br6g in modh, 
* c6 ' fos ainm do cheile cain, * b6 ' ben ocub* * b6 ' adhaigh.' 

53. *Treanadh* ainm gach caoinidh^ dhe, * aime,' 'fed' fuirecht«, 

na ceil,* 
* 6cosc'' cuma,'' nach coir cleith, 'homo' duine gan® dichleith. 

54. 'Deime'* ainm fe8cu[i]r" go fnin, 'teibedh* ainm [gach] 
tinsceduil, " 
'ionnsa' ainm doilghiosa dhi[l],^' 'fobhairt' is ainm d'ion- 
saighin.^' 

[fo. 97^] 55. *Cann' ainm leastair," luai[d]ther libh, 'blosg' 
ainm gach. gotha gl^ghil,^* 
' foidhi'^^ tairm no fuaim go s^, 'saithe' sluaigh" no B6chaidhe. 



H. 2. 12, p. 7. 

56. Tr' deabatdh thall ann gach modh, is 'liacA^uin' ainm do 

fhliuchadh, 
*iarlon»' iarthar ambi biadh, ' inntille ' leastar" no tiagh." 

57. * L6d' soillse ecus'*' *lang' meabhuil, 'lus'*^ lamh, ni lem nach 

meabhair, 
'lioghrodh' tenga thuirmne^ dhe, 'lucA^aire' saobhchoire uisce. 



^ oidhche, H., and 0*£. b.y. ce. D& ainm don oidche, L. 

* ifl, S. and H. 

' adhuigh, S. aghaidh, L. 

* aiCf L. ainm coinnmhe, S. 

* snirech nis ngle, H. fuireachras oidche, 0*C1. gl. s.v. aime. &ime f6t, 
fuireachras gle, L. Perhaps we should read fuirechrus ngl6. 

* eathar, S. eugusc, H. 
"^ comha, H. 

^ nach, H. and L. 

* D6ine, S. Deimhe, H. and L. 

^^ feascrach, L. hut in marg. feascair. 

^^ is teasccoe^ ainm each tionsgnat^h, H. is tasgar ainm tionsgadail, L. 

^^ iodhna ainm doil^heasa dil, L. 

^3 siCf H. f6bairt is ainm d^innsoighidh, L. forba ainm nintamhuin, S. 

^^ leathuir, S. leastar, H. leasdoir, L. 



^' gleghlan, H. gleghloin, L. 
" foithe, H. faitiie, L. 



^^ sleighe, H. sluaieh corrseted into sluagh, L. 
^^ lea^rr, H. intiUe leastor, L. 
^' This and the following ten quatrains are wanting in S. In L. they come 
next after the quatrain numbered 75. 
au is, H. 

" 1ms, L. and O'Cl. 
** thuirmheach, L. 



FORUS FOCAL. 19 

58. 'Mand'^ [is] sen-ainin na hunga,' 'mfadh' airmhidin,' coir 
cumlia, 
' nna'^ uasal, 'm^n' bel blaidhe/ is g^g amWdh suithidhe,* 

69. *Tucht' gach gne, ocus^ 'braon' hochtj^ 'aidhbeil'* cian" 
ocns^ 'ur* olc: 
* mocht ' gach cinin, tonn mara" mb6ir, 'nion' tonn coitchiona 
accetoir." 

60. * Osar * *' eire bhios ar neach, * ainnsi ' [is] * airgsi ' gach 

aithfer : " 
'ong' bron," 'oircne' measan con," *onii'" cloch, ocus' *tort* 
bairghen. 

61. 'Pelait'" righ-tech, coir ros-chom, * puincne, ' is ainm do 

scribol," 
'pont' borby 'pairt' rann ion gach tan,*^ is ' pain'^^ ainm d'aran 
nasal. 

62. * Rus ' roghairme[dh] do ghruaid ghloin," * micedh ' rus," 

* imdherg ' aithfer,** 
is ' sai ' * snabhuis iomaseach, is ' reudaire ' ^ cleir^^h craibhth^ (;h . 



^ ^anxL, L. 

' A srllable wanting. 

s nSaab oirfidin, H. miadh 6ir fidin, L. 

* nnadh, L. 

* menbhel blaidhe, H. men b^lblaidbe, L. 

* is r6 amblaidh socbaidhe, L. 
' is, H. 

^ et panper hoehtf L. 

' Ic^. aidbbean, as in 0*C1. aidh bben, L. in marg. aidbbhean. 
^® aidb bben cbinn, L. 
*^ c^liinthonn mbara, L. 
^ ate, L. nion, tonn accedoir, H. 
*' Osaire, H., but see 0*R. s.v. aaar. osair, L. 
i« ainsi no airgsi gan aitbbbear, L. 
" br6nach, H. 
'• conn, H. 
" on, H. 

" Pfiait, H. Pil&it, L. 
>* sgriboill, H. sgreaball, L. 
»» tani», H. 
«» lAin,L. 
** gbloine, H. 
*' niis, L. 
« aitbbbir, L. 
** 7 saoi, H." et saoi, L. 
* reacaire, 0*R. s.v. reacaire. recoire, L. 



20 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

63/ ' Suithnge ' suilbhir sen-ainm sin, * sfn ' muince,^ ' sin ' ainm 
Blahraidh, 
' searpan ' ' ar geis, is garbh gal, ocas ' ' agh ' ainm daimh 
allaidh.^ 

64. * Ce * talamh, is foirghiol ' fior, is ' teanlach ' teine go ffor, 

* tiiif[e]ad * seimh, * troichit ' • gach ^ corp, ocus ' 'teidm ' gach 

bas bitholc. 

65. * ITr ' olc, ' lios ' deabati^i go dearbh, * usarb ' bas mhilles gach 

dealbh, 

* ascc' dfomus, is deimhin tra, is ' uamhuin ' ^ ainm gacA eagla. 

66. Ni beoloigh/ ni bugbdar ard, oeus ni file fiorgbarg, 

ni 8encba[idb] ag nacb biadb a bfios, foillsigbtt^ feasa fbrus.^** 

FORUS FOCALL. 



The following six quatrains are from the Stowe MS. 
No. III. fo. 97b :_ 

67. Da ainm choitchionna gan coll, *dfth' deiredb is c6ol 'ducbonn,'" 
*neo*"ainm [na] gdoitbe gloine, *uim*" talamh go ttorttftefbe." 

68. * Fuid ' gach fursannadh " go se, * droch ' dfreach, * s6d ' ga<;h 

slighe,** 
* oin ' " cennach,'* * udin * isLS&cht in, * fath ' tea, ocus " * fatb ' ** 
andl. 

1 sin mninche, H. 
' searb&n, L. 
» is, H. 

* eallaiffhy H. 
^ fuirgbioll, L. 

* ta*oicchf(i1iy L. troigh^</h, H., but see Conn. 61. s.y. fothracud. 
"^ gan, 0*R. B.y. tinfeadh. 

^ uamhan, L. 

* heolach, L. 

^0 L. puts this quatrain before No. 65. 

^^ sicy H. duchon, L. S. corruptly, Tri senainm coitcbenn gan col. dith 7 ddr 
is dtichon. 
^» n6it, S. n6id, L. n6, 0, H. 
** uaime, S. umh, H. 

** ttorthuighe, L. tormidhe {i.e. torraighe ?), H. 
**»«?, H. S., corruptly, Idith fCiar sonn. 
" iiCf H. slige, L. 
" siCf H. on, S. 
^^ cneadhach, L. 
" is, S. and H. 
» fat, S. 



"- OK ,,.. ■ ^' ^ 



FORUS FOCAL. ^^ "^ 



69. 'Ailtire' sdor doghnf tech,* 'ailt** ainm tighe, *airt" gaeh 

leth,* 
'aithrinne'* ainm do laogh bo, 'feilisc'* rusg,^ is *meile' bro. 

70. * Bach ' meisge, ' boisgeall '• eilit,* * bir ' tiobra, ' berr'" g<w?h 

ngairid, 
*boigill*** borb, *bach' s^e serbh, 'mormuir,' moin," 
'bothach,' seiscenn. 

71. 'Carson'^ senainm d'arrarA^ de/^ is ainm do chinn 'calb' 

cuimhne," 
*car[th]uid* craibhthech," bdidh go mblaidh," ocus 'cast'" 
ainm do ghenmnaf^ni. 

72. ' Dagh ' maith, ' droch ' olc ocus gann, ' duibheall ' " ainm do 

ga^h udmall, 
*duchus** deabhuidh, *dibheoil" balbh, *daigh' tine, 'dorr' 
gaeh. n-agarbh.** 

The following three quatrains are from H. 2. 12, p. 8 : — 

73. 'Ealg'*' ainm d'Eirln[n], 'ealg' aghaidh, sean-ainm treoin 

'earr,' gan meabhuil, 
* 6 ' bron, caire, ann rochlos, * esconn '^ seanoir, guth cadhus.** 

^ tegh, S. teach, H. tech, L. 
' nCf H. ail, S. 
' tic J H. iar, S. 

* both, S. leac, H. art gach leath, L. 

* nCf H. airgime, S. 

* fioluscc, H. feilioscc, L. 
^ ticj H. nii0g, S. 

* boischeall, H. is boisceall geilt, L. 
»«ih,H. 

*• Dior, S. and L. 

" boitheaU, H. boiteall, 0*B. boitheal, L. 

I) mormoiim, S. mor, muir : moin, H. m6rmh&ir m6in, L. 

" #i^, H. Canan, S. Cur86n, L. 

14 do riarnigh sen, H. d&rsuidh, sin, L. 

1* cuimhnecn, H. cuimhnigh, L. 

^* tit, H. aidedh craibhdeadh, S. 0*R. has ** earait, adj. religions, deyout, 
Jot. foer 

" tie, H. mbloidh, 8. 

1^ caist, S. casd, H. 

» tic, H. dnpeall, S. 

^ tie, H. dnoh cos, S. dtichos altered into duchon, L. 

'> tie, H. dib^l, S. 

^ dfm gaeh ngarbh, S. d6ir gach nagh garbh, H. See 0*C1. s.t. dorr. 
Here ends the Stowe copy of Forttt Focal, H. has three more quatrains, which 
are printed in the text. 

» Ealcc, L. 

** easgan, S. eascconn, L. 

^ go oc&dhtM, L. 



22 IRISH METRICAL GLOSSARIES. — ^MR. STOKES. 

74. *Fi8' taidhbhsi, 'feimhin" gach lond, 'fuid' fuwA^, 'irei'* 

gach n-6ttrom,' 
'glus' solus ocus* *searbh' gaid, 'fdoinell'* is ainm do[n] 
oinmhid. 

75. * Gno[d] ' • is rinn mar dorala, ' grodhan ' ' ainm churaig* 

manti 
'giabair'* meirdrech, *grecli' cnu tra,*® oeus *gibne' adaro 
leagba. 

The following stave is quoted in O'Reilly's Dictionary, s.v. 
Cdise, as from the Forus Focal: — 

76. ' Caisi ' mioscais, ' caisi ' searc, do r6ir na leabhar Uincheart : 
ro sgaoil neart na tromsluagh Dhe, dias dan[a] comhdhual 

c^i." 



II. The Derbhsiur Qlossart. 

23. L. 21, p. 9. 

1. Deirbhsiur don eagna an eigsi, as comide a coimb^dsi, 

si na blatb oir mar eagna ; coir do cbdcb a coimbfreagra. 

2. Dearbbratbair don 6igsi aird, an sencbtM raidbit rigbbbaird, 
a bratboir ni bbf ar forbbas, matbair { don ugbdardbas.^' 

3. A n-eigsi ni bbi gan bblas, as buime { don eolus, 

ni gblacfa SLcht solus gan sal, a dalta an ^' f ohm focal. 

4. An t-6igios gidb be be f^in, airdrfgb ar an 6igsi eis^in, 
egios eadbon e gan cbeas, se gan ansodb ^^ madb 6icce8. 

^ feimhin cancelled and f^ine tcritten under it, L. 
' oic, L. sirsi, H. 
' He, L. n-eittrom, H. 

* in, H. 

^ foinneall no fainneall, L. 

* gno, L. 

^ erod&n, L. 

* do churan, H. church&in, L. 
' giabhuir, L. 

^ chnu chena, H. 

i» Thifl quatrain is also in H. 3. 18, p. 612*, and in O'Clery's Focl6ir, s.t. 
cais. In H. 3. 18 the second line is mar innisit linbair l&ncert, which is 
hypermetrical. The first line re-occurs in the Derbhiiur glossary 60. 

^^ The dh inserted by a corrector. 

" a daltan, H. 

** ansodh, H. anft6gh, L. 



THE DERBHSIUR GLOSSARY. 23 

5. Gach nech 'ga ^ mddh dbrcha dan, as 6 ' bhios a cces chomhlan, 
gaeh neach go brath da mbeith a cces, do ghnath ni ba he 

an t-eiges. 

6. As se an t-eigios s^imh sothal,' d^bbbhrdthair^ na ndnbhfocal, 
as 86 ghealas ^ an glor dubh, madh lor a t'eabhtM d'ollamh.* 

7. Gidh be riocht na^ bfuilim f6in, do dhean glor" solfM soilleir, 

don gblor as doibhe * ar domhan, m6r gach tuile ar ttiomsu- 
ghadh. 



8. 'Nee ' duine bemaoid ga bhratb, do bhearmaoid ^^ do tra tosach, 
ainm dho 'dsd' na deaghaidh soin, 'gnae' ag feraibh na 

foclaiih." 

9. * Fich ' is ' grian ' fearann a bfad, ' tola ' is ' dlumh ' da ainm 

d*iomad, 
' dfu ' ainm do cbian, ni cam soin, oetu * mann ' gach biadh 
bunaidh. 

10. 'Baghadh' gach gealladh, dearbh^' duit, 'gno' oeus 'ealg' 
da ainm d'ordhnirc, 
' dinnis ' ainm do luighe ^ Mn, mar innis ^^ duinn an deghdhan. 

11. 'Eccosg' cuma, oeus 'foath' dealbh, 'einech' aghaidh, ni 
hindearbh, 
' dian ' go^h loath ag dul a blad, oew ' drubh ' ainm do charbad. 

12. 'Grib' ainm do luas, fionnuidh fein, 4ocht' is ^troicchedh' 
dann chnisr^idh, 
sloinn gan tairm,^ gan locht, gtm len, ainm do chloinn ' core ' 
is * ceioel.' 



* aga, H. ga, L. 
' as e, H. se, L. 

* sotal, H. 

* dealbhadoir, H. (Wbhrathair, L. 
^ eheallfM, H. 

' do ughdar, H. 
'a, K. 

* doghen ^loir, H. 

* don ghloir is duibh, H. 
10 dobhemm, H. 

1^ ana focladb, H. 

*^ sie, H. dearbhadh, L. 

" loidhe, H. 

^*' nar innis, H. mar dinnis, L. 

^ sloinidh da ghairm, H. 



24 IRISH MEl'RICAL GLOSSARIES. — MR. STOKES. 

13. 'Bronnadh' is < gleitb ' caithemli c4idh,^ 'grant' ainm do 
Hathadh' Idnbhlaith, 
' tort ' bairghen, is ' caladb ' cruaidb, mar do canadb on 
cb6aduair. 

14. Trace' Idmb, oeus 'luibbne' m6r, *onn* is *ailcne' clocb 

cboimbtbren, 
is' do cbloicb fos as ainm 'ail,' gan dalbb * ocus^ gan docair. 

15. ^Coart' ainm binn do bbrugbuidb, 'bar'* ainm do saoi 

suadbambaily 
' re ' ^ ainm milis do ^ gaeh modb, * i ' inis, is * aill ' uasal. 

16. *Iodbna,' is 'ceis,' ecus* 'cealtoir/ ar " sleagbaibb da 

sfrleantoir, 
' suit ' ga^b datb da mbe fa bbladb, * tucbt ' " gne gae;b bratb 
do bhunadb. 

17. 'Gorm'" ainm d'nrdbairc gan ail, * dfocbmbairc ' ^ goid, 
combrddb cnbbatV^b, 
*dile * leanmbnin, lugbu bladh,** * a ' ard, octta * bri ' ** briatbar. 

18. *Cem' buaidb, is 'cearn* fer fearrdba, 'segbadb* alladb " 

oiregbdba, 
' bra ' ^^ gaoh mala seng sedacb,^^ octta ' earr ' gacb gaisgeadbacb. 

19. * Tiomgboire' iarruidb, nach lag, oetts is " ' cam * gacb combrac,'® 
' rubba ' guin na ^^ crecht corcra, ocm ' ceebt ' gacb cumba^A^a. 



^ chaich, H. 

' no l^the, interlined, L. 

' sie, H. as, L. 

* 1. doilb.e, L. 

* is, H. L. 

* tiCf H. barr, L. 
^ In marg, read, L. 

* tic J H. da, L. 

* is, L. 

>o araa, H. 

" tie, H. cucht uel tucht, L. 

" 1. gormadh. 

1' diochmaire, H. 

1* lugha abladh, H. 

» bngh, H. bridh, L. 

^* «M7 H. L. teemt to have alladh altered into ialaidh. 

" In fnarg, braoi. 

^^ br4e malach, seng sheghach, H. 

*• tie, H. as, L. 

^ tie, H. comhradh, L. carad in marg, 

»i 1. no, L. budh, H. 



THE DERBHSIUR GLOSSARY. 25 

20. ' Cearb ' ciorrbhadh, * cerb * tesgadh tren, * bladh ' ' sldn, 

' bladh ' baile, is binnsg61, 
*guim/* blogbadhy *gunii'' brdighe bhdn, a raidhim' Bimn as 
senradh, 

21. ' Ruannidh ' ^ ainm dilios don d^g, 'iar' ainm don dubh gan 

dluithbhedhg, 
da luadh ann amach 's amuigh, ' flann ' gacb ruadh, ni dath 
duthain. 

22. ' Lang ' ainm da gach meabhail mboir, innister ' is ni beccoir, 
'mucbna' ainm do gbruaim gan gbean, gan stuaim, gan ainm, 

gan aireamb, 

23. < Falc ' ainm do gacb manntaiVHi moir, innist^ is ni beugcoir,* 
' foUan ' ainm do mhaisi amuigb, ni braisi do na beloibb. 

24. 'Raiftinne' is^ rddb gan cbes, ainm don gbairge, glor^ dfleas, 
is* ainm do garg 'lore/ dar lem,*° gan fadhbh," gan locbt mur 

luaidhem." 

25. 'B^agba'" sen-ainm don ghaoith gbrinn, *16th*'* ainm do 

cblumb,^' ni ceilim, 
' ur ' gacb tosacb, tren a bladh, * cria ' " cennach, is * 6r ' 
nasal. 

26. * Son' oeus *fuach' focul feigh, *nith' guin, *nion' liter 

lanreidb, 
' tebeadh ' " ainm do bhuain bhuna»(^, * aoi ' ainm ^^ do go^h 
ealadhain. 



blath, H. 

1. g6n. 

araidhe, H. 

Ruanai^h, H. 

tic, H. umistior, L. 

neH.f L. omits these two lines. 

sic H., Raistine (1. raisde) as, L. 

gairg is gloir, H. 

sicn.. as L. 

sic H.. Kt learn, L. 

tic H. sadhbh, L, hut corrected in margin into fadhbh. 

lnaighsem, H. 
» DeSe, H. 
w loth, H. 
^ don chlaimh, H. 
'• sic H. criadh, L. 
" teibhe, H. 
^^ snaidlmi ainm, H. ainm aoi, L. 



26 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

27. 'Besccna' sfth, 'besgna' b6rla, 'tugaid'* adhbhar a dhenmlia,' 
gin motha na f6agiiiai8 sin, go bfionnmaois ' mur ta a thuigsin. 

28. 'lairchena' osin amach, 'ceachoir'^ ionann i is lathach, 

' annach ' ' as ainm grinn do ghlan, aga ghairm as binn do 
bhunadh. 

29. 'Loc** ainm d'ionadh, sonn' go se, 'aoide' ionann f is oige, 
' c6 ' talamh, is ni dluth " dhamh, ionann ' ur ' octis nasal. 

30. ' Porchaombnagair/' combrddh glan, ionann is gach ni f^btar/^ 
* adcboda ' " dligbedb dieagbair," do siredh " a senleabratbb." 

31. ' Riaradb '^^ do dbutbcbtM as ainm, * feib ' feabbus cacb dba^' 

cbombgbairm, 
'feibb' ionann e oeus amboil, in da gne^^ do gbabbabboir. 

32. Ealadba ^^ is caingen cubbaidb, is dligbedb nae;b duagbamboil,^* 
ainm doibb a ttriur * aoi ' gan ail, a ngnaoi gan liudb ar loeht&ihh, 

33. ' Loscc ' bacacb, ' coscc ' ^ tegusg tenn, ' adb ' dligbedb, is 

«dliitb'« inneall, 
'tdidbe' goid do cbombloit crodb, 'nnall' ainm d'urdbairc 
'na £arradb. 

34. * An ' is * rann,' ^ * ban ' is * binne,' ^ anmonna iad d'firinne, 
atbairm gach laoi ni leagbor,^ agas ' aoi ' ainm d*foircbeadul. 



^ tuig^^, H. tugaidh, L. 

' an deamna, H. 

' interlined 1. do bhermaoid, L. go bfhedmaois, H. 

* ceacar, H. 

B andag, H. annuigh, L. annach, 0*B. 
« Dog, H. 

7 B6n, H. 

8 dlugh, H. 

* Cor caomhna gair, H. 

^^ ionan gach ni feuchar, H. 

" tiCf H. adhcoda, L. 

^' deagair, H. dleaghoir altered into dleaghor, L. 

13 is readh, H. 

1^ eiCf H. shenleabboir altered into ahenleabbor, L. 

16 B&cad, H. 

!• pa, H. 

" 1. an da ghne, L. 

" Ealad, H. 

^' dQaghamhuil, H. 

** eiCfU.. coeg, L. 

»» dl<i, H. dlugh, L. 

»' rtin, L. rann truth, 0*B. 

'* ar is rann bar is binne, H. 

M atairm gach laoi gach l&gor, H. 



THE DERBHSIUR GLOSSARY. 27 

35. ' Gubha ' caoinedb, ciadh do cbar,^ ' techta ' dligeadh, * art ' 

nasal, 
' meall ' aoibhinn, ocus ' fod ' fios, mdoidbim iia^b 6g do airmbios. 

36. ' Gem ' caitbeamb,' ' ong ' teallacb te, is * sion ' catboir nua 

nimbe, 
is * nisadb ' oird^rc re beadb, sdsadb nar chombloit cineadb. 

37. ' Forcbongra ' • farailiomb * f6m, * ellgbeadb ' * adblacadb 

aigbmb^il, 
docbim go gar s^imb seadbacb, ' ail ' * mia, is ' f 6igb ' 
fairgsionacb.^ 

38. 'Sgill' obaniiy gan • taom taoisi, 'suin" clutbor *** no 
caoinmbaisi," 
' gfdd ' ^ iarraiVni no guidbe gbrinn, diambair go^b dnine 
adeirim. 

89. * S6ire ' proinn, oilemain " Idn, degbiear * cur* can^^marcbombrddb, 
'deiltre'" d6e draoidbeacbta in, d'aos nidbeacbta 'na secbrdn. 

40. 'ToarasdoP luigbe gan 16n, is *clannadb' gacb sddb sfrtbr6n, 
' gnaoi ' ainm d'aoibbinn gan ail, * gnaoi' ag feraib ar na focluibb. 

41. Da ainm don cblaoine colati^b ' ciol ' is ' cle ' ^' re a ccanambain, 
's edb aderar re linn * laitb,' a 86nadb " linn as lanmbaitb. 

42. 'Us' ga«b sloinneadb, doirbb an radb, 'ealladb'^" aisgidb 
naeh adbndr,^* 
' dnar ' gacb rann dana no^^b dubb, aga radba ann don ollumb. 

^ da do char, H. 

' $iCy H. cathair, L. 

' For conghair, H. Forchongair, L. 

* foraiWh, H. 

* eiUigh, H. 

* ail alUred into aOghean, L. 

^ do chim go r^idh seimh seaghdha ailmin is feigh faircdona, H. 

* gan dliered into agas, L. 

* ran, H. B&in, in marg, svin, L. 
*« dutar, H. 

^' «t«, H. c&oinmhnisi, L. 

^ g^d, H. g&idhe, in marg, ^dhe, L. 

" ffluillean n^dh, H. oileamain altered into dil leanmain, L. 

1* ddthf^ corrcan, H. 

14 tdUW, H. ; butue 0*C1. s.v. deilltre, 

" cQSs ddth, H. 

17 innsenadh, H. an (altered into a) s^nadh, L. 

1* eallomh, H. in marg. ealladh no eallamh, L. 

1* agfanar, H. adhnair, L. 



28 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

43. ' UiiuiBi ' ionann e oeiM atd, ' neimheadh ' go^^li din da ndemta, 
' cioP bds is ' biodhbha' leatrom, is 'iodhna' gaeh. eineachlann.^ 

44. ' Cuislionnach/ ergna re cdch, as ainm e don feaddnach, 

* siiocht ' do gach c6ill adcluine, is ' cioe;^ ' r6idh gach rionn- 
uidbe. 

45. * Eislinn ' esinneall, ni breag, ionann ' cul ' oeus coimb6ad, 

' athreidhe ' ' sa thairm go brath,' ' ceile ' as ainm do gach 
oglach. 

46. *Datan' ainm gaeh oide finn, 'dathnuid'* ga<?h bnime 

bhelbbinn, 
'fisleadh' annamh' biadh fo bhladh, 'gnia' gae;h mac sethor 
solamh.' 

47. *Coimdhe'^ gach. tigheama tenn, *cealt' ainm d'6dach gan^ 

tuighcall, 
c^cb da' gbairm is " trom an tred, is " ainm * com ' do ga^h 
coimbed. 

48. * Seimh ' beag, * airm ' baile ar bun," * forba ' leadradhf * mal ' 

nasal, 
mar bbfos gaeh dal gan degbail, 'mdl' cios aga cbruinniogbadb.^' 

49. ' Slaibbreadb ' ^ ainm do cboibhcbe is cair, * muaidh ' " maitb 

secb soitbin ^^ siordbdil, 
da ainm don choimbidea^A^ cain, 'sail ' ocus^* 'caoimbtbea^A^'^^ 
cbubbaidb. 



^ eineclann, H. 

^ aithreighe, H. 

' In marg. aithreidhe sa gbairm go brath, L. 

* dathnaidh, H. 

^ aicy H. fis leadh ainm L anamh biad, with the letters and word leadh ainm 
1. cancelled f and underneath are the words fios gacb dath da mbiadh 7c. L. So 
0*R. fis colour t dying f tincture, 

« featrh]ur folamh, H. 

"^ eicB.. Coimhdhe, L. 

8 re, H. 

»ga,H. 

w etc, H. as, L. 

'1 dar liom, H. ar lamb 1. bun, L. 

^' cuimhniughndh, H. chuimhniughudb in marg, 1. cbniinniogbadh, L. 
" slabradh, H. 

^* muaidh, L. but with the i cancelled, 

^* HCf H. In L. soithin U cancelled and oscon written under secb and so. 
" is H. flfui L. 
>^ 1. coimh(thecbt), L. caoimthus, H. 



THE DERBH^IUR GLOSSARY. 29 

50. *Feith'^ ar eladhain' ainm dhi, 'troigh' ar thurghabhail 

ngr6ine, 
* fofor ' ' ainm * do thobar thr6n, * sloighre ' gach cloidhiomh 
cmaidhgh^r. 

51. 'Dron' gach dfrech, breth naeh br6g, oeus *comruir** gach 

coimh6d, 
ba* be an sgel nacb bu gan^ greim, ocus is gacb dirim 'dreim.'^ 

52. * Maime * brath, ni coir a cbeilt, * gno ' is • ainm do *® gach 

oirdbeirc, 
tairm '* fa blath is daingne dbi, * airbbe ' " is ^ * fdtb * da ainm 
d'aisde. 

53. *TJatb' ainm d'uir, diambair an datb," * foilerbadb ' " garb 

bas bronacb, 
ionann 'ceis' is cuairt re a radb, is suairc gach gn6 don 
gliabblan. 

54. 'C» ' is 'dsBy' ciodb diambair in de,^' da ainm tigbe a mbia daoine, 
' taitbmbeacb ' gach scaoilidb go sc6imb, " is maoidbemb 

'aitbnecb' eis6in. 

55. Baidbrecumairce" gancbol/snaidh^^ni,' 'faoisidin,' 'faosamb,'^' 
ionann 'dionn' is*^ maitb mbordlia, d'ardflaitb na mionn 

mblatbordba. 



1 Fddh, H. 

* wtlaghain, L. 

' In marg, fophor. 

* for forainm, H. 

* tieH. coimhrin, L. 

* bndh, H. 

^ gaiM, H. and L. 

* 7 18 dzeim gach dirimb, H. et dreim ann {cancelled) gaeh dirim {altered into 
adirmi). 

* ne'R. as L. 
*• «k H. da, L. 
" atairm, H. 

^* airbhre in marg, no airbhe, L. airbhre, H. 

" ««; H. as L. 

^ tie the eorreetor of L. far d*uir . . . datb the original scribe wrote don 
dath diamair rath. H. hot uath ainm diamhradh an dath. 

^^ eie the corrector in margin : the original scribe wrote failearibadh. H. has 
foletrbhadh. 

'^ »k H. Cs is doe {no dse) diambair dhe, L. 
. " «wr the corrector in margin : the original scribe wrote sgaoileadh go s6imh. 

^ C&idh re comairce, H. 

^ tie the eorreetor ofli,i the original eeribe wrote snaidh«<^ faosamh faoiside. 
H. ha$ 7 snaoidhe faosamh f&oisidhe. See 0*C1. s.y. faoeamh and 0*E. s.y. 
f^eiaemdh and em^him * I protect.* 

» «< H. as, L. 



30 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

56. ' Euan ' ainm do bhrat bbfos fa bhladb, * eo ' ainin do go^b 

maitb moltar, 
' eo ' ainm do gaeh. fert abbos, ' sgeo ' is ^ ainm do gacb agus.^ 

57. 'Daoe" tegbdbuis, deilm* gan on, <batb' is 'ibatb' is 

*diobbadb,'» 
fregraidh^ an triar don bbds bbea<?^, grianfios as gndtb^ do 
cbloisdeacbt. 

58. 'Asccal' iomagallamb* ann, 'uagbbba" togbba re tagball,^^ 
gr6as" nach. cclos a macb go mion, ' mos' bes, octu * buicb"' 

brisedb. 

59. ^ Dremban ' ainm do Dbemban dubb, ' oirdberc ' ainm d'foUus '^ 

nasal, 
'robuist'^^ coimb^d, 'dagb' moitb m&r, 'aidbbbsi' ceol oeui 
crondn. 

60. 'Caisi' miosgois, 'caisi' sere, 'drocb' ^^gack dorcba go dnibbcbert, 
' oibid ' umbla, ' ceo ' sechna, ocus ' beo ' ^* gacb buaincbetbra. 

61. * Tinne ' saill " ocus m6itb mor,^* * acobbar' " saint go sfotblogb,** 
' d6imb'" no * dembal ' dfogbbbail" sin,** sgel go mbrfgb go^b 

dbdil ^ bnnafdb. 



^ 8%e^ H. as, L. 
^ ogos, H. 
» D», H. 

^ «M7 H. delmh n6 delbh, L. 

^ The th o/b&tb and the i o/'ibath inserted by the eorrector, ba oeue bath oeue 
diobhadh, H. 

* HeK. freamiffh eorreeted into freagraidb, L. 
^ ni gnath sa ctiiaU, H. 

B iomagalladh, H. 

* uadha, H. 
i<> tadhal, H. 
" pi^, H. 

1* buith, H., and L. has buith in marg, 

^^ sieH. dollum corrected into d*ollu8, L. 

^^ sie H. robruisd corrected into robuisd, L. 

" drocht, H. 

w b6o ffia, H. 

1^ sic H. Teine Boill, L. 

^B no m6ith mh6ir, H. 

1* acobair, H. 

^ ro ch6ir, H. et sic L., hut sloth 16gh in marg, 

'^ dr^im, H. dreimh, L., but corrected into d^mh. 

ss diabhail, H. diabhal eorreeted into dloghbh&il, L. 

^ om. H. 

^ sgel go mb&odhail iiibr6nach, H. sgel. gach mbridh goch dh&il bunaiih, L. 



THE EGERTOK METRICAL GLOSSARY. 31 

62. * Cndh ' * cenn, oeus * cudh ' * iudhbhairt, agus ' buth ' bith 
blditbiooroic,' 

an^ derbhcbest amblaidb gan ail, labbruidb bHgb' inbbar 
mbiiatbraibb. Deirbhsiur. 



III. The Egerton metrical Qlossart. 

Egerton 90, foL 17* 1. 

1 [fo. 17» 1] is gel *caiiidlech,'« 

' arco * ^ locbor do Dia a dul, oeui * menmarc ' gacb emuaineadb. 

2. 'Kang' msel boidbi lorn gan lag,^ 'lis" faisneis, 'ribar' criatbar, 
' sol * grian, * lugna ' esca and, ' salt ' leim otersa (?) aderam. 

3. * Segamla ' gacb blectas bog, ' samb case araid ^^ na cnmagb, 

* snatb ' " folt,. nocbo scele Hum, * smer ' teini [ocus] * fuaid ' 
fuigbell. 

4. ' Seist * meodbon-lse, is * [sjceng ' lebaidb, ' sorb ' locbt, * sin ' 

croind csemnegair, ' 
' slab ' cumang, oeua * coid ' coill, ' segb ' ^' fiadb istuig (?) mar 
tbuirmim. 

5. ' Satb ' biadb, * sen ' Kn fiadba amacb, ' serr ' gacb ^ n-og is 

gacb n-nallacb, 
'tell' foaim mar doglach goglan, 'td' brat,^^ [octu] 'borr' bmcbtadb. 

6. 'Turba' gacb bniden bindsi, 'fuaim' tobeim, *totb' ban-indsci," 
' tuarad ' cuid adeire dnn, ' buili ' ^* go deimin drocbron. 

^ MCy H. Cudh n6 catb, L. 

» cuth, H. 

' Im marg, ordhmrc. 

*a,H. 

* brigh, H. brldh, L. Then in L. follows this prayer in an Italian hand : 
Oremns. AbBolve qnaesmnus Domine animam famuli tui Hugoni ab omni 
Tinculo Delictorum : ut in ResurrectioniB Gloria inter Sanctos & Electoe tuoe 
resofldtatus reepiret Ver ChriBtum Dominum noetmm. Amen. 

* is gell caimeGh, Eg. 
' erca,Eg. 
•laga,Eg. 

» fit. Eg. 
^ caacaraid. Eg. 
" leg. snuath? 
" 8eth,Eg. 
^ Beirgach, Eg. 
»« tibrath, Eg. 
" toa atmsi. Eg. 
>« boili. Eg. 



32 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

7. ' Bro ' beimneach, oeus * ur * olc, ' baigliu ' * Iseg allaidh' ardrod, 
^ bascall ' geilt, is ' baislech ' dam, ' band ' liathroid, * aisdrech ' 

umal. 

8. 'Bracbt' beoil, *brac" lamb, is 'brann' ben, 'braicbem'* 

damb allaidb armer, 
' bas ' barr nacb fuilgend fingal, is * cuimling ' ' gach 
cait[b]irgal. 

9 acb, 

is cnatb fumaigi re * foir,* is * culmaire ' gacb cairbtbeoir. 

10. ' Ceinntecol ' trialladb • go trie, * consal '' combairlecb 
csBmbglic,^ 

* duar ' rand, * duar ' gac[b] toradb tend, * iacb ' bradan, ' solam ' 

sliseng. 

11. 'Drenn* garb, 'del' sine mosecb, * teim ' dorca, is 'drocbt* 
gacb ndirecb, 
< ai ' a bsBn, nocho n-ecbt ' ecbtga, ocus ' cecbt ' gacb cumacbta. 

12. * Ealga ' in Eire re baicbne, ocus * eidel ' umaigtbe,*® 

' at ' sescba, ni lesg ale, ' gann ' easgra, ocus ' esc "^ nisce.^' 

13. *Icbt'" ceinel, *erc* nim gidb nar, 'flesg' bun fich, [is] 

'Isegb' lind ban, 

* feirend ' cris cengail gu coir, * ffs ' do cennaigb in canoin. 

14. *Focbnadb' lasamnacb IsBgaidb, ' fual' gacb uisci e domain, 

* troced * " corp, [is] * li ' moladb, maitb ni bi omain ardr. . . 

15. ' Gamb ' geimreadh, * pit ' " proind nacb mor, *caidb ' idan nocha 

na nodh, 

* fod ' faidecb, is * les ' letrom, ocus * fen ' ^* gac[b] feiretrom. 

^ bui^lui, Eg. 
* allaigh, Eg. 
' brach, Eg. 

braici, Eg. 

amling, Eg. 

leg. tniailledli P 

comsal, Eg. 

-gleic, Eg. 

necth, Eg. 

umacti, Eg. 

esg, Eg. 

uiflgi, Eg. 

Ucht, ii. 

oroced. Eg. 

fid, Eg. 

fe. Eg. 



THE EOERTON METRICAL GLOSSARY. 33 

16. 'Aig' fiiacbt, 'fol' bonn, is 'gam' ben, 'gabar' solos ga 
slainneadby 
' Ina ' breb, ' gil ' an sugad seng, < gnftb ' gutb munadh nacb 
milfem. 

17. * Grot ' ^ goirt, * gno ' cuidmide ' clsen, ' gnai ' s^gdae;^^,' 

' clamar ' gacb aer, 

* gaire ' gairsecle * go grod, is * bairseaca ' gacb baetb-rod. 

18. * Prann ' tonn mara, maitb in t-ainm, * pui[n]cni ' sgreabull, 

re scfam ^ gairm, 

* pain ' bairgen,' oem * pairt ' penn, * pingur ' [saland],' 

gus aibtnn. 

19. * Eaideran ' inar masedb, * braitsi ' * asan re airem, 

'iar' dubh, 'for' nrlum re bedb, 'rucbt' mer, nf cle da 
coirgeadb. 

20. * Feib ' marsin, ' du ' baili blaitb, ' dumacb ' dorcba gan 

dimbaigb, 
'fecor' brecc nacb rosena sund, 'lang' fled ocus 'cro' cumung.' 

21. 'Eangacb' gres gusmhar go ratb, 'druimcH' gacb legtboir 
lagacb, 
' siUidb ' ben doni tuaicle, nf bi sel aeht saeb fuaicbtbi. 

22. 'Batb' cor, oew 'motbla' msetb, 'cliste''^ urlamb . . . nibsetb, 
dena sloindte gar na slecbt, is ' droingcedul ' gacb naid^^^A^.^^ 

23. * Dlomodb ' fuagra, ' commaim ' ^' ben, ' buas ' a ainm bolg, 

' boindi ' blaitbgel, 

* tap ' oband, * tore ' craidhi coir, * ge[i]s * guidbi, is olc gacb 

* egcoir. 



gnoth Eg., but see Corm. 6L s.y. gniiten. 

gnodh cuidni, Eg., but see Conn. Gl. s.y. gn6. 

8«gdacht Eg. but see LU. 109» 41 (gnae .i. s^gda). 

gairsech Eg. 

resglam Ee. 

baireein, Eg. 

Cf. Pingair .L saland, L.Lec. Voc. 

braidci Eg. 

crumung Eg. 

clisdi, Eg. 

This couplet is obscure and prob. corrupt. 

comuaim. Eg. . 



Tram. 1891-2-3. 



34 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

24. 'Dear' ingen, oeus 'uath' uir, 'buanann' buime go blaith 

ciuin, 
' bracbt ' [buain], ' baircne ' cat,* * bil ' mong, ' brae ' lam, 
' oidsen ' sin sloindem. 

25. ' Ficell ' . . . signum gu se, ' bus ' topais, is ord aichne, 

in sceil crid guma ' cnedhach, ocus ' gr[e]id * gach gaisgedhach. 

26. *Brac* min, 'bf[acb]' faebur ferdha, *mur' gacb (n-imm)ed 

ilardha, 
' base ' gach d^g adeire a ndan, * blim ' sili, ' buidbi ' ballan. 

27. * Briar ' delg 6ir, * bruinnecb ' ' mathair, * cod * * buaid, ' core ' • 

finda fataig,' 

* f aid ' fuacbt ocu8 * croch ' gach n-ard, * cem ' buaid, is * loth ' 

gach langarg. 

28. * Cath ' doaim ' is * ort ' ' orgain, * ciar ' dub, * coth ' biad, blaith 

ordain, 

* cum * gach cis, ' laemdha ' nach lag, 

29. * Going ' lenmain, ' cumlachtaidh ' ore,' * clairiu ' *° fodhail re 

fochmorc," 

* crumduma ' ottrach,** * eel ' nem, ' cil ' clajn, ni mor do 

muinedh. 



' bairci cath, Eg. 

' Read, perhaps, guba. 

' dniinecii. 

* cob Eg. 

* tore Eg. 

^ Eg. adds eongeir or cengeir (the second letter is doubtful). 
' corrupt. 
■ ord Eg. 

* cumlachtaigh arc, Eg. 
**> claire Eg. 

** fachmorc Eg. 

*' crumdauba odtrach Eg. 



35 



Supplement. 

Here follows the part of the prose vocabulary in the Book 
of Lecan, fo. 155* 3 — 156** 1, which corresponds with the 
metrical glossary in Egerton 90 : 

Geal .i. caindleach. Area .i. lochar do Dia. Meanmarc .i. 
emuainead. Cich .i. ger. Ris .1. faisnes. Ribar .i. criathar. 
Sol .i. grian. Lugna .i. esca. Salt .i. leim. Smer .i. tine. Seist 
.i. medonlai. Sgeng .i. lebaig. Sorb .i. lochL Sin .i. cruind. 
Saith .i. biad. Serrach .i. each cnocc* Tnrba .i. buidean. Fuaim 
.1. tobeim. Tot[h] .i. bainindsci. Tuarad .i. cuid. Buili .i. drochrun. 
Bro .i. bemnech. IJr .i. olc. Baidbliu .i. leeg allaig. Bascall .i. 
geilt. Baisleach .i. dam. Band .i. liathroid. Brachta .i. be[oi]l. 
Brae .i. lam. Brann .i. bean. Braici .i. dam allaid. Bas .i. barr. 
Cnibleang .i. cathirgal. Car .i. toit. Cicht .i. fidhidoir. Culmairi 
.i. cairptheoir. Ceindtegal .i. turailead. CotibslI .i. comairleach. 
Bnar .i. rann. Duar .i. each torad. lach .i. bratan. Dreann .i. 
garb. Greann .i. ulcha. Deal .i. sine. Tern .i. dorcha. Drocht 
.i. direaeh. Ceeht .i. cumachta, Feidil .i. nmaidi. Gand .i. 
eascra. Ease .i. us^^. Foehnada lasamnaith. Fual .i. UBce, 
domain. L{ .i. molad. Gam .i. gemrad. Pid .i. proind bee. 
Caid .i. idan. Aid .i. faacht. Fol .1. bond. Gamh .i. bean. 
Gabar .i. solus, Lua .i. breab. Gni .i. gut[h]. Cldmor .i. air. 
Prand .i. tend. Pain .i. bairgen. Pairt .i. peann. Pingair .i. 
saland. Faideran .i. inar. Fuindchi .i. feandoc. Braitsi .i. asan. 
lar .i. dub. Fuinebi .i. sindach dub. Du .i. baile. Lang .i. fleag. 
Cro .i. cnmang. Rath .i. cor. Bus .i. tobais. Gred .i. gaisgeadach. 
Blomad .i. fuacra. Comaim .i. bean. Dear .i. inga.' Uath .i. ur* 
Bracht .i. buain. Baircne .i. cat. Bil .i. moug. Briar .i. delg 
oir. Bruineach .1. mathair. Cod .i. buain no buaid. Core .i. 
oenel no finda. Finit dona dubfoclaib. 



* leg. n-6cc. 
' leg. ingen. 



36 



Arc. gl. 
AU. 

Amra Choi. 



BB. 
Bk. Arm. 

Bk. Feu. 
Bk. Rights. 

Corm. 

Corm. Tr. 

D. 

E. 

Edin. xxxviii. 
Eg. 88. 
Eg. 1782. 
Ff. 

Fel. 

FM. 
GC». 

Gild. 

Goidel. 
H. 2. 15. 
H. 3. 18. 
H. 4. 22. 
Harl. 432. 
Harl. 5280. 
L. 

Laud 610. 



Glossarial Index. 

List of Abbrbyiations. 

AscolijGlossarium palaeo-hibemicum, pp. xvii-cxl. 
The Annals of Ulster, vol. i. ed. Hennessy, Dublin, 

1887. 
The Amra Choluim chilU (from LU. and LB.), 

ed. Crowe, Dublin, 1871, and (from LH.) in 

Ooidtlica, London, 1872, pp. 156-173. 
The Book of Ballymote. Facsimile, Dublin, 1887. 
The Book of Armagh, a MS. in the library of 

Trinity College, Dublin. 
The Book of Fenagh, ed. Hennessy, Dublin, 1875. 
The Book of Rights, ed. O'Donovan, Dublin, 

1847. 
Cormac's Glossary, printed in Three Irish 

GloesarteSf London, 1862. 
G'Donovan's Translation of Cormac's Glossary, 

Calcutta, 1868. 
Derbhsiur don E'cna^ the metrical glossary 

printed supra, pp. 22-31. 
The metrical glossary printed supra, pp. 31-34. 

from Egerton 90. 
A MS. in the Advocates' Library, Edinburgh. 

( Egerton MSS. in the British Museum. 






Laws. 
LB. 



Forus Focaly the metrical glossary printed supra, 

pp. 8-22. 
FHire OenguesOy the Calendar of Gengus, Dublin, 

1880. 
Annals of the Four Masters, ed. O'Donovan. 
Grammatica Celtica, second edition, 1871. 
The Lorica of Gildas, printed in Irish Gheees^ 

Dublin, 1860, pp. 133-151. 
Goidelica^ London, 1872. 

MSS. in the library of Trinity College, Dublin, 
containing unpublished glossaries. 

Harleian MSS. in the British Museum. 

A MS. in the library of the Boyal Irish Academy 

marked 23. L. 21. 
A MS. in the Bodleian, described by Todd, 

Proceedingsof the R. I. Academy, ii. 336-345, 

and by O'Donovan, Book of Rights, xxviii — 

xxxiii. 
Ancient Laws of Ireland, vols, i.-iv. Dublin, 

1865-1879. 
The Ledbhar Breac^ Facsimile, Dublin, 1876. 



LIST OF ABBREVIATIONS. 



37 



LH. 



Lism. 



LL. 

L.Lec. Yoc. 

LU. 

Mart. Don. 

Misc. Celt. Soc. 

MR. 

O'Br. 

O'Cl. 



O'Dav. 

0*Don. Supp. 

O'Flah. 

OMoU. 
O'Mulc. 

O'R. 

RawL B. 502. 

Rawl. B. 512. 

Rev. Celt. 
Salt. R. 

Stowe XIX. 

Three Frags. 

Trip. Life. 
Wb. 

Vind. Wort. 



The Liber Hymnorum, a MS. in the library of 

Trinity College, Dublin, the Irish in which 

is printed in Ooidelicay London, 1872. 
The Book of Lismore, a MS. belonging to the 

Duke of Devonshire, described in Lism. Lives, 

i.e. Lives of Saints from the Book of Lismore, 

Oxford, 1890. 
The Lehar Laignech, or Book of Leinster, 

Facsimile, Dublin, 1880. 
The prose vocabulary (of about 570 words) in the 

Book of Lecan, ff. 155a-157». 
The Lebar na hUidre, Facsimile, Dublin, 1870. 
The Martyrology of Donegal ^ Dublin, 1864. 
Miscellany of the Celtic Society, Dublin, 1849. 
The Battle of Magh Rath, ed. O'Donovan, 1842. 
O'Brien's Irish-English Dictionary, Ist edition, 

Paris, 1768: 2nd edition, Dublin, 1832. 
O'Clery's Glossary. Louvain, 1643. Ed. by 

Arthur W. K. Miller, Revue Celtique, vol. 

iv. pp. 349-428, vol. v. pp. 1-69. 
O'Davoren's Glossary, printed from Eg. 88 in Three 

Irish Glossaries, London, 1862, pp. 47-124. 
O'Donovan's Supplement to O'Reilly's Irish- 
English Dictionary. 
O'Flaherty's Glossary, now in the Bodleian, 

marked MS. Ir. e. 1. 
O'Molloy. Grammatica Latino-hibemica, 1677. 
O'Mulconry's Glossary, H. 2. 16, col. 18-col. 

122. 
O'Reilly's Irish-English Dictionary, Dublin, 

1821. 
A twelfth-century MS. in the Bodleian, described 

by Macray, Catal. Codd. MSS. Bodl. Part v. 

fasc. i. cols. 719—722. 
A MS. in the Bodleian, described in Trip. Life, 

pp. xiv-xlv. 
Revue Celtique, Paris, 1870-1890. 
Saltair na Rann, Oxford, 1883. 
The St. Gall glosses on Priscian, ed. Ascoli, 1880. 
A MS. formerly belonging to the Stowe library 

and now in that of the Royal Irish 

Academy. 
Annals of Ireland. Three Fragments, ed. 

O'Donovan, Dublin, 1860. 
The Tripartite Life of Patrick, Rolls series, 1887. 
The Wiirzburg glosses on St. Paul's Epistles. 

Printed in The Old-Irish Glosses at Wurz- 

burg and Carlsruhe, Hertford, 1887. 
Windisch's Worterbuch, Leipzig, 1880. 



38 IRISH METRICAL GLOSSARIES. — ^MR. STOKES. 

i (MS. a) height J D. 17. So Conn. s.v. arad. A .i. ard, Stowe 
XIX, L.Lec. Yoc. As vowel-flanked b is lost in Irish, d ' height' 
may be cogn. with Lat. asa^ ara : so alius with altaria {altare). Two 
other homonymous words are d 'chariot' cogn. with Skr. ^d» 
*to sit,' and the aw, X€7. d * mouth,' gen. sg. Trip. Life, p. 140, 
I. U, with Lat. o», drw. 

acais curse, Ff. 29, note 12. acais .i. aor no mallacht, O'Dav. 48. 
virulence, gan dgh, gan accais, gan urcoid, MR. 294. tug se acais 
do he cast a slur on him, Coneys. 

accobar N. desire, D. 61. Z'. 222 : ririr accobui* a sula he gave up 
his eye^s desire, Amra Choi. 71 ; from ad-^rcohar. 

adchota personal law ? D. 30. This is prob. a verb,=atchota, LL. 
345^ 38, and L.Lec. Yoc, though it is explained as a noun in the 
Brehon laws cited by 0*R. s.v. adcoda .i. dlighead dligheas neach 
d'faghail no do geibh neach. 

adh law D. 33. So Stowe XIX, L.Lec. Yoc. and O'Cl. ni hada 
.i. ni dlig«(fa, Amra Conroi. A deriv. adae or ada .i. fas occurs. 

aed^rtf, Ff. 5. So Corm. and H. 3. 18, p. 63^. aed tene, LU. 
45* 30, and see infra s.v. smer, ace. amal aed (.i. tenid) tre fithicen, 
LU. 109» 35. Cognate with W. aedd * calor, studium,' Gr. a76o^^ Lat. 
aedes, aestus, A.S. dd. Aed 'eye,' Ff. 39, is doubtless the same 
word. So aed .i. suil, O'Cl. adam aed .i. adam suil 'my two 
eyes ' .i. im chind, LL. 208* 27. mu da n-aed .i. mo da suil, Amra 
Sen din. 

ag stag^ Ff. 63, properly a bovine animal, agh .i. bo, O'Cl. ag 
allaid 'cervus,' a neut. stem in s, pi. nom. ace. aige alta. dual 
nom. it e da n-ag ata cainium robatar inn £iri, Eg. 1782, fo. 
76^. Cogn. with W. ewig 'hind' from *agika, Skr. ctja-s. 

1. di ontf, E. 11: Old-Ir. ae, "adjectivis pronominabilibus each 
(cech), nach (na) aequo adjunctum ac oin unus," Asc. gl. xvii. ' one, 
person, individual,' do cech ai, LL. 254^ 50, do each ae, O'Don. 
Supp. The forms di, ae point to an Old-Celtic aivo-s^Qr, o?o9, 
Cypr. oifo9 'alone,' 0. Pers. aiva- 'one,' Av. aeva-f diva-, Skr. iva 
* alone,' 'only,' and (with a different suffix) eka-s, 'one' just as 
the Old-Ir. din ' one,' W. un, points to Old-Celtic oi«o-«=Gr. 00/09, 
oitnj ' the ace on dice.' Lat. unus. 

2. ai F. sei^ncCf D. 26. 32, ai .i. ealadha, Stowe XIX. oe .i. elada, 
L.Lec. Yoc. art (dan) pp. 2, 3, doctrine, D. 34. Sg. nom. dligid a{ 
astud, LL. 345<^ 32. gen. iar cantain a di .i. a 6icsi, LL. 186^. iar 
fuithir grene .i. iar ndagthir inna ai (.i. inna eicsi) grianda, Rawl. 
B. 502, fo. 60^ 2. This word seems to occur in aefreslige, ai idan, 



01X)SSAHIAL INDEX. 39 

names of metres, and of. a n-aae ina. .i. a nduana 7 a iidrecbt[a] 
7 a n-admolta, Rawl. B. 502, fo. 62» 1. 

3. ai law, D. 32, Stowe XIX.=O.H.G. Swa Gesetz, Norm, Skr. 
tea 'course.' 

4. ai a suit, p. 2. D. 32. Sg. nom. ae camgeo isin tsengaedeilg, 
LB. 238<^ 54. gen. airbert aoi iama hastadh ' to pUad the cause 
after binding it^ O'Don. Supp. ace. inti aighiss in ae, f^. 

aibeis, F. sea, Ff. 15. aibeis .i. muir O'Cl. sg. gen. lebbeann na 
baibheise, Lism. 117* 2. ace. forsin n-aibheis n-anbbail nembfor- 
cennaigb, ibid. perh. from an Old-Celtic ^ahensi-s ex *ahhent'ti-s, as 
•eis infra, from *sensis, *sent-ti'S, Cognate witb Ir. abann, Lat. amnis 
(from ^abnis), etc. A somewhat similar word, abis, is borrowed 
from Lat. abyssus, ind abfs mor inro inclannad dliged circuil (gl. 
circulos abyssi magni), LH. 12^. abisns scientiae .i. abis fessa, 
LB. 196% pi. ace. abissin, ML 51^ 8. In LB. 230^ dibiis is twice 
used for abis : ataat (soil, the fallen angels) hi f udomain dibessi : 
i n-abeis na teined suthaine ichtair iffimn : cf. D. C. Purg. i. 146. j 

aidben (MS. aidhbeil) remote, F. 59. adben .i. etircian, H. 3, 
18, p. 75% and so O'Dav. 65, O'Don. Supp. aidhbhean .i. imchian 
no fada, O'Cl. .i. olc no deoraidh, O'Cl. ond Athain aidben anair, 
from the far-off Athens, from the East, LL. 215» 46. 

aidbse music, Ff. 12, D. 59. So O'CL, Conn. s.v. adann, LIT. 
6^ 8. aidbsi .i. aircetal, O'Dav. 47. (a)idb8e .i. aiom do chful n6 
chroniLn doguiteis ermor fer nEreud immdlle, a name for the musie 
ar burden which the greater part of the men of Erin used to make 
together, LH. 34^ 1 (Goidel. p. 156). aidbsi .i. corns cronain, 
LU. 6^> 5. 

aidmirt, airmirt. See airmit infra. 

aig ^cold^ (rather f^), E. 16. Mathair etha aig, mdthair saille 
snecta, ice is mother of corn, snow is mother of bacon, LL. 345* 9, gen. 
bissi ega icicles, lit. fingers of ice, bommand ega hailstones, lit. 
hits of ice, W. id, 0,'N.jaki, A.S. gicel. 

1. ail ¥, stone, Ff. 27, D, 14. So O'CL sg. gen. ailech, a ^-stem, 
connected by Fick with Lith. uld * Felsen.' Why not with «7\t^ • 
rcV/Ms, Hesychius ? It has been connected with w€\eicv9 and paragu, 
as Lat. saxum with Ags. teax, O.H.G. sahs; Skr. dgman, with 
ax fit/ ; and Ir. art ' stone ' with Med. Lat. artavus. But then we 
should have had *elech, 

2. ail smooth, gentle, D. 37. Compd. ail-menmnach, LU. 87^ 11. 
ail (two MSS. ail) asking, seeking, Ff. 8. So O'CL ail .i. abairno 

dliged no guide, O'Dav. citing ail dame dibhse, gen. ic cunchid ail 






\ 






40 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

no aisCy LB. 216* 31. The length of the a is established by dund 
ail (gl. ad supplicandum) Ml. 40^ 44, the denom. verb atlim .1. 
guidim, Stowe XIX, better dilim 'I entreat,* dih-iy dliss *rogavit,' 
Bk. Arm. 18* 1, 18» 2, as well as by the o of the cogn. W. add-oli 
* to adore.' 

ailcne a {Utile) roch, stone, Ff. 82, D. 14. So Stowe XIX. 
Alccne .i. ail bee .i. digabtach indi is ail, H. 3. 18, p. 74®. A 
dimin. of alec-^ the stem of 1. ail. A form aike also occurs. 

aill noble, D. 15. So L.Lec. Voc, O'Dav. 49 and O'Cl. 

ailt house, Ff. 69. So O'Cl. and H. 4. 22, p. 61*>. sg. dat. ro 
dosuidigthi i n-oen ailt. Salt. R. 5158. alt .i. teach, O'Bav. 54, 
who cites is ae alt conae clit 'it is his house that preserves 
fame,' and toingthi fo ailt neimhi * swear by the house (vault?) of 
heaven.' O.Br, costad alt (gl. aeditui). May come from *{p)altd, 
which has been connected with Lat. palatum * vault, palate.' If 
so, alt neime would be a close parallel to Ennius' palatum caeli. 
But the connexion is very doubtful. 

ailtire architect, Ff. 69. So H. 3. 18, p. 74®, s.v. ailt, O'Dav. 54 
s V. alt, and O'Cl. ©Itaire 7 sair, LL. 29% aeltaire 7 rathbuige, 
LL. 29*^, 39. Cogn. with ailt q.v. and the verb cor' altar mo fert-sa 
lat, LL. 269* 16. 

ain pleasant, Ff . 46. See 2, an infra. 

din rushes, Ff. 46, O'Cl. and now written aoin, Sg. gen. 
Bcena buana afnC) Laws iv. 310, dat. robatar grfanana .... 
essarda do din, LL. 263^, fon ain, fon tuige, Laws i. 140, ace. dia 
taidled in n-uir no in n-ain forsa laiged Martain if he touched the 
mould or the rushes whereon M. used to lie, Bev. Celt. ii. 400. From 
*iatni' cogn. with Lat. iuni-culus, iuni-perus and iuneus from 
*ioinieulos, *ioini-perO'S, *ioini-co-s. 

aincis a eurse, Ff. 29. aingcis, O'Cl. ; a sister-form of acais supra. 

ainf^in unusual, strange, Ff. 33. anfen .i. ingnad, K. 3. 18, 
pp. 63^, 638. dinf6n, LIT. 83» 8, seems a different word. 

ainner a woman, Ff. 24. ander .i. ben, ainder .i. bean, H. 3. 
18, pp. 64% 635*^, aindear, O'Cl. pi. n. aindre 7 ingenrada 7 
maccaeme, LL. 109*^ 10, dat. andrib, LU. 81» 3. W. anner 'heifer,' 
O.W. enderic (gl. vitulus). Windisch compares Gr. dp07jp69, 

ainnse (leg. dinsed?) blame, Ff. 60, ainsed .i. ainmhedh no 
imdergadh, O'Dav. 47, citing n{ ainsid enech ruirech nd ollaman 
t?te honour of a chief or an ollave was not reproached. Cognate are 
dinsem (accusatio), dinsid (accusativus), and the verb tainsither ' is 
reproached, censured,' O'Don. Supp. 



GLOSSARIAL INDEX. 41 

airbbe a kind of poem, D. 52. So O'Cl. Hardly a mistake 
for airhert .i. aircetal, aircetal, 0*Day. 49. 

airbri, ahundance^ p. 1. So in H. 3. 18, p. 611^ airbri .i. 
immady LL. 311^ 12. airbbre .i. eluagb a host, army, O'Cl. (pL 
dat. airbrib, Salt. R. 716. LL. 281* 21. LB. 131» 48. ace. 
Tohairbriu, Amra Cbol. 33), may be the same word. 

airgse (aircse?) blame, Yi, 60. Cogn. with aircis .i. 6accaoiiie, O'Cl. 

airm place, D. 48. So O'Dav. 54 and O'Cl. cia airm sund hi ta 
Cuchulaind ? LIT. 68*. Other exx. in Asc. gl. xxvii. 

airmit (MSS. aidmirt) prohibitum, tabu (geis), Ff. 33. So Conn. 
B.T. Diarmait, and O'Day. 51. 

aime watching, Ff. 53, watching at night, O'Cl. ame Fingin, 
Rawl. B. 512, fo. 109» 2=aime Fingein, Stowe MS. 992, fo. 46% 
feacht n-aen dia mboi F. aidchi tsamna in Druim Finghein i 
n-aime, ibid., 0. Jr. aire, for air, fritJiaire, and the areanos of 
Ammianus Marcellinus, 28. 3, may be cognate. 

airrde (airidin ?), leasnghadh, Ff. 33. Prob. corrupt. 

airt side, direction, point (of the compass), Ff. 69, better aird, 
Wind. Wort, dird, O'Br., Gr. ^a««? 

aisdrech humble, E. 7. Seems wrong, for aietrech (derived from 
auter,^ * journey,' ' error : ' gan aisder, gan oil, Bk. Fen. 242, 
cethri mogaid mor aistir, ibid. 370), means 'unsteady,' 'inconstant,' 
BeT. Celt. xi. 128. aisdirech 'far-travelled,' Misc. Celt. Soc. 
374. 

aithnech boasting, D. 54. Prob. corrupt (leg. mdidmech 'vain- 
glorious ?). Hennessy renders aderim rib gu haithnech, Bk. Fen. 74, 
by / sag to you, knowingly. 

aithrinne calf, Ff. 69. So in H. 3. 18, pp. 63^ and 638. athimi 
.L loeg, LU. 8* 25. aidhrinne, O'Dav. 48. 

1. an truth {true?), D. 34. .i. Or, H. 3. 18, p. 633^ and Stowe 
XIX, p. 30*. 

2. an {ia?) pleasantness (aoibhnios), Ff. 30, 'an adj. pleasant, 
O'K seems the same word. 

3. an water, Ff. 15. So O'Cl. an-bruich [leg. an-bruith] .i. 
nisei 7 broth, Harl. 5280, fo. 11**. Cf. Gaul, anam, paludem, 
Eodlicher's Glossary, Kuhn's Beitr. vi. 227. 

an truth {true ?), p. 1, Ff. 4 : a mistake for I. an? 

ana wealth, Ff. 26, better dnae, Trip. Life, 118, a masc. stem in 
-aio. For exx. see « Asc. gl. xxxv. Seems cogn. with ops, opes, 
o0»^of, apnas, 

^ The Buter-form attar occnra in LB. 7^ 15. 



42 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

an-art Mft^ Ff. 34. See art. 

an-gaa{ (?)/«!</<? D. 32. Prob. corrupt: cf. angid 'wicked.* 

an-main haU^ Ff. 34. leg. anmuin, the opposite of muin = 0.1^. 
munr * mind/ ' love,* or anmain == anmaoin, 0*Br. 

annach clean^ D. 28. andag .i. glan, anannag .i. innglan, Stowe 
XIX, andad .i. glan, anandac .i. nemglan, L. Lee. Yoc., for met a 
n-anannaic, LB. 258** (where it is a subst.). neamh-annach impure^ 
* Don. Supp. annao ffuiUlesSf * Don . Supp . 

annoit church, Ff. 32. So L.Lec. Voc. 0*C1. and Stowe XIX, a 
moM^-church, O'D., whose explanation is supported by the gloss in 
H. 3. 18, p. 74c, Andoit .i. eclais do^t in aile as cenn 7 is tuiside 
(.i. tus). This is andodit in Bk. Arm. 18* 2, andoit, 0*Dav. 71, 
8. v. ceim. Low Lat. antitas antiquitas, Ducange. 

anno year, Ff. 32. Borrowed from dat. or abl. of Lat. annus. 

aprainn (MS. abrann) evil, alas, Ff. 44. appraind LTJ. 45* 5. 
apprinn 0*Mulc. ba abrain[n]. Trip. Life, 190, 1. 6, used as an 
interj. Apraind ni basa for mo nirt de sede, LTJ. 78**, 29. 

ara M. gillie, Ff. 33. And so in the gloss on Dan do Bhrian na 
miirtha. na haraidh .i. na giolladha, Betha Finnchua. Cf. Skr. 
arati'S, Gr. vw-ijpe'njs. Seems the same word as the t-stem ara 
charioteer, sg. gen. arad, LTJ. 64*. Compd. ban-ara maidservant, 
0*Br., daor-ara slave, ib. 

arco (MS. erca) I beseech God, E. 1. arco fuin dom Dia, Corm. 
and H. 3. 18, p. 63*>. Seems=Skr. archdmi. But Windisch brings 
arco from Spared, cogn. with Skr. pragna, Lat. preces, Goth, fraihna, 
frah. 

ard-broenud (gL imber) p. 5. Compd. of ard * high ' and hrdenad 
verbal noun of hrdenaim * I shower,* a denom. from broen * a 
shower* : dech do sfnaib ceo, ferr a brathair broen, LL. 345* 14, 
15. broen derg, LTJ. 90^ 18. 

arg a man, Ff. 24, prop, champion, hero, Corm. and H. 3. 18, 
pp. 63*, 80*, 541. ar argg (.i. laech) niad (.i. trenfer) ropsam 
fartail, LL. 208* 30. argg .i. anrath, LL. 31 1*» 25. etir argaib erritib 
.i. anradaib, LTJ. 47* 9. Cogn. with Gr. apxo^* Hence argani 
sg. gen. deis dr n-argain uais, LTJ. 9, marg. sup. and argdha, O'Br. 

art hard, Ff. 34. art 7 anart cruaid 7 maoth, 0*Mulc. 56. From 
A.S. heard ? Or is it the same word as art * stone ' ? 

art noble, D. 35. So Corm., L. Lee. Voc, Stowe XIX, and 
O'Cl. Eochaid art .i. uasal, LL. 393* 53. 

asc pride, Ff . 65, from *at-ko- ? cf. ataim turgeo. 

ascal conference, D. 58,=axal .i. imacallaim, 0*Dav. 56, axul .i. 



GLOSSARIAL INDEX. 43 

imagallmha, H. 3. 18, p. 628. pi. gen. ba eola axal n-aingel, 
Amra Choi. 47. ace. ranic axalu la arbriu archangliu, ibid. 33. 

at sescba, E. 12, seems corrupt, at * milk,' sesc-ha *dry cows.' 

athreide ? aithreighe? D. 45. Corrupt. 

axal nohl^, Ff. 32. So H. 3. 18, p. 74®. Perhaps for *ascal, 
*at-klo ? See ase, supra. Axal is the name of an angel in Corm. 

ba, hQ& death, p. 3. So H. 3. 18, p. 61 1^ and O'Cl. fit dibad 
7 (ba)th 7 ba 7 teme ic sluinn epilten, LH. 26^ 1 (Goidel. 162). 

bacc {var. lee, bach) crozier, p. 2. So Corm. bacc buana finime (gl. 
ligo), Sg. 62*» 10. Qorn^. fid'loce (gl. arcus ligneus) Sg. 107*' 1. 

bacad, D. 31, note 15. Obscure and prob. corrupt. 

1. bach drunkenness, Ff. 70. So O'Cl. madness, O'Dav. 56. 
Comp. all-bach, LU. 106*^ 31. bach-lubhra yro^-5^wo/w#, Coneys. 

2. bach sea, salt-water ? Ff. 70, seems corrupt. Head loch, and 
ct bochna * sea * ? Compd. boch-thonna, LB. 118**. 

bacht (rar. ko, bach) reaping, p. 2, better bocht, as in 0*C1. 

bagad a promising, D. 10, verbal noun of hdgaim. ** Robdgus," 
ol se, ** ffad Fergus comrac fri Coinculainn imbdrach," LU. 68*' 39. 
Bagais Fallamain ni ragad arculu co hEmain, LU. 78** 6. bagais 
Cuchulaind co ndingiied samlaid, LU. 75^ 25 : a denom. from bag 
.i. briathar word, 0*C1. pi. n. ar ngnima ropsat mara, ar mbaga 
ropsat beca our deeds were great, our words were small, LL. 208^ 34. 

baigliu a fawn, E. 7. So O'Cl. This word occurs in the following 
quatrain (LL. 200% last two lines) : 

Atchonnarc bndchem is bru I saw a stag and a doe 

0CU8 baigliu ettirru. And a fawn between them : 

Bochaide rodech immach, A multitude which looked outside, 

oeus brech ica marbad. And a wolf killing them. 

baircne eat, £. 24. A female cat ace. to Corm., a whife cat, 
ace. to O'Dav. 58 and Laws, i. 150. braicne, O'Br. 

bairsecha foolish talk? E. 17. lairsighe 'brawling, scolding,' 
O'R., who gives a number of cognate words, hdirseach * a scold or 
shrew,* hairseoir * a scold,' etc. bdirseacht * satire,' O'Br. 

baislech doe? stag? E. 7. So L.Lee. Voc.: baislec[h] .i. eilit, 
O'Dav. 57. baisleach .i. dam allaidh, Stowe XIX, p. 30*. baisleach 
.i. damh, O'Flah. an ox, O'Br. 

ban truth, D. 34. So 0*C1. In H. 3. 18, p. 633^ Stowe XIX, 
p. 30*, and in L.Lec. Voc. ban is glossed by fir * true ' or * truth.' 

band, bann, a ball, E. 7. So Corm., L. Lee. Yoe. and O'Cl. Hence 



44 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

perhaps the adj. hannach: suil glas hannach, LU. 11 3^ 3. Hardly 
cogn. with Jj&t.funda * sling,' from */onda^ 

bar a aage^ D. 15. So Corm., L.Lec. Yoc., LB. 101, and H. 3. 
18, pp. 65% 663. Cobthach cloth . . . breo (.i. noem) bar .i. sui, 
LB. 101, marg. inf. bar. .i. sai, LL. 377^ 16. From Hharo-s ? cf. 
Lat. hariolus * soothsayer ' ? 

barann a hlow, Ff. 22. So O'Cl. n{ fuirceba-su and fer rosasad 
... a b6im, a bruth, a barand, LXJ. 58^ 41, lin a baraon mbuan. 
Salt. R. 7934. Cogn. with Lat./mo, O.N. hetja, Skr. hkara 'fight,' 
Lit. harniSf Slav, hranl, 

has top, E. 8. So L.Lec. Voc. has .i. barr, O'Flah. A contraction 
of bathai, haithes * crown ' ? The * bras .i. barr,' of Stowe XIX, 
p. 30*, seems wrong. 

bds M. deaths p. 3, note 2, Wb. 15*^ 28, gen. bdis. Transferred to 
the slain: etir has 7 ergabail loth killed and pruoners^ AU. 912. 
Cogn. with the verb heha * mortuus est ' and the noun bath infra. 
Fick connects Jj&t. fatuus, 

basal pride, arrogance^ Ff. 12. So O'Br. baiseal .i. diomas, 
O'Flah. Derived from bas supra ? Cogn. with Lat. fasttu ? 

base red, E. 26. So Conn., O'Dav. 62 and O'Cl. So also H. 3. 
18, pp. 65^, 624, 560^ OTlah and O'Br. Possibly cogn. with Lat. 
bacca from ^bat-ca. 

bascall one maddened by fear, E. 7, and L.Lec. Yoc. baisgeall .i. 
gelt, Stowe XIX, p. 30». baiscceal .i. geilt, O'Flah. [b]oisgell .i. 
geltan, H. 3. 18, p. 64^. Doghnf boiscill dia bhibhdhaidh, he makes 
a * boscelV of hie foe, Dan do Bhrian na miirtha. 

1. bath death, p. 3, D. 57. So Corm. s.v. baten, and H. 3. 18, 
p. 65, col. 1. morbas bath dom-r^, LU. 119^ 30. Hence bathach (gl. 
moribundus) Sg. 59*. A cognate verb, bathar, occurs in Amra 
Choi. 129. 

2. bath manslaughter, Ff. 25. Cogn. with Gallo-Lat. batuere, 
batudlia, Fr. battre, bataille, A.S. beadu, 

1. b6 night, Ff. 52. be .i. aidche, Stowe XIX, p. 30*. be .i. 
ai[d]che, L.Lec. Voc. Cogn. with Gr. 0aio9 (from *0aiaos?) 
dtMky, 

2. b6 N. woman, Ff. 52. So Corm. s.w. Buanann, Be N6it, 
L.Lec. Voc, Stowe XIX, and O'Cl. be .i. ben ut dicitur B6-bind 
.i. ben find, LH. 16^. cid as messo ban ? ni ansa : Be 



^ So 8unt, avunculus are from sont^ avoncloa. There seems no ground for 
regarding /miu^ as borrowed from ap€yt6yri. 



6L0SSARIAL INDEX. 45 

chaim, LIj. 346^ 30. Be earna and M n-imroma are glossed by 
merdreeh (=meretrix) in L.Lec. Voc. sg. voc. a b6 co mbail, 
LL. 260» 4. 

beo cattle, D. 60. So 0*C1. Identical with hio = Lai, fgjvivtu^ 
Skr. jlvd'S^ Goth, qius : cf . the Eng. expression live-stocky O.N. 
hihendi, Ir. margad bid, margad beo-craid, LL. 216^ 53. beo- 
almai, Rawl. B. 612, fo. 112» 1. 

berr (MS. bior) short, Ff. 70. So H. 3. 18, pp. 64^ and 633*. 
Hence the verb herraim * I clip.' W. hyrr brevis. From a pre- 
Celtic hhersO'8 cogn. with Skr. hrasva ? 

1. b^na peace, D. 27. So O'Cl. bescna .i. sith no bearla, 
Stowe XIX, p. 30% L.Lec. Yoc. 0*R.'s heasenadh. nembescna 
strife, O'Don. Supp. 

2. b^Bcna the language of a people, a nation, D. 27. Cogn. perh. 
with Skr. hhdshd * speech.' This is O'Cl.'s hSscna * every country or 
every land wherein are languages.' He quotes the Calendar of 
Qengoa, £p. 318, athath in each hiscnu. So in the Auraicept na 
n£ce8, Eg. 88, fo. 63% 2 : gach son fordorcha robui in gach 
besgna 7 in gach berla fofrith ionad isin Gaidilc every obscure 
word which was in every country and in every district found place 
m the Gaelic, As to the use of lingua, langue and ^{K&aaa for 
'nation,' see Ducange. belra a parish or district, O'Br. and see 
LB. 132«>. 

biach penis, E. 26. So H. 3. 18, pp. 61% 626, and O'Cl. biach 
dori riastradh oili a penis which will reach another's vexation, O'Dav. 
57. Another example is in the following quatrain ascribed to 
Conall Menu, Eg. 1782. fo. 64» : 

Dia ti Luingsich don Bannda, cona trichait c6t imme, 
gellfaith, cid lebur a bhiach, Cellach Liath Locha Cimme. 
If Loingsech come to the Bann with his thirty hundreds around him, 
Re will make submit, tJiough long his penis, Cellach the Grey of Loch 

Cimme, 

In LL. 43^ 6, da biach bar baill brde is glossed by duo testiculi 
eins circa molam. Hence biachachd priapismus, O'Br. 

bibda, miswritten bidba (gl. reus), p. 4, D. 43. bibdu bdis, 
Wb. 1* 15. pL n. bibdaid (gl. obnoxii) Ml. 134»> 1. bibdid 45»» 10, 
gen. bibdad 43* 12. Hence bibdamnact, Wb. 1^ 15. O.W. bibid, 
M. Bret, bevez, 

bibdanas, MS. bidbanus (gl. reatus), p. 1. So Conn. Tr. 147. 

bil hair, mane, E. 24. So LL. Voc, Borrowed from Lat. pilus, 



46 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

as helleOf hdc 'kiss,' and hrolachy from pelliceus, pdcetn and 
prolofftu, 

bile mouthf Ff. 41. O'Cl.'s hil .i. b^l is prob. an error. 

binne truth D. 34, derived from hinn * true,' 0*Br. 

1. bir (MS. bior) water, Ff. 15. So O'Cl. s.v. hior. 

2. bir a well, Ff. 70. So Conn. s.v. hirar and O'Cl. s.v. bior. 
Prob. identical with 1. bir. 

1. bla (MS. bladh), healtht/, D. 20. bla .i. slan, O'Dav. 60. 
bid .i. slan, 0*C1., from *mla, cogn. with Lat. melior, Gr. ftoKa ? 
In the Laws this word means exempt, 

2. bid (MS. bladh) a place, D. 20. So O'Dav. 66 and O'Cl. 
bligim (gl, mulgeo), p. 6. From *mligim cogn. witb Ir. melg, 

mlicht milk, blichtach milch-eowe, LIT. 68^ 31, Lat. mulgeo, Gr. 
d'fieXr^to, CN. mylkja. For exx. see Wind. Wort. 400. 

blim spittle (of a dead man), £. 26. Hardly cogn. with f^XafiMtv, 
rf\afAvp69, Rather a corruption of hlinn, blind, Corm., and H. 3. 
18, p. 65^, cogn. with pkiwa 'mucous discharge/ fiXet^vot 
* slime,' and fiXeppoi * drivelling.' 

blosc voice, Ff. 55. So H. 3. 18, p. 623. It also means sounds 
noise: roclos blosc-b^imnech a chride, LU. 80* 1. muadh-bhlosg .i. 
mor-thormanach, O'Cl., s.v. muadh. From *bhlozgo-s = Gr. 
0\oc<r/3o9, and cf. Lith. bldzgu, bldtgeti, 

bluch (MS. blach) bacon (saill), Ff. 44 and O'CL fatness, O'Br. 
Cogn. with blong, LB. 2l7^ 77, 79. 

bochna sea, Ff. 15. Hence tarbhochnach transmarine, O'Br. 
V. supra s.v. bach, 

boigill rude, savage, Ff. 70. This seems corrupt. Read *boitill, 
cogn. with boiteall pride, and boiteallach haughty, savage, 

boinde smooth, white, E. 23. brat boinni odarda, LB. 2 1 6<^. 

boisceU a hind, Ff. 70. So O'Cl. 

1. bole F. (MS. holg) gap (bema), Ff. 5. 16im dar boilg, LL. 263^ 
30. W. bwlch, Br. boulc^h, 

2. bole (MS. bolg) cow, Ff. 14. Perhaps from Lat. bucula, with 
metathesis of I, 

borr E. 5, where the gloss bruehtadh is obscure. As an adj. borr 
means 'great,' 'haughty'; as a subst. 'grandeur,' 'pride.' Ri 
betha buirr. Salt. R. 4453, for gablaib crand broach mborr, LL. 
43* 9. in tonn baeth bhorr, LL. 88^ 41. Corapds. borr-bfast, 
borr-buiden, borr-oclaech. Wind. Wort. 401. panathir brecc-borr, 
a spotted great panther, Rawl. B. 502, fo. 45^. From borso-, bhorso-, 
cogn. with Lat. fastus from */arS'tus ? 



6L0SSARIAL INDEX. 47 

botbacli hog^ fm^ Ff. 70. So O'Cl. brugh bothach-mhor, Rawl. 
487, fo. 14^ 2. From *bu-t-ako-. Cogn. with Lat. imhuo {in-huo) 
Iwetf mouten. 

bott Jire^ V. infra, 8.v. tmer. bod .i. teine, L.Lec. Yoc. 

bra eyebrow, Ff. 39. bra, braoi, D. 18. brae, LH. 2^ dubidir 
re bran a brae, LL. 44^ 27. df broi duba dorchaide, LU. 56* — 55*^. 
da bra doile dubgorma osna rosea sin, LB. 219\ Compds. 
bra-dorcha, LL. 266^ 43, bra-dub .i. malacbdub, LL. 266» 22. O.Ir. 
*bnii =, O.-Slav. hruvif Gr. o-ippm, Skr. hhrU, From the same 
root comes the Gaulish hrtva ' bridge,' from *hhfcd. 

1. brae Jlottr (rectius malt), E. 26, Corm. s.v. brocoit, the 
Welsh equivalent of Ir. hraich, gen. bracha, Laws, i. 128. sg. dat. 
cess no da chess lana do braich 7 do arbur biid, LL. 286^ 35. 
aire ucht n-ech do braich 7 biud, ibid. 37. ace. mar miles mulend 
mnad-braich, LL. 86^ 21 =muadmraich, LIT. 106** 34. Hence the 
name Mrachide, ATI. 726. Gtiul. brace, 

2. brae F. hand, E. 8. 24. So Corm. s.v. braccille, L.Lec. Yoc. 
adam braicc .i. adam Idim my two hands, LL. 208^ 28. FroYn Lat. 
hracchium, whence also W. braich, 

1. bracht 8ap,fatne%e, E. 8, and so Corm. s v. anfobracht. bracht 
.i. sug no beoil, L.Lec. Yoc., H. 3. 18, p. 663. daceird bracht cu 
feic (.i. saill furri corrici a fiacail), LL. 208^ 12. Hence the adj. 
hraehtach ' fat.' 

2. bracht a breaking, cutting (reaping?), E. 24. bracht .i. 
bris[e]adh, Stowe XIX, p. 30*. .i. briseadh .i. buain, L.Lec. Yoc. 
From *bhrag-to, cogn. with Lat. fra-n-go, frac-tu-s, Goth, brikan, 

braen (broen ?) poor (?), Ff. 59. Some mistake here : brdinech, 
LU. 106*> 19, braon€tch .i. bronach sorrowful, O'CL, may be connected. 

braichem (MS. braici) stag, E. 8=braicheamh .i. damh alluidh,- 
O'Cl. See quotation at baigliu supra. 

braitse hose, E. 19. braitsi .i. asan, Stowe XIX, p. 30. pi. 
bristfghe, O'Br. From Mid. Eng. breche, 

brann woman, E. 8. So L.Lec. Yoc. Cogn. with brii * womb,' gen. 
bronn. 

brech wolf = Skr. Tj-ka. See quotation at baigliu supra. Compd. 
Brechmag. 

1. bres, F. some kind of noise, Ff. 6. breas .i. gach greadhan 
no gach glor, O'Cl. sg. ace. mar 'tchuala Mfchol in mbreis conruala 
in aluag 'roon tegdais. Salt. R. 5969. 

2. bres great, Ff. 6. So 0*C1. bress, O'Don. Supp. 

bri word, D. 17. So O'Dav. 57 and 0*C1. bri .i. briathar, LL. 



48 IRISH METRICAL GLOSSARIES. — ^MR. STOKES. 

187* 58, L.Lec. Voc, and Stowe XIX, p. 80*. pi. n. conid de 
ataat na br( se, LL. 287**, 9. 

briar a golden brooch, E. 27, weighing an ounce, Conn, and H. 3. 
18, p. 54 1^ briar A. delg, O'Dav. 56, and H. 3. 18, p. 64^. 
Identical with the Eng. brSr, now brter, briar. But which 
language was the borrower is not clear. 

bricht reliquary ? diadem ? p. 2, = breacht .i. mind, L.Lec. Yoc. 

bro cutting, abusive, E. 7. bro .i. beimnech, Stowe XIX, p. 30% 
•i. beimneach, Edin. xxxviii. bro, 0*R. 

bronnad consuming, D. 13. So L.Lec. Voc, 0*C1. bronnadh .i. 
Bcaileadhno cnamh, K. 3. 18, p. 609, andO'Don. Supp. sg.gen. breth 
brondta ithe aidche judgment as to eating com at night, O'Dav. 60. 
dat. do longad 7 do brondad cacha bfd, LB. 217**. The cogn. verb 
bronnaim occurs in Ffacc's hymn 8 (n( bronna), in LU. 100^ 6 (is 
ed robronnad frisna cdic fichtiu bargen sin), and in LL. 344^ 50 
(robronnat na rothrebat). 

broth ^A, Pf. 45. 80 O'Cl. Cogn. with fiopa and Lat. (g)voro. 

bru doe. See quotation at baigliu supra, and cf. the Messapian 
ppivhov ' eXafpoif, Hesychius. 

bruinnech mother, E. 27. So Corm., and H. 3. 18, p. 65®. Na hi 
bruinnech balb {his) mother was not dumb, 0*Dav. 56, ba hi a 
bruin [n]ech ro oilestar Mac De it was his mother that nursed God^s 
Son, ibid. 61. Derived from bruinne 'breast,' or brd gen. bronn, 
* womb.' 

bu death, p. 3, note 2. So H. 3. 18, p. 61 P. 

buanann nurse, E. 24. Buanann muimme na fiann, Corm. 

buas belly, E. 23. Cogn. with Gr. ^va/co and N.H.G. Bauch? 
buas innbea * entrails,' O'Dav. 56. 

buich a breaking, p. 2, Ff. 14, D. 58. So H. 3. 18, p. 615*. 
Probably a verb : buich .i. robris, LU. 5*^ 27. rocehaes gair co 
mbuich, Amra Choi. 39. buich bron cerd Cuind grief routed Conn^s 
division (i.e. the northern half of Ireland), Amra Choi. 130. 

buide chum, E. 26. W. buddai, for muide (=Lat. modius), 
muddai. sg. dat. im-mudiu bic i mbid ass, LL. 286^ 49. dobeir 
a df laim fon mudi, LU. 53^ 12.* So buime, E. 24, LL. 378* 24. 
buimme, BB. 31^ 4, for muimme * nurse.' However, the b of buide 
is possibly right, and the word may be cogn. with lAt.fidelia and 
Gr. wldo^ for *<f>iho9, 

buile an evil secret^ E. 6. So L.Lec. Voc. * an evil design ' 

^ A part of a haip was also called mude. See LU. 8^, 41. 



GLOSSARIAL INDEX. 49 

O'E., as if there was some connexion with fiovXy. Cf. buaile a 
nomine holin .i. consilium, H. 3. 18, p. 80^. 

bus .i. topais, E. 25. Both lemma and gloss are obscure to me. 

buth world, D. 62. budh, 0*Br. If this be right, cf. Gr. 0v<ri? 
(from *0i;-Ti-«), Skn hhu-ti-a * existence,' bhu-mi-a * earth.' 
Bat the dat. dual hathaih in the line maithi uli du dib 
bathaib, aU the nobles of the two worlds, Bk. Fen. 160, where 
it rhymes with rathuib^ seems to show that buth is a mistake 
for bath. 

cadla^oa/, Ff. 17. So O'Mulc. 279 and O'Cl. 

caer candle, Ff. 44. So H. 3. 18, p. 626 (^caer .i. caindell). caor 
« Jlame, Jire, Q'Don. Supp. caer thened hisinn aidche, LL. 207^ 
29. pi. n. caera crethir comraicthe, LU. 91» 20. Cogn. with Goth. 
tkeirs 'clear/ A.S. scir, Eng. sheer, y/ski *to shine.' 

1. cai (MS. cae) house, D. 54. So Corm., and H. 3. 18, pp. 66^, 
633*. Compounded: cerdd-chae (gl. officina) Sg. 51^ Cogn. with 
Low Lat. eaya, cayum domus, Gr. Keifiai, Yedic gayd Lager, 
Kohestatte, \/(i, Goth, heiva-, Eng. hke. 

2. ciipath, p. 3. So Corm., L. Lee. Yoc, and see Laws, i. p. 32, 
where the words gtlla dom-ainic ar cae (a lad came to me on 
the way) are misrendered by * a youth protected me on the way.' 
ar caoi .i. ar slige, H. 3. 18, p. 210. aroen contiagmais ar cae, 
LL. 44^. nir* lensat cai na eonoxc dun primsligtt^ immach, BB. 
407» 4. for oen choi, LU. 65^ 28. Cognate verbs are o shunn cai 
Helesseus ar eel, henee Elisha went to heaven. Salt. R. 7223, and in 
O'Dav. 52 and 71, s.v. cae: aracae osar sinnser the younger goes 
Ufore the elder, and perh. ro-chim, Cf. Corn, ke 'go,' pi. keugh 
' go ye ' ; Lat. cio, eteo ; Gr. kivj, Kivvfiai, Ktveiv, 

caid pure, chaste, E. 15. So Corm. andO'Cl. Frequent in Fel. Cogn. 
with Lat. eastus from *cad40'S. Another caidh .i. uasal, O'Dav. 72. 

cail protection, p. 1 note, cail .i. cornet, H. 3. 18, p. 66^, LLec. 
Toe. and Corm. s.yy. bracille, caile, culpait, cerchaill. comad moidi 
a cail, Bk. Fen. 414, which Hennessy renders 'That the higher 
might be their fame.' 

caimse (MS. cuimsi) F. shirt, Ff. 35. caimmse Corm. caimse, 
H. 3. 18, p. 67*. pi. n. teora caimsi, LU. 94, 1. 14. Borrowed like 
^.eamse. Com. cams, A.S. eemes, from Low Lat. (Gaulish?) eamisia. 
Cognate are N.H.G. Hemd and Skr. gamulya. Whence is W. hejls ? 

caindlech bright, E. 1. Frequent in Fel. rose cainlech glas, LU. 
130^ 25. Derived from eaindel, LU. 89^ 14, and this borrowed 
from Lat. eandela, 

Phil. Traiu. 1891-2-8. 4 



60 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

cais si/Sf Ff. 39. So O'Cl. From *ka8'ti. The Irish cognate 
verb is ad-chiu (*ati-ke8i6) * I see * ; pass. pret. sg. 3 ad-chess : ^kes, 
of which the Bo-called Skr. ^caksh is a reduplication, Windisch, 
in Kuhn's Zeitschrift, xxi. 425. 

1. caise eause^ Ff. 46. cais, O'Br. cas .i. caingen .i. a cdusa, 
H. 3. 18, p. 66*. 

2. caise love^ p. 3, Ff. 76. caisi, D. 60. So H. 3. 18, p. 612», and 
O'Cl. From *^a«-^w)-. cais 'love,* wjtWaw * hatred,' are from the 
same root kas^ perh.= Skr. ^kash * to itch ' ? 

caisse, hatred, p. 3, Ff. 76, D. 60. So H. 3. 18, p. 612% and 
0*C1. From ^cat-tio-, cogn. with Gr. icoTov, or from ^ead-tio, cogn. 
with Goth. hatiSf £ng. Itate. A shorter form, cats, occurs in Amra 
Choi. 51 : Tar cais cam-denam for ?uitred (return) well-doin^f and cf. 
miseaiSf infra s.v. neoit, gen. miscsen, LL. 344<^ 34. The W. cas 
' hateful,' easedd ' hatred,' are the British cognates. 

calad Aarrf, D. 13. SoL. Lee. Voc.O'Dav. 65, and H. 3. 18, p. 663». 
calath, LU. 106**. caladh .i. gand, Stowe XIX, p. 39». Compd. 
caladgela, H. 2. 16, col. 716. W. caled, Br. caht. Kluge connects 
N.H.G. held from *hale]7. The Ir. ealma * brave' is nearer in 
meaning. 

calb (calbh, CR.) hardness, Ff. 22, cognate with calad, 

calb head, Ff. 71. So L. Lec.Voc. and H. 3. 18, p. 615^. do 
chalb re cloich, O'Br. gen. sg. : Deaith doerchrau i crichaib cain 
cailb .i. diaith .i. ni haith doerchra[n]sat mo suile hi caincrichaib 
mo chind . calb .i. cend, LB. 241^. dat. adam aed (.i. adam suil) im 
chalb (.i. im chind), LL. 208* 27. Lat. calva, calvaria? 

cam combat, D. 19. So Corm. and O'Dav. 64, 71, 103. midach 
teora cam a brave of three combats. Borrowed either from A.S. 
camp or (as K. Meyer thinks) from a prehistoric form of O.N. kapp. 
The spelling caam in Bk. Arm. 13* 1 {* lignum contensionis quod 
uocatur caam apud gentiles) has not been explained. 

cann a can, Ff. 55. So Corm. s.v. eseonn. An early example is 
oann airgit, Bk. Arm. fo. 17*2. Borrowed from A.S. canne, F. 

cap chariot, cart, Ff. 19. So Corm. s.v. capall, and L. Lee. Voc. 
hearse, bier, ructhar i capp .i. i carr, LU. 6^ 30. In H. 3. 18, p. 624, 
cap is glossed by ordchar * bier.* 

ckma flesh, Ff. 30. So Corm. and O'Cl. cuirm ecus cama, Corm. 
s.v. ore tr6ith. cdma samaisci do indarggain, LL. mir do chamu, 
Corm. (Laud 610), s.v. imbas forosnai. ni ba lam laich lethas 
cama caurad, LL. 87^ 43. noco n-ernaba cem na cama dit asind ait 
hi tudchad, LU. 86* 20. hi ech, or, argat 7 cama, FM. 1088. 



GLOSSARIAL INDEX. ^^iL<^' 

carna, cuirm, cnoimes, cadla | it e ada na samna, fleshmeat, ale, 
nut'WMstf tripe, these are the dues of All Saints* Day, Bawl. B. 
612, fo. 98^ 2. 8g. voc. a chama cruni, LB. 253*. Compd. : a 
chomla do thirm-chamu, LB. 21 5^ Cogn. with Lat. caro. 

carthoit (MS- caruid) piouSy Ff. 71. So H. 3. 18, pp. 66» and 
634* {earthdtt, cardot), and O'Cl. {cartoit). Bai carthait, (Boi) 
caidait, Amra Choi. 76. The noun cartoit, Conn., like W. cardod, 
is borrowed from an obliqne case of Lat. caritas, 

cast ehasU, Ff. 71. So H. 3. 18, pp. 66% 613% 634^ Bai cath, 
bai cast, Amra Choi. 73. From Lat. casttts, as eastdit, Trip. Life, 
xvii, from eastitdtem. 

castas, p. 1. From Lat. eastus. 

cath doaim, E. 28. Both lemma and gloss are obscure. 

1. ce night, Ff. 52, ace. to H. and 0*R. From *skaiu (cf. Skr. 
chhdjjfa, Gr. ffKia), as caeeh ' blind,' from *skaiko-s, 

2. ce companion ? Ff. 52. So O'Cl. . 

3. ce earth, Ff. 64. D. 29. So O'Cl. Hardly from *qhj6, 
cogn. with Gr. x^<^'' x^^'''^'^* ^^^' kshd? Is it due to a mis- 
understanding of the common phrase for bith chi ? 

cechair a slough, D. 28. So O'Cl. Ace. to O'Dav. 69, the upper 
pert of a eeehair is marsh ; and the lower water ; while in the case 
of a lathach, the upper part is water, and the lower mud. conid 
corastar dar droichtiu' ind liss isin cechair, LL. 117* 35. cobair 
bo hi cech/iir, LB. 204* 19. cobair cethrae ar choin 7 cechair, 
Harl. 5280, fo. 38». ceachair dirt, filth, O'Br. Hence the adj. 
eeeharda: nirpsa grenach cecharda, Bawl. B. 512, fo. 112* 2. 
The Skr. gdkan and Gr. tcoirpo^ may be cognate. 

cecht power, D. 19, £.11. cecht .i. cumachta, Stowe XIX. 
8k. gakti, ^ qak. O.N. Uttr. 

1. ceis spear, D. 16. So O'Cl. From *kesH. Cogn. with Gr. 
Ktrrpov, Lat. castro (from *eastrum knife), Skr. gastra-m. 

2. ceis cuairt, circuit, visit, D. 53, = ceis curit, L. Lee. Voc. 
From *ked'ti, cogn. with the verb ceisim (ro ceis romna n-des .i. ro 
cheimnige i rosomaine na aese, Bawl. B. 502, fo. 62* 2), and this 
irith Lat. eado, eedere, cessare, Gr. K€Kahoino, Zend gad gehn, 
Terlassen, fallen ( Justi). 

1. eel death, p. 3, note 2. So Conn., L Lee. Voc. and O'Cl. fa( 
foa doluid for a cheal, LL. 370^ 23. Dochuaid for eel . . . for 
slicht sen, Salt. R. 3685. not-chuirfe in flaith ar eel, LL. 278, 
marg. inf. Cf. O.N. hel 

2. eel heaven, E. 29. So Corm. and O'Cl. shunn cai Heleseus 



52 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

ar chel eter noebu for noeb-nem, Salt. R. 7185. Borrowed from 
Low Lat. *cHuin, caelum, 

c61e servanty D. 45. So O'Dav. 63, O'Cl. and see O'Don. Supp. 
8.V. ceile D^ * sei-vus Dei,' anglicised ctddee. ceile caich a comhair 
a chomdedh tJ^e servant of every one in his lord's presence^ O'Dav. 63. 
dlegar don ceile denamh drechta im dun na flatba the tenant is 
liable to make a drecht(?) round the ehieftaivCs fttrt, 0*Dav. 78. 
See also LU. 68» 13, 16, and 74^ 9. Compd. soir-chele (gl. 
libertus), Wb. 10* 23. Perhaps originally identical with die 

* socius.' 

celt raiment, D. 47. So Corm., H. 3. 18, p. 66, and O'Cl. 
Anglicised kilt, celt asas treu, LU. 95** 38. imdatuigethar celt 
clithargarb, LU. 95^ 25. dechelt, Corm. Hence celtar^ LU. 79* 
20, celtair, H. 3. 18, p. 66^ and celtach 'kilted,' Bk. Fen. 78. 

celtair spear^ Ff. 36. D. 16. So 0*Dav. 68. pi. n. co mbeodutar 
celtra cat ha Oaier ! Corm. s.v. gaire : dual n : di cheltir ina laim, 
LU. 133* 26. Cognate is diceltar, LU. 133* 44=diceltair the 
shaft of a spear without an iron on if, Corm. s.v. gaire. "W. paladr 

* hastile,* Gr. TreXrrj and ttoKtoi/ shaft j pole, 

cenel children^ D. 12, rather means kindred, and is^W. cenedl 
gens, genus, natio. Hence cenelach, Wb. 23* 32. Cognate with the 
Ir. verb cinim *I spring from, descend,' O'Don. Supp., the verbal 
noun ciniudf and the noun cine (leg. cined ?) • a tribe,' O'Don. Supp. 

cenntecol .i. triallad, E. 10. .i. tt^railead, L. Lee. Voc. 
coinwtegal .i. truailb(^ 'corruption,' Stowe XIX. These glosses 
are due to a misunderstanding of Cormac's ceinticul .i. combrec 
rotruaillned ann .i. cenical (it is) Welsh that has been corrupted 
therein, to witj * centecul.' He explains it as * a name for 
wool whereof they make a blanket.' So H. 3. 18, p. 67* s.v. 
centecul, and see O'Cl. s.v. cinnteagal. The Middle- Welsh form 
is kenhughely Welsh Laws, i. 308, where the translator says, that 
cynnygl (as he modernises kenhughel) is "literally 'wadded'; 
perhaps a gambeson." From a Low-Lat. *contegulum ? 

ceo besides ? D. 60. Sceo 7 ceo 7 neo tr( coraoccomail Goideilggi, 
LL. 26* 1, and see Amra Choi. 73 (ceo ni coirm, ceo ni sercoU). 

1. cerb slaughtering^ wounding^ D. 20. sg. gen. crothid conchend 
catha ceirp, LU. 47* 22. mac in chirb, is e a bes guin, LL. 148* 14. 
Apparently identical with 2. cerb. 

2. cerb cutting, D. 20. So O'Cl. cerp .i. teascad, O'Dav. 63. 
An adj. cerb, cerp 'keen,* occurs in Salt. R. 891, 4767, 
6913, 8087, the verb nos-cerband in LU. 109* 23, and the participle 



GLOSSARIAL INDEX. 53 

eerhtha in LTJ. 56^ 12 (belt cuirp cerbtha, cainfit mnd, lodies will he 
hacked, women will wail), 

1. cem victory, Ff. 18, D. 18, E. 27. So O'Cl. pi. dat. do 
cernaib .i. do buadhaib no do gnimhuibb, Amra Conroi, H. 3. 18, 
p. 49. pi. nom. mor a ceama ^a tuireamb, Dan do Brian na 
muriha. gen. is tu laech na cemd 7 na comram, LU. 100^ 17. Hence 
cemach, LL. 294^ 12. cearnacb .i. buadach vietoriotM, Stowe XIX. 

2. cem a man, D. 18. So O'Gl. 

3. cem act of eomuming^ D. 36. So O'Cl. 

oerr a lopping, Ff. 25. So O'Cl. Cognate with cirrbed, Bk. 
Fen. 216, and the verb ro cirred iarom ind lam sin, LU. 98^ 2, 
cirrfitir colla, LL. 254^ 24. 

cerrbach plunder, Ff. 43. So O'Cl. 

cetnait (MS. and O'Br. ceathnaid) a sheep, Ff. 17. So O'Dav. 
62. roearba rocetnaitt comair a tigi teacht ina lins he mdered a 
Ug sheep near his home to come into her fold, 0*Dav. 72, s.t. comair. 
This seems a dimin. of eet, the dat. pi. of which citoibh (leg. cetaihh) 
is glossed by cairidh [leg. cairigh] finda in K. 3. 18, p. 49. He 
Dce eetaimni A. cairidh [leg. cairigh], Duil Laithne 117. 

ciar black, E. 28. So Stowe XIX, p. 30», and O'Cl. pi. ace. fri 
(alia ciara .i. fri demna .i. elta duba demoniorum, Brocc. h. 91. 

cichlad a lopping, Ff. 25. ceachladh .i. ciorrbhadh, O'Cl. 

cicht (from *qviktu-) engraver, D. 44. So in H. 3. 18, p. 66**. 
debt .i. geibire .i. rindaire. cicht .i. geibiach, Stowe XIX. cichtce 
,L geibiach .i. rannaigh [leg. rinnaighe], O'Dav. 63. Cogn. with 
Pietus * Pict,' and perh. W. pwyth * stitch.' 

1. oil (MS. ciol) wrong, D. 41, E. 29. So Conn., L. Lee. Voc. 
O'Dav. 66, and O'Cl. partial, false, O'Don. Supp. Compounds are 
kth'chil and ciUhreth. 

2. cil (MS. ciol) death, from *celu, D. 43. So O'Cl. co tarrlaic 
iaodb fir cil so that he gave a truly deadly shot, O'Dav. 64. 

dairiu division, E. 29. So Corm. clairiudh .i. roinn, OTl. 
cbiireadh .i. foghail (leg. fodail) no sgaoileadh, O'Flah. 

clamar satire, E. 17. .i. air no escuine, H. 4. 27, p. 67. clumar 
.i. air, Stowe XIX. clamor .i. air, L. Lee. Voc. Borrowed from 
O.N. kldm-or^ * a libel in verse.' Cogn. with cldmaithe * reproach,' 
LB. 228* 11. The Ir. ^/dm(in the phrase gldm dicenn) is either 
borrowed from or cogn. with O.N. kldm * foul language.* 

clandad thrusting, D. 40. So O'Dav. 70 and O'Cl., L. Lee. Voc. 
and H. 3. 18 cited in O'Don. Supp. s.v. elannsad. Verbal noun 
of clandaim — nos-clanna in sciain ind, LL. 269^ 23 ; corthe no- 



54 IRISH METRICAL GLOSSARIES. — ^MR. STOKES. 

chlantais intan bad maidm n-imairic, LIT. 86^ 42 ; clannsad cleatha 
doghra thrid, BB. 31* 26 — borrowed from Lat. planto, whence also 
"W. plannu * plantare, serere.' 

cle wrong ^ D. 41. So 0*C1. Seems derived from oil q.v. But 
perhaps the e is long and cU would then be identical with cU 
* sinister/ W. cledd, Goth, hlei-duma. 

cliste readi/f E. 22. cliste active^ Bwift^ O'Br. laoch cliste, 
O'Don. Supp. B.v. leirg : cliste ar a laimh dels chl^, marcach ain- 
chlisde, 0*Br. A living word : fear cliste * a clever, sharp, smart 
man,' 0*Growney. Cogn. with clisid, O'Dav. 70, s.v. cleth, and the 
8-pret. tf/iMW, LU. 69^ 13. 

cloch, F. stone^ p. 4. Prob. cogn. with calad, q.v. 

coart landholder^ D. 15. So O'Cl. and see O'Dav. 62. coairt, 
Conn, hi coairte .i. i mbriugaide, Rawl. B. 502, fo. 62* 1. 

cob victory y Ff. 30, 50, coph p. 1 , E. 27. So Corm., O'Dav. 63, and 
OTl. So also H. 3. 18, pp. 67% 634». Hence the name Cohthach. 
The Gaulish names Coh-launo^ Cob-nertus, Coh-nerta, Ver-eohius are 
also prob. cognate. Fick connects 0. Korse hap^ Eng. happy, Skr. 
^ag-md * powerful.' 

cobra shield^ Ff. 37. So O'Cl. pi. gen. tri maic TJislend cobra 
ngarg, LL. 114» 35. 

coemthecht (=com-imm-thecht) accompanying , D. 49. sg. dat. 
i coimthecht dagdoine, Wb. 1 6^ 2. Hence coemthechtaid ' attendant.' 

coic a secret, Ff. 47. So Corm., O'Dav. 63, s.v. coig, and 
Stowe XIX. ni cuala coic nuin he heard not an evil secret, H. 3. 
18, p. 61^ coig .i. run, LL. 393*53. coic .i. comairle counsel^ 
H. 3. 18, p. 66^, and coigli .i. comairli, 0*Dav. 63. coicle .i. 
folach concealment, ibid. 61^, may be cognate. 

coice, coidche (O'Br.'s coice) mountain, Ff. 27. 

coidche armpit, Ff. 41 : reading doubtful. L. has caoiche is coll. 
A gloss in a note to Fel. Aug. 4, on the line Molua mace oche, 
preserves another word for 'armpit': ^*ocha apud ueteres ochsal 
dicitur prius," where ocha may be cogn. with Lat. oculus. 

coimmdiu (*com-mediot-) lord, D. 47. So O'Dav. 72 and O'CL 
sg. gen. coimded, Wb. 9«, Ml. 26^ dat. coimdid, Wb. 8^ 27«. ace. 
coimmdith, Sg. 29*. Cogn. with Ir. midiur, Goth, mita, Gr. 

coindelg counsel, Ff. 10. So O'Cl. and O'Don. Supp. sg. gen. 
fer condilc ftr, Trip. Life 210. ace. contuaset aes na bnidni uli a 
condelg, LIT. 93, Kne 15. 

coing a fallowing, E. 29, *a going together,* O'll. 



OLOSSARIAL INDEX. 55 

coit a wood, E. 4. So Corm. s.v. Sailchoit. Borrowed from 
W. eott, now coed. Cogn. with Goth, haithi, 

colg F. 9word^ Ff. 36. So Corm. and O'Cl. nf for braigtib 
dam na bo cloithir colg (.i. claideb) mo ruanado, LL. 277^ 28. cole 
oc mo cboilc-se, LTJ. 6^ 10. The spelling <?<i/^ (=W. caly veretrum, 
Br. eaUh) seems more correct, calg d^t, LU. 87^ 18. 

coll head^ Ff. 38. So O'Cl. and O'Br. gen. cuill: daceird bracht 
en feic a cnill .1. a carpait, LL. 208* 12, whence it seems that 
coU means ' jowl.' The dat^ and gen. pi. seem in the phrases dicetul 
do chollaib cend, Kawl. B. 512, fo. 114^ 1, dic«tal di cennaib coll, 
Laud 610, fo. 57^. Hardly borrowed from A.S. ceafl. 

aoM food, Ff. 20. So Corm. s.v. asgalt, and H. 3. 18, p. 615^. 
cen cholt .i. cen biad, LU. 8^. eter cond sceo colt .i. enech no biad, 
LB. 24 1\ Literally, perh., iomething cooked, cugn. with Lat. ealeo, 
Skr. erta, Lith. niltas, 

com-etaid (gl. custos) p. 4, cometid la mmaccu (gl. paedagogus) 
Wb. 19«. From corn-it * servatio,' GC.» 793, 871. Compounds 
are ciU-comMd and fos-chorndtaid, 

comm covering, shelter, protection, D. 47. sg. dat. i com argit 
gil, LB. 233^. ace. tabar com (.1. coimhed) dun. Bawl. B. 512, 
fo. 35\ A sister-form coimm occurs in FM. 1599, p. 2140 
{ia coimm ria gcioih dosum), the dat. sg. of which is frequent 
in the expression f6 a ehoim, LU. 68* 10, and see Lism. Lives, 
2025, 2027, 2398, 2396. It is the Irish reflex of Gr. ico/i/3o9, and is 
cogn. also with Lat. cingo, 

com-maim wife, lit. yoke-fellow, £. 23. So L. Lee. Yoc. and 
O'Day. 70. commam. Trip. Life 14. sg. gen. comaime, LB. 198^, 
dat. da chommaim, Bk. Fen. Hence commdmue * matrimony,' sg. 
gen. commamsa, O'Dav. 70. From eom=Jja.t, cum and mdim, cogn. 
with mam ' jugum.' 

commar nose, Ff. 40,»commor .i. sron, O'Cl. Meaning doubtful. 
In LL. 108* 30, it seems to signify some other part of the body: 
corntar ecnaig a thruim 7 a glainene for ettegail dar commur a 
chraes 7 a bragit. So in LU. 15^ 23: ar cend caid comarli, ar 
commor moradbal. 

oomrair F. caee, D. 51. comrair taiscedai, Bk. Fen. 12, better 
comrar (gl. capsa) Sg. 36% 92^, LU. 114*32. ata comrar chloche 
i mbi and hi talam, LU. 134^ 3. a muintorc argit for a chomrair, 
ibid, L 6. comrar conga each cethra, LL. 293^ 19. Metaph. 
comrar dana, LL. 187* 15 and Bawl. B. 502, fo. 62^ 2. is hi tra 
comrair chrabuid leosum, LB. 10* 15. 



66 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

conn sense, Ff. 27, 45. So O'Cl. and Stowe XIX. arnacha ndecrad 
a chond nach a cbiall, LU. 79* 35. Bg. dat. asa cund, LTJ. 48^ 11. 
bid crad da ce[i]Il is da cond, Bk. Fen. 246. ace. tncussa duib 
sere mo cbride 7 mo cbond, LL. 302^ 34. is fin romedair in sluag 
CO mbatar cen cbond cen cbeill, LL. 282* 1. Perbaps cogn. witb 

Gr. KOVVCtO. 

conruiter, p. 3=conr6iter, Amra Choi. 13, 43, where it is glossed 
by eain rdfitir, no rochomet. The con is cogn. with xatvo^, Lat. 
re-eens, 

consal M. counsellor, E. 1 0. So Stowe XIX. pi. gen. Ni cumangar a 
rim a ngradaib 7 a ceimendaib ar imat a consal ... 7 a ndiuice 7 
a centure it is impossible to rehearse them in ranks and in degrees 
because of the multitude of their consuls . . . and their dukes and 
their centurions, Eawl. B. 512, fo. 87* 1. ace. consala, LB. 157^. 
From Lat. consul. So Ir. irchdnsal, LB. 180^, from Lat. proconsul, 
as Erpoint, LL. 222^, 232*, from Propontis, 

1. core clan, D. 12. So O'Cl. and H. 3. 18, p. 66«. O.Ir. 
corcu: in populo Korku Reti, Adamnan*s Vita Col. ed. Reeves, 
p. 89. de genere Corcu-Chonluain, Bk. Arm. 11^ 1. in regiones 
Coreu-Temni, ib. 13^ 2. de genere Corcu-Theimne, ib, 15* 2. 
Corcu Dubni, LL. 292* 47=Corco Duibne, ib. 277^ 22. in chland 
Corco Laigde, ibid. 210* 51. in fines Corcu Ochland, Trip. Life 94. 
mace Rime noeb Chilli Chorcu-Roide, ib, 138. i tir Corcu-Themne, 
ib. 122. 

2. core (MS. tore) hair, E. 27. So L. Lee. Yoe. Cogn. with 
Gr. Kap, Lat. crinis ? or with xepKo^ * tail ? ' 

cose instruction, correction, D. 33, sg. gen. coisc, cuisc. cose .i. 
tegose, Stowe XIX, p. 39^. dat. cose: teit-seom cosin n-araid 
dia chose, LU. 64*. Hence ticosc, tinchosc * instruction.' ro-coscad 
(gl. correptus est), W. cosp. From con+y/seq. 

cot {a cutting) victory, E. 27. cod, L. Lee. Voc. and 0*Br. 
Form and meaning doubtful. Cotan .i. laoch *hero,* Duil Laithne 
38, may be derived, and Gr. tcevreto, /coVto?, may be cogn. 

coth/00^, E. 28. So Corm. and 0*C1. coth cibum .i. biadh ar in 
berla amuii asb^rar cothugt^ na n-indile, Harl. 432, fo. 3^ 1, and 
Bee H. 3. 18, p. 634*. 

cria buying, D. 25. So L.Lec. Voc. and H. 3. 18, p. 633^ But 
cria is a verbal form from crenim=S\iT, krindmi, and means 
* emas ' : ni ria, ni cria do dodhamnu, thou shalt not sell, thou 
shalt not buy, to or from an unqualified person^ such as a thief 
or a little child, O'Dav. 79 and Laws, iii. 58. 



GLOSSARIAL INDEX. 57 

crich F. (gl. finem) p. 5, now erioch. WiDdisch connects this 
word with KpUo^ and circus. 

crimthann (MS. criomhtan) fox, Ff. 61. So L. Lee. Voc, 0*C1. 
and Mart. Don. p. 160, where the author says that criomtann, 
in the hard Gaelic ( Gaoideilg crtsaid), is the same as ' fox.' 

cro narrow, £. 20. So L. Lee. Yoc. cro .i. timargain, K. 3. 18, 
p. 51\ H. 4. 22, p. 62*. 

croch, high, £. 27. croch gach n-ard 7 gach n-inn, K. 3. 18, 
p. 67*. Cogn. seems croich .i. nachtar hainne * cream,' 0*Dav. 68. 

cmm-duma dunghill, E. 29. So Corm. who cites cin chon 
crumduma a crime of a dunghill dog. crumdub .i. ottrach, O'Dav. 63. 
The literal meaning seems * worm-heap,' from cruim == W. pryf, 
^kr. krmi, Lith. kirmi-a, and duma cogn. with Gr. Grifiwv, Owfidv, 
£ng. dam. 

cuanna (MS. cnadhna) hill, Ff. 27. So O'Cl. and O'Br. Cogn. with 
O.X. hiinn * knob *? The W. cwn, cynu, erchynu, and the Old-Celtic 
euno- and *Ap-tcvt/ta {oprf) are from the same root, with which 
Windisch (Kuhn*s Beitr. viii. 40) connects O.H.G. hUn * giant,' 
pi. hikni. Or is it from yjqeup, whence Lith. kaiipa-s * heap,' 
O. Pers. kaufa ' mountain ' ? 

cuar crooked, Ff . 44. So O'Cl. ras-tarraing da corranaib cruaidi 
coara, LL. 230*. cuar-sciath, Eawl. B. 602, fo. 47* 1. pi. n. 
cuar-sceith, LU. 87^ 17. Hence the verbal noun cuarad, LL. 
236* 13. 

1. cud (cuth?) head, D. 62. So Corm. s.v. descud: cud .i. 
cenn, Duil Laithne 4. cudh no cuth, O^Cl. lea hanail ho chud, 
LU. 128^ 10. caut .i. cenn, H. 3. 18, p. 67*. cuth, O'Br. for- 
cuth, ib. 

2. cud (cuth ?) an offering, D. 62. 

1. cuib dog, Ff. 16. .i. cii, O'Dav, 71, s.v. ciunes, and O'Cl. 
ma chuib (.i. mo chu) asa ho, darchinnius (.i. darlecius) co dfan 
mo chuib .i. mo choin, LL. 208* 12. In the Amra Conroi, 
H. 3. 18, p. 49, cuim .i. cu is perhaps the same word. 

2. cuib taord? Ff. 30, or is it a sister-form of cob * victory,' q.v.? 
cuime protection, Ff. 31 — leg. cuimne, as in O'Br. = 0. Ir. 

coemna, caomhna, O'Cl. 

coimleng F. combat, £. 8. gl. agon, Gild. 19. cuimliug .i. comrac 
H. 3. 18, p. 624. sg. dat. i comling, Trip. Life 666, ace. doroine 
caimleng, note to Fel. Apr. 19. pi. ace. cumlenga, LL. 64* 6. 

cuinsel face, Ff. 40. cuinnseal, O'K. A sister-form is cuinnei 
•i. aghaidh, U. 3. 18, p. 639. coinso, LU. 87^ 11. cuinsiu chorad, 



58 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

LIT. 106»7. coindse .i. drech, H. 3. 18, p. 67% and O'Don. Supp., 
and see O'Dav. 62, s.v. cucht. 

cuislennach /7t]p^r, D. 44. pi. n. cuslennaig, LL. 261^ 30. gen. 
pi. cuslennach, LU. 88^, 97**. Derived from cuisli .i. crand ciuil, 
L. Lee. Yoc. and Stowe XIX. gen. pi. cuislenn: bindfogur na 
cuislend, LU. pi. ace. na cuislenna (gl. Tenas), Gild. 222. So 
avpif^^ is used of any duct or channel in the body. 

1. Q\A protection, p. 1. Ff. 30, 50. D. 45. So H. 4. 22, p. 67, 
and Stowe XIX. cia beith do iar cul, though he might have it under 
his protection t Laws, i. 1 50. A sister-form culu occurs : doberind 
culu (.i. comet) ar gart (.i. ar einech) Find, LL. 208^ 31. These 
words, like eulaid 'raiment/ are cogn. with N.K.G. hulle, 

2. cul chariot y p. 1, Ff. 19. 50. So Corm., H. 3. 18, p. 60% 
O'Cl. cul, L. Lee. Voc. culu (.i. cul .i. carpat) tria neit (.i. 
cath) LU. 6^ 24. fonoad (i. ro immanad) col carpait dun and, 
ol se, LU. 122^ 38. atchlunim cul carpait, LL. 83^ 11. Compd. : 
culgaire na carpat, LL. 109^ 23. O.Slav, kolo * wheel, circle.' Lat. 
eolus, Gr. w6\o9, 

culmaire chariothuilder, E. 9. So H. 3. 18, p. 66^ and O'Cl. 
Also charioteer: is culmaire .i. is cairptech, LU. 109^ 40. cul- 
maire .i. cairptheo[i]r, Stowe XIX. culmairi .i. cairptheoir, L.Lec. 
Voc. culmhaire wheelwright, O'Br. 

cum a vessel? E. 28. Used topographically: ingen DinQ o 
Chum D(nil i crich Corco Duibne, LL. 277^ 22, where cum is 
=W. cwmm ' vallis, convallis.' Gr. Kv/ifiif, KVfifiov, *cup,' N.H.G. 
humpen. 

cumlachtaid pig, E. 29, is rather a sucking-pig, Corm. com- 
lachtaid, O'Dav. 62, where seven other words for * pig * are given. 

cum-rech N. (gl. ligo), p. 1. cumrech, sg. dat. cuimriuch, pi. n. 
cuimrecha, is a noun, not a verb, and means * bond, fetter.' The 
cogn. verb is conriug, pres. ind. sg. 3. conrig (gl. alligat) Ml. 23<' 2. 

cunn sense, Ff. 27, a sister-form of conn q.v. Cogn. with cunnlacht 
* wisdom.' 

cur M. hero, champion, D. 39, also caur, LU. 85^ 32, 114^ 41, 
LL. 106^ 9. sg. gen. da suil churad i cind in chlaim atas-cim 
a heroes two eyes I see them in the leper^s head, LL. 303^ 14. 
caurad, LU. 77^ 33. pi. n. cauraid, LL. 106^ 11. curaid, LL. 
256* 16. Compd. curath-m(r. Hence caurata, LU. 80» 28. Cogn. 
with W. cawr gigas, Com. caur in caur-march (gl. camelus), Skr. 
gavira, gura, Gr. tcupo9, xvpio^. 

curson a sage (arracht de image of God?), Ff. 71. So H. 3. 18, 



GLOSSARTAL INDEX. 59 

pp. 66*, and 634*, H. 4. 22, p. 67, and 0*C1. curson e a mbeasoibh 
brughadh, Dan do Bhrian na miirtha, 116. Tbe oaraon of the 
Amra Senain, LB. 241% is perhaps a mistake for curson, 

cusal strength, Ff. 23. courage, 0*Br. Hence perh. the adj. 
cossalach, LIT. 96*. 

dae house, D. 54. So 0*C1., who quotes rioghdhae {rkg-dae) 
'palace.' daoe, D. 57. Possibly cogn. with Oatfio^ * oUta, Hesychius, 
Qse. fdmaum * to dwell/ Skr. dhaman, 

dag good, p. 3. Ff . 72, D. 59. So Corm. s.v. aingel ; droch do 
drochaib, dag do dagaib, Corm. Tr. 61 and H. 3. 18, pp. 68*, 634^ 
Boir dam .x. ratha (.i. lis), .x. treaba ( i. tighe) dagha (.i. maithi 
no cona ndilius feibh), Amra Conroi. Derb dag i mba, Amra 
Choi. 46. W. da. Can it be cogn. with Gr. raxvv (from ^Oaxvsi), 
compar. Oaaawv, as (ace. toGollitz) ^vv is = Skr. dyit'i Generally 
used as a prefix, another form of which is deg, 

daig/rtf, Ff. 72. So Stowe XIX and H. 3. 18, pp. 69*, 635». 
a ^stem, sg. gen. mnir ndaiged. Salt. R. 910. Moiling lassar 
daiged, LL. 305* 25. sg. ace. argain in daigid cech n-aidchi, 
LL. 107* 21. pi. dat. go ndaighthibh .i. go dteinntibh, OTl. 
Skr. y/dah from *dhegh, *dhogh, Lit. degti *to bum.' Goth. 
iags. A.S. dag, the bright, warm part of the twenty-four hours. 

daUn foster/ather, D. 46. So 0*Dav. 73. dadan, 0*Don. Supp. 
Oeugus an Procchai, datan Dermatai, Udth Beinne Etair, 70. 
daitean .i. oide, O'Flah. datiucdn, LL. 279» 13. datan, O'Br. 
dimin. of *dat, a child's word, like W. tdd * father,' Eng. daddy, 

datnait (MS. dathnuid) foster-mother, nurse, D. 46. So O'Dav. 
73, B.v. datan ; dathnaid, O'Br. ; but dadnait, O'Don. Supp. is 
perhaps the right spelling. If so, cf. Gr. ti^^ grandmother, from 
*&flOfl, and Lith. dede, 

de division? Ff. 13. Cogn. with the form assumed by the 
numeral two as a prefix. 

dega (dede?) wind, D. 25. The only similar word with this 
meaning is daghar .i. gaoth, O'Cl. and OTlah. daugar augrach ben 
daire the oakwood*s wife is a warring wind. Bawl. B. 512, fo. 
62» 2. daghar, O'Br. 

deichen dimsion? Ff. 13. Cogn. with dechur 'difference, dis- 
tinction,' Ml. 26* 1. 

deiltre (deilltre?) is explained, D. 39, by "gods of wizardry 
for travellers astray." 0*C1. has only deiltre .i. dee draoidhechta. 
O'Br. wrote di-Utre, 

deim taking away, D. 61. deim .i. onf is demo .i. digbaim, K. 



60 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

3. 18, p. 614^ Mack, want,' O'R. cuil deim de eot, cuil deim 
de formut, Amra Choi. 105. From de+emi-. 

deime evening ^ Ff. 54. deme, Corm. do deime .i. dorchadas, K. 2. 
16, cited O'Don. Supp. s.v. diorna. deimhe .i. dorchadus oidche 
* darkness of night,' 0*C1. Derived from deim * dark.' The demithir 
cited hy Windisch as perhaps compar. of deim, is a mistake for 
deinithir, compar. of dian q.v. 

del Uat, E. 11. .i. sine bo, Corm. deal .i. sine, Stowe XTX. * 
of. oc a diul sucking her : Gr. 07f\y, L&t.fellare (Jelare\ O.H.G. tHa 
f r. ^dela, 

demal a taking away, D. 61. See deim, but qy. if this is not 
a mistake for dirndl .i. demon, K. 3. 18, p. 69^. 

der F. daughter, girl, E. 24. So Corm. Tr. 61, Stowe XIX, 
H. 3. 18, p. 69^, and 0*C1. Helech der Fabthaire find, LL. 164^ 4. 
Petronilla der Petair, LB. 85. Corapd. leis-dhear stepdaughter, 
O'Br. 

1. derc F. ege, Ff. 30, Ff. 39, p. 2. So H. 3. 18, p. 81*, and 
O'Cl. pi. nom. tuilsiter mo derca .i. mo siili, LL. 208* 7. dat. en 
CO nertaib nual dar dercaib sluag saer .i. us^^ dar suilib soercland 
ocom chainiud, water over the eyes of nobles bewailing me, LU. 
119^ 21. Compds. cammderc (gL strabo), Sg. 63» 4. fliuch- 
dercc (gl. lippus), Sg. 24» 8. Cogn. verbs occur. Cf. Skr. drg * eye,' 
Gr. ^epKOfiai, SeSopKa= -dare in Ir. ad- eon-dare conspezi. 

2. derc a mote in a sunbeam ? p. 2. 

dfan swift, D. 11. dian .i. oband, Stowe XIX : a common word, 
compar. deniu and diinithir (combo deinithir broin mulind) which 
is miswritten demithir in the facsimile of LU. 111*11. denithir ML 
67° 12. Cogn. with Gr. hieaOat, Skr. digdmi, 

di-bad death, extinction, p. 3, D. 57. So Corm., H. 3. 18, 
p. 68° and O'Cl. fo dfbad .i. maith a epiltiu, Amra Choi. 31 : cen 
d(bad, Colm. h. 44. do dibad innti na tol collaide, LB. 168^. co 
dibadh n-aurdlighidh, O'Don. Supp. Compd. erdibad, LU. 109^ 
112*. 

di-beoil dumb, Ff. 72. So O'Cl. dibeoil .i. cin urlabra, H. 3. 18, 
pp. 68^, 634^. The dibidul .i. balb of Stowe XIX seems a 
mistake for dibeoil, 

dfchmairc theft, D. 17. So O'Cl. Eather seems to mean any- 
thing taken from the owner without his permission. Dic[h]mairc 
.i. cin athcomairc 'without asking,' Corm. Tr. 60. diam dichmairc 
.i. can fiarfaigt(/ d'fir bunaidh without asking {leave) of the owner ^ 
Harl. 432, fo. 11* 2. each dichmarc .i. each errach berar o neoch 



OLOSSARIAL INDEX. 61 

cin athcomarc, thus rendered in Laws, ii. 353, ''every forced relief 
or loan which is taken from one without asking permission." sg. 
gen. omun dichmairc, LB. 260^ 86. 

dighail (gl. demitus) p. 5 = di>gabdil, inf. of d(-gabaim. gen. fis 
dfgbala in uilc-sin, LB. 246^. 

dile a following y D. 17. Verbal noun oi*do-lenm. The simplex 
lenii 'adhaerent' occui's in Wb. 29* 23. the perf. sg. 3 Itl in LU. 
68* 41. pL 3 ro leltar, LU. 114^ 15. W. di-lyn, can-li/n, sequi. Cf. 
the fonns from the Skr. ^li 'cling' in Whitney's Roots^ etc. 
p. 148. 

diniath helmet, Ff. 37, diniath, O'Br. Perh. = dfn niath a 
champions protection'^ For din cf. tair dar ndin, a Muiri, LL. 
308^ 41 : ba d{n do nochtaib, Amra Choi. 85. For niath^ see nia 
infra. 

1. dinn (dind?) cain each cnuasaigh p. 1. dinn .i. aibhinn, 
delightful^ 0*Dav. 79. .i. aibind, Stowe XIX. uas domun dind, 
LU. 50»2. 

2. dinn hiU, Ff. 49. So O'Dav. 79. Used metaph. in Amra 
Choi. 27 : bai dind oc libur leig. gen. denna, LU. 9^ 33. pi. dat. in 
dinnib .i. i telchaib, Fiacc's h. 44. Compd. dinn-senchas. 

3. dinn (dind?) prince^ p. 1, D. 55. A metaph. use of 2. dinn 
hill. 

dinnis oath, D. 10. So H. 4. 22, p. 60^ O'Dav. 79 (dinnus) and 
O'Cl. 8ena[d] iar n-aititiu, leith-fiach la dindis a sodain denial 
after acknowledgment^ half fine with oath \i% incurred"] for that, 
Laws iii. p. 108, 1. 8. 

direch straight, direct, p. 1, from dfriuch, df-riug (gl. rectum), 
p. 1, *de-|-regu-8. 

dith end, Ff. 67. So O'Dav. 79. literally death ; destruction : 
iar ndith Conaire, LU. 46» 8. dith co haen * death of all save one.' 
rena dith .i. ria ec, Rawl. B. 512, fo. 104^ 1. cen dith n-6ighe. 
Bawl. B. 512, fo. 69** 1. Lat. letum, from *delum, and dileo, from 
^di'dio ? Skr. ^dd abschneiden. 

diu long, p. 2, Ff. 30, D. 9. So O'Cl. Borr. from or cogn. with 
Lat- diu, diu-derc nder long look of tears occurs in LU. 7^ 16. 

dlomod act of manifesting, proclaiming, E. 23. dlomud .i. 
fogradach, Stowe XIX. Verbal noun of dlomhaim .i. foillsighim, 
O'Cl. dlomas (gl. ait) Ml. 30^ 19, imper. sg. 2 dlom .i. abair no indis 
no raid, L. Lee. Voc, O'Dav. 75, 77. Fel. Aug. 7. Sep. 13. 
Oct. 12. dlomnais .i. labrais, Stowe XIX, seems an error for dlomais, 

dlum abundance, plenty, D. 9. So L. Lee. Yoc. and O'Cl. 



62 IRISH METRICAL GLOSSARIES. — ^MR. STOKES. 

dlumh .i. umad, Stowe XIX. dluim .i. iniadh, ihid. dluim .i. 
imad, O'Dav. 73. Perhaps this occurs (spelt dlom) in LL. 147* 31 : 
in tan atracht in mac cona dlom ferge fair. 

dluth a warp, D. 33. dluth (gl. stamen), Sg. 14^; certle dliitha 
a hall of thread. The expression dluth agm inneach 'warp and 
woof ' is still, I am told, living. 

dobnr water, YL 15. So O'Cl. dobur Conn, dobur .i. dorca 
nc nisce, Stowe XIX. L. Lee. Voc. W. dwfr, Gaul, duhron, whence 
Douvres (8eine-et-Mame). Compd. dobar-chii * otter '=W. dw/r^ 

gi. 

doctus learned, p. 3. Borrowed from Lat. 

doe, dae human being, D. 8. doe .i. duine, Corm. Tr. 61. dae, 
O'Cl. daoi, O'Br. ropo doi n-eit, Rawl. B. 602, fo. 62^ 1. dae 
uais .i. fear uasal, Dan do Brian na miirtha, 

doit lutnd, Ff. 42. So O'Cl. sg. gen. fail mo lama is mo doitti, 
Bk. Fen. 400. dual nom. nirbdar dermaill a df doit, LL. 43» 
last liDe, and see Corm, s.v. Cerball; pi. nom. doti, LB. 190^, 
dat. doitib (gl. manibus), Gild. 164. dobertha ailge arda foa doitib 
CO ru scaich slaide na ngennti, high stones were put under his hands 
till the slaughter of the heathen had ended, LB. 259*. Compd. 
doit-gel, LB. 218*. 

duibh wizardry, Ff. 13. So O'Cl. gen. ni hi deog cen damna 
nduilb, Aided Finn, 61. ace. tre dolb draidheachta, Battle of 
Ventry, 576. Cogn. with dolhhaim * fingo, ' whence laeg doilbthe, LL. 
210» 38, 43. naithir doilbthe, LB. 121*. in nath cetha dolfe 
[dolbthe], Rev. Celt. i. 40, and doilbecht .i. draighecht, Stowe XIX. 

dorr rough, harsh, Ff. 72. So 0*Br. Still living, applied to a 
person of rude manners. 

drecht chant, Ff . 8. pi. nom. drechta .i. duana no laide 7 roscada, 
0*Dav. 72. canitar drechta, LU. 43* 32. gabtha a nduana 7 a 
ndrechta 7 a n-admolta doib, LL. 263* 1. gen. fir denma drecht 7 nath 
7 admolta Ulad, LL. 109* 12. a duaraib drecht, 0*Dav. 72 s.v. duar. 

dreim F. a company, a party, D. 51. Also dremm. sg. gen. 
fodaig na dreime, Kawl. B. 512, fo. Ill* 1. ace. cor-romarbsat 
dreimm moir, AU. 1018. 

dreimne valour (gal), Ff. 17. "RsitheT fury, fierceness : sg. dat. for 
dremniu na farrce, LU. 84* 21. i ndremni in drecain, LL. 86* 28. 
Derived from dremun * fierce,' * mad,' monur ndremun, Petrie'sTara 
178, dreaman .i. dasachtach, Stowe XIX. Hence O'R.'s ^^ dreimne 
s. a cock," the Latin gnllus having some resemblance to the Irish 
gal, CO dremna, Salt. B. 8282, co ndremnai 8346. 



OLOSSARIAL INDEX. 63 

dremnn the Devil, D. 59. Seems inferred from sach passages as 
the following from LB. 176, marg. inf. : 
Trede dremun is mo col, Three mad things whose sin is 

greatest, 
doman, Deman oetu ben : World, Devil and Woman : 

dpe nos-cara ar bith che Whoever loves them in the 

present world 
hie Mac D6 niscon-ta nem. Hath not heaven with God*s 

Son. 

1. drenn quarrel, Ff. 9. So Conn., O'Dav. 73, and H. 8. 18, 
p. 541^ drenn .i. depaidh, ut est drennach .i. deptach, H. 2. 15, 
p. 82. Hence dreandad .i. deabaid, Stowe XIX. Gompd. drenn- 
galach, Salt. R. 944. 

2. drenn trouble, affliction, Ff. 9. buan in drenn, Bk. Fen. 366. 

3. drenn rough, E. 11. So Corm. drenn each croda, O'Dav. 63. 
drenn .i. garbh, H. 2. 15, p. 184. dreand .i. garbh, Stowe XIX. 
drend .i. garb, ibid, dothaet sunda budni drend do chosnam insi 
Herenn, LL. 109*. Hence dreandud .i. gruamacht, Stowe XIX. 

1. droch etraight, Ff. 68. So Corm. s.v. drochet, O'Dav. 73, 
H. 4. 22, p. 61*, and O'Cl. 

2. droch bad, scanty, Ff. 72. droch .i. each n olc, Corm. W. 
incg. Mid. Br. droue, 

1. drocht straight, £.11. So H. 3. 18, p. 653, L.Lec. Yoc. and 
Stowe XIX. 

2. drocht black, dark, Ff. 11, D. 60, note 15. drocht .i. dorca, 
Stowe XIX. Derived from droch .i. dorcha, D. 60 and 0*C1. 
(borrowed from A.S. deorc ?), or is it inferred from Mid. Ir. edrocht 
* bright, shining ' = 0. Ir. Hrocht, q.v, ? 

dron straight, D. 51. So. L.Lec. Voc, O'Dav. 79, and 0*C1. 
droing-cedul, E. 22, = dron-chetul. Cf. Aed atnoi ule oU-doine 
dron-chetal, Amra Choi. 115, dron-cherdach .i. am diriuch im 
eladaio, LL. 187* 7. The compounds (/ron-wa*, dron-mind, dron-oll 
occur in Salt. R. ; but here the first element may be dron *firm,' 
dron .i. daingen, LL. 311^ 18, — as in for mur dron Dune Delga, 
LL. }54^ 12. 

drubh chariot, D. 11. So O'Cl. and O'Br. The latter has a 
" irubhdir cartwright or coach-maker." 

drucht a rising, Ff . 8. Seems a mistake for driucht=driuchd, O'Br. 

draimcli a reader, E. 21. This is the seventh and highest order 
of wisdom; see O'Cuny's Lectures, p. 495. It is probably a 
metaphorical use of druimmchlk (gl. laquear) Sg. 54\ 



64 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

1. du place, p. 1, du Ff. 50, E. 20, Stowe XIX. dii .i. baile 
L. Lee. Voc. and Stowe XIX. co du * ubi,' Trip. Life, p. 4. L 5. 
ata Dia in each du, LL. 281^ 38. Perh. hSb. 

2. du meet, proper, fit, due, p. 1, Ff. 60. So O'Cl. Nf dii duib 
fornert for tuathi dabar fulang, LL. 238^ 19. As a subst. doberar 
dam CO tinniu doib cona du di chormaim dia feiss, LL. 247^ 6. 
Borrowed from O.Fr. rfti=Lat. dehutus for dehitus. Hence dual, 

1. duar/r«i^, E. 10. So L.Lec. Voc. and Stowe XIX. 

2. duar quatrain, D. 42, E. 10. So Corm., L. Lee. Voc. and 
Stowe XIX. 0*Dav. 72, H. 2. 15, p. 184, H. 3. 18, p. 68% 
and 0*C1. Da mbeth an tis (.i. an fer) ag tinm dhuar (.i. tuicsin 
focul no rann), H. 8. 18, part 1, p. 210. 

duchonn mmie, Ff. 67. So O'Dav. 63 (loinniucc no ceol), 
dnchann, O'CI. na hingena . . nobitfsic amran 7 icduchund, Togail 
Troi^ 1086. do gabail a ndrecht 7 a nduan 7 a ndnchonn, Oided 
mae nUmig, 7. 

duchus a quarrel, Ff. 72. Cogn. with O'Cl.'s ducJumn .i. cogadh. 

duibell restless, Ff. 72. So O'Dav. 73 and O'Cl. But in H. 
3. 18, p. 68^ duibell is explained as * lightning * {saiffhn4n). 

duirbh sickness, Ff. 26. 

dumach dark, E. 20, is prob. a mis-spelling of duhach: sam 
dubach .i. dorcha. Bawl. B. 502, fo. 62* 2. 

dun N. stead (baile), Ff. 28, gen. dune, properly walled fortress. 
W. din, Gaul, dunos, duno-n, A.S. tun, Eng. town, N.H.G. zaun. 

1. 6 sorrow, Ff. 73. e sad, p. 2. huile bith ba hoB he the whole 
world which was his [was"] sorrowful, LU. 8** 34, glossed by he .i. 
truag. e mo seel, Goid. 165. Originally an interjection = N.H.G. 
ei. As a subst. oncoin ainble .i. e 7 eit, Harl. 5280, fo. 74**, marg. inf. 

2. e salmon, p. 2. hi, LL. 12^ 42. Better eo, q.v. 

e-c6ir {unjust), evil, E. 23. ecoir, LL. 57^ 33. sg. gen. do dfgail 
ar n-ecora, LL. 220* 30. From the neg. prefix an- and c6ir = 
coair. Ml. 48* 8. 

ecosc/orw, shape, Ff. 53. B. II, .i. cuma, O'Cl. habitus, Z^. 67\ 
sg, gen. ecosca, LL. and LU. passim. 

eic (MSS. eige, eag) moon, Ff . 20. eig, O'Cl. eag, O'Br., who also 
has idireig * change of the moon.* May perh. be from *pengi-, cogn. 
with Skr. pdfas glanz, lichtschein, Gr. 0€77o<f (from *<nr€<y<yo^, 
Bugge, K.Z. XX. 39), mod. Gr. (pe^^api moon, moonshine, and O.H.G. 
funcho, now Funke. 

eidel prayer, E. 12. edel, Corm. eidil, H. 2. 15, p. 184. 

tinech. face, D. 11. So Corm. s.v. gaire. gen. enig. Compds. 



GLOSSARIAL INDEX. 65 

eneclann, enech-gris, enech-log, enech-ruice. Old-Corn, en^ (gl. 
faciem). Cogn. with Qr. i»f tvvi^. 

eislinn unsafe^ D. 45. eslinn, Sanct. h. 15. 

elar sait, Ff. 44. MS. and O'Br. ealar. See salar infra. 

1. £lg Ireland, Ef. 73, and bo O'Dav. 81. Elga, E. 12. Ealga, 
O'Cl. sg. ace. aon cairde fon Elg n-aragar one treaty of peace is 
uUblUhed (lit. fastened) throughout Ireland^ O'Day. 81. gen. 
morthimchell insi Elgga, LL. 207^ 2. Hechtair Elgga .i. Herend, 
LL. 45* 28. rigan iarthair Elgga, LL. 8 1^ 41. etir fini find-Elga, 
LL88*» 12. 

2. elg (leg. elc ?), face, Ff . 73. 

3. elg (leg. elc?) nohle, D. 10. .i. oirrdric, L.Lec. Voc. ealg 
.1 oirdric, Stowe XIX. ealg .i. oirdheirc, O'Cl. 

ellad or ellam gift, D. 42. Treidi ara carthar escara : gnas, ellad, 
eriabra three things for which a foe is loved i use, gift, eloquence, H. 
2. 17, p. 184*. O'Cl. explains ellam (eallamh) by * a dower {coibche) 
which is got in hand ' : ellom roguid ben Qedhe for a ceili the dowry 
vkich G:s wife asked of her husband, BB. 251» 3. 

ellara a wonder, Ff. 22. wonder, astonishment, O'Br. 

ellged burial, D. 37. So O'Cl. eiUgheadh. 

engach a vehement attack, E. 21. In trath erges Aed engach, 
Bk. Fen. 374, where Hennessy renders engach by 'valiant.' O'Clery's 
inghach ' noisy, talkative ' (LB. 222^ 5) must be a different word. 

englonn danger, Ff. 25. A doubtful word : S. explains it by 

fohha 'smith.' 

1. eo M. pin, brooch, Ff. 5. So Conn. s.v. Emain, and O'Cl. : 

eu 6ir and, a brooch of gold therein, LU. 68** 8. int-eo oir, LL. 
8P 2. cia aithem eo? what is the sharpest of pins ? O'Dav. 81. 

2. eo M. salmon, Ff. 5. So O'Cl. rot-bia eu .i. bratan, LU. 
67» 26. Corruptly 6 : Am h6 il-lind, LL. 12^ 42. sg. gen. iach, 
Brocc. h. 72. ace. ich. pi. n. liogit ich bricc speckled salmon leap, 
RawL B. 512, fo. 119^2. 

3. eo yewtree, Ff. 5. So O'Cl. Also tree. In Fel. March 3, 
and O'Dav. 81, eo is glossed by lignum .i. crand; and see Corm. 
8. v. ubalL Kop eo uasind fid, ropo rfgda ind rail, LL. 147* 32. 
eo Mugna, LL. 200* 14. Cogn. with, or borrowed from, A.S. 
e6w= O.H.G. iwa, 

4. eo a grave, (fert), D. 56. ^o, O'Br. 

5. eo good, D. 56. Old-Celtic an'-? Gaul. Avi cantos = 0, Br. 
Eueant, Or. €V9 from iFt^, Goth, avi in avi-liud x^p^^* evxapurrla, 
Fick, Bezz. Beitr. i. 58. Lat. avere seems cognate. Skr. ^av. 

Phil. Trans. 1891-2-3. 5 



66 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

eochair ton^fue, Ff. 41. So O'Cl. 

er nobk, D. 25. So O'Dav. 47, 81, and O'Cl. er .i. uasal, L.Lec. 
Voc, Stowe XIX. is 6u othair 6r Emna, LU. am sruith uasal her, 
LL. 4^ 25. ^r-ellam .i. adbul ellam, gloss on Colm. h. 49. Seems 
abstracted from the intensive prefix in ir-chian^ etc. 

1. ere hee^ Ff. 51. So O'Cl. Compd. eirc-bheach wasp^ O'Br. 

2. ere hon$y, Ff. 51. So O'CL 

3. ere heaven, E. 13. So Corm. Tr. 67, O'Dav. 81, and O'Cl. 
Arm. erkin * heaven.' 

err M. champion^ Ff. 73, D. 18. err (.i. trenfer) faebur (.i. 
claideb) fland, LL. 43* 10. eirr (gl. curruum princeps), sg. gen. 
gnfma erred, LU. 114* 21. pi. nom. errid .i. trenfir, LL. 312* 8. 
Hence erredacht, LIT. 113^ 29. Cogn. with Gr. a/xri/j/, Zend arshan. 

esc water, E. 12. So Corm. s.v. inesclonn, StoweXIX and O'Cl. 
8,v. eascra. Compd. esc-ong, esc-ongon eel, lit. watermake. The 
ace. sg. of a cognate word occurs in a note to Fel. June 17 : oc leim 
do dar aroile escai i Luachair Dedad as he was leaping over a certain 
water in Z. D. Esc is = Ptolemy's river-name ^I<rira. From 
^fpjid'ha, perhaps cognate with wlha^, wlhyta. So O.W. uisc may 
come from ^fpjeidka, 

esconn an old man, Yi, 73,=easconn, O'Cl. qy. a dotard (es+conn 
q.v.). 

es-slabra generosity, Ff. 20. Tri aib adannat serg : gnuis, aslach, 
easlabra, H. 2. 17, p. 183**. bochta co n-eslabrai, Rawl. B. 512, 
fo. 37* 1. easlabra (.i. enech) Guaire, Eawl. B. 512, fo. 60^ 1. 
ar easlabhra is ar aoide, Ban do Brian na murtha. 

es-saeth (MS. easaoth) health, Ff. 26. So O'Cl., who writes 
eassaoth. The contrary of saeth * sickness,' which is cogn. with 
Goth, sair, A.S. sdr, Eng. sore, 

etriad, p. 1. Meaning obscure to me. 

etrocht(MSS. 6drocht) ]t?ttrtf, light, Ff. 11, is rather shining, bright, 
pi. nora. aurchiche aumochta etrochta, bosoms naked, shining, LXJ. 
107* 1. dat. rug-etrachtaib (gl. praeclaris), Ml. 87<* 3. Hence the 
abstr. etrochta, LU. 27^ 38, 29» 19, 29^ 3, 33^45, 34* 11. sg. dat, 
etrachtai, Ml. 84^ 1. From ^etar-rog-to-, cogn. with Skr. rajata 
'white.' So innrocht .i. nemfollus, O'Dav. 100, is from *a»- 
rog-to'. 

6-tiiaichil not astute, Ff. 11, where it is erroneously explained 
as a substantive : ettuachail .i. aimhghlic, O'Cl. The opposite of 
titaichil, q.v. 

faballa /w, Ff. 13. So O'Cl. and H. 2. 15, p. 182 ; but it rather 



OLOS8ARIAL INDEX. 67 

means a fable or romance, gen. sg. i cend na faible, LB. 217». 
incipit do fabull [leg. tabaill] ibid. Borrowed from Lat. fdhula, 

fael (MS. faol) wolf, Ef. 16. Lucifer is called in fdil feochair 
fir-thnachaill, Salt. K. 1 670. sg. ace. amal fael f 6 cbairib, like a wolf 
mwng $heep, LL. 258** 10. Compd. fael-chii 'wolf/ with which Rhys 
would equate W. gweilgi (i.e. gtoael-^ci) * sea.' A cognate ^-stem 
occurs in Irish : amal foelaid etir chaircha, Tog. Troi' 1433. Arm. goil, 
failte »hield, Ff. 37, from *val-tio-, cogn. with Skr. -v/pa/, valaU^ 
Mid Gr. F^Xtnpop. 

fainell (foinell ?) fool, Ff. 74. Hence O'Cl.'s faoinnealach .i. 
oinmid, corrupted in O'Davoren's feanelach .i. oinmit, and O'Br.'s 
*faoinedlach, adj. /oo/i«A, «t%.' faindelach .i. oinmit. Conn. Tr. 
81. Dligid foindelach fuacra, LL. 345** 24. Cognate is foindel 
'wandering.' dat. sg. for fainiul, LU. 4» 16, for foindiul 7 
•echran, LB. 175». 

faithe (foithe, foidhi) sounds noise, Ff. 55. sg. ace. corro alsat 
iaithi fiangaiscid impi, Rawl. B. 512, fo. Ill* 1. 
falc (leg. failc ?) gap f D. 23. failc .i. manntaighe, O'Cl. 

1. fath breath, Ff. 68, fath, O'Br. urkelt. *va-to. Cogn. witYifeth 
'aura ' and Gr. a-Fyrq^ wind. Skr. vuia. 

2. fath heat, Ff. 68. So O'R. faith, fath, O'Br. I have not 
fwind the word elsewhere. 

fiih a kind of poem, D. 52. So O'Cl. ar mo fath .i. ar mo 
aiste, Rawl. B. 502, fo. 61*^ 1. is tre fir fiatha facaib each dam 
mochta inna suidiu iama stethur co fdthaib fiss fri forcetul 
fethamail, LL. 293^ 27-29. W. gwawd * panegyric' Cogn. with 
A.S. ir<5lS, O.N. d^r, eong, poetry, metre, Ir. faith * prophet,' Lat. 
tdtte, Goth, vdde * mad.' 

fee F. tooth, Ff. 40. So O'Q. Sg. dat. and ace. feie, Wind. 
Wurt. 538. Ir. fee (now feae) a kind of spade, and Lat. vanga 
'mattock ' may be cognate. 

fecur ? speckled, E. 20. Prob. corrupt. The line in which this 
▼Old occurs is hypermetrical. 

1. feib, goodness,!^. 31. feibh .i. maith 'good,' O'Cl.; but it 
IB explained as a subst. in H. 3. 18, p. 667*: feib .i. marsin no 
feabas. co ro molaim rig as each feib imbf, LU. 124» 23. co feibh 
ndelbha, FM. 1004. Hence feha, fehas 'goodness.' Cogn. with 
Or. ir/iiy» and Lat. vegeo, vigeo. 

2. feib as, D. 31, E. 20. * So O'Cl. and H. 3. 18, p. 70^ 667*. 
feib .i. marsin, StoweXIX. .i. marsin no bindis, L.Lec. Voc. feib 
Mdech roboi, LL. 149* 1. Cogn. with Goth, evasve. 



68 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

f6id, science, D. 50, W. gwydd * knowledge ' (Pughe), from ^vidi^ 
^vid, cogn. with Skr. vedas * knowledge.' 

f6ig keen-sightedy keen ? D. 37. feigh .i. ger sharpy O'Cl. corop 
feig rose for n-anme that your souPs eye may be keen^ Wb. 21*^ 9. p]. 
ace. fri faebra f6gi, Brocc. h. 97. Compd. feg-briathra, Rawl. B. 
612, fo. 113» 1. 

feimin vehement ? Ff . 74. Frob. a vox nihili, as it is cancelled in 
L. and feine written under it. The adj. femendeB, LU. 85» 19, 
130* 46, f6menda, LU. 113* 40, applied to horses, may perhaps b6 
meant. 

f6i8 piffy Ff. 17. gen. sg. iomnocht feise .i. croiceann muice 
a eaw'e ekiny O'Curry's Children of Tuirenn, 198. Com. yuie (gl. 
scroffa), Br. ywh, ywSz. 

felisc (filusc?) hark {of a tree), Ff. 69. Seems a genuine word, 
but I have not met it in the literature. Cogn. perhaps with Lat. 
vellus as Lat. cortex with Skr. krtti ' hide.' 

f6n a hicTy E. 15, (gl. plostrum) Sg. 21*. fen dar crfnach, LU. 
84* 1. rosiacht corp a hathar-si don ciU dia adhnacal 7 fean for 
sesrigh ag a iomchor, O'Don. Supp. "dat. atconnaicset ... da dam 
allaid rempu co fhen etarru 7 in corp and, LB. 29*. pi. gen. fengal 
na fen, LL. 218** 42. From *vegno. Cogn. with W. cy-wainy Gaul. 
eo'VinnuSy Eng. wagony wain, and with Lat. vehoy aBferetrum with/<?ro. 

1. ferbh N. tvordy p. 4, Ff. 14. So Conn, and O'Cl. Borrowed 
from *vervumy the British pronunciation of Lat. verhum. pi. gen, 
ferb nDe verhorum Deiy Conn. Tr. p. 72. buaid ferb, Salt. R. 4341. 
dat. ferbaib, LH. 34* 1 (Goidel. 164). pi. ace. amail rochuala 
Domnall tra inna ferba follscaidi, Bawl. B. 512, fo. 113* 2. ace. 
faig ferb fithir, Amra Choi. 52. 

2. ferbh coWy p. 4, Ff. 16. So Corm. and O'Cl. gen. sg. ferba 
LU. 109* (see infra s.v. mdta). pi. n. teora ferba fira. Laws i. 64. 
ferba .i. bai, LU. 125* 20. gen. cona chathchris do colomnaib ferb, 
fffith his hattle-helt of hides of kine, LU. 79*. Cf. Lat. vervex f 

3. ferbh hlistery hloteh, p. 4. So Corm. and O'Cl. fearb .i. briathar 
no bo no bolg no bole, L.Lec. Voc. turgbait ferba fora [g]ruaidib 
iar cilbrothaib blotches arose on his cheeks after (delivering) unjust 

judgmentsy LH. 34* 1 (Goidel. 164). Cf. Bret, gwerbl * bubon,' Lat. 
verhera f 

ferenn M. girdle, E. 13. So Corm. and O'Cl. sg. ace. ferenn, 
Bk. Arm. 5** 2. pi. ace. femu, LU. 58* 9 and Trip. Life, li. 

fet recounting, Ff. 8. fed .i. innisin, 0*C1. Abstracted from the 
verb adtet relates, feid .i. aisneidh, LL. 393* 52. 



GLOSSARIAL INDEX. 69 

1. fi had, p. 1, Ff. 4. So Corm. Tr. 79 and O'Cl. The same 
▼ordas/i * poison,' LL. 46^ 16, 18,=Lat. virus, Gr. /o«, Skr. visha. 
Used as an interj. ff dom-tanic c61miiine, Eg. 1782, p. 34» 2. 

2. ff dUohedience? p. 1, Ff. 4. 

^cell ? E. 25. ficell a vigilia, O'Mulc. 535,=figeU a vigilia .i. 
frithaire H. 3. 18, p. 70». 

fich land, D. 9, is; like W. gwtg, borrowed from Lat. vieus, 
villa, LB. 35» 2. sg. gen. ainm in ficha, r61. p. cxxxiii. dat. sg. 
oc airitiud a cethra hi fich slebi cille, LB. 189^. aibiDd sund 
amne i fich Maige Murthemne, LB. 107^ 48. 

fidrach, fidrad increase, Ff. 43. ffodhrach, 0*Br. for fidrad 
n-aes .i. ar each aes inn-araile, Rawl. B. 502, fo. 61 » 2. 

fidrad emiom, Ff. 37. a Emain . . . asa fidrad adfeidim, LL. 
21»6. 

finnell shield, Ff. 37. So O'Cl. Also finden, pi. n. findne gela 
'nal-laim LL. 276» 4 : im biat faibra £ri faibra 7 findne fri findne, 
LL. 276^ — 277*. Derived from find * white,' the colour of the 
chalked shields. 

firsi unheavy, light, Ff. 74. Some mistake, probablj, here, as 
frnmQ9Ji%' strength, Corm. Tr. 80. O'Dav. 87. 

fis colour, D. 46, note. The ' fisleadh ' of the text is obscure. So 
isthe'ffs' of E. 13. 

fis, F. vision, Ff. 74. sg. gen. for slicht nafisi sin, LU. 119* 12, ace. 
fis, LL. 256* 20. pi. n. fisi. A sg. nom. f'lsi occurs in LL. 208* 10 
(adbolffsi armoth^) and/imu in Salt. R. 3356. All from Lat. vlsio, 

ro fitir (gl. nouit) p. 5. Boot vid, 

1. flann red, D. 21. So H. 3. 18, p. 663, and O'Cl. Faelchad 
file faebur fland, LL. 43* 36. fin flann, Three Frags. 150. Other 
wx. in Wind. Wort. 

2. flann blood, Ff. 21, 45. So Stowe XIX, L.Lec. Voc. and 
O'Cl. Exx. in Wind. Wort. Identical with 1 . flann. 

flesc? E. 13. The line in which this word occurs is corrupt. 
Perhaps \A^ flesc is=fleasc .i. fearann * a land,' H. 3. 18, p. 51*. 

1. fo good, p. 3. Ff. 4. So Corm. Tr. 79, s.v. fochen, and O'Cl. 
^r, aigde Fiada fo, I fear, fear thou the good God, LL. 278* 33. 
fii fo mac rue mathair isin taig-sea innocht, LL. 254^ 37. fo dibad 
•i. maith a epiltiu, Amra Choi. 31. fo (.i. maith) mo cerd laechdachta 
Lr. 75* 30. fo each seel .i. is maith cech sc^l atchuas anuasana, 
LL 188* 48. nip fo lat .i. ni maith leat, 0*Don. Supp. The 
compar./do occurs in Eg. 1782, fo. 75*: ni thainicc nam tegluch 
foo. Compd. fo-amsera, LU. 83\ f6-s6n, LL. 254* 6. Skr. vasu ? 



70 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

2. fo honour, p. 1, Ff 4. gan fo, gan forithint, O'Br. 

foacht (=fo-facht) asking, Ff. 8. Abstracted fromat-pret. ^veq, 
such as iarmifoacht, Rawl. B. 512, fo. 109* 1. 

fobairt attack, Ff. 54. So 0' Dav. 91. Verbal noun offuapraim 
(fo-od-b). 

iochnad Jlrewood, E. 14. So H. 3. 18, p. 70*, and Corm. B. 
fochnod, and see Corm. Tr. 73, s.v. fochonnad. The fachnadad 
lasamuin of St owe XIX seems corrupt. 

foessam (MS. faosamh) safeguard, Ff. 31, D. 55. for foesam rig na 
ndula, ib. 52. for a [f Joessam .i. fora[f Joessitin, Colm. h. 2. for 
a foisam dun ar talmain, LL. 364, marg. sup. for foisam iiDe di^^m, 
ol si, LL. 286« 3. 

foessitiu protection, D. 55, where it is written faoisidin, the dat. 
or ace. sg. for a [f joessam .i. for a [f ]oessitin, Colm. h. 2. 

foir awaiting (?), E. 9. roforbad a foir, Rawl. B. 502, fo. 43*» 2. 

fol hose ? foundation ? E. 16. fol .i. bonn, Stowe XIX. sg. gen. 
dubithir leth dub-folach, LU. 113* 13. ace. for folaig n-athloisc[th]e 
na hecailsi, Be v. Celt. ix. p. 458. Perhaps from *svolak, cogn. 
with Lat. solum, 

folaid ahle, competent, p. 2. This seems the meaning oi folaith 
in LU. is folaith do Diaanisin, 113^ is folaid Dia, 113*. 

folerbhad death, D. 53. So O'Cl. i fanu folerbad fal romiad .i. 
is fan i mbfd immad na fer romiadach i mbas It is a slope whereon 
many most honourable men are lying in death, LL. 187* 52. 

ioMkn beauty ? , D. 23, = fallen, O'Br. It is generally an adj. 
da .L. ban find foUdn, LU. 50» 14, which O'Curry renders by * twice 
fifty women, fair and healthy.' The corresponding adverb occurs 
in LU. 6* 22, cepe nod-geba co follan (i. etir ceill 7 fogur), and 
in H. 2. 16, col. 700 : ciabe gabas each dia Amra Coluim co fallan. 
cuairt nemhfalldin ' an unwholesome visit,' Misc. Celt. Soc. 332. 

fonn a cantred ? (tricha cet), Ff. 28. dar each ferand, dar each 
fond, LL. 81*. flaith na fond, LL. 131» 8 : ba faenchrom a bh-fonn 
na sean an coll, Misc. Celt. Soc. 340. A oighre a n-easbaidh an 
buinn, ib. 342. tur gach aon fhonn d'iath Ealcca, Ban do Brian 
na miirtha. Borrowed from Lat. fundtis ? 

fophor, fofor a well, D. 50. Now the place-name Fabhar * Fore.' 
See sopur. 

1. forba a country (tuath), Ff. 28. forba .i. fearann, O'Cl. 
Eachatsa, em, d'iarra[i]d forba 7 feraind doib, Bk. Fen. 178. 
sg. gen. im roind a forba, LB. 124*. cosnum foirbe re Mac nDe is 
ed dogne lecae ban winning a heritage from God's Son, this would 



GLOSSARIAL INDEX. 71 

make the eheeie tchite, Rawl. B. 612, fo. 62^ 3. O.Ir. orhe^ orha 
with prothetic /. Cogn. with Goth, arhi, 

2. for-ha ilaying^ rending, hacking, D. 48. Cognate with for- 
henim (for-da-rubai, LU. 20» 27) ; Gr. Oelvu), iTretpvotf ; Skr. y/han, 

ior-coem-nacair factum est, accidit, D. 30. .i. tarla, Trip. Life xlviii. 
See G.C 451. A deponential redupl. pret. of an imperfect verb 
of which the pres. indie, was prob. /or<;Atf mcim (Jbrehuimsed 'fieret,' 
Wb. 4*). The 2d. sg. is in Salt. R. 1644, where Adam says to 
Eve : cid mor do locht ... is dom chorp f orcoenmacar. Root nank, 
whence also Lat. naneiseor, 

forcongra injunction, D. 37. So O'Cl. triasin aine a mbui Moysi 
tria forcongra De fair, LB. 259». A sister form of for-con-gur, 
cogn. with gdir^=7N. gawr, Gr. ^rjpv^, etc. 

forcthe (gl. doctus), p. 4, part. pret. pass, of far-canim * I teach.' 
The part. fat. pass, bed foircthi, leg. foircthf (gl. imbuendam, 
studiis) occurs in Ml. 132* 4. The contrary, an-forcthe *indoctus,' 
is io LB. 66* 64. 

1. fot watchful, cautious, E. 16. So Corm. and O'Cl. It is rather 
vatehfulnees, Ff. 68. fo duib fatchius 7 fot, LL. 67^. ri fot 7 ri 
foraire 7 ri freccomas, LL. 171** 30. 'com (f)6it 7 'com forairi, 
LL. 263^. The contrary is anf6t 'heedlessness,' LL. 126^ 66 and 
Eel. July 30. 

2. fot knowledge, D. 36. So 0'R./d<^ ' art, skiU, sense, knowledge.' 

1 . fracc a woman, wife, Ff . 24. So O'Dav. 92 and O'Cl. frag, 
mccthar i capp in diaid phill do[f ]racc, a scdil, may thy wife he 
carried in a hearse behind a horse, hero! LH. 34^ 2 (Goidel. 
158). fri fraicc .i. fri cumail, Broc. h. 71. "W. gwrach. Hence 
the dimin. /ro^mv^n 'girl,' LL. 262*^ 3. 

2. fracc hand, D. 14. So O'Cl. frag. In L. Lee. Yoc. we have 
metathesis of r: fare .i. lam 7 bean, douce a frac amach, Vath 
Beinne Etair, Rev. Celt. xi. 131. 

fraic shield, Ff. 37. So O'Cl. fraig. 

fraicc hair, Ff. 46,=W. gwrych * bristles.' 

fnach word, D. 26. So Corm. Tr. 66, L.Lec. Voc, Stowe XIX, 
and O'CL cia nach laigfedh (.i. nach cuirfedh) siansa (.i. ciall) a 
fuach (.i. a focul), H. 3. 18, part i. p. 210. From ♦fo-fech ? t/veq ? 
Cf . foacht supra. 

foaim reproach, E. 6. So L.Lec. Yoc. fuaim pi. n. fuamand, 
nsufdly means ' sound.' 

foait remnant, E. 3. So Corm. s.v. smerdit, and O'Cl. 

fual water, E. 14. So Stowe XIX. and O'Cl. fual usuaUy 



72 IRISH METRICAL GLOSSARIES. — ^MR. STOKES. 

means * urine.' sg. gen. ar galar fuail, G.C.' 949. tabairt a fuail, 
in-imechtur in dunaid to make his water outside the eamp, LIT. 67^ 34. 
From ^vog-h, cogn. with Gr. vr^-po-^ (from Foy-po-) and O.N. 
vdkr. 

fuan N. mantle, Ff. 35. D. 56. So O'a. fuan (gl. lacema), 
Wb. 80* 19. Bg. nom. fiian cdin coir, LU. 81* 25, fuan corcor-gonn 
im suide, LIT. 113^ 4. ace. dobert fuan corcra cortharach taris, 
LL. 108^ 16. for fuan n-argit .i. etach co n-argut, LL. 187^ 24. 
W. ffum, Eng. goum. The Old-Er. gone, Ital. gonna, seem to come 
from a Gaulish cognate *v6nd from ^cos-nd, 

fuath shape, image, D. 11. fuath .i. dealb, L.Lec. Yoc., Stowe 
XIX. A common word. 

fuider word, Ff. 10,=fuidhir .i. briathar, O'Cl. Cogn. with Skr, 
vad, vadati * to speak, call, sing,' Gr. vSw, vSetv. 

fuideran tunic, E. 19: dimin. of fuidhir .i. brat, O'Cl. Cogn. 
with oOoui] and A.S. wdd, 'Fidan .i. inar/ H. 3. 18, p. 70^, seems 
a corruption of this gloss. 

fuigill word, Ff. 10, note 17. 

fnince fang, talon, Ff. 49. So O'R. but fuinche, S. and L. 

1. fuinche fox, Ff. 51. a black fox, L. Lee. Voc. is tomgmall 
[leg. togmall?J i foir foinchi .i. en toghain fo sinnach, Amra 
Conroi, H. 3. 18, p. 49. 

2. fuinche (MS. fuindchi) royston crow, L. Lee. Voc. So O'Cl. 
Perhaps the gen. sg. of this word occurs in LL. 154^ 21 : duibi 
deoir funchi feidil culchi each coin imthemin. 

fuirmed lowering, Ff. 7. So O'Cl. Rather pressing or laying 
down. Verbal noun offorimim, end. fuirmim. 

1. fuit cold, Ff. 74, E. 27. fuit .i. uacht, H. 3. 18, p. 67«, s v. 
culpait. Fuit co brath, LL. 208» 41 = Fuitt co brath, Rawl. 
B. 502. Dr. E. Meyer suggests that this fuit may come from 
0.K hvitr * white.' 

2. fuit (MS. fuid), blazing, kindling, Ff. 68. This is 0'R.'s/«iei 
' lighting, kindling.' Can it properly mean to excite, stimulate, 
and be borrowed from O.N. hcetja or A.S. hwettan? A third 
Middle-Irish fuit occurs in LU. 59** 42 ; fuit Dia do bethu — and 
seems the Old Irish interjection uit, G.C* 750, with prothetic/. 

fur ready, prepared, E. 19, is rather preparation, making ready. 
fur .i. leatha no urchill, L.Lec. Voc. fur .i. ullmhughadh, O'R. 
fur na long, LL. 401^ 45. A deriv. f&rad occurs in Salt. R. 5885 : 
fri furad na ngruam nglorach. Compd. aran rem-fuir nemdesctha, 
Salt. R. 4352, 4356. 



OLOSSARIAL INDEX. 73 

furaaide (MS. fdrnuigheo) delay, Ff. 25. So O'Cl. Ni roleic 
imorro in t-aingel fuirech na furnaidhe do i maighin eile co 
riacbt Magh Rein, Bk. Fen. 112. on fethem 7 fumaide frisin 
eochair, ibid. 176. on fuirech oem on furnaidhe doronsat na clerchi 
[leg. clerich?] frisin corp, ibid. 180. This is, with prothetic /, 
the same word as umaide, imaidey ernaide * waiting.' 

1. gabar horMy F£ 19. So O'Cl., who gives two spellings, gahhar 
tnd gohhar. Ace. to Cormac gahnv is * goat,' while gohur is * horse.' 
Is alaind feras alluagh(?) gabar Baetan riasin sluag, H. 2. 16, col. 
B73=ls alaind feras in luadh gabair Baedan riasin sluagh, 
Tig. 561. sg. gen. brunni gabra Diarmato, LIT. 117* 14 = LL. 
277» o5. dat. os gahur gil, LL. 154» 47. Doluid for a gahrai 
gluair, Salt. R. 4781. ace. cor-rucait namait a chend, a gabair, is a 
dobcend $o that foes may carry off his head, his horse and his sword, 
LU. 13*, pi. gen. rotbia limsa . . . L. gabur ndubglas, LU. 130^ 
43. tricha gobar luath leironech, Bk. Fen. 366. dual nom. a da 
gabair cona n-allaib oir his two horses with their reins of gold, Rawl. 
B. 512, fo. 116* 1. Hence there seem to be two words, gabar and 

^iiir, from two stems, *gahrO' and *gahri'. 

2. gabar light (lux), E. 16. gabhar .i. solas, O'Flah. pi. n. taitnit 
)ra tar mag lir, H. 2, 16, col. 396 A. gabur ainm do grein a name 

for the sun, H. 4, 22, p. 61% seems the same word, the sun being 
regarded as the steed of heaven ? 

gaet (MSS. gaodh) a wound, Ff. 23. gaet .i. guin, Stowe XIX. 
(VClery's gaod .i. guin. pi. ace. fordacorsatar goeta .i. gona no 
ro gonad, Brocc. h. 66. A cogn. verb ro-gdet (ro-gaod .i. dogonadh, 
O'CL), is used as a pret. pass. sg. 3 of henim, rogaet and do gae, 
LL. 154» 46. gaitis .i. gonus .i. marbus, H. 3. 18, p. 70^. Cf. 
gacdad .1. guin, L. Lee. Voc, con-goite (gl. conpunctus), Ml. 58° 
17. Lith. iaizda 'a wound, hurt.' 

gdid an asking, a prayer, D. 38, goidh .i. guide, O'Dav. 95. 
Abstracted from ro-gdid, perf. sg. 3 of -Ir. yutWti = Zend 
jaidhyimi, Teut. hid/an, which Bezzenberger connects with Gr. 
ToOitM} and 0ia<raa6ai, 

gaire shortness of life, E. 17. So Corm., citing the satire MaiU, 
haire, gaire Caieur (evil, death, short life to Caier !), etc. Derived 
from gor .i. gairit, 0*Dav. 95. Fel. prol. 59. 

galann, an enemy, Ff. 22. So O'Cl. doringned guin galand 
desium and sin 7 rodichend Feradach h6, LL. 258» 13. Cognate 
with W. galanas * inimicitia, homicidium, pretium homicidii.' 

1. gam winter, E. 15. So O'Gl. rofaeth sam, snigid gam, Amra 







74 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

Choi. 63. gen. mf gaim Novemher, Corm. s.v. gamuin. in-aidchi 
gaira, Kawl. B. 512, fo. 102* 2. From ^ghyamo-s^ Windisch, in 
Curtius' Studien, vii. 375. Lat. hiems and Gr. x^^M-^^ ^^^ their 
Celtic reflexes in 0. W. gaem, later gayafy Com. goyf. The Ir. gem 
(im gem-red, gem-fuaeht^ gem-oidche) i8=Skr. hima^ Lat. *himO'S in 
Mmu8 from ^hi-himo-s, 

2. gam wtfe^ E. 16. So Corm. and L. Lee. Voc. Cogn. with 
Gt. 7a/i09, f^a/iietVf Lat. geminuB. 

gann a vessel^ E. 12, a jug or pitcher , gann .i. easgra, O'Flah. 
may he cogn. with xavOapo^ (a cup, a kind of hoat), if this he from 

*Xap&ap09, 

gart hospitahty, Ff. 13. So Corm. and 0*C1. gart .i. tidnacul no 
eneach, H. 3. 18, p. 615*. sg. gen. richis garta .i. einech, LU. 123* 
11. CO lin garta, LU. 47» 21. 

g6g, Ff. 58. A scribal error for ri, q.v. 

geis prayer, E. 23. So Corm. and Stowe XIX. pi. n. gessi, 
gesse, LL. 220*» 26, 32. From *ged'ti, Cf. geesid (gl. supplicem) 
Ml. 40** 22. From ^ged-ti-ti, Cogn. with guidiu * I pray' v. supra 
s.v. gaid. Ceie .i. guidhe, O'Dav. 69, is either a mistake for geie or 
a loan from Lat. quaeetio, 

gen sword, Ff. 36. So O'Cl. and O'Br. The latter has also the 
compound gen-chrioe * sword-belt.' Ki ba eallma bias in gen i 
n-Ard iar n Dubh da inbher not in readiness shall he the sword in 
Ardy after Dub-dd-inber, Three Frags. 90. Cogn. is genam .i. 
claidemh, O'Dav. 96 =y^ii/«, LL. 166** 1, 208» 8. Compare Lith. 
genu die aste am baume behauen oder beschneiden (Nesselmann). 

genmnaid (gl. castus) p. 4. is sf in glan genmnaid, LU. 49^, 5. 
Cogn. with genas, genma and genmnaidecht * chastity.' 

geognad a wound, Ff. 23. sg. ace. dogeba geognad is guin. 
Aed engach 'sin irgail, Bk. Fen. 376. Compare certain forms of 
the perf. sg. 3 of henim, viz. geguin, LIT. 70^ II, ghguin LU. 65^ 5, 
geogain, LU. 72^ 23, geogna, Rev. Celt. v. 202, etc. 

g^sca light, Ff. 29. ghca 'branch' may possibly be the same 
word. Cf. Amra Choi. 62 : raith rith la grein ngescaig, t.*. (writes 
the glossator) "therefore g6dcach {branchy? luminous?) is said of it 
(the sun), for from it there is light unto stars and to human 
beings' eyes." 

giabair harlot, Ff. 75. So O'CL, giabur Corm., but ciabar .i. 
salach no merdrech, O'Dav. 63, tre coiblighi ciabhair through im^ 
pure copulation, H. 2. 15, cited by 0*Curry, Lectures, p. 462. 

gibne cupping-horn, Ff. 75. So Corm. 



GLOSS A RIAL INDEX. 75 

gil leich, £. 16. So Corm. s.v. gildae, who cites from the 
Bretha nemed ; doglen gil teogaid leech sticks to tongue. M® Alpine 
giyes a Highland giol F. leech, giol-tholl horse-leech. W. gel 
saogoisaga, pi. gelody Com. ghel. Possibly cogn. with the Hesy- 

chian xafiXdei' KaraTriyei and fiXe-rve^' at fiieXXai. Eick also 

connects Lat. gula. 

glann shoulder, Pf. 42. gland or glang, Corm., glang, O'Cl. cona 
chreit . . . clang-dfrig, LIT. 80* 28. 

1. gl^jniTtf, Pf. 9. So O'Cl. 

2. gle bright, clear, Pf. 9. Wb. 12* 4. gle la each, LIT. 69^ 19. 
Identical with 1 gl^. 

gleith consuming, D. 13. So O'Cl. 'feeding, grazing,' O'R. 
ac gleith in feoic, feeding on the grass, Laws ii. 238, 1. 23. ar gleith 
ind feoir, Trip. Life 228, 1. 18. Mili do gabraib fri gleith. Salt. 
K. 6299. One of the infinitives of gelim, 

1. glinne cows, p. 3. This and the following two words may be 
cognate with Gr. rfoKa, f^aXa07jp69. 

2. glinne calves, p. 3. O.-Ir. glonn calf, gen. gluinn, Bk. Arm. 
16^ 1, may be cognate. 

3. glinne milk, p. 3. 

4. glinne lead, p. 4. Cogn. perhaps with Gr. 70X71/1; ' plumbago.' 
gluss light, Pf. 74. So Corm., O'Dav., 94, and O'Cl. do-glus .i. 

droch-soillsi .i. glus soillsi, H. 3. 18, p. 68°. so-glus, Bawl. B. 
512, fo. 52* 2. Probably connected with Eng. gloss, Norse ghssi, 
N.H.G. glosten. But the etym. of all these words is obscure. 

gnae, gnai woman ? D. 8. wife, D. 40. gnae a woman, O'R. This 
goes with Vedic gnd, Gr. 71/1/17, etc., while the ordinary ben goes 
with Boeotian /9ai/a. 

1. gnai stately, E. 17. gnoe .i. s6gda, LIT. 109* 41. gnoe imorro 
each segda, Corm. s.v. gno. Probably the same word as 2. gnai. 

2. gna{ (gno( ?) pleasant, D. 40. gnoe, Corm. s.vv. foi and gno. 
gnaoi, 0*C1. 

gnfa a sister's son, D. 46. So O'Cl. A corruption of niae, gen. 
niath, G.C 255, 256,= Lat. nepos, Skr. napat, A.S. nefa, the g 
coming from 0. Ir. gnia * servant,' sg. ace. gniaid, LU. 123» 28, 
compd. fem-gnia, O'Dav. 86, which is cognate with do-gniu 
* facio.' 

gnfth voice, E. 16. Cormac's gnfd, gufd-gal: guioth shout, 
uproar, O'R. pi. dat. perhaps co ngnithaib fiad na slogaib, Salt. 
K. 8118. The gni .i. guth of Stowe XIX seems an error for gnith, 

1. gno distinguished, D. lO, famous, D. 52. So L. Lee. Voc, 



76 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

O'Dav. 94 and O'Cl. Compar. gno som sui .i. urdarca som cech sui, 
Amra Sen&in. noud cerda .i. aurdarcaigim elathna, Bawl. B. 502, 
fo. 61* 1. The 6 oi gn6 is probably from 691 cf. Lat. gndvuB^ 
gnavare^ A.S. endwan ' to know,' from ^knivan, 

2. gno jeering f mocking ^ derision^ E. 17. So Conn., O'Dav. 94, 
O'Cl. 

gnod pointy Ff. 75. gnod a cono, conum est summa pars galeae, 
O'Mulc. 67. 

goithneyaptf/in, Ff. 48. goitbni .i. gaoi, Dull Laithne. A dimin. 
of goth .i. ga, O'CL (pi. nom. goith tentide inal-lamaib, Bawl. B. 
512, fo. 44* 1. Compd. gotb-snecbta, LB. 115 marg.), whence 
also gothnad (leg. gothnat) ibid.=gothnath, LU. 79** 8. pi. ace. 
ragabsatar ... a n-ocht ngothnatta neit, LL. 84* 51. 

gorm excellent, D. 17. .i. oirdheirc, O'Cl. .i. urdairc, O'Dav. 
94. .i. ord[r]aic, Stowe XIX. Compd. gorm-rig .i. na rig erdarcu, 
F^l, Prol. 233. Perhaps a participle passive from a root gor=. 
Vedic gir *to praise, to honour' (Grassmaon), whence gurtd, 
giirya. The 0. Ir. adj. gor 'pius' may be cogn., as well as 
Gr. 7€/)a9 and Lat. gratus. 

grant making grey, D. 13, is rather greg: grant .i. liath, O'Cl. 
and Conn. s.v. crontsaile. Conall grant hua Cemaig cruaid, LL. 
185^ 26. 

grech nut J Ff. 75. Corm. Tr. p. 90. mac greche .i. eitne cn6 
kernel of a nut, Amra Conroi, H. 3. 1 8, p. 49. In Harl. 5280, 
fo. 56^, grech is (erroneously ?) glossed by caeeh^ ut est cna grecha. 

greit champion, £. 25. So O'Cl. greid .i. geraid, L. Lee. Yoc. 
greid .i. gaiscidach, H. 3. 18, p. 537. greit rfg, LU. 106^ 5. con- 
greit rig, Fel. June 1 7, where greit is glossed by anroth, 

grenn, F. heard, Ff . 40. So Corm. Tr. p. 90, and O'Cl. sg. ace. 
greind, LL. 186** 9. Hence the verb grennaigim * I beard,' * I 
challenge,' verbal noun grennugud, adj. grennaigthech, Welsh and 
Bret, grann cilium, palpebra. The Span, grena seems from an 
Old-Celtic ^grennd. Cognate is the Albanian krande. 

grfan land, D. 9. So O'CL, H. 3. 18, p. 625, and L. Lee. Voc. 
gainem a grian, LB. 214*. O'Don. Supp. gives the gen. sg. 
as grkn : fine grin * the original tribe of the land : ' fer grin ' the 
owner of the land.' aco. coto-melat ar mur 7 grfan, LU. 67^ 16. 
Prob. identical with grkan * gravel ' = W. graian, ar uir ecus 
grfan, LU. 106**. itir ur ecus grioan, Harl. 5280, fo. 66^ corbo 
r6ill in grian 7 in gainem in mara, LU. 26* 8. is gat im ganem 
na im grfan, LL. 88* 17. Compd. : murgrian amal mil, LB. 215*. 



OLOSSARIAL INDEX. 77 

1. grib 9wt/tnes9j Ff. 21, D. 12, is rather sm/t, adv. co 
gribb, Mael fsu, cited in Fel. clxxxv. comor 's go gripp, Bk. Fen. 
218. Hence gripe stcijtness, Rev. Celt. iii. 183. 

2. grih prohibiiiony hindranee, Ff. 43. So O'Cl. 

1. grinn stronghold. Ff. 9. So O'Cl. 

2. grinn decency, Ff. 9. So O'Cl. But it rather seems an adj. 
as to the meaning of which manj guesses have been made. See 1 . 
grind in Wind. Wort, and add is ccoiggad ingen ngel ngrind, Bk. Fen. 
48, which Hennessy renders by * with fifty fair sprightly maidens.' 

1. grith sun, Ff. 20. But grioth, O'Br. Apparently the same 
¥ord as grith 'ardour': grith slegi, LL. 267*^. From *ghpti, 
cogD. with Vedic gbpnd ' Sonnenglut.* 

2. grith knowledge, Ff. 45. So O'Gl. Hence gritheach learned, 
O'Br. 

grot hitter, E. 17. So Corm. s.v. gruiten. Seems a sister-fonn 
of goirt (gL acidus). Another grot * active ' ? often occurs : ro 
garb-gles go grod a geir-ingne, Eg. 1782, fo. 34* 1. Gac(h) egin 
grott a mbi in dream, Bk. Fen. 241, where Hennessy renders 
grott by * sudden ' : mana t{ go grod 'na dail, ibid, where he 
renders go grod by ' quickly.' dimiad . . . dom liubar co grod, ib. 
220. 

grotan (MS. grodan) boat, Ff. 75. So O'Cl. 

guaire Atftr, Ff. 41. So O'Gl. Occurs in Lism. Lives 2212, 3798. 

guba wailing, D. 35. So L. Lee. Voc., Stowe XIX, and O'Cl. 
gaba suspiria .i. osnad, Corm. Tr. 89 and H. 3. 18, p. 70^. oc 
estecht fri guba 7 fri golgaire na n-anmand oc troge, LIT. 30^ 39. 
agair a gubse, LXJ. 69^ 36. 

gulba mouth, Ff. 41. So O'Cl. In Corm. Gl. it means mouthful, 
Bg. ace. doepetar gulba da each ferand. The n-stem gulba 
* rostrum' seems a different word. 

1. gunn a breaking, D. 20. So O'Cl. 

2. gunn neck, D. 20. So O'Cl. O'Reilly, mistaking brdige 'neck' 
for braige 'hostage,' has *^ gunn s.m. a prisoner, a hostage." 

homo human being, "FL 53. Aithne dam homa (.i. duine) re haei 
(i. re heladain), H. 3. 18. Borrowed from Lat. homo, 

humal (gl. oboed[i]en8) p. 4. So Corm. Tr. 167. Borrowed 
from Lat. humilis. So W. ufyll, Br. vueL Hence huimle, Salt. 
H, 7300. The cogn. subst. umalddit is from humilitdtem. 

I island, D. 15. So OCl. hi .i. inis, L. Lee. Voc, Stowe XIX. 
Prom O.N. eg = A.S. 6g, kg, 

fach salmon, Ff. 14, E. 10. So L. Lee. Voc, Stowe XIX. gen. 



78 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

• 

iaich, LIT. 16** 39, 40* 16 : a vocalic stem, cognate with the 
^•stem eOf gen. iach = Lat. esox^ "W. eog, Com. ehoe, G.C.* 123. 

iairchena thenceforward^ D. 28, iarchena .i. anegmais, ' besides,' 
L.Lec. Yoc. iarceana .i. anegmus, Stowe XIX. This is archena 
LIT., 28* 37, 30^ 30, 31'* 37, archiana .i. o sin amach, O'CI. and 
arcMana .i. anecmais, H. 4. 22. p. 59^. 

far black, D. 21, E. 19. So L.L.Voc., Stowe XIX, O'CI. and 
O'Flah. Is uime goirthear Lughaidh lar-dhonn de, ionann iar- 
dhonn 7 dubh-dhonn, gona tre folt dhubhdhonn do bheith air 
rdinig Lughaidh lardhonn d' forainm air, Haliday's Keating, p. 336. 

iarlonn the west, hack part, Ff. 56. iorlann .i. iarthar tighe mara 
mbi biadh, O'CI. Can the iarluih of the Voyage of Mael Duin 
(Rev. Celt. ix. 474, I.e.) be a scribal error for iarlonnaib? 

1. fath hell, pp. 2, 3. So Corm. s.v. bachall. aoth a hell, 
O'Br. 

2. iath.famou8f p. 8. 

3. fath eowl, p. 3. 

4. fath (iat ?) reliquary, p. 1. * etag ' 7 * iath ' anmann fethal la 
gentiu, O'Mulc. 469, = * ettech '7 * iat ' anmann fethal la gentiu, 
unde dicitur dotong darsna hiata-so / swear hy these reliquaries^ 
H. 3. 18. p. 81, col. 1. 

5. fath N. land, p. 1, Ff. 4, Ff. 28. So O'CI. iath n-Anand .1. 
Eiriu, H. 3. 18, p. 635°. sg. dat. in sudigud bias in iath (.i. hi tfr) 
Sion, LU. 8» 41. ri dosn-uargaib os cech iath, Salt. R. 7445. ace. 
rdnic iath in nad adaig aiccestar, Amra Choi. 34. cu iath nErend, 
Annals of Boyle, 1014. dat. in-iathaib aidbli aniuil, Salt. R. 3526. 
dos-fuc a hiathaib Egipt, ibid. 4426. Compounds are nim-fath, 
rfg-fath, fath-maige. 

ibath death, D. 57. iobadh, 0*Br. Perhaps an old misreading 
of .i. bath. 

icht children, D. 12, and E. 13 (where the MS. has ueht). So 
Corm. s.v. Eoganacht, and Corm. Tr. 98, where cinn should probably 
be cinil, icht may be cogn. with N.H.G. eeht * genuine,' urdeutsch 
*ahii. Hence the adj. ichtmar, Le rugadh an Righ Keamhdha an 
O'gh fchtmar oirea[gh]dha, Misc. Celt. Soc. 348. 

idna weapon, spear, D. 16. ri hidnae nethes nemthigetar, Corm. 
s.v. nith. O'CI. explains this word by sleagha no arm, PI. nom. 
m'idnu (leg. m'idna?) airgdide, LL. 206^. dat. for idnaib an 
dnruth, LL. 232^ 33. taithniomh oir ara n-iodhnoibh the glittering of 
gold upon their weapons, H. 3. 3 cited in Petrie's Tara 166. pi. ace. 
bruid idnu buden, LU. 47^ 23. O'roghabh a iodhna, Dan do Bhrian 



OLOSSARIAL INDEX. 79 

na miirtha. Hence the adj. idnach warlike, abounding in weapons. 
The root is yudh Ho fight/ whence also many British names 
beginning with lud. Or. ha-fju'i/rf, Skr. yiidhyati^ yuyddha, 

idna honour-price, D. 43. ace. cen idna nglan, Salt. K. 1395. 
Derived from idan .i. glan, O'Cl. 

immderg hlame, Ff. 62. Hence the verb immderyaim, Verbal 
noun immderyud. 

indless yoodneu, p. 2. innlus gach tinnlscra fri deirge dligiic^ o 
rechtaib commamsa on leaving {her husband ?), she is entitled by the 
Uwt of matrimony to the increment of every bride-price, 0*Dav. 70, 
B.T. comaim. 

innsa trouble, Ff. 54. So 0*C1. A cognate adj. existed in O.Ir. 
ni insa, S. Paul, v. 5, the Mid. Ir. form of which was andsa, 
compar. andsu, Rev. Celt. viii. 358, 1. 2. 

inntile a vessel, or ease, Ff. 56. So O'Cl. In Corm. Tr. p. 98 the 
vord is explained as ' a small vessel wherein drink fits.' 

ir=Lat. Ira, but in Ff. 56 explained by deabaidh contention, 
a ir .i. a ferg, Corm. Tr. 116, s.v. mer. ir .i. fearg, O'Cl. Ir .i. 
within (?) no ferg, O'Dav. 98. ir .i. fearg, Stowe XIX. ir . . . 
wgnificaret Hibemis . . . iram, O'MoU. 29. gdir g6r gribi, hfr 
18 ferg, Salt. R. 921. buith co n-fr, LL. 43» in marg. la Demon 
CO n-ir, LU. 114^ 30. cen chomairb, cen f-ir, LB. 261^ 17, cen 
fodord, cen hfr, 262* 9. Hence the adj. krach. Salt. R. 4086, etc. 
The W. ir-llawn, ir-llonedd show that ira was borrowed also by 
the Cymry. 

irchaill doorpost, Ff. 6. earchaill, O'Cl. do chndmaib elefinnte 
tra doronta na doirsi 7 na hirchoUa, LB. 209^. O'Br.'s ircilt * the 
side-post of a door,' seems a corruption of this, ercheallan ' a pole, 
itake,* O'Br., may also be connected, 
irsi adj. light, Ff. 74, a doubtful word. 

itropa head, Ff. 38. itropa, O'Cl. A doubtful word. L. has 
w tropa and possibly itropa may be nothing but an old misreading 
of .t. tropa, as ibath, q.v., of .i. bath, 

laba eyebrow, Ff. 39. Better laupa=lauba, O'Br., cogn. with 
lupaim * 1 bend.' s-pret. pi. 3 rolupsatar, LL. 86^ 45. 
ladg snow, Ff. 5. So O'Cl. ladhg, O'Br. 
laeg water, liquor, £.13. Perhaps a mistake for laith, q.v. 
laemda, £. 28. Meaning doubtful, ronasc go Isemda a lipadha, 
Eg. 1782, fo. 34* 1. St. Fechfn's mother is called Lasair laomdha 
lanlebur, BethaFechin, O'Br. has a loom * a blaze of fire,' and 0*R. 
a komh * strong, powerful.' co ti sil Luigdech in loeim, Bk. 



80 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

Fen. 226. The compd. laem-scel occurs in Bk. Fen. 340, where 
it is rendered by * famous story.' 

laith liqitary D. 41. So O'Cl. laith .i. cuirm. laith .i. ass, 
O'Flah. Compd. tri laith-linni, SP. v. 16. Com. lad (gl. liquor), 
Lat. latex, 

laithre cow^ Ff. 16. So O'Cl. Derived from laith 'milk,' Laws, 
i. 64, 66 : laith find for tellraig .i. as na mho iar8a[n] talroain. 

1. \ang feast, E. 20. So fl. 4. 22, p. 13, and L. Lee. Yoc. 
Cognate with the verb longaim * I eat.' 

2, lang fraud, Ff. 57, D. 22. So Corm. s.v. Gaileng, O'Cl., 
O'Flah., H. 3. 18, p. 635^ and Eg. 1782, p. 26. Hence the 
verb hngaim ' I cheat, betray : ' ro lance in sennin fort. Rev. Celt. 
xi. 131. Gr. €*-\€7xo» 'reproach, disgrace,' may be cognate. 

lathar hidden meaning, Ff. 10. So O'Cl. In Wb. 5M6 we seem 
to have the dat. sg. dont lathur diasndfsiu roboi hi runaib inna 
deacte oc tuiste duile hi tossuch to the unspeakable hidden meaning 
which was in the mysteries of the Godhead when creating (the) elements 
at first. Compd. erlathar, LB. 56<^ 41. Cogn. with Lat. lateo, 

1. ler M. sea, Ff. 16. So O'Cl. the ocean ace. to O'Dav. 100. 
ler tondban for talmain. Laud 610, fo. 90<^ 2. sg. gen. gainem lir 
fo longaib, LB. 28*. mac lir [.i.] mac mara, Corm. s.v. Manannan 
mac lir. dat. sg. liur, ace. mo ler cona lantoradh, LL. 386^ 44, 
Bk. Eights, 196. fer co n-ilur gnfm dar ler, LU. 47^ 30. ri rethes 
ler, LU. 40» 18. imma curchaib iarsin ler, LB. 33». pi. ace. liru 
(gl. pontias) LH. 12» (Goidel.* p. 69). W. llyr mare, equor. 

2. ler abundance, Ff. 21. Seems used as an intensive prefix in 
ler-glor, Ur-thol, ler-mdr, ler-ol. Spit. K. 366, 611, 2722, 4443. 
ler-thinol, LL. 296» 1. Either a misspelling of leir * industrius,' 
or the same word as 1. ler: so the gen. sg. of diliu (borrowed 
from Lat. diluvium) is used to intensify the meaning of the 
governing word, e.g. dam dilenn, 

1. les bladder, E. 15. So O'Cl. sg. dat. mar andil il-les, LL. 86** 
35. Is leges lega cen 16s it is the leeching of a leech without a bladder 
(in which he kept his medicaments), LH. 34^ 2. les lasgtha a 
clyster, O'Br. les each mbolg imbi lind every bag wherein is liquor, 
Corm. s.v. lenan, spelt lesan in H. 3. 18, p. 72*. 

2. 16s light, Ff. 57. 16os, O'CL, leus, O'R. So L.Lec. Yoc. Leos 
and Us, Eg. 1782, p. 26. teora bliadna boi cen les Colum ina 
dubrecles, LH. 34* 2 (Goidel. 161). 16s-boire, sg. gen. 16sboiri, 
Wb. 25* 3. dual nom. da 168paire mora, O'Don. Gr. 352. The 
forms containing o or u may be loans from O.N. Ijos. 



GLOSSARIAL INDEX. 81 

Upraise (?), £. 14. This meaning seems inferred from the U a 
mokd 'splendour his praise/ in Conn. s.v. fili. /i = W. lliw 'colour.' 

lia M. sUMf p. 4. lia laime, LL. 393^ 50. sg. gen. liac, dat. 
luic: compd. nerUia, LL. 255^ 16. 

liachtain moisture^ Ff. 56. ^Uig. Cogn. with \eifiw, \oi/9i/i 
Lai hhatiOf de-lihutus. 

ligrad tongue, Ff. 57. lioghra, 0*B. Derived from ligur * tongue/ 
Com., and cogn. with X€<xa;, li-n-go, ligurto. So in Duil Laithne 
11, ligair .i. tenga. 

lit quarrel, Ff. 65. So L.Lec. Yoc. Borrowed from Lat. Us, as 
V. Uid 'anger' from litem. See Corm. s.y. lesmac. 

loc place, D. 29. So G.C.* 69. log, L.Lec. Voc. sg. gen. luic. pi. 
^t. locaib imechtrachaib mundi (gl. ab hibemis locis), O'Mulo. 700. 
Compd. mac-loc wamh, LL. 273^ 26 and BawL B. 502, fo. 61« 2. 
Borrowed from Lat. {st)locus. 

\wi fierce, D. 24. So O'Gl. lore .i. angbaid no laind, H. 3. 18, 
p. 537. in chrott arpeit Labraid Longsech lore, LH. 34* 2. 

lo6c lame, D. 33. So Stowe XIX, L.Lec. Yoc. and H. 3. 18, 
p. 663^. sg. ace. Ni chuitbe nach sen ciarbat ooc . . . na lose ciaso 
loath Mack not any old man though thou art young, nor (any J lame 
■M though thou art sunft, LL. 344* 32. pi. ace. luscu .i. bacuchu, 
Place's h. 32. lose (gl. dodus) O'Mulc. s.y. coUud. Gr. \ofo9, 
lat luxus. Another lose (=Lat. luscus) means ' blind ' : Domrigne 
low len, LL. 147* 30. 

1. loth plumage, D. 25. cf. luathan .i. en bird, Duil Laithne, 
123. From *(p)luto. or *fp)luta. 

2. loth fierce, E. 27. From *luto, cogn. with Or. Xvtrtra ' rage,' 
irom *lutja, Lith. lutis ' storm,' Ch. Slav. IjutH ' vehement.' 

lothar raiment, Ff. 20. So 0*C1. co ro dubai fri gr6in 7 gdith i 
ccrtaib 7 lothraib, LL. 274^ 1. From 1. loth? 

Id smallness, Ff. 47, is rather small : lu each mbec, Corm. lu 
ibeag, O'Cl. lulaegh .i. Ia6gh mbec. Eg. 1782, p. 26. Hence 
loan .i. mac, H. 3. 18, part i. p. 210. is dal ena tar lua[n] (.i. uisque 
tar uaeidin), Amra Conroi. Compd. leas-luan stepson, 0*Br. 

lua/00^, Ff. 42, kick, E. 16. O'Cl. gives these two meanings. 
lu.L preb. Eg. 1782, p. 26. Atacomcussa com lau, LU. 114^ 10. 
mo da lua .i. mo da choiss, LL. 208* 24. dobert a luie frisin 
comlai col-luid a chos trethe, LU. 19^ 19. tobert Cuchulaind a lue 
a&idiai, ibid. 22. tuc Cerball lua da choiss friasi, LL. 52^ 1 1 . In Old- 
Irish it seems to mean ' heel ' : sal no lue (gl. calx), Sg. 50* 20. 

loan hound, wolf, Ff. 16. So O'Cl. a greyhound, O'Br. 

na TriBS. 1891-2^. 6 



82 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

luba hodff. y. infra, s.v. tethra. Idhha, Idhknaeha or liiihhneacha, 
O'Br. 

luo (Inch?) heUyy womh, Ff. 41. locc mbecc a dimpUy LB. 
148^ 

luchtaire whirlpool^ Ff. 57. So O'Gl. luehtaire also means 
'lanista,' Ir. Gl. No. 10, and, like the Lat. luctor * I wrestle,' comes 
from a root lug meaning < to twist.' 

lugna moan, E. 2. So L.Lec. Voc. Seems borrowed from Lat. 
lana^ with g inserted to indicate the length of the preceding 
Yowel. Cf. lun 6 luna, LB. 24 1<^ 20, a gloss on the fo-lun 
lainderda, etc., of the Amra Sendin. 

1. luibne finger^ Ff. 42, D. 14. So Conn. s.v. deach. luibhne 
.i. meoir^ny^*, 0*C1. 

2. luibne tpear^ Ff. 36. mam luibni .i. fam sleig, LL. 208'. 
Another meaning for luibne is ihuld. So in LIT. 55* luibne gela 
foraib tchtte shields upon them ; but this seems a mistake for ruibni. 
see LL. 208* 7 : mo ruibni .i. mo sciath. 

Imgaej'avelinf Ff. 48. This is luibhne in H. and L. 

luis hand, Ff. 57. So O'Cl. and DuU Laithne, 17, O'Dav. 101, 
H. 3. 18, pp. 71^ and 636, and Eg. 1782, p. 26. Hence luiseag 
'the haft of a knife or sword, the small iron part that goes 
into the handle,' 0*Don. Supp. From ^loc-si- ? Cf . Lat. al-lex, poUlex ? 

lulgach warrior^ Ff. 23. a soldier, O'Br. 

macht death, p. 3, note 2. Cogn. with machtaim I slaughter : o ra 
machtait issin mormaidm, LL. 195* 56, machtad slaughtering, LL. 
193* 25, Lat. macto. 

maen dumb, Ff. 34. So O'Cl. (maon). Asbert in rigan : ' maen 
rue,' BB. 251*, better moen, LL. 269* 29. Cogn. with Lat. mutus 
from *moi'tO'S, 

1. mdin good, F. 43. So O'Cl. (maoin). Cogn. with main 
'treasure' (nf fil mo mdin fen acht a nim, LB. 216^), pi. ace. 
madii 'precious things,' LL. 271^ 16. Lat. munus from *moinos. 

2. mdin love, Ff. 34. So O'Cl. (maoin). 

maime treachery, D. 52. Cogn. with maim, maimed 'betrayal,' 
maimim ' I betray ' (rom aimet nad aincet, LL. 344° 52), mamtid 
' traitor,' LL. 282* 2. 

1. mal soldier, Ff. 23. So O'Cl. Prob. identical with 2. mal. 

2. m&lalord,noble,I)AS, So 0*Dav. 106, O'CL, L.Lec. Voc.,0'Flah. 
Cormac explains it as king, and so in LL. 311^ 33 mal .i. ri : in LL. 
307^ 5 : mad beo mac in mail .i. in rfg, and in H. 3. 18, p. 82^ 
cu mail .i. righ. Qorbo mal each maige moir, Salt. K. 3431 : is 



GLOSSARIAL INDEX. 83 

he mal na slog, ibid. 4427. Voc. cing mil A. c^mnig a uasail, LL. 
186* 26. pi. nom. mail, LU. 40* 17. gen. o chath Mucrima nam- 
ing, LL. 131». Ifn a mdl, Salt. R. 6629. Old-Celtic *maglo-B, 
W. fMel in Maelgvm = Maglocunoa. 

3. mil rent, tribute^ D. 48. So O'Flah. and O'Cl. mal. gen. sg. 
n moradh a mail by increasing its tribute, Bk. Fen. 240. W. mdl 
'moneta' (Davies), 'tribute' (Richards). From A.S. mdl, Eng. 
(black)fnat7. 

maiiais ^ear, Ff. 48. So O'Cl. manais lethanglas, LU. 55' 15. 
113* 8. manais lethanglas Ifmtha Lochlannach, Selg 8Ube na 
tnBan Finn, cited in Battle of Ventry, p. 82. dia notairle mdndis 
... as mo laim sea, LU. 62^ 1, where mdndis and sleg are used 
&s sjnonymous. pi. n. noi manaise, LU. 93, ' 1. 25. dat. bar ar 
mWsib, LL. 85' 3. 

1. mann F. food, D. 9. So O'Dav. 105, and O'CL, who also 
liave mann .i. cruithneacht * wheat.' cin mann cin biadh without 
foddiff or food, Laws ii. p. 18, line 6. Is tre fir flatha mesrada 
mora for fedaib atat at manna milsi blassaigter, LL. 293^ 16, 
346* 1. maind, LB. 121*. Borrowed from manna. 

2. mann ounce, Ff. 58. So Corm., O'Dav. 104, L.Lec. Voc, 
H. 3. 18, p. 636«, Eg. 1782, p. 26, O'Cl. and O'Flah. pi. n. 
numna 6ir forloiscthi Corm. and H. 3. 18, p. 72». Giiterbock 
i^gards this mann as borrowed from m^na fii/a. But mina would 
in Irish have become *men. Siegfried's etym., mann from *manvd, 
oogn. with Gr. fiouvo9, fi6vo9 {jiovFo^), as Lat. unda with unus^ 
seems more probable. 

marc (MS. mairc) horse, Ff. 19. So Corm. and O'Cl. Cogn. 
With W. march, the Oalatian fiapxav (ace. sg.) and Tpi-fiapKial^, 
Paosan. x. 19, and the 0J9.G. marach. The difPerence between 
a mare and an eeh is, ace. to H. 3. 1 8, p. 639, that the former is 
female and the latter male (eich .i. firenna, mairc lathracha, leg. 
laracha 'mares'). But O'Dav. 104, has marc .i. ech no lair horse 
or mare, and in LU. 119^ 28, the gen. pi. mare is glossed by ech. 
In the Amra Conroi marc is declined like a fem. d-Btem : Is 
menn mairce murgeire .i. searrach eich fo ron mara. 

mata (MS. mada) pig, Ff. 17. Sg. gen. curadmfr ferba (.i. b6) 
brachtchi (.i. methi) brothlochi sceo (.i. acus) mdtai (.i. mucci) 
moogthi, LU. 109* 30, and see LL. 118% 48, 50. A sister-form 
^t in Corm., mait, O'Dav. 105, s.v. main. 

meile quern, Ff. 69. So O'Cl. pi. ace. ranirusa immorro deich 
meile, LU. 83'' 4. Cogn. with melim * I grind,' act. pres. ind. sg. 3, 




^^^ c^v^v.^v 



84 IRISH METRICAL GLOSSARIES. —MR. STOKES. 

mar gall mblooic melid broe .i. broin, LL. 43* 7. Lat. tnoh, Gr. 
fivWoff Ooth. malan. 

mell pleasant^ D. 35. So L. Lee. Voc. Mag mell Fairyland, 
O.-Ir. meld: ba mor meld a acaldam, Ml. carm. 1. Lat. mollis^ 
trom^molvis, *moldvii. 

mellach good, Ff. 43, also in LU. 24» 18, 74», 114*22, for 
meldach, G.C.» 18, is rather *gratus.' ba mellach in bag, LU. 114* 22. 

memmur, N. pmU, Ff. 46. So O'Cl., lit. member. Thus memur 
laime no <;oMf means * a finger or toe,' O'Dav. 106, pi. n. oire ntmdem 
membur uiH du Dea, quia sumus membra omnei Deo, Gamb. G.G.' 
1005. Borrowed from Lat. membrum. 

men mouth, F. 58. So Conn., H. 3. 18, p. 72^, and Eg. 1782, 
p. 26. m6n mara .i. b61 na mara, O'Cl. Hence menogud *• hiatus,' 
Sg. 40* 8. gen. ar immgabdil m^naichthe ' ad yitationem hiatus,' 
Sg. 8* 1 . W. min. 

menmarc thought, reflection, £. 1, rather means desire, darling. 
ba h6 menmarc a n-ingen 7 lenndn a n-6cban, LL. 271* \,hs was 
the darling of their daughters and the lover of their young wives, 
sg. gen. cluinte a hosnaid iar ndul a menmairo uathi hearken to 
her sigh after her darling has gone from her, LL. 269* 17. 

miad respect, honour, dignity, Ff. 58. So O'Cl. is miad mor ind 
apstalacht, Wb. 13* 5. pi. dat. miadaib, LTJ. 52» 25. O.Bret. 
a muoed (gl. fastu). 

mi-scaith a curse, Ff. 29. 80 Conn, miscaid, O'Dav. 104. 
miscath .i. mallacht, L.Lec. Yoc. scath .i. beanuacht, ibid, 
sg. dat. fo mfscaid bretheman bratha, LIT. 31* 21. ace. eirgg dot tig, 
ar se, 7 heir miscaid, LL. 272<^ 30. doberat trfst 7 mfscaid 7 berait 
a n-ordit uadib, LB. 258* 52. Hence the adj. miscadach ' accursed,' 
Salt. E. 2392, 2422 (fri claind Cain miscadaig). 

mo good, p. 3. mo-ling 'bene salivit:' mo-genair. For mon 
(= Lat. mantu * good ' ?) or m^ (=maith), which often occur com- 
pounded with verbs: mo-genar, LB. 146°. mon-genar, ma-tuluid 
*bene ivit,' F61. July 12, ma-lodmar, LL. 45». ma-lodsaid, LU. 
65» 15=mad-lodsaid, ib. 64* 7. 

mocht gentle, quiet, mild? Ff. 59. Can this be=W. mwyth 
'mollis,' cogn. with mwythan et moethan delitiae? 

monur N. work, deed, Ff. 9. So O'Cl. monar .i. obair. The O.Ir. 
gender appears from monar n-gle, Colm. h. 37. is mor in monur, 
LL. 234* 32. Hence monorugud, LB. 261 marg. 

mormuir bog, 'EL 70. mormhuir L. Seems a compound of 
m6r ' great ' and miiir borrowed from A.S. m6r or £ng. moor. 



6L0SSARIAL INDEX. 85 

mo8 i. melody, p. 3. Prom *mod-to. Cogn. with Lat. modus, 
'measure, due measure, rhythm, melody/ etc. modulor, modulatus. 
iL mos ehh, iii. mos flood, p. 3. These meanings (which are not 
hdigt) seem to come from the regular occurrence of the phenomena 
in question. 

mos euBtom, pp. 2, D. 58. So O'Cl. each Bob6s, p. 3. immda gun 
cia (.i. fer) sin mos (.i. bes), ff (.i. olc), H. 3. 18, part 1, p. 210. 
ranic maige mos nad genetar ciuil, Amra Choi. 36. Borrowed 
from Lat. mas. Hence mdsaeh, O'Br. 

mothla Moft, E. 22. moist, O'E. Cogn. with Lat. mustus 'fresh,' 
from ^mut'to-s ? 

muad ffood, Ff. 43, D. 49. So H. 3. 18, p. 652, and O'Gl. muad 
.i. nasal no airmidnech, Corm. mac muad Muire, LB. 213^. sg. 
gen. masc. : roselai delai £r muaid moinig, LL. 186^ 31. ace. fern, 
la Macha mudid, LL. 21* 45. 

muchna surly, D. 22. Corm. Tr. 115 s.v. mue, and O'Cl. write 
this word muena. But in LB. 255^ 70 it is muchna. 

mur abundance, p. 1, and Ff. 50, E. 26. Dia mor m'anacul de 
mur theinntide dfu derc nd^r great God to save me from the flery 
abundance of long looks of tears I Amra Choi. 4. So O'Cl. So H. 
3. 18, p. 76<^, s.y. m^r. a mur (.i. a himmed), chluime, Fel. 
ProL 126. Probably cogn. with the second element in vXrjfi-^vpa, 
rXi^fi'fMvpU, w\rifi'fM,vptM)y vXfffi-ftvpetv, Ebel's connexion (Kuhn's 
Beitr. iL 163) of v\fffipvpi9 with Lr. muir ' sea,' W. mor, Gaulish 
wtdri, seems impossible. 

nisad famous, D. 36. So O'Cl. pi. gen. comsid na ndem n^ad 
n-an, the guardian of the famous, splendid saints, LIT. 40* 36. 
Hence nisadach, gen. sg. m. nasadaig .i. erdarcaig, Ooidel. 173. 

neid, neit, neo wind, Ff. 67. n6idh, O'E. The right spelling 
of this word is doubtful. It may have lost initial p, and be 
connected with irveto, irvevfui, ttvoi^. 

n^it battle, Ff. 21. So O'Cl. culu tria neit .i. oath, Amra 
Choi. 2. neit ba hainm don chath nobrisind, LH. 34^ 2 (Goidel. 
158). iar do neit, ibid, neit .i. cath, LL. 393^ 2. n6it .i. 
gain, LTJ. 7* 7. iar do n^it .i. iar do guin, LTJ. 6^ 29. Neit is 
glossed by dia catha 'a god of battle,' in H. 3. 18. pp. 73% 637*. 
So Cormac, and see Eev. Celt. i. 36. 

nemed art, D. 43. neimheadh .i. gach d^ no gach ealadba, 
CCl. each d^ a nemed, Aibidil Cuigni, Book of Lecan, fo. 1 76<^ 
2. nemed * privilege,' seems the same word : pi. ace. ro ordaigset 
dano fir Herenn a nemthiu andsin, LIT. 1 18^ 6. 



86 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

nena thumb, Ff. 42. O'Cl. has nean .i. ordlach. But the gen. 
pi. nena occurs in LL. 208* 14 : triucha ueua (.i. ordlach) Find 
'na feic (.i. 'na fiaccail). 

neoit penuriotu, scanty , Ff. 13. nfb neoit, brecach, LL. 360, 
marg. inf. nirba neoit be not niggardly ^ LB. 101 in r. margin, ar is 
each lesc, lend, etaid, suanach, neoit, dedith is miscais De 7 doine, 
LL. d44\ In the Amra Choi. 103 neoit is a subst. 

nes an earthen stronghold {jkiih), Ff. 18. Perhaps neas .i. cnoc 
' hillock,' O'Cl. a hill, or fortified place, O'Br., or a mistake for mess : 
meitis ri mess .1. commeit ri tolaig as big as a hill, LL. 208V 

nia champion, Ff. 23. 80 Conn., H. 3. 18, p. 77*, and O'Cl. 
In Old-Ir. a dissyllabic ^-stem. sg. gen. niad, Brocc. h. 71. 
pi. gen. forrach niad, LL. 44^. niath, LL. 217^. For another 
nia v. supra, s.v. gnfa. 

1. nin letter, D. 26. So Conn. Tr. 126, H. 3. 18, p. 77^ and 
O'Dav. 108. nin .i. delb no litir, L.Lec. Voc. nion .i. litir, O'Cl. 
Specially the letter n : anamain etir da nin inso .i. nin i tossuch in 
moltai 7 nin ina deriud, .i. JVi disceoil [the beginning of the Amra 
Choluim chille] 7 iVimuain, LU. 6* 34. secht nen-adman (septem 
litterarum conexiones) Ml. 2*^, 2. dar ninu Nede, LL. 186^. Seems 
borrowed from Old-Welsh nihn, and this from the Hebrew nUn, 

2. nin wave, Ff. 59. So Corm. s.v. Ninus : nen .i. tonn, Corm. 
Tr. 126. nin .i. tend, L.Lec. Voc. nion, O'Cl., O'Br. 

nfth mortal wounding, D. 26. So Corm. and O'Cl. comtis nert- 
menmnaig fri each nith, LL. 219^. ni rubaim nfth n-erred n-dn, 
LU. 77* 19. Also means * battle : ' arm fri nith. Laws i. 122. oc 
erlud in nftho 7 in eggnamo frisin fdal, Trip. Life, p. 92, 1. 8. 

noe human being, D. 8. nae O'Cl. nai, L.Lec. Yoc. noe, Corm. 
fo chaid oc noe .i. is fo chataid biim ocon duine ica mbiim in am- 
sa, Bawl. B. 502, fo. 61* 1. Hence naindn dwarf, O'Br. 

nofnnell valour, Ff. 22. naoineal prowess, chivalry, O'Br. Hence 
prob. noenellach, LU. 125^. 

nua noble, Ff. 58,=nua .i. laidir, O'CL, who cites do throid se re 
nia nua he contended against a strong (noble ?) champion, sg. nom. 
in cuire noem nua, LB. 262^ 37. gen. comde nimi nui, LL. 307* 
14. aes each dana nui fo nim, LL. 197* 9. do throit ra nfa[i]d 
nu8^ LL. 83^ 27. From *gnua =Ir. gnd. Lat. gnavus, i-gnavus. 
The verb noud .i. erdarcaigim, LL. 187% seems cogn. 

nuall conspicuous, famous, D. 33. So L.Lec. Yoc. and O'Cl. 

6 ear, Ff. 40. So Corm. Tr. 131, H. 3. 18, p. 77^ and O'Cl. 
mo duals imm 6 .i. mo dom im chluab, LL. 208* 9. O.Ir. au, sg. 



GLOSSARIAL INDEX. 87 

gen. aue^ a neut. «-8tem identical with Gh. Slav, ucho, gen. uiesef and 
cogn. with Lat. auris from *ausiSf atM-cultare, Gr. vapava, ovara, 
Compds. 6i-derg red-eared, LL. 295^ 26, au-nasc earring ^ Conn. 

6-brat helmet^ lit. ear-ftiantle'i Ff. 35. This is eo-hhrat A, 
ceannbhar, O'Cl. .i. 6adach bis air chenn, O'Flah. eo-bhrat, O'Br. 
Cf. .X. eo-barr .i. barr bis am cenn in rfgh, Amra Conroi. 

oeth (MS. doth) M. oath, Ff. 4. So Conn, (oeth) and O'Cl. 
[uUh). oeth la each n-eric, Laws ii. 60. pi. ace. fri oethu, 
LIT. 46^. Old-Welsh ut in anutonou (gl. perjuria), Goth, aiths, 

oibid ohedience, Ff. 31. D. 60, is rather obedient: .i. umal, 
O'Dav. 109. Boi huasal, boi obid . . . cerbo huasal ropo humal, 
Amra Choi. 80. The gloss oidh .i. obuidens, H. 3. 18, p. 73^, is 
a corruption of oibidh .i. obediens. 

oidsen? E. 24. 0*E. has * oidsen a surname,* sed qu. 

oin (leg. oin ?) huffing, Ff. 68. This word is glossed by iasacht 
'lending,' Corm. Tr. 132, O'Dav. 109 and O'Cl. is fiach forcraid 
fomalta for oin there is a fine for excessive wear of a loan, Laws i. 
168, L 10. 

oircne M. lapdog, Ff. 60. So Corm. s.v. Mog eime, and 
O'Dav. 83, s.v. esrecht. im oircne rigna .i. indiaid orcan na rigna 
bw .i. in mesan, Karl. 432, fo. 10* 2=Law8 i. 152 (where in is 
erroneously omitted). A dimin. of oire .i. cu beag no measchu, 
O'CL orci, LU. 39^, 40^, is a sister-form. 

oir-derc manifest, conspicuous, D. 59. 0. Ir. airderc, erdirc, 
irdirc. Tri hirrdraici H6renn, Lem Chonculaind, Dun Cain, Srub 
Broin, H. 2. 17, p. 183^. Boot derk, whence also Gr. hipKOfiai. 

oitiu (MS. aoide) F. youth, D. 29. aidig .i. oigedus, Stowe 
XIX. aide .i. oice, L.Lec. Yoc. oetiu cen sendataid, LU. 33^ 39. 
gen. i sliab n oited .i. i n-ard na oited .i. oclachas, LL. 187^ galar 
n-oeted, Windisch, Ir. Texte, 145, 1. 11. sg. dat. oitid. oitith, 
Wl 75d 10, Sg. 63^ 6. From *juventilt', cogn. with W. ieuant 
'yoQth' and Lat. inventus. Hence the adj. ditidaeh, LL. 267^. 

oil great, Ff. 6. So Corm. s.v. ollam, O'Dav. 109, and 0'C1.= 
6r. voXKo9. The cogn. verb is ollaigim, no ollaiged (gl. ampliauit) 
ML ei*' 6. In the line Infer foil, Bk. Fen. 226, we have prothetic/. 

ond, onn N. stone, p. 4. Ff. 60, D. 14. So Corm., L.Lec. Voo. 
and O'Dav. 109. sg. gen. uinne. Compd. cloth-onn oc buaid, 
Anna Choi. 77. A neut. «-Rtem, cogn. perhaps with Lat. pondus, 
1. ong hearth, D. 36. fire, hearth, O'Br. Seems cogn. with 

Sb. angara ' glowing coal,' Lith. anglis * kohle,' and perh. Eng. 



88 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

2. ong grief, p. 2, Ff. 60. So Corm., H. 3. 18, p. 73* and O'Cl. 
ni ong oen tige, LB. 240^, where ong is glossed by ueh. Compd. : 
ong-[g]alar, Salt. E. 1453. 

ordit (MSS. oraoid, oirbidh) a hleiting, Ff. 29. orait .i. oratio .i. 
aurnaigthe, H. 3. 18, pp. 73*, 637*. oroit, Conn. Tr. 129. rom- 
bith oroit cet, a Maire, Sanct. h. 23. sg. ace. berait a n-ordit 
uadib, LL. 268*. From Lat. oratio. The opposite anorait occurs 
in Bk. Fen. 142, where it is rendered by 'evil prayer,' 'reproba- 
tion.' 

ort mamlaughter, Ff. 25. destruction, E. 27. death, p. 3, note 2. 
.i. orgain .i. bas, 0*Dav. 109. Seems abstracted from the ^-pret. of 
the verb orpim ; but may possibly come from *org-to. In H. 2. 
15, p. 182, ort is glossed by hsgadh 'burning.' The ortaih cited by 
Windisch, Wort. 725, as an example of this word, is for 
ordaih, pi. dat. of ord 'sledgehammer,' ordd (gl. malleus), Sg. 
49*, 4, ordin * mallet,' O'Br., cogn. perhaps with Ordo-vicei, 

oss M. deer, Ff. 51. So O'Cl. sg. gen. cethruime each ois rogab 
cuithech. Laws i. 272. basa chu-sa gabdla uis, LU. 114* 23. ace. 
gabait in n-oss ba nessom do, LU. 62* 40. pi. n. oiss alta, 
LU. 62* 32. uiss 7 altai, LL. 265<^ 5. gen. forrecat alma 
n-oss, LTJ. 62* 31. i ndelbaib oss, LU. 64* 20. ocht fichit 
OSS n-allaid, LU. 57* 10. In Old-Irish this noun also meant 
a wild hoar, and was neuter : fo tuaith do[s]sephain a n-os 
.i. in mucc allaid, Brocc. h. 57. Batar dano da n-oss, LL. 246* 
39. Compd. 088'feoil * venison,' ose-gamain ' fawn,' sg. gen. ir-richt 
os-gamna allaid, LL. 210% o*«-fo^Aar ' deerskin,' LU. 79*. From 
^ukso-B cogn. with W. ych^ pi. yehain^ from *uk8en, Aso. gl. cxxiv. 

ossar a burden, Ff. 60. a burden which ie on a human being, O'Cl. 

othar wages, Ff. 27, note 7. So H. 3. 18, p. 51*, H. 4. 22, 
p. 60*, and O'Cl. pi. dat. do othraibh 7 do thuarustlaibh Uladh, 
Lism. 103* 1. 

pdin bread, Ff. 61, £. 18. So Corm. and L.Leo. Voc. inid 
go caisg. ... do acht madh p&in agas biolar, Cumine. From pdni^. 

1. pairt j9(ir^, Ff. 61. So Corm. pi. dat. fot-dailfind i n-ilpartib 
ochianaib, LU. 69*. Often used for particle, as in pairte cro 
clots of gore, Chron. Scot. 166. From Lat. pars, partis, 

2. pairt j9m, £. 18. So L.Lec. Yoc. This seems a mistake, 
paitric bridle, halter, Ff. 19. O'Clery's /j^a^aw?. Seems formed 

from a prehistoric form of O.N. fjQturr * fetter.' For the change 
in Irish loanwords oi fi^ p cf. piscar-carla, plat, putrall. 

patan hare, Ff. 51, is rather leveret, patu is 'hare,' Corm., 



6L0SSARIAL INDEX. 89 

puttu, H. 3. 18, p. 6S7\ pata, 0*C1. pi. gen. is do thimnaib rechta 
Moysi nemthomailt feola maco 7 patan, one of the commands of 
Mo$es* law it not to eat Jleah of swine and hares, LB. 183^ 37. 
Hence O.Ir. patnide (gl. leporinus), Sg. 37^ 7. 

pauper /M>or, Ff. 69, note 8. 80 O'Cl. each pauper bid rf, Rawl. 
B. 502, fo. 61*2. pi. n. puipir do biathad, Laws iii. 18, 1. 19. 
Borrowed from Lat. pauper. Hence the dimin. pauperdn, Eel. 
Ep. 408. 

pelait F. palace, Ff. 61. piolait, O'Cl. nara chumaing Gonchobar 
n{ doib acbt in phelait ir-rabatar d'fdcbdil leo, LL. 263* 7. ba 
hirgna in phelait rfgda, LL. 256* 45. No doubt from Lat. paid- 
Hum ; but the e and the gender are surprising. 

pell horse, Ff. 19. So O'Cl. gen. ruccthar i capp indfaid phill do 
[f]racc, LH. 34* 2 (Goidel. 158)=LU. 6^ 9. da n-o pill fair .i. da 
chluais capaill fair, H. 2. 16, col. 690. Cormac has another form 
fill. Both seem borrowed from some cogn. of Eng. foal, Goth. 
fda, which Kluge refers to a pre-Oermanic peldn-. 

pet playing musie, Ff. 18, peit, L. Abstracted from the verb 
srpeitim, inf. airfitiud, ar-us-pettet a n-des ciuil, LU. 57* 20. 

pingur salt, E. 18, =pinguir, L.Lec. Voc. Borrowed from some 
Bomanic word cogn. with Fr. fane, fange, Ital. fango. For the 
connexion of ideas cf . Com. hdl * a saltmarsh.' 

pit a small meal, E. 15. So Gorm. s.v. fogamur, and L.Lec. Yoc. 
Compd. leth-phit, terc-phit, Eel. Sep. 8 = terc-cuit, LB. 260^. 
Prom W. pSth=lT. cuitt. 
pent rude, Ff. 61. So H. 3. 18, pp. 73^, 637*, and O'Cl. 
prann sea-wave, E. 18. prand, L.Lec. Yoc. prann, O'Br. Bor- 
rowed from some cognate of N.H.G. hrandung. 

poincne scruple ( = three pinginns), Ff. 61, E. 18. So Corm., 
H. 3. 18, pp. 73c, 637®, and O'Cl. A dimin. of pone. 

raiftinne fierceness, D. 24, = roptene, LL. 164* 49, .i. gairge, 
H. 8. 18, p. 73*. robtine .i. garg no ailgin, O'Dav. 110. Fiachu 
Boiphtine, RawL B. 512, fo. 102* 1. This seems O'Br.'s *'raiftine 
laughter, laughing." Did he mistake gairge for gdire? 

ramat road, Ff. 18. Hather a high-road, rdmut mo oldas rot, ace. 
to Cormac s.t. J^t, and O'Cl. ar each ramut ar bith che ria sluag 
namat conar-tf, on every road in this world against a host of foes 
My Ee come to us ! LL. 308* 7, fer tri ramata, O'Curry, Lect. 
506. 

rang baldness, E. 2. range, ace. to Corm., is where the temples 
are high. 



90 IRISH METRICAL OLOSSARIKS.— MR. STOKES. 

rann truth, D. 34. So O'E. ran .i. firinne, O'Cl. = ro+aii? 
See 1. an supra. 

rath a surety^ E. 22. r^th (gl. sequester medius inter duos 
altercantes), Leyd. 59*. rath .i. urra, O'Cl. rdth seeuritt/, 
guarantee, O'Don. Supp. Am rath-sa dia raith-sium, Fel. Ep. 165, 

re (ret?) manner (modh) ? D. 15, where L. has read in marg. 

re (MS. re) a multitude^ Fl 58, note 6. ind re, Conn. PruU. 
W. rhai. 

recht=:Lat. rectus, p. 3. In the literature, recht is generally a 
substantive, sg. gen. rechto, Wb. 21^ 13 or recta, Wb. 21° 1. for rout 
recta .i. for sligi dirig, LL. 3 1 6c 12, ace. conroiter recht robust, 
LU. \0^ 35. But in BB. 355^ 13 it occurs as an adj. Eismeach 
in rf recht. 

r6daire a cleric, Ff. 62. retaire .i. legthoir reader, H. 3. 18, 
p. 640^ reataire, O'Br. For this word, obviously borrowed from 
A.S. rddere, L. has recoire, and O'R. reeaire * reciter,' which occurs in 
Lism. 152^ 1 : Do fiarfaigh in doirseoir in raibi dan acu do righ 
Laighen. ' Atd,' ar in clamh, 7 is mfsi is reeaire do the doorkeeper 
asked had they a poem for the King of Leinster ? * We have,^ says 
the leper, ' and I am its reciter.* This is &om A.S. receere. 

reisi, reiside, Ff. 12, see rfss, rfsside. 

r6n a span {^airtOafirj), Ff. 42. rean, ren, O'R. reon, rian, O'Br. 
from *regno- connected with rigim * I stretch out,' as opyvia, 
opoffvta, opef^fUL with opeffu), 6p€yvv9» The mod. riise F. is from 
*rexsid: cf. 0'p€^i9. Cognate with both words are Lat. rego and 
Qoth. uf-rakfan, 

reos, p. 3. Borrowed from Lat. reus, 

1. rfad running, Ff. 8, is rather going. So Corm. s.v. arad, and 
O'Cl. Hence the verb riadaim. Cogn. with A.S. ridan, Eng. to 
ride, N.H.G. reiten, O.N. ri^a. Kluge connects Gaulish rSda 
* chariot ' and Gr. i-pt0o9 messenger, servant, 

2. rfad authority, law, discipline, Ff. 8. So O'Cl. taming, 
subduing, O'Br. Hence the adj. riata applied to a trained horse. 

T\BXdA(^) hereditary right, D. 31. A doubtful word. H. has 
hdcad. 

rfbar sieve, Ff. 47, E. 2. So Corm., L.Lec. Voc. and O'Cl. 
riobhar meala honeycomb, 0*Br. Seems borrowed from Lat. 
cribrum (from ^kreithro-n, Jx, criaihar) ; but the absence of initial 
makes this doubtful. 

ridnacht a giving, Ff. 21. Abstracted from do-r-idnaeht, Fel. 
Nov. 12, the t-pret. sg. 3 of tidnacim * I deliver, offer, give.' 



GLOSSARIAL INDEX. 91 

rise, itory, tale^ E. 2, corruptly reisi, Ff. 12. riss, Corm. ris .i. 
£u8Deis, L.Lec. Yoc. ris re aisneid Colum cen bith cen chill, Amra 
Choi. 8 : pi. n. dU rig rissi redi, LH. 26» (Goidel. 159), LL. 
187* 37. ace. fochlus int sentonn rise nde, Uath Benne JEtair 7. 

risedde iiory-teller^ Ff. 12, where it is, corruptly, reisidhe, risighe, 

robuist protection {?)f D. 59. So 0'C1.» but it seems rather an 
adj. borrowed from Lat. robustus. conroiter recht robust he kept 
tktfirm law, Amra Choi. 43, LU. 10* 35. 

robustus p. 3. From the Latin. 

rodLn tunic, Ff. 35. tunic or cowl, O'Cl. Formed on A.S. rocc 
(=N.H.G. rock), or Med. Lat. roccus, whence Ital. rocchetto, Eng. 
rw^t, 

rosal a judgment, Ff. 12. So O'Cl. Perhaps from ^rodh-tlo-. 
Cogn. with Teut. ridan, N.H.G. raten, A roMcel .i. brath (leg. 
brath?) occurs in H. 3. 18, p. 636. 

rose unfkrstanding, Ff. 10. So 0*C1. and O'Br. (rosg). Perhaps 
only a metaphorical use of rose * eye.' pi. n. roisc, LL. 210^ 14. 

ruanaid red, D. 21. So Corm. s.v. ruam, and O'Cl. Euanaid 
atberthe cosse frisseom ar m^t a naire, LU. 115^ 27. 

raba a mortal wound (guin), D. 19. So O'Cl. Li Laws 
1 106, 160, ruha in the phrases fuba 7 ruba, na tri rubs^ is rendered 
by 'services of defence.' Hence the verb conid-rubaim trenocu tria 
nert gaile, LU. 124* 14. From the account of the fight between 
Cucholainn and Ferdiad rubad seems to mean 'thrusting with 
spears,' as distinguished from slaide * slashing with swords.' 
O'Curry, M. and C. iii. 444. 

nicht 9wi/t, quick, E. 19. ruchd eudden, vehement, O'Br. 

ruice (MS. ruicedh) blush ? Ff . 62. So L. Lee. Voc. amdip 
rucce doib, Wb. 30* 3. ruice rebuke, reproach, O'Br. 

ruicet (MSS. ruicheat, ruichet) raising up, Ff. 7. ruiceat, 
O'Br. ruiceadh, O'Cl. The cogn. verb is exemplified in O'Don. 
Sapp. ni ruicer aire a thing {which) is bestowed upon her, 

rumra sight, Ff. 47. romhra, O'Cl., O'Br. What is romra, 
in LU. 40^ ? m' oenuran im romra ro. 

russ cheek, Ff. 62. So O'Cl. russ .i. agaidh /«<?«, Corm. Tr. 146. 
rts .L aigid, O'Dav. 110, and see Wind. Wort. s.v. 2. russ. gen. 
ronma rossa .i. romna aigthe ic aerad, LL. 1 87* 17. 

1. sab strong, p. 2. each soabb, p. 3. Seems taken from LU. 
^ 34 : ba so-abb i suthemlacht cech b^rlai co clethi. sab ' princeps, 
ioitia,' G.C.* 255. pi. n. sabaidh, O'Dav. 114. 

2. sab successor, p. 3. 



92 IRISH BfETRICAL GLOSSARIES. — MR. STOKES. 

1. s&egloxm judge f p. 4. H. 3. 18, p. 78^. O'Dav. 115. 

2. saeglonn old man^ p. 4. So O'Dav. 115. Derived from saegvi^ 
and this borrowed from Lat. Boeculum, 

8. saeglonn kingy p. 4. 

4. saeglonn /»t7^, p. 4. So O'Dav. 115. 

sdi tntldnesi, gentleness, Ff. 62. saidh, O'R. 

sdil aeeompanyingy D. 49. So O'Cl. a sail suad .i. a comaitecht 
suad, LL. 186^ 33. Hardly cogn. with O.H.G. sod (now SaaT), a 
house or hall, '' serving especially as a place for social union" 
(Kluge), whence Ge-selle, 

saillim (gl. condio) p. 4,=saillim (gl. sallio) Sg. 187V condio 
.i. sallim .i. intl nosailled o f orcetul br^ntaid ar cinad, LU. 8^ 4. From 
*saldid, cogn. with W. halltu. 

saithe M. a multitude, Ff. 55. So L.Lec. Yoc, H. 3. 18, p. 663% 
O'Dav. 116, and O'Cl. co Crist cechaing saithe, F61. Jan. 25. 
c6t-saithi a ndire na saileach, O'Don. Supp. s.y. saithe. 
deich mbeich[8]luaig (.i. saithe bech 7 lestra), H. 3. 18, p. 49. 
pi. nom. in tsaithi (gl. examina) Ml. 90^ 7. dat. rodamnad co 
sdthib slog, Salt. K. 6731. W. haid. 

salar (MS. ealar, et sic O'Br.) salt, Ff. 44. salor, O'Cl. Cogn. 
with salann, W. halen, Com. haloin. 

salt leap, £. 2. So Corm. and L.Lec. Yoc. isin bliadin sin hi 
cuiretar salt, Cr. 32^ 11. gen. slan dliged salto, Cr. 3°. Borrowed 
from Lat. saltus. 

sam summer? £. 3. The line in which this word occurs is 
corrupt. Perhaps the first three words should be samh each 
samradh. So O'Cl. sam sogar .i. dagthorthech in samrad, LL. 
188» 33. sam taurfrossach, LL. 293^ 37, 346*. is grian etrocht 
imbi sam, LL. 284* 40. Compd. uide se sam-la a journey of six 
summer-dags, Lism. 48^ 2. Vedic samd * year,' Zend hama * summer,' 
W.hdf. 

siihfood, £. 5. So Corm. and O'Cl. bai B6im sdth, Amra Choi. 
Different from sdith * sufficiency, fill,' sg. gen. do cathim a satha 
dfa[f ]e6il, LU. 46* 19. ace. ni thormalt sdith no s6ire, LL. 371^ 19. 

1. scdl warrior, hero, Ff. 32. So O'Cl. Scdl Balb, LL. 9*. in 
scdl sciathach, LL. 45* 25. gen. Mungairit meic Scdil Bailb, LL. 
206*. Gleann an scdil, Leac an scail, Lochan scdil, O'Don. Supp. 
s.v. sedl. Probably the same as 2. scdl. 

2. scdl a man, Ff. 24. So O'Cl. voc. a scdil, LH. 34* 2 (Goidel. 
158). airddithir a sciath ri scdl his shield as high as a man, 
LL. 44*. Hence sgalog (0. Ir. scdldc ?) * homunculus/ O'Moll. 94. 



OLOSSARIAL INDEX. 93 

sceng h^df E. 4. So Gorm. and O'Gl. sgeng .i. iomda, Dull 
Laithne 196. ard in sceng, LIT. 40* 1. im scing bic .i. biim il- 
lepaid immalle fri rig, LL. 187*. Perhaps borrowed from A.S. 
uuemg : on 9€Bec%ngum * in grabatis.' 

1. sc^ underitandtng, Ef. 12. So O'Cl. (sg6o). Cogn. with Lat. 
wio, ieius. 

2. Bc^o and, D. 56. So O'Cl. sc4o 7 neo, 7 ceo tri comaccomail 
Ooideilggi, three eonjunctiom in Gaelie, Amra Choi. 10. fodh 
macaib sceo ingenaib, O'Dav. 86, s.v. foth. immad ffna sceo meda, 
LL 343*. a muirib domnib sceo moraib, LL. 293^ 24. 

8ciU iuddeny D. 38. So O'Dav. 116, and O'Cl. (sgill). sciUa .L 
obann, L.Lec. Voc. 
8cip (MSS. scibh, scib) hand^ Ff. 42 : a sister-form of cib, O'CL 

I[8] 8(n 

teit in mal ina thech rig, 
i ndegiult cen cassair trit, 
CO nduibciond ' ina dag-scip.' 

Tkui goes the lord into his palace^ in a garment without a brooch 
through it, with a sword in his right hand, LB. 240*. 

scuirt shirt, Ff. 35. So O'Cl. (sguird). roghabh scuird-leinidh 
BToill, Battle of Ventry, 474. Borrowed from O.N. skgrta, F. 

sec honey Ff. 27, better siie, as in O'Cl. or seie, O'Br. 

s^g M. deer, £. 4. seg .i. oss allaid, Conn, (who quotes the ace. 
^segu), .i. agh allaid, O'Dav. 116. segh bos, O'MoU. 36. sed 
i 068, L.Lec. Voc. 

segach goat, Ff. 17. Derived from s4g? Or is it a mistake for 
^hach wolf, O'Dav. 117? 

segad (?) excellency 'i D. 18. A doubtful word. 

segamlae milkiness, E. 3. So Corm. Derived from seghamail 
'nilky,' O'Dav. 116, and this from segh .i. lacht, H. 4. 22, p. 67«. 

seim little (rather slender, ^ne, subtile), p. 3, D. 48. So 0*C1. 
seim (gl. exile). Ml. 139* (gl. macer) Sg. 37*. pi. n. s^ime, LB. 195* 
38. dat. semib, LIT. 35^ 34. Compar. semiu (gl. exiUor) Sg. 14^, 
(gl. tenuior). Ml. 19*. Hence the fem. abstr. sSme : sg. dat. semi, 
lU. 35* 42. ar mu s^mi-se (gl. pro ipsa mei adtenuatione) Ml. 
22* 1. Compd. s6im-tana (gl. exilem) Sg. 14* 8. 

Beire a meal, D. 39. .i. proinn no din^r, O'Cl. scire .i. feoil, 
H. 2. 15, p. 182. nochar* chaith saith no s^ire, Mart Don. 

^ .i. cloidem. 
' 'na dealam. 



94 IRISH HKTRICAL GLOSSARIES. — MR. STOKES. 

p. 188. nib airbirech fri s^ire, LL. 860, marg. inf. = nir' bat 
sereracb fri sere, LB. 101, r. margin, is ann roraid fri lesau in 
aliab do seilg co tucad sere do, 7 co tartad som a bendacbta dosom 
fora muUacb ar in sere sin then he told Eeau to hunt on the mountain 
andhnng himfood^ and that for that food he woM give his blessing to 
him on his head, LB. 113^ 38, caith in s^re, ol se, ib. 51. adbar 
sere detsiu, ib. 54. nati s^re, LB. 260*'. 

s^is learning f p. 3. So O'Gl. rofes ruaim, rofes s^is, his burial- 
place was knoum, his learning was known, Amra Choi. 44. LIT. 
10» 89 = LH. 27* 1 (Goidel. p. 163). adgenammar a s^is (gl. non 
ignoramus cogitationes eius) Wb. 14^ 28. bid glan a 86i8, LL. 
297» 8. sg. gen. fogroll sese, LL. 187* 4. feal ai 7 seis, unde dicitur 
fealmac .i. mac seasa, O'Dav. 86, s.v. fealmac. ace. cen dula dar 
s^is no smacbt, Salt. B. 2393. pi. ace. a druim re seisi, BB. 
293^ 13. From *sent'ti'. Cognate with, but not borrowed from, 
Lat. sensus from *sent'tU', Another s^is musical strain, is from 
*8end-ti-, cogn. with the verb sendim. 

seist mid-dag, E. 4. From Lat. sexta (hora). So Conn., L.Lec. 
Voc. and O'Cl. Spelt sest in the Palatine MS. 68, fo. 30». 

s^n a net for catching deer, E. 5, for catching birds, Conn, and 
O'Dav. 117, for catching deer or birds, O'Cl. sen fuirmither 
dichmairc a bird-net which is set without leave, 0*Day. 89, s.v. 
fuirmedh. The cognate W. hwyn-yn or hoen-yn means a hair of 
the tail of a horse, ox, etc., a springe or gin, 

Be6it property, especially cattle, Ff. 13. seoid, O'R. Cf. seoit 
taurclotha, Corm. s.v. aicillne : one of the noms. pi. of set ? ace. co 
rucsat a bu 7 a seotu, Bawl. B. 512, fo. 109^ 1. 

serbh theft, Ff. 74. 80 O'Dav. 117, but searb (with hard b), 
O'Cl. foserba petty thefts, larcenies, O'Dav. 117. O'Cl. has also a 
siorbhai * theft.' From *stervd cogn. with Gr. ffrepeto. 

sercoll flesh, Ff. 30. So O'Cl. sg. ace. ni sercol [.i.] sechnais 
sercol, Amra Choi. 73. pi. n. sercla (gl. irritamenta gulae) Sg. 
63% 11. batCr 6 a sercla: grut bruithe etc. LL. 117^ 23. cosin 
sercoll sochen61 mid, LB. 219\ So Diarmait says to Grainne (Rawl. 
B. 502, fo. 58» 2 : cf. O'Br. s.v. searcall). 

Is maith do chuit, a Grdinne, Oood is thy share, Ordinne I 

is ferr duit inda rfge : It is better for thee than a kingdom : 

serccoU na cailech f eda the flesh of the woodcocks, 

la banna meda mfne. with a drop of smooth mead. 

Originally a delicacy, relish ? Derived from sere * love ' ? 



GLOSSARIAL INDEX. 95 

serpan (serban ?) wan, Ff . 63. So O'Cl. searp^, O'Br. 

serr everything young and haughty, E. 5. sen* each n-uallach 7 
each n-ogla. Conn. s.v. seirach *foal.' aearr eolt, O'Br. cf. serr- 
graig a herd offoaU, LL. 103^ 16. 

1. serrda edge, Ff. 49. So O'Cl. But it is a participial adj. 
meaning provided with sickles, carpat 8er[r]da " currus falcatus." 
See Wind. Wort. s.v. serda. 

2. serrda cut, lopt, Ff. 49, the same word, with a slight 
difference of meaning. 

sescenn /J?», Ff. 70. salach rus .i. seiscenn, 0*Dav. 115. co 
sescen in da cor, a sescunn in da cor, Bk. Arm. 17* 1. pi. dat. 
sescnib, LIT. 28^ 19. 

set path, Ff. 68. ''semita nnius animalis,'' Gorm. s.v. rot. 
8g. gen. seta. pi. n. s6ti and seuit. W. hynt 'journey, way.' Goth. 
nnth 'journey.' 

sethnach side. See infra s.v. tethra. 

sillid a woman who performs tuaicle, enchantment ? E. 21. This 
word is identical with siUid * looker,' and probably means one who 
has the evil eye. Cf . Conn. s.v. miUiud. 

sin (MS. sin) necklace, chain, Ff. 63. So H. 8. 18, p. 73^. 
Cogo. with Or. ^vta. Sin round, E. 4, seems the same word : cf. 
Corm. s.v. sin. 

sioQ = Zion, 'Stwv, city of heaven, D. 36. In faith De dede Sion 
snidioth, Amra Choi. 11. fordonsndidfe Sione .i. non-snaidfe co 
siiab Sion .i. co cathair nemda, ibid. 140. 

sirsi, adj. light, Ff. 74, scribal error for irsi? 

dab narrow, E. 4. Cogn. with slahar 'narrow,' Corm. s.v. 
ilabrad. Perhaps the line in E. should be emended thus: 'slabar' 
cmnang, is ' coit ' coill. Hence es-slahar ' wide,' whence esslahra q.v. 

slabrad, slaibred (leg. slabra, slaibre?) D. 49, a kind of dowry 
composed of kine and bridles (ainm do bo-cethraib is d'echsrianaib, 
H.3. 18, p. 608*»). slabra Used of horses, LL. 85» 27, 33. slaibhre 
.1 coibhche, O'Cl. 

ilicht sense (ciall), D. 44, the O.Ir. sliucht ' cognitio,' Sg. 200\ 
Hence intsliucht, Sg. 201*, intliucht, and the adj. intliuchtach. 

sloigre sword, D. 50. slaighre, O'Br. For the suffix cf . hligh^e 
'a milker,' O'Don. Supp. sloighreadh, O.R. The root may be 
«^, whence the O.-Ir. perf. ro-selach (for ro-seslach), and Goth. 
fUhen. Slacc .i. claideabh, Duil Laithne 25, seems cognate. 

smerjire, E. 3. So Corm. s.v. smeroit, H. 3. 18, p. 637*, and 
O'Mulc. : Aod 7 tnu 7 smer 7 bott (.i. beoait) 7 tene, quinque 



96 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

nomiua ingnis, H. 2. 16, col. 90. In the gloss 6m6r .i. tene, 
H. 3. 18, p. 7S^f the mark of length seems wrong. Cogn. with 
Gr. fjuttpa, the dog-star^ futpfialpw^ fmpiew, a stone that takes fire 
when water is poured upon it; Lat. m^rus. The s is prob. 
prothetic. 

BTL^dedi proteeticn, D. 55. A sister-form of snadhadh .i. comaircei 
O'Cl. 0. Ir. snadud, verbal noun of sndidim. W. nawdd, nodded. 

snath hair? E. 3. So Corm. But perhaps we should read 
snuath = snuadh .i. folt, O'Cl. ced imda (.i. fada) a snuadh (.i. 
folt), H. 3. 18, part 1, p. 210=gidh fada a snuadh, O'Br. 

sn^id little, p. 3 = sn6idh .i. each nf beag, O'Cl. hii sa^gul 
sneid, Amra Choi. 24, ag seng sn6id, Bawl. B. 502, fo. 60^ 1. 
saiget Saxan sneid, Bawl. B. 502, fo. 47* 2. 

so good, p. 3. This is probably the laudatory prefix su-, so-, W. 
hg- = Skr. su-, Zend hu-, and perhaps Lat. su^ in su-hucula * a 
sacrificial cake made of spelt, oil and honey,' a different word from 
suh-uoula 'an under-garment.' See so-nirt, sopur, suithnge, 
infra. 

sol sun, E. 2. Borrowed from Lat. s6l, W. haul, however, 
shows that the word existed in Celtic. The genuine Lrish cognates 
seem sul A, grian sun, gen. sulut (leg. sulot), H. 3. 18, p. 74*, and 
sHil * eye.' 

solam sliseng? E. 10. The gloss is obscure to me : solam usually 
means quick, .i. so-ellam, L.Lec. Yoc. 

son M. word, D. 26, and L.Lec. Voc., = son voice, noise, O'CL 
son a gotha, Amra Choi. 41. sg. gen. suin and other cases in G.C 
981. Cognate with, or borrowed from, Lat. sonus. 

Bonirt (gl. robustus) p. 4=so + nerti-s. W. hy-nerth. Cogn. 
with Sab. nero, nerio, Gr. d-v^p. 

sopor a weU, D. 50. Siltair sopur na segsa for topur na trenoensa, 
LL. 156* 18. Sopor somma .i. am topur co n-immud eolais, I am 
a well with abundance of science, LL. 187* 5. Prom so-od-hur, as 
topur * well ' from to-od-hur, and/o^r * well ' supra, from/o-W. 

sorb fault, E. 4. So Conn., L.Lec. Voc, H. 3. 18, p. 74*. .i. 
lochd no salach filthy, O'Cl. foul, dirty, 0*Br. Possibly cogn. 
with Gr. <rvp<i>09, <n;/>0€TO9, and Goth, svairhan * to wipe.' 

stiall, girdle, apron, (fuathrog), Ef. 35. 

suan odar error, p. 2. This lemma is obscure to me, as is also 
the adj. siuin cech slemon, LL. 344* 50. 

suba blood. See infra s.v. tethra. 

1. Bmnacover? D. 38, of. suin .i. oumdach, O'Dav. 115. In 



GLOSSARIAL INDEX. 97 

the Tecosca Cormaic siiin is an adj. Ni cLuitbe nach sen ciarbat 
ooc . . . na nocht ciarbot suin, mock not an old man though thou art 
pvng^ nor a naked man though thou art clad, LL. 344^ 29, 31. 
2. sum heauty ? D. 38. 

saithnge eloquent, weU-tongued^ Ff. 63. So O'Dav. 117 and 
(yCL Sulpicc Bothnge suabaisy Fel. Jan. 13 (Bawl. 505), sui slan 
fiothnge sutbain, F6I. Sep. 30. From tbe prefix «o- and tenge. 
The opposite would be dothnge\ but for this we find dothenga: 
coeaitecb cecb dotheng[a], LL. 344*^ 11. dligid cacA dotheinga 
dige, LL. 294* 7. dligid cech dothenga miscais every eviUtongmd 
imrvee hatred, LL. 346^ 33. 

suit colour, D. 16. So O'Cl. This is perhaps the meaning in 
vmia tanie a eult ocus afeth, LIT. 129* 5. 

Buth milk, p. 1. So Corm. s.v. uth, L.Lec. Yoc. and H. 3. '18, 
p. 663. suth nime ticed don draic sin a flow of poison that came 
from that dragon, LB. 180*. Cognate with Gr. t/et it rains, as 
tvik .i. clann, H. 2. 15, p. 182, pi. ace. suthu, Ml. 39® 22, is 
cogn. with Gr. v/o*. 

tabartha wages, Ff. 27, literally (something) given, the pret. 
I>art. pass, of -tahraim the enclitic form of do-biur. 
taf (tof ?) deaf, Ff. 34. So O'Cl. and O'Br. (taoi). 
taide theft, D. 83. taide .i. gataide, Stowe XIX : no hailed in 
mac and sin i taide, Maegnimartha Find, § 4, where it means stealth, 
*weey. So in Fel. Prol. 71, cen taide .i. cen folach. And so in 
O'Don. Supp. mac doirche .i. mac dognither i taidiu sech fine a 
Mn of darkness, i.e. a son begotten in secret outside the tribe, oc laige 
la nrnii Find hi taide, Corm. Gl. s.v. ore treith,=il-liugu la mn^i 
Find hi taidiu lying with Find's wife in secret. Laud 610, fo. 82^. 
a topur na tonn taide, LL. 157^ 38. Derived from tdid (gl. fur) 

8g. 47* 9. Cogn. with Gr. rrfraofiai. 

tairdm (gl. fero) p. 5. is rather 'affero,* *impertio,' Asc. 
gl xcvii. : do-dircim is the orthotonic form, duairci (gL efficit), 
^ 61*. Verbal noun tdirciud, Wb. 13« 9, Ml. Ill* 5. 

taithmech a breaking, D. 54. .i. sgdoileadh, O'Cl. analysis, Corm. 
Tr. 156, B.V. triath. sg. ace. doniat a cotuch cen taithmech tria 
Hthn, LL. 303* 2. The right spelling is taithbech or taithbiuch. 
tathbinch carat. Bawl. B. 512, fo. 40^ 1. do taithbiuch to 
*irogate, Laws i. 18, 52. taithbech rudartha, LB. 101, marg. 
ioi oc taithbiuch a fuilt dia f olcud loosening her hair to wash it, 
H. 2. 16, coL 716. ^-pret. of the cogn. verb : fobfth to-n-aidbecht 
&Q0 a sid because he destroyed their fairy-mound, LU. 99*. pass. 

nSL Train. 1891-JM. 7 



98 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

pres. 8g. 3, taidbegar, Trip. Life, 160, 1. 19. y/lheg^ Skr. y/hhaj^ 
hhajyate, 
tap iudden, E. 23. So 0'Cl.=top, Conn, to-ud-h ? 

1. tebed cutting^ D. 26. teibeadh .i. buain no tarraing, O'Cl. 
tepedh, MR. 286 ; but tebe, teibe .i. buain, L. Lee. Voc., Stowe 
XIX. dobretha Fergus tepe forsin ngabul, LL. 61* 18. adbul- 
teipi. Laws i. 202. The cognate verb occurs: ri ro-thepi . . . 
asin ch6t-adbar . . . talam, Salt. B. 22. rotheip a meid is a 

• 

muin indlais a lama 'na fuil, LL. 154^ 30. From to-aith-he. 

2. tebed (tescad, tascar) a beginning, Ff. 54. A doubtful word, 
tec hone, Ff. 27. Another doubtful word, as one MS. here has teach 

and another sec. But O'Br. has tecy a bone, and O'B. teceach bony. 

techta N. law, D. 35 and L. Lee. Voc. = techte, Sg. 1 17* 5. 

teidm death, Ff. 64, is TQ,\heT peitilenee^ Corm. Tr. 139. pi. nom. 
ticfait iarsein tedmand ili ancride, LL. 188^ 53. Hence the adj. 
tedmnach. Salt. B. 946. A cogn. verb is no-tedmais (gl. 
tabescebamus), Ml. 131^ 4. 

teim dark, E. 1 1 . So Corm. and Stowe XIX. tern, L.Lec. Yoc. 
Skr. tamas, timira, 

tell sound (fuaim), E. 5. Seems inferred from Corm. s.v. tailm 
sling, which he explains etymologisingly as tell-fuaim. O'BeiUy's 
teall 'a noise, sound,' rests on the supposition that fuaim 'noise' 
is here a gloss on tell, 

teme death, p. 3, note 2. melg-theme, Corm. Tr. 108. Skr. ^tam, 

temel death, p. 3, is rather darkness, shadow, concealment, .i. scath 
no folach. Cogn. with Lat. temere, tenehrae, Skr. tamisra-m, 

tenlach^ra, Ff. 64, is rather hearth, Cormac explains it (more 
suo) aa=tene 'fire' and lige 'bed.' The n is usually assimilated, 
and we have tellach, gen. tellaig, dat. tellug, pi. n. tellaige with 
passage from the o- to the « -declension. 

tethra aroysUm crow (badb), Ff. 14. This seems a mistake for 
Tethra (a Fomorian king, see Corm. s.v.), the husband of the lamia 
Badb. eter triunu Tethrach .i. eter na treono Tethrach ; .i. ainm 
rfg Fomore, LL. 187^. Thus in Mac Lonain's stave (LIT. 50, upper 
margin) : 

Mian mna tethrach (.i. badb) a tenid (.1. gae 7 arm), 
slaide sethnach (.i. taeb) iar sodain, 
suba (.i. fuil), luba (.i. corp) fo lubaib (.i. fo feraib), 
ugail (.i. Buli), troga (.i. c6nd), d(r drogain (.i. fuach). 
The desire of Tethra^ s wife {i.e. Badb) are her fires {i,e, spear and 
weapon), 



OLOSSARIAL INDEX. 99 

Sluhing of sides thereafter^ 

Bloodf a body under bodies {i.e. men), 

jSyM, head, a just word. 

tf F. mantle, E. 5. So Corm. Tr. 156 and O'Dav. 121. pi. n. 
tii dubglasso col-luibnib corcraip impu, Eg. 1782, fo. 72. The gender 
appears from the compound : fortf choir imbi, LU. 87^ 8. co forti 
lethanchlaiss orbhuide tar a formna sechtair, Uath Benne Etair 68. 

timmgaire a seeking, D. 19. So O'Cl. tanic timgairi do tichtain 
dia crich fesin, unto him came a request to go to his own district, 
LU. 125^ 6. ba menic didiu a timgaire 7 al-Ifn, LL. 270^ 3. ba 
menic a timgaire, Eawl. B. 512, fo. 122<^ 1. See also O'Don. Supp. 
8.T. tiumgaire. A cognate ^-pret, timgart, O'Dav. 122. 

tinfed slender, Ff. 64. Bather perhaps slenderness, thinness, 
(-/«/=Br. -gued, see G.C 890), tinfed .i. tiniugud, Stowe XIX. 
tinfed .i. tinad, L.Lec. Voc. Cogn. with Lat. tenuis, W. ienau. The 
tinfeth ^aspiratio, spiritus' is quite another word, from *to-in'Ve'tO'. 

tinne baeon, D. 61. a hog killed and salted, O'Don. Bk. 
Rights, p. 121, note j. tindi .i. batun (leg. bacon), L.Lec. Voc. 
dam bnithe dano 7 tinne forsind Idr, LU. 23^ 38. sg. gen. amra 
tinne (.i. saille) senastar, a marvel of bacon she sained, Broc. h. 45. 
pi gen. tricha tinne, tricha b6, LU. 115^ 29. gurub fiu tri tinne 
logh bo eile, O'Don. Supp. s.v. logh. 

tir domaisi mountain, p. 3. Here domaisi seems the opposite of 
wnam, LU. 79» 10. 

tiacht the earth, Ff. 27. So O'Cl. Either from *tal-acto-, cogn. 
with tdam, or a metaph. use of tiacht 'raiment,' O'Dav. 119, 

from *tlagto, root l^gh, whence {'r)Xaxvo^i (j)\axyri and Eng. flock, 
O.R.G.fhccho from *]^lukken, Bezz. Beitr. xvii. 165. 

tlas a fair, Ff. 18. 80 O'Cl. 

Mfire, Ff. 5. So H. 3. 18, p. 615% O'Mulc. and O'Cl. et 
T. supra s.v. smer. tnu tene, LL. 393* 50. ba tarb tnu fri gleo, fri 
cath, Salt. R. 3895. Cogn. with Zend tafnu * heat.' 

tochell a going, journey, Ff. 6. So O'Cl. toichell richid, LU. 
^* 6. A cognate verb occurs : is f toichled Erinn i n-oenl6, LL. 
115^ 11. mairg toichless i ndagforcetul 7 doeall hi sseibe, LB. 12^. 

toichim a going, Ff. 6, note 13. tochim a charpait, LU. 105^ 32. 
From *to-c^im. 

tola abundance, flood, D. 9. tola .i. iomarcaidh excess, O'Cl. tola 
^9ci, LB. 25*. col-Hnad tola 7 lia husque less a muime, so that a 
P>od and spate of water was filling his foster-mother's garth, Trip. 



100 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

Life, p. 10, 1. 11. tola nsci, ibid. 434. tanic tola didirme do 
biastaib, LB. 141*^ 32. sg. dat. di tholu sechtrann et niLmat et 
geinte, Beicbenau Bseda, no. 167. ace. la tola n-echtrand 7 
dae8car8lua[i]g, LL. 188« 49, Rawl. B. 602, fo. 62» 2. Prob. 
eogn. with ttnrdol, fordil, derdily and intdla, Asc. gl. cxv. 

tore heartf E. 23. So Conn., O'Dav. 121, and O'Cl. So, too, 
H. 3. 18, p. 76*, 8.Y. Ion. sg. gen. tuirc. pi. nom. (used for ace.) 
dobend^is tnirc 7 tromchaepa a taebhaibh 7 a torc-asnach, a cheili, 
£atil0 of Ventry, 883. 

tort a eake, Ff. 60, D. 13. So Conn, and O'Cl. Hence the 
diminutives tartins, Corm. — da toirtine do thara, O'Don. Supp. s.v. 
tarrai, — and tuirtin, Laws ii. 242, 418. W. torth. All from Lat. 
torta (diuisit uniyersis tortam panis, Faralip. 16, 3). 

toth feminine gender, E. 6. So Corm., Stowe XIX, and O'Cl. 
Compds. toith-ghiobhair, toith-leannan, toith-8ear[r]ach, O'Br. 
Windisch connects toth (from to-sutd ?) with tdud * gignere,' from 
*i0'f0'8utu, ^8Uf whence also Ir. stUh fetus, Skr. suta ' son,' 
6r. V169, Goth, eu-nus. 

trenad lamentation, Ff. 53. So 0*C1., who has a cognate treana 
Tailltenn explained as a clapping ofpalme or lamentation which used 
to he at Teltown, 

trethan foot^ Ff. 42. So O'Cl. tricha treten (.i. traiged) dam 
comnaib, LL. 208^ 13. tren for trethan i nEgept, Salt. E. 
3644. ar rouir, ar tir mor a trethan, Bk. Fen. 228. atchuala 
tdirm 7 trethan in aen-oclaigh chuigi, Lism. 148* 1. 

1. triath king, lord, Ff. 2. So Corm. triath tiri Tethrach .i. 
Muiredach, Rawl. B. 512, fo. 110^ 2. sg. gen. tr§ith, LL. 187^. 
(in oenuch tuirc thr6ith). dual nom. da thrfath, LL. 232*. pi. ace. 
triathu, LB. 205^ 16. Compd. triathgein .i. gein rigda he, LB. 101, 
marg. inf. Ferhaps cogn. with the first element of Lat. trlt^vue, 
strit-avue. 

2. trfath wave, Ff. 2 : the gen. sg. is said to be trSthan, Cognate 
is trethan * sea,' gen. trethain. Cf. Tptrtvp and ^Afitpi-rptrff. 

3. triath hoar, Ff. 2 : the gen. sg. is said to be treithe. But the 
nom. pi. is treith : deich treith tire .i. sentuirc .i. tuirc bite for 
faithche, Amra Conroi. 

4. trfath hill, Ff. 2. 

trogan raven (brainf iach), Ff. 14. trodhain or troghan, O'Br. 

troichit hodg, Ff. 64, troced, E. 14. So Corm. s.v. fothrucud. This 
should perhaps be trocit, troieit, as in H. 3. 18, pp. 74, 638. 
troieit corp, Duil Laithne 2. Corn, trogeh Perhaps Lat. truncue. 



GL08SARIAL INDEX. 101 

troig tunrise, D. 50. trogh, O'E. A doubtful word. In 
Com. Gl. 8.V. trogein 'sunrise,' trog is said to mean 'bring 
forth:' cf. trogais .i. tusmis, LIT. 128^42 ; and in Conn. Tr. p. 162, 
tnigli, trag is glossed by clann 'children.' In LL, 186^ 37, 
i^^gM is said to be a name for the red rising of the sun in 
the morning. 

troiged children^ D. 12. cf. trog .i. eland, Corm. Tr. 162, s.v. 
tnigli and H. 8. 18, p. 650% and the verb trogais (.i. tusmis) 
di lorchuire (.i. dd serrach) {the mare) hrottght forth two coltSf LIT. 
128» 42. 

troniath helmet^ Ff. 37. A doubtful word ; spelt trdithiath^ L. 
troiath, O'Br. 

tropa (see itropa supra) may be a mistake for the troga .i. c^nd 
of LU. 50*, cited supra s.v. tethra. 

troll head, Ff. 38. So 0*C1. Cogn. perhaps with Lat. truHeum 



tuaichil cunning, astute, Ff. 11. tuachil, Sg. 60^ 7. fir-thuachaill, 
. E. 1670 : compar. tuachliu (gl. sapientior), Goid.^ 68. 
Oompd. fir-thuachaill. Salt. E. 1670. Hence the abstract noun 
imichle (misspelt tuaithle) .i. glicus. Trip. Life, pp. Ivii, 256, 1. 27. 

toarad a share, E. 6. So L.Leo. Yoc. tuaradh .i. cuit, Stowe XIX. 

toarastol wages, D. 40. sg. dat. nf dia thuarustul, S. Mart. 11. 
ace. na gebaind a thuarastal, LL. 109*. From ^to-fo-ar-es-talo-. 
cognate with Ir. taile (gl. salarium) and Gr. t€\o9 ' tax, duty, toll.' 

tacait cause, D. 27, L.Leo. Yoc. tuccaid, O'Cl. tucait (gl. 
cauaa) ML 58" 13. tucait a denma, LL. 186*. 

tucht/om, shape, Ff. 59. D. 16. So H. 3. 18, p. 609% and O'Cl. 
noantais eter each dd trdth in tucht sin, LIT. 133^ 15. tucht ara 
Ddalfar-sa, Cf. perh. rvKo^, rvxiXio,^ ^tuq. 

1. toirigin king, Ff. 3. So Corm. and O'Cl. 

2. tmiigm judge, Ff. 3. So Corm. and O'Cl. 

3. tuirigin (leg. tuiridin ?) tower, Ff. 3. So Corm. and O'Cl. 
a droim frisin tuiridein, Salt. E. 4520. Derived from turid. 

4. tuiiigin (leg. tuiridin ?) tongue, Ff . 3. So Corm. and O'Cl. 
den-fac for a tuiridin, Udth Benne Etair 60. 

tul countenance, Ff. 44. So O'Cl. o thul co aurdomd, LU. 79» 19. 
CGnecmaing a tul immon n-all, LIT. 109*^ 22. tul dreiche. Laws 
i. 66. tul i tul, LL. 225» 39. Compound : tul-lethan, LU. 105»» 37. 

tarba a hand, £. 6. So O'Cl. Corruptly turhaid .i. buighen, 
Stowe XIX. Deich [n]-exercitus . . . tiagait ... in turba, 
I)eich turba . . . iss ed tiagait i n-agmen, Salt. E. 771-774. 



102 IRISH METRICAL GLOSSARIES. — MR. STOKES. 

Borrowed from Lat. turha. According to a note in Egerton 1782, 
fo. 49^, there were 10,000 in a legion, ten legions in a ^nta (cuneus), 
ten cunias in a mares (myrias), ten mareses in a caterva, ten catervas 
in an exercitus, ten exereitm in a turha, ten turhas in an agmen, 

turbuid protection, exemption, Ff. 45. turbaid chotulta sleepless- 
ness. Common in the Laws : re na turbaide .i. 198 : i turbaid, i. 
1 94. See O'Don. Supp. s.vv. turbadh, turbaidh. 

uagba choice, election, D. 58. uaghbha, O'Br. Perhaps we have 
an obi. case of this word in tri embaid uagboid, LU. 94, 1. 22. 
In BB. 351^ 4 — bad athlum aine im-uaga — the last word seems 
a corruption of uagha. From ua+gab- ? 

uain a lending, Ff. 68. oin eich ag &r for bid, O'Don. Supp. s.v. 
bla. sg. dat. oc uain 7 airlicud, Wb. 31® 5. Hardly cogn. with 
Gr. usv-q * buying.' 

uamun (MS. uamhuin) M. fear, Ff. 65. in t-6mun, Ml. 42*9. 
sg. ace. lat graain 7 t'omun, LU. 98^ 9. ar omun lathe bratha, LL. 
281^ 26. pi. ace. 6mnu, Amra Choi. 16. But dat. and ace. sg. 
uamuin are found. Hence the adjj. omnach: (is mana der co 
homnach, LL. 257^ 22), and immuamnach : imuamnach cech cintach, 
LL. 344c 9. W. ofn, M. Gaulish *obnO'S in Exohnus, 

uath mould, clay, D. 53, £. 24. So L.Lec.Yoc. and O'Cl. sg. gen. 

uatha, Conn. s.v. audacht. Compd. : ualh-onn : in-adbai uath- 

uinni .i. i n-adbai ure 7 chloche, .i. uath uir 7 ond cloch, LL. 

187* 32. 
uathad N. solitude, Ff. 21. .i. beagdn, O'Cl. a few, conna 

torchair acht uathed mbec im Conaire .i. nonbor nammd, LIT. 98^ 39. 

is lor uathad dib for desmbirecht, LB. 219°. In grammar the 

singular number and the first decad ; gen. aile uathai^^ esci Martai, 

LB. 90, lower margin, cethramad uathat^^ 6sci luin, ibid, hi 

coicid huathid on the fifth of the first decad, Cr. 33^, as distinguished 

from coiced deac fifteenth, lit. fifth of the second decade, coiced fichet 

twenty-fifth, lit. ^ fifth of the score.^ o coiced uathaid. Bawl. B. 

512, fo. 52* 1. Uathad is also used as an adj. rop uathad i 

sanasaib let him be solitary in secrets, LL. 343* 20. giarb' uathed 

do dam malle, LL. 87* 25. rob uathad mo ddm, LB. 214\ Cogn. 

with Lat. pau-cus. Got. favai, Eng. few. 

ugail eyes. See above, s.v. tethra. pi. n. of *ugal, borrowed 
from Lat. oculus. 

uim earth, Ff. 67. So O'Cl. Compd. uim-chrith earthquake, 
O'Br., of which arm-chrith, Trip. Life 46, 1. 5, seems a corruption. 
Borrowed from gen. sg. of Lat. humus ? 



GLOSSARIAL INDEX. 103 

\Md a pack-saddle, Ff. 19. ulaidh, O'Cl. 

umla F. humility, Ff. 31. umla (gl. obsequium), Eg. 88. sg. 
nom. umla cen fodord, LL. 37 1^' 29. in umla-sin dorat Euagair 
doThatha, LB. 147*. ace. humli, Amra Choi. 80. ni thucc an 
mac ba so umhla don mac ba sine, FM. 1228. Derived from umail 
or umal (bat nmal cor' bot uasal, LL. 345<^ 19), which, like W. 
«/e^, is borrowed from Lat. humUis, 

Tm&Y. famine, "Ft, 26. SoStoweXIX. dith f or finibh no plaigh 
no una no duinebhath, Bk. Eights, p. 184, 1. 5. A corruption 
of niLtia (maita la nuna .i. gorta, LL. 188° 60. nuna 7 gortai, LB. 
IH*), and this perhaps of O.-Ir. ndine, sg. gen. di phlagaib tened 
etnoine et gorte, Reichenau Baeda, No. 167. W. newyn, M. For 
loss of initial n cf . «im»r=numerus. 

unse here is, D. 43. So O'Cl. Uinsi, O'Dav. 124. Undse sund 
tali, or Fergus, LIT. 69^. undseo col-luath dot-bia in tuath ra 
togais, LL. 45^ 33. undsea, LL. 100^ 51, 101^ 1. unse a ben lasin 
ng; ondat a bai issin tfr ar far mbelaib there is his wife with 
il^ king, here are his kins in the land before you, LL. 252^ 14. TJindsi 
tliall he .i. aici thall he, L.Lec. Yoc. Huinse Conall Cemach sund, 
lA^h as dech la TJltu, LL. 252^ 31. Ise, issi, iss^, uinnse, unnse 
[leg. uinnsi], onnar a urlunn indsci, BB. 330^. Isse, issi, iss^e^ iar 
macaib Mil^^. TJindse, uindsi, ondor iar Feraib Bolg. Mod 7 tod 7 
tmeth iar Tuaith J^e Danonn, BB. 327^ 49. Ondar here is, LIT. 
62* 12=undar, LU. 65** 41, seems cognate. 

1. or heyinniny, D. 25. So Corm. Tr. 166, s.v. urla, L.Lec. 
Voc., Stowe XIX, and O'Cl. 

2. or noble, D. 29. So O'Cl. This and 1. ur seem nothing 
but the prefix air-, er-, ir-, ur- ^ Skr. pari, Gr. »•€/>/, Lat. per, 
So^r .i. mor, O'Dav. 81, citing the adj. erderg very red. 

w had, Ff. 59, 65, E. 7. So Conn., Stowe XIX, and O'Cl. Vr 
•i. olc, Duil Laithne, 136. From */ti-ro-«, cogn. with vveu, vvov, 
Iat|?iu, Qoth. fdls. 

U8 declaration, narrative, D. 42. So O'Cl. go roib a hus 7 a 
imthtMa 7 a deired na Tanad gonici sin, so far the narrative, and 
the events, and the end of the Tain (W Cualnye), LL. 104^ 4. From 
*^tu, *vad'tu. Cognate is immthiis=^imm-to-iis. 

tissarb death, Ff. 65. So Corm., who quotes H rodet do-ussairb 
i n-UUtaib, a kiny who suffered^ an evil death in Ulster. 

^ TO'tkt^ the ^-pret. s^. 3 of danuUm, The quotation is gi?en in LL. 179^ 
» ' lig rodet roussarb nUltaib.' 

[Read April n, 1891.] 



104 



II.— THE CELTS AND THE OTHER ARYANS OF 
THE P AND Q GROUPS. By John Rhys. 

[Mead February 20, 1891.] 



I. 

It is a commonplace of Celtic philology that the Celtic 
languages of modern times divide themselves into two groups^ 
namely Goidelic and Brythonic, and that the Ooidelic group 
consists of the Gaelic dialects of Ireland, Man, and Scotland, 
while the Brythonic group is made up of Welsh, Old Cornish 
and Breton. It is equally well known that Brythonic speech 
is characterized by the use of the consonant p (liable accord- 
ing to its position to become b and ph), while inscriptions in 
early Goidelic show qu, which is simplified in the later stages 
of Goidelic pronunciation into c (liable to be modified accord- 
ing to its position into ch and g). Now the question suggests 
itself whether the Continental Celts of antiquity had not 
likewise two languages similarly characterized by p and qn 
respectively: in other words, whether, besides the Gauls, 
whose language is known to have had p for original qu, there 
was not on the Continent a Celtic people whose language 
was like early Goidelic and resembled it particularly in 
retaining qu. I have been gradually convinced that there 
was such a people, and I should call it Celts as distinguished 
from Gauls, but that the modem meaning attached to the 
word makes this inconvenient. One has therefore to fall 
back on the name given by Pliny to their portion of Gaulish 
territory, namely Celtica} and call them from it Celticans, 
just as we speak of the inhabitants of Africa and America as 
Africans and Americans. Celtica was the country of the 
people who according to CaBsar called themselves Celtm in 
their own language. He regarded Gaul (exclusive of the 

^ Nat, Hiit. iy. 105: ''Gallia omnia comata uno nomine appellate in tria 
popnlomm genera diyiditor, amnibua maxume distincta ; a Scalae ad Seqnanam 
belgrica, ab eo ad Oanmnam Celtica eademque Lugdunensis, inde ad Pyrenaei 
montis excunum Aquitanica, Aremorica antea dicta.'' 



CELTS AND OTHER ARYANS. — J. RHYS. 105 

ProYince) as consisting of three distinct regions, one of whicli 
situated beyond the Garonne was inhabited by the Aquitani, 
who were wholly or mainly non-Celtic. The other two 
peoples were the Celtse and the Belgae. As to the CeltsB he 
says, that they were separated from the Aquitani by the 
(xaroDne, and from the Belgae by the Seine and its tributary 
the Mame. That is to say, one is given to understand that 
Caraar's CeltcB, whom it is here proposed to call Celticans, 
occupied all North-western, Central and Southern Gaul 
outside the Roman proyince ; and within the limits of that 
province itself they probably formed the bulk of the Aryan 
population there, at least before the Allobroges were annexed 
to it.^ Moreover they had also penetrated into Spain ; for 
^e possess important evidence to their presence there in 
the well-known name of the mixed people of the Celtiberi. 
I^tly, as the Celtican element found its way to the heart 
of the Iberian peninsula, it may have also possibly reached 
the North of Italy ; but this must depend for its proof 
on the interpretation to be put on the slender facts of 
language to be presently brought forward. 

Let us begin with the instances of q names found in the 
Ancient inscriptions of the countries here in question, and, for 
tbe sake of convenience, let us take first those of Spain 
and Portugal Here we have Alluquius, Arquius, DoquirnSy 
^me9U8 and Quarquerni, together with some others about 
which there is less certainty.* 

Arquius, Alluquius. These two names occur in one and 
the same inscription ^ at Valen9a on the Minho in North 
Portugal, while Alluquius occurs elsewhere in an inscription 
at Paimogo* in the west of the old province of BaBtica, 
near the river Guadiana; and somewhat less certain is its 

^ I make this qualification aa the name Alhhropes wonld seem to mean a people 
" of other marches : *' possibly they belonged to another branch of the family. 

^ Sach as the dative Omio Laquiniesi on a stone from Caldas de Yizella near 
voiinaraens in the North of Portugal : see volume ii. of the Berlin Corpus 
j*«n>. Lat. No. 2405. Such also as Atlondut MaquiaeMUt Sunnae J*., of 
wnibtfol reading : see No. 4980. 

* No. 2465: Bis . Manibus | AUuquio . Anders:! . F. | Aeturae . Arqui . 
'• I Macro . AUuqui . F. CI | utimoni . Alluqui . F. Civi | £na 

* lie. 961 : Glaucus . Aluquii . F. | H.S.E 



106 P AND Q GROUPS. 

presence in an inscription at Arroya del Puerco^ not 
very far from Caceres or the ancient town of Norba in 
Lusitania. 

The origin of Alluquiua is obscure ; but I cannot help 
regarding it as possibly one of the forms which Arquim 
took in Iberian mouths.' This latter name, besides occurring 
with Alluquius in the Yalen^a inscription, is to be met 
with in others, at Trujillo, the ancient Turgallium, in the 
east of Lusitania, at Monteagudo between Tarazona and 
Gascante in the ancient Tarraconensis, and at Astorga; the 
ancient Asturica Augusta, to the south-west of Leon.' Also 
at Calderuela near Soria between Tarazona and Osma,^ and 
at Carri9a in the vicinity of Oporto,* besides Arquia in an 
inscription from Condeixa a Nova, near Coimbra.® The name 
Arquius probably meant one who had to do with the bow, 
that is to say, an archer, and was derived from a Celtic 
word^ cognate with the Latin arquus and arcm,^ a bow or 
arch. Whether the adjective Erquesf« ^ should be considered 
as in any way related to Arquius is extremely doubtful. 

Doquirus, Docquiricua, We have Doquirua ^® from Trujillo, 
and Docquirus ^^ from Idanha a Yelha, the ancient Igaeditani 
in central Lusitania, and a probable Doquira^^ from Soure 
in the neighbourhood of Alfeizarao in Lusitania, also a trace 

^ No. 737 : Amona, AUuoi, F., where the reading Alluqui has been suggested 
by the editor Dr. Hiibner. 

^ Here my colleague Prof. Nettleship reminds me of Cicero's words when in 
his oration Fro Arehia he speaks concerning **natis Cordubae poetis, pingue 
quiddam sonantibus.'* 

3 Nos. 632, 2990, 2633 respectiyely. 

* No. 2834. 

B No. 2373 : other instances will be found in Nos. 2458, 2433, 2435, all from 
localities in the ueighbourhood of Braga, the ancient Bracara Augusta. 

• No. 377. 

"^ We have probably a trace of the word in the Welsh arffed^ *the lap or the 
abdomen/ since arqu must become arph or arff in that language : as to the 
meaning compare the German scham^M^. On Teutonic ground the word is 
implied by the Gothic arhvazua * an arrow,' A. -Saxon earh^ the same. 

*) The modem Welsh arch is the Latin areut borrowed, but in the colloquial 
the English arch (with palatal eh^ is usoally substituted for iL 

^ It comes from Alcala del Rio north of Seville, and, purporting to be the 
name of a centurion, it occurs in company with such other names as Beiesiif, 
Aryaborestf, IsinesM, Isurgutofto, etc. : see No. 1064. 

10 No. 624. 

»i No. 448. 

" No. 364. 



CELTS AND OTHER ARYANS. — J. RHYS. 107 

of the name in an inscription at Alfeizarao itself.^ One may 
add Docquiricm or Docquirinus,^ from Freixo de Nemao on 
the Douro in Lusitania, Docqumcua^ from M^rida, the 
ancient Augusta Emerita in the south-east of Lusitania, and 
a doubtful instance ^ from Lara de los Infantes, south-east 
of Burgos in Tarraconensis. The origin of Doquirus ^ and 
ita derivatives is not certain, but it cannot be Oaulish ; so 
it may be regarded as here in point, since it has the appear- 
ance of being an Aryan word. 

Eqmbona. This is the name of a place in the Itinerary 
of Antoninus,^ and it was on the route between Lisbon 
and M^rida on the Ouadiana. It is unmistakably Celtic, 
and recalls such other Celtic names as Yindobona and 
Bononia, but in Gaulish it would have doubtless been 
I^pobona, like Eporedia and the like, the first element in 
the compound being forms of the Celtic word which is 
in Latin equu% ' a horse,' in old Irish ech * horse,' and in 
old "Welsh ep-aul, now ebol, *a colt.' The derivation of 
*ww in JEquabona, Vindobona, and the like, is unknown ; 
it may be of the same origin possibly as the Welsh bdn 
*the stem or trunk (mostly of a tree),' Irish bun " Wurzel- 
atock," also probably the German buhne^ which refers 
father to the boards made out of the trunk of a tree ; but 
the Greek word fiowo^ * a hill, height, heap, mound,' would 
86em to suggest an easier explanation of the Gaulish place- 
iiames ending in bona. It has been hinted by M. d'Arbois 

* No. 360. 
'Ko. 431. 
'No. 651. 

* No. 2862. 

* It lemindB one of Gartnait Diuperr, or Oartnaich Diuberr of the Pictisli 
Chronicle, whose name is rendered in other chronicles Garnard Divea and 
Gtrnard le riehe: see Skene's Chron. of the Picts and Scots, pp. 6, 28, 172, 
200. If JHuperr or Diuberr he a loanword from a Brythonic dialect, it might 




P"^ as Doquirut and Liuperr, in case they are to he analysed as here assumed. 
In a note from Dr. Whitley Stokes I learn that he gi^es the preference to the 
ipelling IHupeiry as he connects it with the Welsh pair 'a cauldron,* and 
ttTokfis the parallel of an Irishman who was called * a cauldron of hospitality.' 

•See Parthey & Pinder's edition, No. 416 (p. 197) ; in the Index they 
»«i% Equahona with Couna, Coyna. 

^ See the fourth edition of Kluge's Diet. 



108 P AND Q GROUPS. 

de JubainvIUe that the first part of Equabona is purely 
Latin. Of course, names like Juliobona and Augustobona 
existed in Gaulish, together with others into which the 
Latin proper names Augustus, Csesar, Claudius, Drusus, 
Flavins, and Julius entered ; but at present I cannot 
recall an early instance involving a Latin appellative like 
equa, 

Equam and Quarquemi. These were the names of two 
of the peoples who formed the so-called Conventua Bracar- 
augu8tanu9 as enumerated in an inscription ^ found at Aqusd 
FlaviaB. The peoples of this Conventus dwelt between 
the rivers Minho and Douro, where the principal towns 
were Bracara or Bracaraugusta, and Aquse Flavise, now 
Braga and Chaves respectively. The Equaesi are so called 
also by Pliny;' but the Quarquemi's name is given as 
Querquerni^ by him. Here Querquern- is probably identical 
with the Perpern- of such personal names as the masculine 
Perpema and the feminine Perpernia, both of which occur 
in Spain, namely at Tarragona and Barcelona ; ^ also in 
Cisalpine Gaul, to wit, at Padua ;^ but the most note- 
worthy instance, perhaps, is an inscription found at Martos, 
the ancient Tucci, not far from the eastern boundary 
of BaDtica : it reads ® — D . M . S | M . Perpema Gallicanus | 
Annor .L.H.S.E.S.T.T.L| Huic . Mer . Fil . Et . 
Nep . Fee | . Querquem- and Perpem- look like reduplica- 
tions of a stem represented in Irish by crann'^ and in 
Welsh by prenn, now written pren, *a piece of timber, 
a tree,' which are presumed to be of the same origin as 

» The Berlin Corpus, ii. No. 2477. 

> Nat. Hist. iii. 28. 

» Ibid, 

« The Berlin Corpus, ii. Nos. 4301-2, 4393, 4547, 4555. 

^ Ibid. T. Nos. 3004 and 2856, which is of doubtful reading. The name C. 
Perpemius, which occurs in a list of gladiators found at Venusia in Apulia, 
probably belongs here: see the Corpus, ix. No. 466. The relation between 
Perperna and Ferpenna is a question of considerable difficulty, which I am not 
prepared to discuss. 

« No. 1709. 

"^ The form in early Celtic must hare been qy^fnon of the neuter gender, and 
the reason why the reauplication should yield, not q^erq^rarm-y but q^«rquem'y 
is to be sought in the too great accumulation of consonants the former would 
occasion. 



CELTS AND OTHER ARYANS. — J. RHYS. 109 

the Latin quercua ' an oak/ quernus * oaken, of oak ' : com- 
pare Sabrann, an old name of the Irish river Lee, Welsh 
Hafren *the Severn/ and Irish salann *8alt/ Welsh halen, 
as illustrating the same treatment of rn and In in Irish 
and Welsh respectively. 

Let as take next the other outlying portions of the Celtic 
world looking towards the south, namely Gallia Cisalpina or 
I^orthem Italy and a portion of the Alps. Here we have 
a variety of names which may have belonged to a Celtic 
dialect ; but several of them are too obscure to be of use 
to us, such as Quassaunaf^ also given as Cusonia,^ both in 
inscriptions at Verona; and such as Querra^ in another 
inscription at Verona. Then we have a people called 
Qoadiates on the Cottian Arch at Susa : ^ they belonged 
to the Cottian Alps, and were probably the same as the 
Qaariates supposed to have left their name to Le Queyras 
in Switzerland. It is scarcely probable that they were the 
Germanic people of the Quadi of whom we read in the 
Oermania of Tacitus ; far more likely is it that they were 
either Celts or Ligurians, if we may judge from the other 
peoples enumerated in the same inscriptions.^ Putting aside, 
then, such names as the foregoing, we find remaining in the 
inscriptions such forms as Equasia, SquilUuSy and Veiquasiua. 
The stem of the name Equasia found at Trieste^ might be 
either Latin or Celtican, but the formation of the word seems 
to resemble that of VeiquasiuSf the Celtican origin of which 
i« not improbable. This, together with Vequasim and Vequasia, 
occurs as a personal name in inscriptions^ from Polenza and 
neighbouring places in Piedmont. The variety of the 
spelling shows that the vowel of the first syllable was long ; 
80 that we may have here in fact the same stem as in the 
Irish names Fiachna and Fiachra. The genitive of the latter, 



1 Berlin Gorpiu t. No. 3463. 
« No. 3916. 
» No. 3597. 

* No. 7231. 

* See Mullenhoff*B Deutsche Altertunakunde, toI. ii. 249. 

* No. 694. 

f Nos. 7694S 7680, 7682. 



110 P AND Q GROUPS. 

Fiachrachf appears in a late Ogam as Veq^mcc} while that of 
the other, Fiachnay is found in an Ogam of the earlier sort as 
Veqoanai} The later forms with their ia show that the 
original vowel was ei or e and the Irish common noun fiach 
' a raven ' suggests a possible interpretation of this group of 
names. 

Lastly we come to the names SquiUim ' and SquilimntM ^ 
in inscriptions at Verona; to these may be added Squeillanius^ 
from an inscription found at Narbonne, and possibly a 
genitive Squeiioles^ in a Christian inscription at Marseilles. 
The variety of spelling here also suggests that the vowel of 
the firs£ syllable was possibly long ; and in that case the 
length was probably the result of the elision of a consonant, 
the original stem being presumably squetlot a neuter squetldn, 
meaning ' a piece of news, a tale.' The Goidelic language of the 
insular Celts retained the t for centuries later, as we find on 
one of their monuments in Glamorgan the compound scttii^ 
rissi, the genitive of an earlier aquetliviaao- ; but the point of 
importance here is that the Brythonic treatment was quite 
difierent, seeing that the Welsh word for a story is chwedi, 
from an earlier snetlon^ to which aquetlon had been levelled,^ 



^ The stone comes from Monataggart, in tlie parish of Donoughmore, in the 
county of Cork, and is now at the Royal Irish Academy. 

^ 'i'he stone is at Cooldorrihy, in the parish of Eilmichael, in the same county. 
The of Veqoanai is unusual and meant probably for the h (of the qj* ordinarily 

represented by the single Ogam character 1 1 1 1 1) : in another instance the 1 1 1 1 1 is 
followed by the Ogam for w or v^ namely m, but the Ogam for qu is never con- 
founded with that for e or k. 

» Berlin Corpus, ▼. No. 3335. 

« No. 3401. 

• Berlin Corpus, xii. No. 6972. 

« No. 491. 

' For other instances of iu for squ see my Celtic Britain ', p. 92. The dis- 
carding of the q was here probably due to the syllable being unaccented ; so also 
in Welsh peunydd ' every day, daily * for qjtoundu- from an earlier quoqnn-dii'. 
Compare the old Irish eethir four, the initial of which suggested the symbol 1 1 1 1 
for e in the Ogam alphabet ; so the word was probably at one time eetu6re» 
from an earlier quet^res, corresponding to the Welsh petguar^ now pedtcar. 
This accentuation is also that of the Sanskrit catvSras and it is implied by the 
d of Gothic Jiditor^ and possibly by the a of the Latin quatuon see Mr. Wharton's 
paper in the PhU. Trans, 1888-90, pp. 46-9. On the other hand, the former 
initial of the Irish e6ie^ e6ig * five * suggested the Ogam symbol 1 1 1 1 1 for qu, 
which the fifth numeral must have retained in Irish Gaelic : in fact Manx GaeUo 




CELTS AND OTHER ARYANS. — J. RHYS. 

and the same remark probably applies to Gaulish. Such a 
name then as Squillius would mean a news-bringer or a 
scout, and perhaps a story-teller. 

The ancient inscriptions of lUyricum present nothing very 
certain as regards the present question : we get a personal 
name Quordaio^ from the neighbourhood of Salzburg, but 
it is obscure to me as is also the genitive Quitai,^ which comes 
from Kov&csi, north-west of Buda-Pest in Hungary. It 
is probably more correct likewise to regard q names, such as 
Quermy^ from the Roman town of Aquileia and its vicinity, 
as Latin rather than Celtic. So we now recross the Alps to 
Transalpine Gaul, where we have already found the name 
SqueillanitM and others in point. 

A word must now be said as to the relation between 
the two Celtic peoples of the Continent, and first from 
the geographical point of view. If we take for granted, 
as we safely may, that the earliest population of Gaul was 
not Celtic or even Aryan, and further that the Celts of 
ancient Gaul did not originally come thither from the 
direction of Spain or of Italy, we are entitled to conclude 
that the Q Celts arrived in the west before the P Celts, 
as they are found occupying the furthest parts of the Celtic 
Area, namely north Italy, Spain, southern and western Gaul, 
Ireland, Man, the Scotch Highlands and Islands. The 
conclusion is scarcely to be avoided that the later comers, 
the P Celts, came as invaders and conquerors constituting 
themselves the ruling people wherever they could vanquish 
the other race. This seems to have been the case in most 
of Gaul, and some of the results of this state of things 
^ indicated in CsBsar's Commentaries/ as, for instance. 



^ retains it in its quetg ' five * as contrasted with kegeesh (pronounced kegith or 

^riA) ' a fortnight,* in Irish eoicdigis or edicthighis, with 6i owing to the 

«"»logy of edig. 

^ See the Berlin Corpus, iii. No. 5523, from which it appears that Conginna, 
^Qghter of Quordaio, nad married the grandson of a man bearing the name 
'^^tvai'i, which looks like Celtic. 
' No. 3621, but Quita's daughter bore the apparently Celtic name of 

'VoLt. No. 1270. 
* Books L 4, Ti. 13. 



112 P AND Q GROUPS. 

in the multitudes of clients and dependents forced to attach 
themselves to the Gaulish chiefs. This would also explain 
the scarcity of data as to the language of the earlier Aryans 
there; for Roman and Greek writers would come more directly 
in contact with the ruling race and hear Celtic names only 
in the Gaulish form which the ruling Gauls gave them ; 
and, to judge from similar cases elsewhere, the subjugated 
race must have acquired at an early date the habit of 
translating its own names and forms into those of the 
conqueror's idiom. We seem to have an instance of this 
in an ex- veto to certain Nymphae Augustae in the neighbour- 
hood of Yaison in Gallia Narbonensis, where the name 
applied to them is Percernes,^ in which we seem to have an 
imperfect translation into Gaulish of the qiierquern- of the 
proper name Querquemi ; so the epithet may be supposed 
to have characterized them as nymphs of the forest or wood- 
lands. But be this as it may, that is the sort of process 
always going on in dialects brought into contact with a 
dominant dialect. 

Nevertheless there are one or two well-known names which 
for some reason or other successfully held their own, and 
those are Sequana and Sequani, and Aquitani. How the 
name of a river ^ like the Seine should have been accepted 
by the invading race is not very difficult to understand, but 
the difficulty is greater when we come to that of the people 
known as the Sequani, as there are no data to decide whether 
they were Celticans or Gauls. Their geographical position, 
however, among the mountains of Switzerland, and the rdle 
played by them in Gaul at the time of Caesar's advent 
and for some time previously, are not inconsistent with the 
possibility of their belonging wholly or in great part to the 
earlier race. Caesar represents * the ^dui and the Celtican 
people of the Arverni as the heads of two rival federations ; 

1 Vol. xii. No. 1329. 

^ It occurs also as a woman's name, in an inscription from the neighboarhood 
of Trevisio to the north of Venice : see the Corpus, v. No. 2129. Most probably 
Sequana was in the first instance the name of a goddess (see Rey. Celt. vol. iii. 
p. 306) ; but a compouud name Sequanoiotuot occurs on coius ascribed to the 
Sequani. 

* Bk. i. 31. 



P AND Q GROUPS. 113 

and that of the Arvemi proving the weaker, it was they 
who, together with the Sequani, invited Ariovistus and his 
Germans to come into Gaul. 

The presumption is that both Sequana and Sequani are 
words of Celtic origin, but with regard to the name of 
the Aquitani we are left in doubt whether it was Celtic 
at all or not ; if the word is Celtic, it may be supposed 
to inTolve a Celtican word for water, of the same origin 
as the Latin aqua, and that the first bearers of the name 
AquUani were in that case described as a people dwelling 
near the water, meaning the sea, and this would agree well 
enough with the little that is known of the history of the 
word. Strabo ^ maintains, and probably with perfect justice, 
that the people beyond the Garonne, whom Caesar called 
Aqoitani, were more Iberian than Celtic. From them the 
province of Aquitania, constituted by Augustus so as to 
extend to the Loire, took its name. Pliny makes the Iberian 
and original Aquitania into Aquitanica, adding, in the passage 
already cited, that it was at one time called Aremorica. So 
it is possible that we have here to do with a term Aquitan 
or Aquitanic, admitting of being rendered into Gaulish by 
the adjective Aremoric, which meant * maritime, belonging to 
the sea-coast.' The former may in that case have been 
derived from a Celtican word of the origin already sug- 
gested. Thus Ph'ny's Armoric Aquitanica coincides with 
the information which led Caesar to confine the name 
-Aquitani to the Iberians on the further side of the Garonne. 
Pliny traces the name there to a single tribe bearing no 
other designation than that of Aquitani : their exact position 
IS not given, so that one is at liberty to suppose that they 
dwelt on the coast somewhere between the Garonne and the 
-^dour. Though originally synonymous, the later usage had 
the effect of severing the terms Aremonc and Aquitanic. 
Armoric or Armoric became associated with the Armoric 
I^gue, and shifted with the shifting fortunes of the states 
oonstitating it, which Caesar describes as civitates qum 

^ See Meineke*B ed. It, 1. i. 
Pliil TnuiiL 1891-2-8. 8 



114 CELTS AND OTHER ARYANS. — J. RHYS. 

ArmoriccB appellaniur,^ On the other hand, the other 
adjective was associated by Augustus with the province of 
Aquitania, including the spacious region from the Garonne 
to the Loire ; and it is there that the name Aquitania has 
ever since had its home, and has come down to later times 
in the French form of La Ouienne. 

Here must be mentioned the term Ckortonicum, which 
occurs, among other geographical names, in the Wessobrunn 
Codex in Munich, a manuscript written before the year 
814. The names of Celtic interest in it are the following : 
Hyhemia . scottono lant, i.e. the Land of the Scots. Oallia . 
uualho lant, i.e. the Land of the Welsh. Chortonicum . auh 
ttualho iant, i.e. also the Land of the Welsh. Equitania . 
uuascono lant, i.e. the Land of the Vascones or Gascons. Dom- 
nontam . prettono lant, i.e. the Land of the Britons. But the 
name of prime importance here is Chortonicum, a spelling 
which may be regarded as standing for Cortonicum in the 
text which the writer of the Old High German glosses had 
before him, and Pott with his usual insight perceived that 
it was to be interpreted by means of the words Cruithne, 
Cruithneach, *a Pict.'^ Cortonicum would thus have to be 

^ Blu. T. 43, vii. 75. In both parages the spelling is Armorieas, nor does 
Holder mention any variant like Aremoricae, which, etymologically speaking, 
would be the older and better form of the word. Fhrof. Kettleship reminds me 
that the prevalent form in the poets is scanned AremorieM. 

' £t^. Forsch. II. ii. 899 : see also Windisch in Ersch & Gniber*s Ency- 
clopeedia, s.v. Keltitehe tpraehen, I copy the glosses from Graff's Diutiska, vol. 
ii. pp. 370-1, in the hope chiefly that some one will inform me to what text 
they belong : it is needless to point out the curious questions which some of them 
suggest : — 

Fol. 6 la. Heo nomina de variis prouinciis. 

Hybemia . scottono lant. 
„ 616. Gallia . uualho lant. 

chortonicum . auh uualho lant. 
Equitania . uuascono lant. 
Uacea . uuascun. 
Germania . franchono lant. 
Italia . lancparto lant 
Ausonia . auh lancparto lant. 
Donmoniam . prettono lant. 
Bruteri . prezzun. 
Araues . sarci. 
Ispania . benauentono lant. 
Cyuuari . suapa. 

Pannonia . sic nominatur ilia t^rra meridip danobia . et uuandoli 
habent hoc. 



P AND Q GROUPS. 115 

explained as an adjective formed^ probably in Latin, from 
the name of a people called in Celtic QurUiones or QurHtoniif 
of the same origin not only as Cruithne ' a Pict/ but also 
as the Welsh Piydj/n, the part of Britain inhabited by the 
Picts of the North, and Welsh Prydain, the name of the 
whole island. So Corionicum may have meant all Gaul, or a 
part of Oaul, or else it may very probably have been used in 
both senses ; and to discover approximately the part of Gaul 
it could have specially referred to we cauDot do better than 
take the synonymity of Cruithne and Pict in the British Isles 
as our guide on the Continent. This clearly suggests that 
the Cortonic district par excellence covered the whole or 
part of Poitou or the land of the Pictones, whose name, 
together with that of their town now called Poitiers, 
cannot, etymologically speaking, be severed from the name 
of the insular Picts. Thus it may be presumed that the term 
Cortonicum referred to the region so called, whether all 
Gaol or only a part of Gaul, as the land of the non-Aryan 
aborigines ; and it goes to show that the early Celts regarded 
the latter as one and the same race, whether on the Continent> 
or in the British Isles. That is too large a subject, however, 

[Note oontmiied from previous page.] 

FoL 62a. Arnoricas . peigiro lant. 

Istri^ . peigira . later . danobia 

Sclaaus et auarus . honi et nuinida. 

Palestina . indeono laat . hoc est circa hieroflolima. 

Unandali . huni . et citta . aut nuandoli . 

Auriliana . sic nominatur ilia terra abi roma stetit. 

Pentapoli . sic nomina^r illia patria . ubi rapana stat. 

Tharcia . ilia patria . nbi constantinopoli . Btetit. 

Cjnocephali . canini capita. 

Amazones . hoc .8. uir^ines. 
,y 62b. Thebaida . ilia patria mde fait manriciiu . Argi . g^reci. 

Ethiopia . patria mauri . 
De Ciuitatibiu. 

Lnetona . Liutona. 

Argentoratensis . strazpurao. 

Nimitensis ciuitas . spira. 

Unangiaoninm . ciuitas auormacip. 

Agrippina . cholonne . 

Constantinopoli . costantiniises punic. 

Neapolis . ciuitas noua. 

Nonca . reeanespuruc. 

Allofia . raaa8p6nsa. 

Betfagia . pazauua. 
„ 63«. Ualuiottla salzpuruc. 



116 CELTS AND OTHER ARYANS. — J. RHYS. 

to be discussed here ; let it suffice for the present to have 
called attention to the form of the word Gortonicum^ as it 
proves to be derived from a Celtican source, not from 
a Gaulish one. Its history, nevertheless, is incomplete ; 
for it is not known in what author it was found by the 
writer of the glosses in the ninth century ; so one cannot 
say whether the text he re£id is still extant, or on the 
other hand whether it did not carry him back a con- 
siderable distance of time towards, let us say, the fourth 
century. I say the fourth century, as we have evidence 
in the dialogues of Sulpicius Severus that not only Gaulish 
but Celtican likewise was in use in his time : his words are, 
Tu verOy inquit Postumianus, vel Celt ice aut, si mauis, Oallice 
loquerei dummodo Marti num loquaris} The words which 
Sulpicius puts into the mouth of Postumianus deserve all the 
more attention as Sulpicius is said to have belonged to 
an influential family in Aquitania, where Celtican speech 
may not unnaturally be supposed to have been last heard. 

The word Prydain, which is the Welsh name for Great 
Britain, is not etymologically related to the latter name 
from the Latin Britannia^ which was formed from the Latin 
name of the people whom the Romans at first called Britanni. 
For after the Romans conquered a part of Britain and became 
better acquainted with its peoples the term Britanni seems to 
have gradually fallen into disuse in favour of the kindred form 
Brittones, which is also the pronunciation of the name to 
which the Bry thonic Celts bear testimony by their use of such 
forms as Welsh Brython * a Briton or Welshman,' Brythoneg 
* the Bry thonic language, Welsh,* Old Cornish Brefhonec 
*the Brythonic language, Cornish,* Breton Brezonek 'the 
Breton language.* Where then did the Romans learn their 
first name for the Brythons? Probably in the Roman 
province in the South of Gaul. At any rate it is remarkable 
that the old Irish word for Brythons, which had also to do 
duty for Britain, was the plural Bretain, genitive Bretan, 
dative Bretnaib, accusative Bretnu, which corresponds to the 

> Halm*B ed. Dial. i. 27, 4. 



P AND Q GROUPS. 117 

Latin Britanni, or, more exactly, Brittani, Hence it would 
appear that the Romans must have learned the name 
from a people who spoke a language resembling Goidelic, 
that is to say from the Celticans of Southern Gaul or the 
Boman province. This is corroborated by the fact that 
Britanni and Brittanni resemble the Greek form Bperravoi, 
which reached Greece probably through the Greeks of Mar- 
seilles, who doubtless learned the name in the same district 
as the Romans : perhaps it would be more accurate to regard 
the Britanni of the Romans merely as the Latin rendering 
of the Greek Bperravol, and to suppose the Greeks to have 
learnt the latter from the Celts nearest to Marseilles. At all 
events it was not the form which the Brythons themselves 
gave to their own national name. It probably represented 
nther the pronunciation which the Q Celts of the Continent 
gave it, as it must have done, with tolerable accuracy, that 
of the Q Celts of Ireland, judging from the forms which 
the name exhibits among them in later times. 

Such are the traces on the Continent of a Celtic language 
resembling Irish as contrasted with Welsh, and though they 
seem to me to render its former existence there highly 
probable, I leave others to settle how far it is probable. 



II. 

The question of classifying the Celts into two groups 

characterized by the use of P and Q languages respectively 

IS not to be dismissed without reference to a similar grouping 

<>f the Aryans of Italy and Greece. Thus to begin with the 

former, the Romans used qn just as the ancient Irish did, 

bat in so doing they stood well-nigh alone in the Italy of 

historical times; for the Osco-XJmbrian dialects replaced 
qu hjpi 

J^H6i8a|»iaii and poeiiblj other dialects of the south are to be regarded as more 
™ to Greek than to the other languages of Italy : see Mommsen's Unter- 
^f^iuckm JHtUtkit, p. S5. 



118 CELTS AND OTHER ARYANS. — J. RHYS. 

The Latin-speaking Italians of the beginning of Roman 
history occupied a comparatively small area, their original 
territory in the peninsula having probably been narrowed ^ 
by Umbrians and Sabines, by the latter of whom the legends 
of Rome represent the nascent city hard pressed. On the 
other hand the scanty remains of some of the dialects of 
ancient Sicily are supposed to show close similarity to Latin, 
and this suggests that Latin or kindred dialects were once 
spoken over the whole of the eastern coast of the Peninsula, 
from Latium to the Straits of Messina. Their disappearance 
from all that tract was probably due to the conquests made 
within historical times by the Samnites, and other tribes 
of Oscan stock, and in part also to the Greek colonies 
planted on the coast. In the rear of the Oscans came the 
Umbrians, occupying not only the district which bore their 
name in Roman history, from the Adriatic to the Apennines 
and the Tiber, but also the adjacent country westwards 
to the sea opposite Corsica. In other words they owned, 
if not the whole of what came later to be Etruria, the 
south at any rate and the east of Etruria, so that they 
completely barred the north of Italy from sea to sea 
behind the peoples of the Siculo-Latin stock. I cannot 
discuss here the extent to which the latter had been deprived 
of their territory by the Osco-XJmbrians, or the fact of these 
last being in their turn deprived of so much of their territory 
by the Etruscans, that Pliny ^ found the Etruscans to have 
taken possession of no less than 300 towns of the Umbrians. 
The one thing to be borne in mind for the present is, 

^ The remains of the Faliscan dialect show clearly that the people of Falerii, 
for instance, belonged to the same stock as the Latins and not to the Umbrians 
or Oscans : see Deecke's * Falisker* (Strassburg, 1888), where, pp. 135, 166, 
193, he ^ves as Faliscan ^que or -cu«s Latin -^Mf, euando ^^hutm quando, and 
other similar eyidence which cannot be OTorthrown by the occurrence in a Faliscan 
inscription of Fuponio equated by him with Fomponiutt at the same time that 
he suggests equating Popia with rublia : see pp. 153, 187. 

' See iii. il2 : the passage is highly instructiye, especially the following 
words — *' Ab Ancona Gallica ora incipit togatae Galliae cognomine. Siculi et 
Liburni plurima eius tractus tenuere, in primis Palmenseni, Praetutianum 
Hadrianumque agrum. Umbri eos expnlere, hoe Etruria, banc Galli. Urn- 
brorum gens antiquissima Italiae existimatur, ut qnos Ombrioe a Oraeds pntent 
dictos, quod inundatione terrarum imbribos supeifuissent. Trecenta eorom 
oppida Tusci debellasse reperiuntur." 



P AND Q GROUPS. 119 

that the Italians of the Q group appear to have entered 
Italy before those of the P group, that the Siculo-Latin 
race had already settled down when the Osco-XJmbrians 
arrived. 

Mach the same kind of remarks may be applied to 
Greece, where the Q group is most obviously represented 
hy the dialect of Herodotus,^ with such forms as kw and 
WT€ for the iri^ and ttc^tc of the other dialects. The 
historian was a native of Halicarnassus in Asia Minor, and 
his Oreek belongs to the group of dialects called Ionic. 
As there is no reason to suppose that he invented the 
language of his writings, he must be supposed to have 
heard it at Halicarnassus or in some part of Asia Minor, 
or else, as has been sometimes thought more probable, in 
the island of Samos, where he is said to have passed some 
years of his life. Now Halicarnassus is spoken of as a 
Doric colony, namely, of Troezen ; but this by no means 
proves that it was not also Ionic. That it was so is seen 
from the fact that the Ionian deity Poseidon was worshipped 
At Halicarnassus with peculiar zeal,^ and the connexion 
itself of Halicarnassus with Troezen is by no means in- 
consistent with this view, but rather the contrary; for 
Troezen was a mixed city. Originally Ionic, it had been 
conquered by Dorians. The name Ionian is applied early 
to the inhabitants of the north-east of the Peloponnese, 
including Troezen and other places; also to the district on 
the north coast, subsequently known as Achaia. In what 
iiumerical proportions the Dorians and the lonians lived in 
the towns conquered by the former, it is impossible to say ; 
^ut in Sicyon, for instance, the old inhabitants formed a fourth 
Mhe, whereas at Corinth they consisted of five tribes, while 

^Herodotiu is selected as the representatiye of the dialect, for one learns 
^ tiie editions of such antiiors as Hippocrates and Heraclitus that the 
j^^Bveripts narrow the domain of the k as compared with the usage of the 
''J'torian. It was donhtless very natural for Greek scrihes and editors to normalize 
ue £alect hj giving its forms the w with which they were familiar in Attic 
P|[^; but the longer the writings of an author were, the more resistance 
wey offered to this process of assimuation. 

' Bvelyn Ahhott^s Greece, vol. 1. 12d ; Miiller's Dorier, L 407» ii. 105 ; and 
u« Berlin Corpus of Greek Inscriptions, No. 2655. 



120 CELTS AND OTHER ARTAKS. — J. RHYS. 

the conqueriDg Dorians only made up the three remaining 
tribes. The conquest of the Feloponnese known as the 
return of the HeraclidaQ was efiFected by Dorians helped 
by -^tolians, and the former can be traced back to a district 
between Oeta and Parnassus, where their name survived in 
Doris : at a still earlier period they dwelt in the Thessalian 
district of Hestiaeotis.^ When the Dorians and their allies 
conquered the Feloponnese, the ancient lonians and kindred 
peoples were subjugated or driven into corners, such as the 
north-east of the peninsula, together with Attica, also 
Euboea, the other islands and Asia Minor, whither the 
Dorians in due time followed them. Eventually ensued 
a Vision of races, which it would be useless to attempt 
to analyse. But it will suffice for the present to bear in 
mind that the Dorians and kindred inv£iders found the 
lonians in the Feloponnese before them, and that the Ionian 
dialects contained among their number one, at least, which 
belonged to the Q group. 

The principal questions which these remarks raise may 
be comprehended under two headings, the fusion of F and 
Q Aryans and the common descent of the F Aryans. Let 
us take the latter first : is the modification of qu into p 
enough to prove the nations speaking languages of the 
P group to have spread from a single centre, whether they 
are found in Celtic lands, in Italy, or in Greece P As the 
physiological change here in question might, so far as one 
can see, take place in any language having words with qu, 
that change would be a precarious foundation for such a 
conclusion.^ But it makes a very great difference to find 
that the change has been resisted in the outer area alone, 
that is, among the Goidelo-Celtican group in Celtic lands, 

^ Abbott, vol. i. 61 ; Herodotus i. 66, Tiii. 43. 

' Most of tboee wbo notice this, speak as if they consider the chanee of gu into 
p a very common phenomenon, whereas, besides the case of the Aryan languages in 
question, I have not succeeded in eliciting an instance from ihends acquainted 
with non-Aryan languages. One is generally asked to take Roumanian as in 
point ; but that will not do as an independent instance. For who is to say 
timt the legionary ancestor of the modem Roumanian who says pairu for the 
Latin quatuor, was not an Oscan or Umbrian, or 6?en a Gaul, inheriting tiie p 
pronunciation P 



P AND Q GROUPS. 121 

the Siculo-Latins in Italy and some of the lonians among the 
Greeks. In fact we have only to make a simple calculation 
of chances, which may be put thus: let the three P groups 
of the inner area be represented by three white balls, and the 
three Q groups of the outer area by three black balls ; then 
suppose the six balls placed in a bag, and the question is, what 
the chances are, in drawing three at a time, that those three 
will be the white balls. Now the number of different triplets 

that can be formed of six balls is J^^=20 : that is to say 

the chance of drawing any given triplet is -j^* or> i^ other 
words, the odds against it are 19 to 1. Suppose we leave 
out Greece on account of the evidence being less clear there 
than in the case of Italy and Celtic lands, our calculation 
is then made with four balls instead of six, and the number 

of pairs possible is |^=6: that is, the chance of drawing 
any given pair is ^ ; in other words the odds against it 
we 5 to 1, which is a preponderance of probability by no 
means to be despised, and from which one might proceed 
to assume the same state of things in Qreece as in the other 
two linguistic regions. It thus appears unreasonable to 
suppose the change of gu into p among the Aryans of the 

• 

inner area to have been a mere matter of accident, and 
the alternative explanation which alone deserves considera- 
tion is, that the P Aryans issued from a common centre 
where the phonetic change in question had been accom- 
plished once for all in their common dialect. 

Here may be added the fact that the consonantal change 
w not the only characteristic of the P languages, as they 
show another change, namely that of long u into long ii or 
even 1; but as this will come under notice later, let it suffice 
at this point to remark that it vastly increases the strength 
of the argument.^ 



\yoi instead of limply drawing the three white balls once, the chance of 
vbieh taking place was shown to be only ^, one has to suppose them (replaced 
f^) ioccessfmly drawn a second time. The chance of the compound event 
Ml\rX|Ar=iiv : that is, the odds against it are 399 to 1 . If Greece be omitted 
weebsQ^ of the compound event is ixi=^, or 35 to 1 against it so far as 
^^'^tma the Celtic ana Italian areas. 



122 CELTS AND OTHER ARYANS. — J. RHTS. 

Irish, Latin and the Greek dialect in point must so far as 
concerns the consonantal change which interests us here be 
regarded as having remained in a sense on the level of 
Teutonic, as a single instance will suffice to show : take, for 
example, the English interrogative whOf Gothic hvas^ and 
compare with it the initial of the Irish da ^who,' Latin 
qui and quia, and the Herodotean Koio^y contrasting with 
them the Welsh pwy^ Oscan pis * who,* and the common 
Greek iroio^. As no philologist supposes the change to have 
been from p to qu, Irish, Latin and Herodotean Greek must 
be regarded as having in this respect remained on the ancient 
level, the reduction of qn to p being a later phenomenon. A 
similar remark applies to the vowel change to which allusion 
has been just made. So the two changes agree in signalizing 
for us the rise of a new Aryan dialect, and this translated into 
ethnology can hardly have any other meaning than that 
Aryan speech had been adopted by a non- Aryan race, which 
introduced into its adopted language the habits of pronun- 
ciation prevalent in its previous language. The question 
next comes, where the conquest of non- Aryans by Aryans 
took place, and looking at the distribution of the Aryans 
of this mixed origin as represented by the P dialects of 
Gaul and Britain, of Italy and Greece, also possibly by the 
languages of peoples of the Thraco-Phrygian stock both in 
Europe and Asia Minor, one cannot help thinking that it must 
have been somewhere in the region of the Alps. 

Granted this, we are enabled to explain a variety of the 
features which characterize Celtic, Italian and Hellenic dia- 
lects : in other words, we now come to the other question to be 
noticed, that of the fusion of the later or P Aryans with the 
earlier or Q Aryans. But here, as before, I only attempt 
to deal with the races indirectly, through the linguistic 
data which they have left us, and Latin proves of prime 
importance to the inquiry. The history of that language is 
a very remarkable one, for, after having its domain narrowed 
to a comparatively small area, it begins to conquer all the 
dialects around it, nor desists till it becomes one of the great 
languages of civilization; but what is particularly to be 



P AKD Q GROUPS. 123 

borne in rnind is the fact, that the antagonism between the 
makers of Rome and the peoples around them seems to have 
been so intense that Latin remained comparatively unaffected 
by non-Latin dialects up to the time of its classical period. 
Thus not only does Latin agree with the rest of the languages 
characterized by qu in the matter of that combination, but it 
farther abstains from the corresponding change of gu to b, 
which happens in Irish and Qreek. It retains gu or reduces 
it to f : take, for example, the word ungtw *I smear or anoint/ 
unguen *fat, ointment/ Oscan umen (for *umben), Irish imb, 
genitive tm6^ (like aintn 'name/ genitive anme), Breton amann, 
Welsh yw^yn 'butter.' Take also the Latin vivua 'living* (for 
Vlf«w«), Oscan bivua " vivi/* Irish 6^0, Welsh bi/w 'quick, alive.' 
A remarkable exception in Latin is bos ' an ox,' which, had 
it been native Latin, ought to have had some such form as 
^f and not boa; but the very fact of the latter standing 
pretty nearly alone speaks clearly for the comparative freedom 
of Latin from any considerable admixture of Oscan.^ Another 
J^wpect in which Latin remained pretty nearly on the old level 
u that of long u, which in the P languages shows a narrow- 
Uig towards French u or even i. Thus while Latin had ana 
'a sow,' Greek had its S?, pronounced with a vowel like the 
French u of the present day; and the accusative singular 
<md plural of this same word were in XJmbrian aim and ttif 
J^spectively. TJmbrian had also pir and /r if, corresponding 
to the Greek mp ' fire,' and Latin /rti^^« ' crops.* Similarly 
I^tin tU, English ihou, appear in Irish as ta, but in Welsh as 
^h and Irish cQ ' a hound,' is in Welsh ci, not to mention 
that such a word as Irish ttmth, in Gaulish touta ' a tribe or 
people,' is tiid in Breton, and in Welsh ttit and tOd, with a u 
which in North Wales is much narrower than French fi, 
while in the rest of the Principality it becomes mostly i. 
■Hiis maybe regarded as a Brythonic extension of the domain 
oi the tendency to narrow or unround the vowel. 

The comparative freedom of Latin from the changes her6 
^ question suggests the conclusion that they were phonetic 

^ If iot be oot Otcan or Umbrian, it may be of Gaulisb orig:in, or else a loan- 
void Irofli Qreek, which tome are inclined to maintain. 



124 CELTS AND OTHER ARYANS. — J. RHTS. 

characteristics of the P dialects, that is to say, all three 
of them, the labializing of qu into p^ of gu into 6,^ and the 
modifying of u towards !.' Granted this, it follows that one 
must suppose the linguistic influence of the later Aryan 
comers to have varied considerably in force as measured by 
the changes it effected in the phonetics of the Q dialects of 
the Celts, the Italians and the Greeks respectively. In Itcdy, 
as already stated, it may be said to have been a minimum 
quantity, but it was quite otherwise among the Celts and 
the Greeks. Let us begin with the latter: the majority 
of the Greek dialects are known to have used ir for the 
qii of the labalizing Aryans, but they did not treat that 
combination always in the same way. Before the vowel o 
they made it into tt as in iroio^; ' qualis,' but before € and t, 
it became t, as in -re and t/9, with which are to be 
contrasted Latin -que and quis, while before or after v the 
guttural remained, as in kvkKo^ ' a circle,' A. -Saxon htceohl, 
English tvheel. Under similar conditions gu became /8, S 
and 7 ; also ghu became ^, and p^ respectively. No such 
limitations of the phenomenon of the guttural plus the u 
yielding a labial are known to have systematically existed 
in the P dialects of either Italy or the Celtic world,^ and 
certain Greek dialects seem to have somewhat defied them, as 

^ The change of ohu into/ or ^ has been omitted, partly as being of minor 
importance here, and partly as contained, so far as cbncems Celtic, under that 
of ffu into b. 

* My friend Mr. Wharton, in the paper to which reference has already been 
made, gives it as his opinion, p. 64, that *' the influence of dialect on the Koman 
long vowels, and especially on the diphthongs, was much stronger than on the 
short vowels, as conversely that of accent was much weaker ; ** but from among 
his instances of u becoming f, I should be inclined to remove such as divinus, 
amicus, formido, vopiscus, as contrasted with opportiinus, cadiicus, testftdo, 
coruscus, and the like ; for the use of a termination 'nii«, for example, rather 
than unu8f raises questions as to the thematic vowel, and it is open to the suspicion 
of being due to analogy rather than to a mere phonetic change. Among 
Mr. Wharton's other instances may possibly be some which were due to the 
influence of Oscan and Umbrian. 

> But Prof. Zimmer in his Keltische Studien in Euhn's Zeitschrift, vol. zxx. 
pp. 134, 140, regards Irish guth * voice* and Greek jSo^ as of the same origin, 
and treats fftgon * I have wounded,' as the perfect of a verb which in its other 
tenses has b, such as bmim * I strike.' So here we have puo becoming gu or go^ 
and some such a rule must have applied in the case of Sequana, supposing it to 
be— against the probabilities of the case — Gaulish and not Celtican. Add the 
form SEQVONIVS, which, if correctly read, is deserving of notice in this con- 
nexion : see the Hevu$ Ctltique, voL ilL p. 307. 



P AND Q GROUPS. 125 

when the Thessalians said /ctv for rk 'who/ Then as to 
the dialects characterized by the unrounding of long u into 
long tt, similar treatment is found to have been extended 
to short u, which was changed in the same dialects into ii ; 
and this still remains the sound of v in certain dialects 
of Greek, such as the Tzakonian. But the ancient dialects 
retaining u and u are said to have been Bceotian and Laco- 
Bian, probably also Arcadian, Cypriote, Pamphylian, Chal- 
cidian, and Lesbian ; ^ and such instances may be mentioned 
as the Bceotian name Evdovfio^ for the Attic EvBv/xo^, 
and the Laconian verb fiovalBSei for what would have been 
in Attic fivOifyt in the sense of X/iXet ' speaks.' On the 
whole the P dialects may be said to have had the upper 
hand in the Greek world, and the fusion, not to say the 
oonfnsion, of the dialects of the two groups must here be 
regarded as of such a nature that Greece cannot help us 
much in the attempt to distinguish between peoples of the 
P and Q groups : rather have we to assume that the 
oonclosions already drawn as to the corresponding peoples in 
Celtic and Italian lands may be applied in principle to 
Greece likewise. 

On Celtic ground we have no data to enable us to judge 
^th any precision of the attitude of Gaulish and Celtican 
towards one another on the Continent, but we seem to have 
a distinct trace of the influence of the former in the French 
pronunciation of Latin long u^ a pronunciation which not 
only characterizes the French language, but is found also 
in modem Proven9al, in the Engadine, and in Lombardy.^ 
Irving the Continent, we have in these islands the facts 
of Brythonic and Goidelic pronunciations to draw upon. 
In the first place it may be mentioned that the modifying 
of ii towards u or I is unknown in the Goidelic dialects, 
u is also the reduction of qnto p\ but, on the other hand, 
^ for gn and ghn is as general a rule in them as in the 
brythonic dialects. How then is this to be explained? 
To some extent, probably, by the geography of the British 

^ Brugmann, §{ 48, 56. 

* Dim, Oram, tkr romaniaehen Spraehm (Bonn, 1870), yoL i. 426. 



126 CELTS AND OTHER ARYANS. — ^J. RHYS. 

Isles ; for the case would be met by supposing the Q Aryans 
to have first conquered Britain as far, sooner or later, as the 
sea separating it from the sister island, and the P Aryans 
to have eventually arrived in such force as to establish 
themselves in the east and south-east of Britain, preserving 
their own language at first unmixed, while influencing 
that of the Q Aryans to the west of them more and more 
as time went on, and reducing their territory. In fact 
we may suppose the same thing to have taken place 
here as in Gaul, namely, that the P conquerors everywhere 
made themselves the ruling class, even where the Q dialects 
continued to be spoken. It was the speakers of these 
mixed dialects, that is to say, of Q dialects modified under 
the influence of those of the P group, that probably sent 
forth men to conquer Ireland; they set sail, let us say, 
from Anglesey and landed on the opposite coast, somewhere 
between Bray and Dundalk. In Ireland the Q language 
of the invaders became stereotyped against the further 
influence of the idioms of the P group, while the Q dialects 
left behind in Britain continued subject to that influence 
until they were assimilated out of all separate existence. 

It is needless to say that while Brythonic made this sort 
of conquest, it could not itself escape being phonetically 
modified, namely, by the previous habits of pronunciation 
of those who adopted it instead of their own language. Add 
to this the subsequent influence of the extensive conquests 
made in Wales and Dumnonia by invaders from Ireland 
about the end of the Koman Occupation. 

Such at any rate would be a possible account of the rise 
of Goidelic as it was when it spread from Ireland to 
Man and Scotland; but with regard to the influence of 
the P dialects on it, only one instance has been specified, 
namely, the change of gu into h. This must, however, not 
be regarded as standing alone : there is probably one other 
which must be supposed to date quite as early : it belongs 
to the history of the treatment of Aryan p in Celtic, which 
must by no means be confounded with the treatment of 
Aryan k or qu whether changed into p or not Thus it has 



P AND Q GROUPS. 127 

been for some time known to philologists that Aryan p 
wholly disappears in the Celtic tongues, that is, it has 
either given rise to another consonant or it has simply 
left; its place empty. Take for example the Irish word Idr, 
Welsh liaivr, which means floor and is of the same origin, 
or the Old Irish word aihir, which means father and is in 
feet the same word with/ather and Latin pater, with their con- 
geners; similarly the Irish verb lenim *I adhere, follow,* 
rednpHcate perfect (3rd singular) ///, (3rd plural) leltar, future 
^lit 'sequentur,' Welsh di/^n, can/yn *to follow,' and erlj/n 
'to prosecute,' are of the same origin as the Lithuanian 
ft'wptt * I adhere,' preterite lipau, Old Bulgarian prilina (for 
prilipna) ' I cleave to or remain hanging.' ^ This consonant 
has disappeared even in Gaulish, as may be seen from such 
words as Aremoricus, where the prefix are, Welsh ar *on, 
open/ is of the same origin as the Ghreek irapa and the 
whole adjective may be rendered irapaJBaXdaaiOf; ; take also 
the name of the Belgic people of the Remi, who have left 
It to the town of Rheims, the ancient Durocorforum Remorum, 
^tere Remi is of the same origin as the Welsh rhxcyf * a 
king or ruler,' Latin primus : compare English first and 
German Jiirst * a prince,' and the Latin princeps, whence, 
through French, the English word prince. In other Celtic 
^ords the original p is represented by a guttural, as in the 
Irish iecht * seven,' Welsh saith, seith for *8echt, or the Irish 
Word uasal * noble, high-bom,' Welsh uchel * high,' which was 
in Gaulish ucsef- as in UxellodxinTim meaning High Town. 
This adjective is parallel to the Welsh isel *low,* and the 
^0 are to be traced back to the prepositions which have 
yielded the Welsh yn (Irish i n-, i) ' in, into ' and *up,^ 
rf the same origin as the Greek word ir^Xo^ * high ' and 
English aver, German Hber, Sanskrit upari, ' over.' In all 

, Bm^Buumn, L f 3S. 
This etjmologj was niggested tome years ago in my Otitic Britain? p. 310, 
^ in older etymology is accepted, among others, by Bmgmaxm, who, in his 
fl^T^rin der pergl. Orammatik, vol. L § 434, connects Uxello- with Greek oi^^v 

^ ^i^crette/ and lith. dtJcnttu * high.* If this were correct, one wonld rather 
?^ the word to hmre beinm in Gaulish with ou,iu, or mu, which howerer it 



128 CELTS AND OTHER ARYANS. — J. RHTS. 

these instances it would suffice, phonologically speaking, to 
regard p as having in the first instance become h (in the Celtic 
dialects with p for qu)y just as the same j? is known to have 
done in Armenian in such words as hair ' father/ the etymo- 
logical equivalent of that word and of the Irish athir, Latin 
pater, etc. S before vowels would disappear on the way 
down to the Neoceltic tongues, while before t it may be 
supposed to have helped to produce the favourite combination 
X^f retained in Irish, but modified in Welsh to Uh, as in 
Old Irish secht * seven,' Old Welsh aeith, now mith. Before « 
the analogous combination %a was liable to undergo the 
changes illustrated by Irish uasal and Welsh tichel already 
cited.* 

The theory here propounded leaves the Goidels nearly 
related to the ancient Latins in harmony with the striking 
similarities between the Irish and Latin languages, at the 
same time that it represents the same Goidels as inseparable 
from the Brythons by reason of manifold mixing. So we 
should still be justified in speaking of these two peoples as 
Celtic, and not merely as Goidelic and Brythonic, or as 
groups only distantly related to one another, which is the 
utmost one could have said of them when they first came 
in contact with one another. The same kind of remark 
applies to the two sets of Aryans in Italy and in Greece. 
But who were these races distinguished as those speaking 
Q and P languages respectively P The former, suffice it for 
the present to repeat, were of the same Aryan stock as the 
Teutons, Slavs, Albanians, Armenians, Iranians and Hindus, 
that is to say in so far as those nations were not merely non- 

* The case of initial sp making in Welsh/ (written/") and in Irish t has heen 
omitted here as probahly representing an original Aryan sp-h or sp aspirated : 
compare Welsh ffer * the ankle ' with the Greek a^vp6y of the same meaning, 
and Welsh ffon * a staff/ ffon dafl * a sling,' Irish aonn, with the Greek a'ptyB6yii 
* a sling.' In such Irish words as ionn the $ has probably replaced a preTious 
/*, as is the case in such borrowed words as Irish man = Latin /r^num, and suistss 
Latin fastis. Some more instances of initial / in Welsh will be found brought 
together by me in the Rev. Celt, vol. iL pp. 335-7. Sp-h or sp* became ^ or/ in both 
Brythonic and Goidelic, and in Brythonic the/ has remained while it has become 
s in Goidelic. In any case I see no room for the «v, which Dr. Whitley Stokes 
suggests in his CsUie Declension^ p. 26. 



P AND Q GROUPS* 129f 

Aryans Aryanized by Aryan conquest. In Italy they appear 
to have been not only the Latins of history, but the race 
niore widely represented by an early civilization, of which 
traces occur from Etruria to Sicily, as shown by recent 
archcDological research. In Greece they were possibly the 
mysterious people called the Pelasgi with whom Herodotus^ 
identified the lonians. But the Pelasgi, though Aryans, 
were not exactly Greeks, however readily they became 
Greeks in the sense of losing their national identity in the 
ganglion of peoples which went to constitute the Hellenic 
world of history. 

In answer to the question who the P Aryans were, one 
may first say that they appear as a second stratum, so to say, 
covering a part of the area previously occupied by a Q stratum, 
namely, from the North 8ea as far perhaps as Asia Minor, 
^th an overflow into Britain, Spain, Italy and Greece. So 
their area of dispersion appears to have been included within 
the other area. Further, since the language of the P Aryans 
lA to be regarded as a modified form of the older Aryan 
speech, it may be asked to what cause the modification is to be 
traced. One might of course answer that all languages are in 
a permanent state of change, more or less rapid,^ so long as 
they are living; but this would hardly be a sufficient answer in 
the case of changes so definite as those here in question. We 
have accordingly no alternative but to suppose, as already 
^ggested, that the dialects of the P group arose from the 
coming of speakers of the older Aryan dialects, namely, those 
of the Q group, in contact with a non- Aryan race, which, 
conquered or otherwise powerfully influenced by them, 
•<lopted Aryan speech without having got rid of the non- 
^an characteristics of its inherited pronunciation. This 
^apposition of a very considerable absorption of non- Aryan 
elements makes the P Aryans a mixed people talking what 
°^^ght be termed Neo-aryan. .This view which derives counten- 
^^'^ in this country from the fact that archsDologists find the 

Book, L 56. In the chapter following; he sormises the dialects of Creston 
ttd Placia to have been Pelasgian in his own day. 

m. Ituifl. 1891-2-8. 9 



130 CELTS AND OTH^R ARYANS. — J. RHYS. 

round-barrow Brython of the Bronze age to have been pre- 
dominantly a broad-skulled man, though it is believed that 
the original Aryan was long in this respect rather than broad 
or round. Perhaps one might venture to regard the lake- 
dwellings of Switzerland as the homes of the mixed population 
of the Neo-aryans. At any rate, we have an instance of the 
possibilities of the Alpine region so late as the time of Geesar, 
when the Helvetii set out from their country en masse to 
seek a home elsewhere ; and, but for the intervention of the 
legions of Home, they would doubtless have succeeded, as 
many similar migrations from the Alps had probably done 
before. It is interesting to note the destination of the 
Helvetii : they set out for the territory of the Santones in 
the West of Gaul : in other words, theirs was the march of 
Aryans of the P stock to conquer the territory of Aryans 
of the other stock, or else of a people perhaps more correctly 
described as not Aryan at all. 

Having endeavoured to show that the Aryans of the P 
stock emanated from a common centre, I have nothing to 
add except a word as to the wider classification suggested 
by Brugmann's treatment of the consonants of the palatal 
and velar series.^ Compare, for example, the Gothic hvas 
'who,' with its Lithuanian and Sanskrit equivalent has, 
which gives no indication of its ever having been kuas or 
quas : similarly in the case of velar g and gh, the eastern 
group of languages show, according to Brugmann, no trace 
of H as the mark of labialization.^ On the other hand, they 
agree in dififering in their treatment of palatal k, g, and 
gh from the languages of the western group, namely, in 
that they reduce those mutes mostly into spirants.' The 
Western Group consists of the following languages : 
Teutonic, Celtic, Italic and Hellenic, and the eastern 
group of Letto-Slavic, Albanian, Armenian, Iranian and 
Sanskrit. So our P languages take their place as a sub- 



» Grundrias, vol. i. §§ 380-466. 
2 Ibid, § 417. 
s Ibid, § 380, 



P AND Q GROUPS. 



131 



diTision of the Western Branchi whicli may be shown thus 
at a glance : — 



I. LiBiALiznro Lanouaobs. 

I Those ebaracterixed, some time or 
<***». ^ 9fft Wi 9^tff namely : 

1. Teutmtic, 

2. a. Goidelic, 
b, Latin. 

e. HerodoUan Oreeh, 

fi- Thoie with the jf combinatioiis 
ndneed tothe labials p, b^ ^, namely: 

1. Brythonie. 

2. OstO'Umbrian, 
8. SUmdard Grtek. 



II. ASSIBILATINO LaNOUAOES 

Lettoelayic. 

Albanian. 

Armenian. 

Iranian. 

Sanskrit. 



It is right to state that Brugmann, while distin- 
guishing between the Aryan languages which labialize 
uid those which do not labialize, hesitates ^ to draw the 
conclusion that the Aryan parent speech had split up into 
tvo dialects. In his discussion of the consonants, however, 
lie is obliged to divide the whole family into the two groups, 
which have been termed respectively Western and Eastern, 
ui spite of Albanian having somehow straggled out of the 
direction indicated by the geographical position of the other 
inembers of its group. It need scarcely be added that he 
deals with the Aryan languages and not with the ethnology 
of the nations by whom they have at any time been spoken. 

Since these conjectures of mine were written an important 
article has been published in the Nineteenth Century by Prof. 
Huxley on *' The Aryan Question and Prehistoric Man " ; 
^d I may say that my view is decidedly favoured by 
his conclusions, though I cannot now enter into details, 
as my paper is already longer than it was intended to be» 



1 Ibid, i 417, and note 1. 



132 



III.— NOTES ON ENGLISH ETYMOLOGY. By the 

Rev. Prof. Skeat, Litt.D. 

(Eead at a Muiifig ofth$ Fhilologieal Society <m Friday, April 18, 1890.) 

N.B. — ^A brief abstract of tbis paper appeared intbe AthetuBum^ April 26, 
1890 ; and another in the Academy, May 3, 1890. 

Aflkaimces. This difficult word, meaning 'as if,' occurs in 
Chaucer; see Kew £. Diet. In my belief, it is made up of 
E. as, and O.F. quameSf * as if,* a word given in Godefroy's O.F. 
Diet., with refereuces to Romania, xviii. 152, and Forster, Cligw, 
1. 4553 (note). Thus the sense is, literally, * as as if,' the £. as 
being tautologically prefixed. Gf . Lat. quasi. 

Bedene, forthwith, together, etc. The etymology is unknown. 
I suggest that it is from M.£. he, hi, by, and the pp. ^dene, repre- 
senting A.S. d&t, done ; from this there may, perhaps, have arisen 
the sense * by the time that it is done,' immediately on its com- 
pletion, or immediately afterwards. Cf. Mididone (below). The 
A.S. dhi occurs in ge-din; * synna aSr gedenra,' Cynewulfs 
Christ, 1266. The O.Northumbrian form is doen {z=.den); cf. 
'bi% gidoenj' gloss to 'agitur;' Durham Eitual, p. 113, 1. 20; 
cf . ffidoe tve, faciamus, id. p. 4, 1. 2, etc. ; ^at ^u ne gedoe, ne 
feeeris; S. Mark, x. 19, in the Lindisfame gloss. I am told that 
deen for * done ' occurs in M.£. ; the reference is probably to 
y-dee, i.e. done, in St. Editha, ed. Horstmann, 1. 290, where y-dee 
rimes with fle, Cf. deand, doing ; St. Cuthbert, 3227. We find 
d^n for ' done ' in O.Friesic ; see Richtofen, s.y. dtM. 

Cant. From Lat. cantum, ace. of cantus, a song. Formerly 
cantum, * I pray yow, telle me what was wreton vnder the mares 
f ote ? What was it, prose or ryme, metre or verse ? I wold fayn 
knowe it, I trowe it was cantum, for I herde you synge ; ' Caxton, 
Beynard the Fox, c. 27 ; ed. Arber, p. 63. 

Chete, Cheat, a thing. This is an old cant term for 'thing;' 
see Cheat ^ sense 3, in the New £. Diet.; and see Gloss, to 



»0TE8 ON ENGLISH KTYMOLOQY. — ^PROF. SKEAT. 133 

Barman's Caveat. "We find, indeed, tlie entry: *Jterum, ceatta' 
^ Wright- Wiilker, Voeah. 606, 28 ; but it is a general opinion 
that this is not to be relied upon, and may be due to some blunder. 
Perhaps, for seeatta, 

Cockroach. The Span, euearaeha also explains the variant form 
««^A (Nares). 

Compame, in Chaucer, C.T., A 3709, certainly ought to be 
f9» ha me, i.e. come kiss me. At least four MSS. have this 
'Siding. (It was subsequently discovered that this suggestion 
was made by Mr. Ellis in 1870. See his E. E. Pronun. p. 715, 
note.) 

Cowl, a tub. From A.S. eufi; not in the dictionaries, but the pi. 
^^ occurs in Birch's Cartul. Saxon, iii. 367. Cf. cyjku, in 
^glia, ix. 264, col. 1. M.E. euvel, O.F. cuvele is the same word ; 
^m Lat. eupella, dimin. of eupa* In Wright's Yocab. 677, 10, we 
find : * Ctipa, anglieey a cupe, or a cowle.' 

CrneiUe. Possibly a coiruption of crassipulum or eraBsihulum, a 
pot for holding grease ; cf. cresset. Observe the entry : ' Crassi- 
P^um, Crass^fularium, Crueiholum, angliee, a cresset;' Wrt. Yocab. 
576. 9. 

Cury, cookery ; as in M.E. ' form of cury.^ So also in * Petty 
Cory, Cambridge,' i.e. cook's quarter, or quarter for eating-houses, 
irom O.F. queurief cookery (Godefroy) ; from O.F. queu^ a cook ; 
^. ace. coewn, 

Kcker, Daykyr, a lot of ten hides. From O.F. daere (Qode&oy), 
Sesame; Low Lat. dacra, decora; from Lat. decur%a\ from decern. 
Cf* Swed. d&ekeTy * a dicker ; ' Widegren. 

Bnllor. This prov. E. word, signifymg * a moaning noise,' or as 
> verb, * to whimper, moan, with pain,' is spelt phonetically. The 
^ofd eciour shews that it stands for dolour, i.e. grief, anguish ; 
whence the verb. 

Feekless, useless, etc. In Lowland Scotch ; see Jamieson ; 
merely short for effectless, void of effect. * A fectlesse arrogant 
^nceit of their greatnesse and power ; ' K. James I., Basilikon 
^>ron, paragraph 17. 

Albert. I have shown that it should rather be philhert. The 
^B. form is actually philhert \ see Britton, ed. Nichols, i. 371, 
Wte5. 

Gtift, a slate-pencil (Essex). Formed, with added -t, from O.F. 
r^i a pencil (Godefroy); Low Lat. graph%um\ from Qreek 
V^iy. QLQ.griffel. 



13f KOTES ON ENGLISH ETYMOLOGY. — PROF. SKEAT.' 

TfilrKng ; of lindiscoyered etymology. But it is solved by noting 
that E. often turns (M.E.) $n into tn. Hence inkling is for mkling\ 
a substitution for enUin or mclin. From F. enelin, Cotgrave has: 
' Enelin, an inclination, dispositioni addiction, ... or humor unto.' 
Also ' EneUn, incliDed, bent, prone, given, addicted, or disposed 
unto.' An inkling of a truth is an inclination towards it. The 
substantival use is lost in French ; see enelin in Oodefroy. This, of 
course, is but a guess at an unsolved puzzle. 

Liuoious. M.E. lueiue ; as in ' with luciui drinkes ; ' Bobson, 
Three Met. Rom. p. 17; a yariant of lieiue, as in 'With lieiue 
drinke ; ' id. p. 38. And lieitis is short for delieiotu ; see Wedg- 
wood. 

mdidone, forthwith. It occurs in Weber's Met. Bom. iii. 54, 
57 ; Seven Sages, 11. 1368, 1442. Eeally two words ; mid idane, 
with its being done, i.e. immediately afterwards ; from A.S. mid, 
with; and ged6n, done. Weber's comic explanation — 'at mid- 
night'— seems to be founded on the fact that F. midi means 
* midday.' 

Pawn, at chess. Littr6 is wrong in connecting it with paon, 
a peacock. See O.F. peon^ also j?«on, a foot-soldier, in Gk>defroy. 
"E, paten : OJF.peen :: 'E.fatan : O.F./tfon. 

Pie, a pasty. Beally from Low Lat. piea, lit. magpie ; perhaps 
from the miscellaneous nature of a pie's contents in medieval times. 
At any rate, in the Babees Book, ed. Fumivall, ii. 36, 1. 51, we 
find Lat. pi. piee in the sense of ' pies,' in close connection with 
' PastUli,* i.e. pasties. 

Plash, a pool. O.F. plaseq, pheeie^ a pool (Oodefroy) ; of Low 
O. origin. Hezham has; ^een Floe, ofie Plaeeh, a Plash of 
water.' 

Pony. The etymology of this difficult word may be found in 
Oodefroy. It is from O.F. pouUnet, a little colt ; dimin. of poulain, 
a colt ; from Lsit.' pullut. The / is lost before n, as in Colney Sateh, 
Lineoln ; but we find Lowland Sc. poumey (for polneg), like etown 
for stolen, gowd for gold. 

Bail, a bar. Not from Low O. regel, as in my Diet., but 
rather from O.F. reille, a rail (Oodefroy). 

Boaeh, a fish. M.E. roehe. From O.F. roehef the name of a 
fish (Boquefort). Of Teut. origin. 

Sleigh. M.E. ecleyef Mandeville's Tray. ed. Halliwell, p. 130. 
This answers to an O.F. *eseleie, which would be regularly formed 
from Low 0. elede, whence £. sled, sledge.- Cf. E. Friesie eledei 



K0TB8 ON ENGLISH ETYMOLOGY. — PROF. 8KEAT. 135 

eommonly shortened to sli (Koolman). But Dr. Murray tells me 
that oar present sleij^h is modem, being in 1806-9 an American 
word, borrowed from Dutch colonists ; i.e, from sUe^ short form 
of iieii, (The spelling imitates that of nei^h.) 

Snore. This is usually derived from A.S. Mora (Bosworth). 
l^ut there is no such word. The A.S. word is really fiwra. 
'Stenmtdeio, Aiora;' Wright's Yocab. 48, 14; 200, 9; 213, 21; 
^77» 26. This became snare because its root-yerb fniosan became 
M^«v; perhaps, too, it was associated with miort In Chaucer, 
^•T., B 790, the liSS. have inarethf fiwrUthf morUth^ and even 
fnmUth (!). 

Stodge. From O.F. estochier, to stab, to stop; cf. Walloon 
M^MT, to fix, fill. Of Germanic origin; cf. O. steelcen, 8ee 
^Mier or estoqtUer in Gknlefroy. Cf. M.E. ttoken^ to stab, in 
^^cer, O.T., A 2646. A form Hoehen seems to be established 
^ HaUiwell, who gives : * Stoehe^ a stab ; ' Yorkshire. 

leanii. Of unknown origin. I draw attention to the Low 
I^t tmar^ the palm of the hand ; from Gk. Oipop^ the same. Cf . F. 
j^itpaume. The Low Lat. tenor occurs in Wright's Yocab. 158, 
1^: ' Ubla, uel tenor ^ uel ir [i.e. x^V]) middeward hand.' (Probably 
^i^Dg; but all the guesses are futile.) 

Vearish. This word occurs in Nares ; the right sense seems to 
^ * pimpled.' Cf. ' CaUtu, wear;' Wright's Yocab. 863, 30; 
'0^, wearras, ibs;' id. 363, 5; < CaUoei, wearrihte;' id. 374, 
^^1 etc. ' WeoTy a hard pimple on the face ; ' Cockayne, A.S. 
J^echdoms^ iL 409. 



NOTES ON ENGLISH ETYMOLOGY. By the 

Rev. Prof. Skeat, Litt.D. 

[Read at the PhiMagiedl Soeiety'e Meeting, June 9, 1891.] 

Alaui, Alftunt Alaun, in Chaucer, C.T. 2150 (or 2148) means 
* ^e mastiff, or a wolf-hound ; see New E. Diet. It is ex- 
Puuned as being from O.F. dlan\ cf. Ital. and Span, dlano. The 
^^mate source is not given. Cotgraye has : * Allan, a kind of big, 
'^i^gt thick-headed and short-snouted dog; the brood whereof 



136 NOTES ON ENGLISH ETYMOLOGY. — PROF. SKEAT.' 

came first out of Albania (old Epirns).' Pineda's Span. Diet, has t 
*AlanOf a mastiff-dog; also an Alan, one of that nation.' The 
Alani were a race of warlike horsemen, first found in Albania; 
Smith's Classical Diet. I suggest that the Low. Lat. alanus orig. 
meant 'an Alanian dog;' which gives us the etymology. In 
Higden's Polychronicon, i. 144, the great size and strength of the 
Albanian dogs are enlarged upon ; they could attack and master, 
he tells us, not only a bull, but a lion. Smith's Diet, further tells 
us that Alanus orig. meant ' mountaineer ; ' from the Sarmatian 
word ala (mountain). Observe that the Molossi were also a tribe 
in Epirus; and the Lat. molossus means (1) a Molossian, and (2) a 
masti£P-dog. 

Beggar. Dr. Murray shews the high probability that this word 
is nothing but a special application of the name Beghard or Begard, 
a synonym of Beguiny and originally applied to a lay brotherhood 
who took their name from a certain Lambert B^gue. There is a 
passage in the E. version of the Eomaunt of the Rose, which is 
valuable in this connexion, and gives support to his suggestion. 
At 1. 7256, and again at I. 7282, we find: 'But Beggars with 
these bodes wyde ; ' and again : ' Who may that Beggar wel 
excuse ? ' In both places, Beggar should be spelt with a capital 
letter, because the French text has Beguin, Here is positive proof 
that E. Beguin and E. Beggar were used as convertible terms. The 
passage is the more remarkable, because Jean de Meun here uses 
Begum as a term of reproach. He seems to mean, not the Beguins 
themselves, but the Franciscan friars. The passage is difficult, and 
should certainly receive attention. 

Cole. In Weber's Eang Alisaunder, 813—816, we find : 

' King Phelip, that was his lord, 
Gurd him with a god sweord, 
And gave him the toU aryght, 
And bad, he scholde beo god knyght.* 

The Glossary has : * Tole^ the tool, instrument, i.e. the sword. The 
Bodley MS. reads perhaps better eoUre,^ This shews that t and e 
have been confused. The right reading is, of course, eoU or eoUe^ 
with the same sense as acolee^ an accolade (New. E. Diet.). See 
numerous examples in Godefroy, s.v. eoU$, I regret to say I did 
not discover this in time for insertion in the Dictionary ; but there 
is an excellent example of the word in Barbour, which Dr. Murray 
has duly quoted. This new example is a little earlier. 



HOTRS ON ENGLISH KTYMOLOGY. — PROF. SKBAT, 137 

Derring-do. This is given, in the Century Diet., s.v. Daring-do. 
The accoant is partly correct. * The word was adopted by Spenser 
in the erroneous spelling derring-dOf which from him and his 
imitators has become familiar in literature from Chaucer; M.E. 
^yng don, during do, etc., a peculiarly isolated compound, from 
^rrynjf^ mod. E. daring, pres. pt., and infin. don, to do. The 
BMociated phrase to dorre don, consists of the infin. do depending 
on the infinitive dorre, durre, dare, and is not, as some think, a 
compound verb.' This leaves the real error untouched. The fact 
iS} that Spenser or his editors misunderstood the matter, and the 
supposed compound derring-do is not really a compound at 
^ It is curious that the editors of the Cent. Diet, should 
^^e seen the right construction in one case, but not in the 
oto. But we have only to look at the original passage in Troil* 
T. 835 : 

• • * Troilus was never, unto no wight. 

As in his tyme, in no degree secounde 

In durring don that longeth to a knight, 

Al mighte a geaunt passen him of might, 

His herte ay with the firste and with the beste 

Stood paregal, to durre don that him leste.' 

iQ the last case, to durre is dat. infin., governing the infin. don^ 
^^ there is no composition at all. So in the former case, durring 
^ not by any means a present participle, but a verbal sb., mod. E* 
*'''*V» It is followed by the infinitive don, by an elliptical con- 
•Action. The proper form would be, in full, *In the durring 
^^Q/ where don = to do ; and in modem English we should 
P'obably say * In daring to do,* though the shorter form *In daring 
^ Would be idiomatic and permissible. My point is, that durring 
^ is not a compound at all in Chaucer ; and if Spenser chose to 
^i^der it so, he was wrong in so doing. It would be ridiculous 
^ talk of ' daring to do it ' or 'daring to go' as a compound ; and to 
^ of * daring do it ' or ' daring go ' as a compound is equally 
'idicnlous. 

^k. I give an early instance of the use of dirk in 1661 ; 
^^^^r forms are dork, durk ; Dr. Murray has ' Two Scotch daggers 
^^ dorkt' in 1602, and, in Eitson's Eobin Hood, p. 78, <a drawen 
•urt* The mod. Irish word is duire, but I do not know whether 
that u borrowed from English or not. I venture to compare the 
^* Ush deig^ given in Windisch ; it nteans ' a thorn, a pin to 



138 NOTES ON ENGLISH ETYMOLOGY. — PROF. SKEAT^ 

fasten a brooch.' The mod. Irish form is dealg^ 'a thorn, a 
skewer, a pin, a bodkin, a prickle.' Of. Shakespeare's use of 
bodkin in the sense of 'dagger.' We also find A.S. dtdo^ a 
brooch-pin; M.E. ddlke, a brooch-pin, in the Catholioon, a.d. 
1483. But we want more light. 

Fewte, Feute, a track. In my gloss, to Wm. of Paleme I give 
*feute^ scent, track,' and qnote from Morris : ^fewt^ a trace of a fo^ 
or beast of chace by the odour.' In the gloss, to Sir Gawain, 
Morris has- ^fewU or odour,' s.t. VewUra, I wish to point out 
that, etymologically, fewte is simply ' track,' and that the notion 
of ' scent ' or ' odour ' is unoriginal. It is possible that Strat* 
mann's Dictionary is correct in deriving it from the F. fwU, lit. 
* flight ; ' hence, ' a track.' The sense ' odour ' was imported into 
the word; hence we find: '/w^, odowre' in the Prompt. Parv. 
Of course Way is wrong in connecting this wiiikfeui»r0r \ in fact, 
he only suggests this as a guess. 

Oofish. This is really a ghost- word. It occurs in Troil. iii. 584, 
but only in the black-letter editions, which read : ' For to be war 
of goJUh peples speche.' Tyrwjiitt explains it as ' foolish, from the 
F. goffe, dull, stupid.' This is impossible for two reasons: (1) the 
F. goffe (see Littr6 and Godefroy) is not known before the 16th 
century, and appears to be merely borrowed from Ital. goffo, stupid, 
a word of unknown origin. Secondly, words in -Uh are formed 
from sbs., not from adjectives ; the exception fool-ish is accounted 
for by the fact that the word /oo/, properly an adjective, was 
commonly used, in English, as a sb. We might add a third reason, 
viz. that Chaucer would not add the suffix -ish to an unfamiliar 
word. 

When we turn to the MSS., the Campsall MS. has go&glgehe, by 
some mistake ; for the line will not scan with this reading, nor 
does it give any sense. The Camb. MS. omits the word. But the 
Harleian MSS. have gooeUh^ goeieshe, though Morris's edition 
unluckily has ^oo/Ef A for the MS. reading goosish. Beyond question, 
the right reading is gdeish, and the sense is goose-ieh, i.e. goose-like, 
silly. Dr. Murray finds no other example of the word, but it is 
quite intelligible and legitimately formed. It is also quite in 
Chaucer's manner ; we may compare Pari. Foules, 568 : — ' Lb 
here, a parfit reson of a goos ' ; and 586 : — * For sothe, I preyse 
noght the gooees reed.' I find that Chaucer uses the words 
mannishf eh%ld%$h, eherlishf and rammish; and Wyclif has doggish. 
The original forms of ehildisk^ churlish^ folkish, heathenish, and 



NOTES ON BNGLISH ETTMOLOGT.— PROF. SKEAT. 139 

man^ more, occur in A.S. As for ^qfishf it is the old story of 
misneadiiig a long « as an /; cf. eftureSf in Malory, for estrea. 

IcLLc, Tdle, an isle. The form ffdle, with the sense ' isle/ occurs 

KfewLtiedlj in King Alisaunder, ed. Weber, 11. 4840, 4856, 5040, etc. 

I find no notice of it in Stratmann or Kalliwell or Godefroy. I 

vifib. to point oat that it is formed quite regularly. In A.F. si 

beconciefl sdl, as explained in my Eng. Etyni. 2iid Ser. p. 236. 

Tbns meiUf a medlar, became *meidle, whence, by loss of «, E. 

»Me^ the fruit of the medlar-tree ; also mesUry whence ^mssdleTf 

nd tlie E. verb to meddle. So also A.F. isle became ^tsdle, and by 

loss of 9, *idl0 ; regularly. I find that this form is duly noted in 

Mitzner, but he does not account for it quite correctly. He asso- 

<^te« it with yet another form tldSf which he correctly explains as 

iormed with an excrescent d after /, just as we find vilde for vile in 

Tudor-English. There is, however, this distinclioQ, ihat idle is due 

^ an excrescent d after «, which e afterwards, as in other cases, 

^ped out. Thus Ude and idle were really formed in rather 

Cerent ways, and should be dissociated from each other to that 

extent 

I^igne, a leash for a hawk. This word occurs in the Rom. of 
^9 Rose, 1. 3882, where the original has lan^. This is the mod. 
^*kngt^ in the sense of 'tether,' also spelt loi^ne in O.F., which 
'^^onts for the form here used. The Low Lat. form is hngia^ a 
^er (see Bucange), which is merely a derivative of hngus^ long. 
Hence the Century Dictionary merely gives a wild guess, in suggest- 
^ that loigne is another form of line. There is, of course, another 
^•P. higne, £. loin, mod. E. hngsy in the sense of ' loin ; ' from 
^ Inmhea. It thus appears that the original sense of loigne was 
'^y ' a long piece ' or ' a length ; ' and this result is remarkably 
<^Qfinned by another passage in the Romaunt, 1. 7050, where 
^ I*, text has : ' U aura de corde une longe ; ' and the E. version 
"^: 'He shal have of a corde a loigne^^ i.e. a length of cord, 
^ongh cord to bind him and lead him away to be burnt, as the 
•^>»text shews. 

I>1UIM, a hawk's jesses. Cotgrave gives 'a hawk's hme, or leash,' 
^ one of the senses of F. longe. This is the mod. F. longe^ a tether, 
^^ Lat. longia, a derivative of longm ; see Loigne, 

I snspect that lune is merely a variant spelling of the M.E. 
^»yjM, a hawk's leash. Godefroy gives the spellings loigne^ longne ; 
^ I think the form longne, which occurs in Froissart, is sufficiently 
^^. I may add that, according to Godefroy, the distinct O.F. 



140 ^0T£8 ON ENGLISH ETYMOLOGY. — PROF. SKEAT^ 

loigne^ a loin, was sopietimes spelt lu%n$y which seems to shew thai 
luine is a poteible variant of loigne^ a tether. This brings us very 
near to lune, Cf. M.E. moyl^ a mule. 

Lyngell. This curious word occurs twice in Libeaus Disconus, 
in Bitson's Met. Eomances, ii. pp. 37, 54, 11. 861, 1274. In the 
first case, we find : Hyng$ll and trappure: ' in the second, ^yngell, 
armes, trappur was swich.' Drappure means ' trappings ; ' and the 
reference is to heraldic display. I think lyngeU represents O.F, 
lineel^ a linen vestment ; it may have been confused with the O.F, 
adj. lingBy made of linen; we also find O.F. lir^etU^ linen cloth 
(Godefroy). If so, lyng$U means 'linen cloth,' which goes very 
well with trappings. The O.F. lined is from Lat. Unteum, made of 
linen ; all from Lat. linumf flax. * Lingel^ a shoemaker's thread,' 
in Halliwell, does not suit the context. 

Mistery. The usual glossaries to Chaucer do not notice his use 
of mt'steris in the sense of * ministry ; ' but an example is given in 
the Century Dictionary, s.v. Myttery. There is another example 
of it in Morris's Chaucer, iii. 348, 1. 4, where it translates the Lat. 
offieium of the Vulgate (1 Sam. ii. 12, 13): 'nescientes officium 
sacerdotum ad populum.' The Century Diet, explains it as formed 
by O.F. tneslieTf with added -i> or -y. I regard it as formed from 
the Low Lat. misteriumf a shortened form of mintsiertum* The 
ultimate origin is the same either way. 

Moysoun, measure, size ; Rom. of the Rose, 1677. Not con- 
nected, as suggested in Bell, with matsson, harvest, but the same 
as O.F. moisan, size, Lat. ace. menaumem. See moison in Strat- 
mann. 

Oubit, a hairy caterpillar ; also spelt iootthit, wohat, wowhet (for 
woubet); see Jamieson. The right M.E. form is wolbode^ which 
occurs in Wright's Vocab. 706. 16 (cf. 766. 28), misspelt welbode^ 
and explaining Lat. hie multipes. In the Shrewsbury MS. of the 
Ortus Yocabulorum, it is written toolhede^ in which the second 
vowel is phonetically weakened. The etymology is easy, viz. 
from A.S. wuly mod. £. wool\ and an A.S. form *hoda or *buda, 
closely related to A.S. htidda, & beetle. Cf. £. bawd, a weevil, and 
botf a worm or maggot. Thus the sense is ' woolly worm,' i.e. 
hairy caterpillar. Of course, toool becomes ^oo* in Lowland Scotch. 
(N. and Q. 7 S. x. 324). 

Parodie. This is among the words in Chaucer which T3rrwhitt 
says that he could not explain. The passage is in Troil. 
V. 1548. 



VOTES ON ENGLISH ETYMOLOGY. — PROF. 8KBAT. I4l 

' Among al this, the fyn of the parodie 

Of Hector gan approchen wonder bly ve ; 

The fate wolde his sonle sholde unbodie.' 

I think it is certain, from the form of the word, that it can be no 

other than the modem parody^ from Lat. parOdia^ Gk. Trapt^^ta, The 

lit. Bense is * a song sung beside,' and we might take it in the simpler 

Wi«« of <Bong.' I think Chaucer took it to mean 'story;* the 

senac is : — * the end of the story of Hector was quickly approach- 

^^S-* We may observe that Chaucer uses 'tragedie' to mean 

'lamentable story ;' and we may note Dante's use of * Cammedia/ 

^ note in Bell's Chaucer explains it from the Gk. vapoho^^ a 

P^^^^age, but used in the technical sense of the first appearance of a 

Greek chorus in the orchestra. But there are two fatal objections. 

^^''at, the Gk. vap6Bo9 would only have given 'parod;^ it could 

not easily have produced a trisyllable. Secondly, Chaucer knew 

no Oreek to speak of, and he certainly had no copy of Liddell and 

Scotrt in which to hunt up the meaning of a technical term of the 

^1^ Greek theatre. Bell's note adds: * parody^ in the modem sense, 

1^*^ quite a different derivation;* which is precisely the point 

whieh I dispute. 

tentacle. There seems to be little doubt that pentaele answers 
^ t;lie M.£. pentangel^ for which see Gawain and the Green Knight, 
'* ^^7. The fact that a pentaele usually meant a six-pointed star- 
^'^Pcd figure, whereas a pentangel signifies ' five-angled ' is easily 
explained by two considerations. The first is the similarity of the 

^^o fgores; and the second is, the partial ignorance of Greek in 

T^^land in the fifteenth century. As to the figures, their similarity 

" Pi^)ved thus, 
^lie pentaele was formed by two equal and equilateral triangles, 

^^^ ^hove the other, disposed thus : O 
-'^e pentangel is described in * Sir Qtiwain * as being * five- 

^^led,' and also as being an ' endless knot ; ' i.e. the lines forming 
^ere continuous. Hence its shape was this : ifr 
-^l^e substitution of the six-angled figure was natural enough. 
'^^as more easily connected with astrology, as it represented six 

^^ '^ilie aspects called * trine ; * whereas it was hardly possible to 

^^^^ect the pentangel with astrology in any way. 
^wldote, Perydote, the name of a precious stone. The pi. 

^^^^^tee occurs in a list of precious stones in Emare, 1. 155 ; in 

^^^^«on, Met. Rom. ii. 210. Godefroy gives O.F. peridot, with five 
tsxiants, and nine examples. He describes it as being of a 



142 NOTES ON ENGLISH ETYMOLOGT.-^PROF. SKEAT. 

yellowish green, and less hard than rock-crystaL But what the 
etymology of the O.F. form is, I do not know. 

Pomet. In Bitson's Met. Brom. ii. 55 » in the Bomance of 
Liheaus Disconus, 1. 1295, we find a mention of 'pomet touris,' 
where ' touris ' is * towers.' It should rather be ponud ; and the 
sense is, that the towers were finished ofE with pomes^ i.e. apple* 
shaped ornaments, which were usually gilt. Cotgrave has : 
^Pommeau d^un tour, the ball of a tower, or middle of the top 
thereof : that part whereon the weather-fane or weather-cock is 
planted.' Godefroy, s.v. pamsl, the diminutive of O.F. pome, has 
several examples of tents and towers being thus ornamented. Cf. 
£. pommeL Similar ornaments or knobs may still be seen at the 
summit of the poles of a cricket-tent. 

Posset. I have given this as of Celtic origin ; and perhaps it is 
so. But we must not overlook the O.F. potsette, cited in Palsgrave 
as equivalent to £. poiseL The Prompt. Parv. has poasot as the 
M.E. form; it appears also as poshoot and poatet in Wright's 
Vocabularies, but does not seem to be older than the 15th century. 
The history of the O.Y.poasette does not seem to be known. It 
seems to have meant a drink that is curdled, being explained by 
Lat. eoagulum. 

Pray, a flock, troop. This word is nowhere correctly explained^ 
Weber's glossary to his Metrical Bom. has: * Pray, Alls. 2595, 
press, crowd, rhythmi gratia ; ' which shows that he fancied it was 
a licentious form of the word ' press.' Halliwell copies this in the 
form: ^ Pray, press, crowd; Weber.' The right explanation is 
something different. The lines are : ' Of his people theo grete pray 
Laste twente myle way ; ' i.e. the vast host of his army extended 
for 20 miles. Pray is precisely the same word as the mod. £. prey, 
but in a very different sense. The fact is, that the Low Lat. 
prada, O.F. proie, was used in the sense of a flock of sheep ; 
evidently because a flock of sheep was a very convenient thing 
to make a prey of. See Ducange and Godefroy. Hence, in 
this passage, it means ' flock,' or * multitude,' or * host,' or 
* troop.' 

Prepense. In the phrase ' malice prepense,^ it might seem that 
the etymology is from the F. pmaer^ to think, with a prefix due to 
Lat.jpr^, beforehand. But this is not so certain. The prefixes j9r<8, 
per, and pro were remarkably confused in French ; and it is a fact 
that 'premeditated lying in wait' appears in the Laws of William I. 
as ' agwait purpens6 ; ' sect. 2. Godefroy gives the O.F. phrase as: 



ROTES ON ENGLISH ETYMOLOGY. — PBOF. SKEAT* 143 

'de malice ptntrpemee,^ Cf. 'felonie purpmse^ in Britton, i. 15 ; 
and the long note in Elyot's Governor, ed. Croft, ii. 875. 

ftoert Stratmann gives : ' Quf&rt, whert, adj. safe and sound ; ' 
and ' Qiurt, sb. sound, health.' Ritson, in the gloss, to his Met. 
Bomanoes, collects several examples, shewing that the phrase in 
f^t is common ; I should explain this phrase by ' at rest,' or ' in 
peace and quietness,' or ' in security.' An attempt has been made 
to explain it from the Er. euer, heart, as if the final t did not 
matter, and as if in quert meant ' in good heart,' which is a very 
forced explanation. I am clear that it is not a French word at all, 
bot Scandinavian. It has the characteristic adverbial sufix -^, 
originally the sign of the neuter ; and I have shewn (Etymology, 
li 467) that we cannot explain the final t in the words atkwar-tt 
^^t, tof't^ ioan4, and toigh-t (adj. active), in any other way. So 
^^re, the real origin is seen in an old Scandinavian form *kwert, 
neuter of ^ibff^. In mod. Icelandic, the adj. is kyrr, but the older 
form is hirr^ which, as Yigfusson notes, is common in Norse MSS. 
The Dan. form is qvidr, quiet, silent, still ; and the Icel. word like- 
^se means quiet, still, at rest. In Swedish, it appears only in 
pbraaes, as in lufffa qvar, to stay, remain ; qvar-hlifva, to remain, be 
left; par-hallaf to retain, lit. * to hold safe ; ' qwir-Ufvay remainder, 
'^Qe, etc. That -^ is a suffix, appears from the Catholicon, 
P- 297, where we find : * to make quar-full, prosperare ; ' quarful- 
^ prosperitas; the same as hele [good health].' We also find 
*i'*^fuUe* in the same, and even ' quartt/fuU$,^ with the senses, 

^inpofi, prosper, sospes.' I take it that quert was first an adv., 
tbenan adj., and lastly, a sb., with the successive senses 'at peace,' 

Peaceful,' and ' peacefulness.' We have further cognates in the 
^' ^e, calm, and the Goth kwairrua, gentle, whence the sb. 
^^irrei, gentleness, meekness. Eluge gives *kwer as the form of 
^^ Teutonic root; which he writes qer. See the examples in 
*^fu86on, shewing that vera kyrt meant ' to be quiet : ' eiija urn 
'y^^i to sit at rest, not to stir ; cf . kyrr-ligr^ calm ; kiftT'laika, 
^^uiUity ; kyrr-eata, a living at rest. The notion of * tranquillity ' 
^^ all the E. examples very well. This etymology also explains 
^ variant form whert. 

^t Notice the M.E. quelde-poynte^ a quilt, exactly re- 
P^eaenting the Lat. culeita puneta. It occurs in Gawain 
^ the Grene Knight, L 877; but the explanation in the 
^loiBary is incorrect. It does not mean a 'hassock,' but a 
'counterpane.' 



144 NOTES ON ENGLISH ETYMOLOGY. — PROF. SKEAT. 

Bankle. I have shewn that the A.F. form of the yeih *i6 
rankle is rankler. I have also said that it seems to be connected 
with the Lat. raneidtu, and I daresay it may really have been so 
connected in the popular mind. But the real etymology is very 
different, and past all guessing. The fact is, the word has lost 
an initial d, as shewn by Godefroy. Godefroy gives draoncle, 
draneldf ra&nele, raanele, ranole, an eruption of the skin, or, as we 
should say, a rankling sore. The verb is draonckr, raoncler, 
raneler, to suppurate. It is obvious that draonele must be, in Low 
Latin, draeunculusi and we find accordingly, in Ducange, that 
dracuneulus, also spelt dranetdus, by contraction, is a Low Latin 
term for a kind of ulcer, or, as we should say, a rankling sore. 
But draeuncfdus, as Ducange says, means, literally, neither more 
nor less than * paruus draco,' i.e. a little dragon, a diminutive of 
draeo. It is thus quite clear that our verb to rankle is a derivative 
from the very Gk. sb. which we spell dragon. Dragons were 
supposed to be venomous ; and draeunetdus probably meant, at first, 
a small venomous thing, and was applied to a poisoned or sup- 
purating wound or swelling. The Low Lat. draeue, properly 
a dragon, also meant a demon; and Ducange quotes the phrase 
fa le draCf to play the devil ; as well as the O.E. drage^ a sorceress. 
In Picardy, as Godefroy remarks, the initial d is still retained; 
he quotes ' j'ai le doigt draneU,^ I have an inflamed (lit. rankled) 
finger. 

Beheten, to cheer. Given in Stratmann; add a reference to 
Bom. Rose, 6509. From O.F. reheter, rehaiter, rehaHier, to cheer, 
encourage ; as said in Stratmann. The etymology of the O.F. verb 
is difficult. The most likely solution is that given by Diez, that it 
is from the O.F. sb. hait^ pleasure, wish, whence F. eouhaiter ; and 
that this sb. is of Teut. origin. The Goth, ga-hait means a 
promise ; Icel. heity a vow ; and Diez notes that, just as Lat. uotum 
combines the ideas of < vow ' and ' wish,' so the Icel. heit^ a vow, 
may be represented in the O.F. phrase a ton hait, according to his 
wish. Hence we come to the idea of pleasure, cheerfulness, etc. 
8chade takes the same view ; under M.K.G. heiz, a command, 
promise, he ranges O.F. hait, wish, pleasure ; O.F. haitier, to 
cheer, and F. souhait, a wish ; and refers us to Diez. 

Senoiail. I only know of two examples of this rare word. It 
occurs in Morris's Old English Miscellany, p. 92, 1. 70, and p. 96, 
1. 106. We there find mention of * robes of russet, ne of rencyan ; ' 
and again, ' vouh, ne gray, ne rencyan* It was therefore the name 



KOTBS ON BKGLISU ETYMOIXKJY. — PROF. 8KEAT. 146 

^' Bome sort of garment. Morris explains it by 'a robe of a roan 

^loor/ but gives no reason, and it looks like a guess. The real 

^iise is ' a robe made at Eheims ; ' and the etymology is from the 

O-f . adj. raeneim, given by Godefroy as an adj. formed from that 

pi^ce-name. The suffix -ten is adjectival ; Lat. -ianus. 

' ^fiaipiee. BeBpiee is given as the name of an unknown wine in 
*^^ Squier of Low Degree, 1. 756 ; ed. Ritson. I should guess it 
^ Qtand for nupieey and to be allied to raspure, given in Godefroy 
•^ cxcurring in the phrase vins de raspure, Cf. rape (for rdpe, i.e. 
^^^•ipe) in Clotgrave, as the name of a thin wine. Perhaps allied to 
^- roip-heny, 

SUdeled, gathered, pleated, Rom. of the Rose, 1235. This is a 
^^»"i) formed from M.E. ridel, O.F. ridel, a curtain ; see Stratmann. 
^y^^ sense is that the garment mentioned was pleated at the neck, 
^^>« a surplice. Halliwell refers to ReliquisB Antiques, i. 41, 
ere we read of ' filettis, and wymplis, and rydelid gownes.' It 
not mean ' riddled with holes,' as suggested in Bell's Chaucer. 
I give, in my Dictionary, a quotation from Gower about 
e eeaUe of a fish ; ' and another from P. Plowman which 
^^^xitions ' the $caU (or shale) of a walnut.' I have not made it 
^l^s^r, however, that the form Bcale must be of F. origin, viz. from 
^- P. &$eaie, because the A.S. sceale or scale would only give the 
f^X"^xi shaU. The ultimate source is, of course, the same, because 
^^ F. eecaU is from the cognate O.H.G. neala, but it makes a 
^>^ttcrence phonetically. There is a good example of O.F. eacale, a 
"^^^U, in the Gloss, to N. Bozon, Contes Moralisit, 

8oak. I note, in my Diet., that the A.S. form should be ioeian, 
^^^ that it is unauthorised. But it occurs in Cockayne's Leech- 
doo:^ twice, in the phrase ' 1st socian,' i.e. let it soak : ii. 240 ; 
"»^- 14. 

faut Spelt tausfht in Phillips, 1706. M.E. toht in Stratmann, 
spelt to^hi in Chaucer, C.T., D 2267. Pp. of to^hen, from 
^*^^X. io^a^ to draw, draw together ; a secondary verb from tfit^^a, to 
"'^►^ cognate with G. tishen, Cf. E. taw, verb; practically, a 
^*^^T)let of Unced. 

^Iranter, a carrier. Given as a dialect- word in Halliwell ; it 

**^^«2^^MB in Hardy's novels. I think it refers to the old time when 

^^^'iner's carts went at a foot-pace, and the carrier walked slowly 

"^^ide the horse; or (as Dr. Murray suggests) to a still older 

^^*^**^e when tranUre trudged along, carrying their packs on their 

0^^ shoulders. See trant, tranten, in Stratmann. Hexham's 

FUL Trans. 1891-2-3. 10 



146 NOTES ON ENGLISH ETYMOLOGY. — ^PROF. SKBAT. 

Da. Diet, has: ^trantekn^ or trantm, to goe lazely, softly, or a 
soft pace.' Also : ' een Tranl, a march, a pace, or a step.' Ulti- 
mately allied to E. trend, trundle, 

Trayeres. This word occurs in Rich. Coeur de Lion, 4785 ; in 
Weber's Met. Rom. iL 188. The line is — 'Berges, schoutes, 
trayeren fele.' The Glossary has : ' Trayeree, long boats, resembling 
trays or troughs.' This is all pure invention. But it is copied in 
Halliwell, who has : * Trayeres, long boats : Weber.' But the &ct 
is, it is a ghost-word. By the ordinary mistake of t for cr, it is a 
misprint for erayeree, a well-known word, discussed by me before 
this Society, June 7, 1889. 

Vewter. In Sir Gawain and the Grene Knight, 1. 1146, the 
word vewtere occurs ; it is explained in the Glossary by * men who 
tracked deer by ihefewte or odour.' To this there are two obvious 
objections ; (1) men are not usually educated up to such keenness 
of scent ; and (2), fewte does not necessarily mean odour. I deal 
with this second objection under the heading Fewte, 

The fact is, that / never becomes v in this poem, as far as I 
know; the West-Midland Dialect hardly admits of it. Hence I 
take vewtere, if correct, to mean a kind of dog,' which in O.F. was 
ventre, and in Cotgrave is vatdtre, and in Dante appears as veltro. 
But the context suggests that vewtere refers rather to men, in which 
case it is a mere corruption of vewtrere ; cf . Low Lat. veltrariue, a 
man in charge of veutree, I have fully explained all about this in 
my article on FetUerer, read before this Society on Nov. 6, 1885. 
But we may go further ; vewter is no error, but simply a corrupt 
form of vewtrer, the former r being dropped. This will account 
for the entry : * Vewter, a keeper of hounds * in Halliwell, given, 
unfortunately, without a reference. This shews that the right 
explanation might have been found in Halliwell. Moreover, I can 
supply a reference. It occurs in the Glossary to Dr. Eumivall's 
Edition of the Babees Book, with a note on the word, which is 
correct throughout. We can now explain the passage: 'To 
trystors vewtere yod, Couples huntes of kest ; ' i.e. Men with grey- 
hounds went to stand beside the tristore, or men who kept the 
stations, and the hunters then cast ofE the couples, i.e. removed the 
leashes from the dogs. 

Wayz-goose. I find this term in Bailey, ed. 1731. He says: 
* Wdyz-yoose, a Stubble Goose, an Entertainment given to Journey- 
men at the beginning of winter.' Ko doubt the entertainment is 
named from the fact that a * stubble-goose ' formed a principal dish 



KOTES ON ENGLISH ETYMOLOGY. — PROF. SKEAT. 147 

it the feast. I have said, in Not$9 and Querteff that this wa^z is 
the same word as that which appears in Baily as ' Wayz, a hundle 
of straw/ though the latter word is more commonly spelt tra«tf, and 
its commonest sense is a straw pad, for carrying a weight on the 
head. I have heen told, also in Notes and Queries, that I have not 
proTed my point, as to the identity of wayt, stuhhle, and wase, 
ft straw pad. Bat I helieve it is quite capahle of proof, and that 
the connection will appear to any one who consults the Swed. Dial. 
Kct of Rietz and other authorities whom I shall mention. More- 
over I take the word to be of native origin. From the A.S. 
^^thoHf to writhe or twist, which is a strong verb, we have the 
•^" tPTdim, a chain, lit. a twist, for which see Greiu; This would, 
'^^Sularly, have become wrdse in Mid. English ; but we find what is 
^abtless an allied form in the M.E. wrase, also wane, noted in 
^^tmann. Thus, in O.E. Miscellany, ed. Morris, we have : * Of 
®^e wrase of thomes, he wrythen hym one crune,* i.e. of a twist or 
^*^th of thorns, they wreathed our Lord a crown. I think this 
'^'^ answers to an A.8. *wrSthe, which would regularly become 
*^«w, by vowel-shortening. Then the r shifted, as shown in the 
^ttolicon, which has the entry : *A wareey fasciculus;' at p. 426. 
"^e should particularly notice that, though spelt wa, it is entered, 
^Pliabetically, as if it began with wr. Of course, this is the same 
^^^^ Next, we may note that Mr. Herrtage, in his note on the 
^^i>d, tells us that worse is * probably a slip for wase,'' because wase 
^'^^'^^^rs in M.E. with the sense of a pad of straw ; it also means 
'^^^ds,' or a wisp or bundle of hards for stopping up a hole with. 
'"'^t; the fact is just the contrary, viz. that wase is a corrupted form 
^' hoarse (the equivalent of wrase), due to the loss of the r. In this 
'^y, we see that wase, a wisp, twist of straw, is formally connected 
^^^ii wrase, and is ultimately derived from A. 8. wrlthan, whence 
^^^^> E. wreath. In fact wase and wreath have the same sense, and 
*"*1^ differ in shewing different suffixes. 

X^ remains to show the connection with stubble. To begin with, 

"^^^e twists were usually made of straw, and the connection 

^"t^een * straw ' and * stubble * is sufficiently close. But it appears 

^^*i^ clearly in this. The Swed. dial, vrase is explained by Eietz 

^ Uiean precisely the com which, in reaping-time, was not bound 

"^^ in the sheaves, but was left by the binders, and raked up into 

heaps afterwards. This is just the very thing out of which the 

80^^ would pick up its living; I take a stubble-goose to mean 

T^'^^^ly a goose which is turned into the corn-fields after the com 



148 NOTES ON ENGLISH ETYMOLOGY. — PEOF. SKEAT. 

is reaped and carried. Further, Eietz says that vroM or vaie also 
means the lowest layer of hay in a hay-loft, though tiiis may be 
a different word. Hence we may equate the form vra9e with 
the common Swedish vase, a sheaf of com; Norweg. vase, a 
wisp or bundle of thread or of hair, a bundle or heap of brush- 
wood ; Dan. dial, vase, a sheaf, a bundle of straw with the com in 
it (Molbech). Then, again, Hexham gives Du. ufoee, a torch, 
clearly because a torch was a Hwist,' as the derivation of torch 
shews. I think this is enough to shew that ivrase and warse and 
tDoee are the same word, and that the senses were : (1) a twist, 
hence, a wreath, wisp, bundle, esp. of straw or faggots, and (2) a 
heap of straw, and hence, the leavings of com in the fields, on 
which geese were fattened. The Du. Waae, turf, is a difEerent 
word; see Schade, who distinguishes A.S. wrdten, a chain, twist, 
from M.H.G. wrase, G. Masen, turf; and shews that they are 
probably from different roots. I mention this last point, only to 
shew that I have not overlooked it. Lastly, I would observe that 
a similar loss of r occurs in dace, formerly darce, derived from O.F. 
darSf a dart ; cmd again, in pash, formerly ^arce, from O.F. garseir^ 
to scarify, to lance. K.B. I find an instance of itom, a wisp of 
straw, in Lydgate's Troy-book, ch. 34 ; fol. t6, back, col. 1. 

Yuly. Halliwell notes the wonderful word *^yuly, handsome,' as 
given by Eitson ; but thinks it is an error for ynlAf, There is no 
doubt about it ; inly often means ' closely ' or ' narrowly ; ' the line 
is — * He behelde ynhf hur face ; ' £rl. of Tolous, 337. The sense 
' handsome ' is wrpng. 



149 



IV.— ON THE BODLEIAN FRAGMENT OF OORMACS 
GLOSSARY. By Whitlbt Stokes, D.O.L. 

[Jteaa Dteembtr 4, 1891.] 

• 

CoRMAc's Glossary is a mediaeval Irish Etymologicum, full 
of absurd attempts to trace words to their sources, but 
yaluabloy partly as explaining many obscure Yocables, 
partly as containing extracts from ancient law-books, 
whereof some are now lost, and partly as preserving sundry 
traditions and legends which interest the student of Celtic 
history, folk-lore and romance. It is ascribed to Cormac, 
the king-bishop of Cashel, who was slain in battle in the 
year 903 ; but its language proves that it was not written 
much before the eleventh century, when the change from 
Old- to Early-Middle-Irish had set in. 

Two complete copies of this Glossary are known. The 
elder is in the Lobar Brecc, a fourteenth century MS. in 
the library of the Royal Irish Academy, pp. 263 to 272 
of the fac-simile, Dublin, 1876. The younger is in the 
Tellow Book of Lecan, a vellum in the library of 
Trinity College, Dublin, cols. 3 to 87. This part of the 
codex was written in the beginning of the sixteenth 
century. Three fragments also are known, one in the Book 
of Leinster, a vellum of the middle of the twelfth century, 
p. 179 of the facsimile, Dublin, 1880, another in a Bodleian 
MS. of the fifteenth century, and a third, containing only 
the articles PruU and Mug ^e, in fE. 75*-76* of Harl. 
5280, a vellum in the British Museum. The Lobar Brecc 



150 



MR. STOKES— ON THfi ;B0DLEIAN FRAGMENT 



copy and the fragment in the Book of Leinster were 
published in Three Irish Glossaries^ London, 1862, and a 
translation of the former by John O'Donovan was printed in 
Calcutta, in 1868. The copy in the Yellow Book of Lecan 
has not yet been printed ; but a photographic facsimile of it 
and the rest of the vellum in which it is contained will be 
published next year by the Royal Irish Academy, with funds 
supplied by the British Government. The Bodleian fragment 
now for the first time appears tff extenso} It is taken from 
the MS. marked Laud 610, which has been described by 
Dr. Todd {Proceedings of the Royal Irish Academy^ ii. 336) 
and by Dr. O'Donovan {Book of Rights^ pref. xxviii, xxxiii). 
This fragment begins at fol. 79* with the last four words 
of the article iarmbilre,^ and it ends at fol. 86* with the 
article turigein, the nine articles beginning with u being 
omitted. The scribe of this part of the MS. was Se&n 
Buidhe O'Cl^righ, and his colophon, written in 1440, runs 
as follows : 



ISh6 annalu inTig^ma inudir 
doscribai in sanasdn so na Salt- 
rach .i. mile blia^n 7 ceithri 
cet hliadan 7 tri bliaina d6c 7 
da Jkhitf in cuiced H domi l^ebra 
7 intoehtmeA li donesca. misi 
Seaan Bvidhi oGl^iny doscrib 7 
d£mann Bvillter mac Rieterd 
doscriboi sanasan Saltrach Cor- 
maic so. 



This is the year of the Lord 
when this little Glossary of the 
Psalter hath been written, to 
wit, a thousand years, and four 
hundred years, and two score, 
the fifth day of the month of 
February, and the eighth day of 
the moon. I am Yellow John 
O'Clery who wrote this, and 
for Edmund Butler Mac Richard 
hath been written this little 
glossary of Cormac's Psalter.* 



1 The articles Imhoit forotfuti, Modehrotky Mug Sme, y68 and Pdtraic were 
published from Laud 610 in the Tripartite Life of Patrick, Bolls Series, pp. 568, 
670. 

' d.h. die unbetonten Worter, die fiir die Alliteration nicht mitzahlen, 
Thumeysen, Insche Texte, 3** serie, 1. heft, s. 116. 

3 i.e. the compilation called the Psalter of Cashd, as to which see Petrie, 
Ecclesiaetical Architecture of Irelandj pp. 38, 39, and O'Curry, Leeturee on 
the MS. Materials of Irish Mistorj^f p. 11. 



OF OORMAC'S GLOSSARY. 



151 



The history of the MS. is continued by the following entry 
on the upper margin of fo. 110^ : 



Salttair moic Ruisd^ Buitil^ 
.L 'Emmm Bnitiler, in tsalttair 
860 nogtf dtaca[d] maidm Baile 
in fPoill ar iarla Urinaman 7 
ar mac BuisdM^ le iarla Desmu- 
man .i. Tonuw, 7 do bained in 
leabor so 7 Leabor na Carruigi 
as f uasgZo^ mefc Huisderd, 7 isse 
in Mtfc 'SMuderd sin do cbufir na 
leabair sin da scribad do fein no- 
gur bain Tomas iarla Desmuman 
amach iad.^ 



This Psalter (was) the Psalter 
of Mac Richard Butler, to wit, 
Edmund Butler, until the defeat 
at Baile in Phoill (now Pilltown) 
was inflicted on the Earl of 
Ormond and on Mac Eichard by 
the Earl of Desmond, to wit, 
Thomas.' And this book and 
the Book of Carrick were taken 
in ransom for Mao Eichard, and 
it is that Mac Richard who caused 
those books to be written for 
himself, until Thomas Earl of 
Desmond took them away. 



0*01ery must have transcribed from a very ancient 
manuscript, for the language and spelling of this Bodleian 
fragment (Z.) are far more archaic than those of the copy 
in the Lobar Brecc (LB.)^ the oldest complete copy known 
to exist. The comparison of a few forms will justify this 
assertion. The corresponding forms in the Yellow Book 
of Lecan (F.) have been added. 



Laud 610. 

atopair 

cotagair 

himmaig 

isintig 

tarsa muUag 

forsa crand-sin 

conammelt 

condaig 

na semmand (ace. pi.) 

ond anmuim sin 

don chrand 

forrumither 

banbragid (nom. pi.) 

bertatar 



Lehar Brecc. 

hidbraid 

gutagair 

amuig 

istig 

darin mullach 

forsan crand sin 

coimlid 

cuindchis 

na semanda 

on ainm sin 

do crand 

furmither 

banbrugadha 

inrucsat 



Yellow Book of Leean, 

adodpair 

cotagair 

imaig 

isintigh 

tar mullach 

forsin crandsin 

conamailt 

condaig 

na semand 

on anmain sin 

don crund 

forruimther 

banbrugaid 

bertatar 



^ There is an incomplete and inaceurate copy of ihlB entry in the Froeetdingt 
0/ the Royal Irish Aeadetnyj ii. 338. 
> The eighth Earl, who was heheaded at Drogfaeda in 1467. 



152 MR. STOKES— ON THE BODLEIAN FRAGMENT 

In the Bodleian fragment the neuter article is still in nse 
(an-iy a tenm l&ida, s.y. imbas forosnai, an-etag, an-^tach, 
s.y. kgam, a c^tnae, an-aithesc, s.y. kthech, a main, a nduiu, 
a mbraut, s.y. modebroth, al-lathi-sin, s.y. mairt, tar-sa- 
mallag, s.y. tnalland, for-sa-crand-sin, 8.y. nescaitf i-sal-leth 
n-aill, s.y. prull). Neuter stems in n are declined as in 
Old-Irish: thus we haye sg. gen. anma^ s.y. Morann, dat. 
anmuim, s.y. Mumu, dual nom. dd n-ainm, s.y. Morann, pi. nom. 
il-anmann, s.y. rout, ace. aemmend^ semmand, gen. semmend, 8.y. 
nescoitf dat. anmannaib, s.y. nemnall, rSmennaib, s.y. triath. 

Comparatiyes in -thir are frequent : buidithir, camtdir^ 
cuirrithir, duibithir, glaimdir^ luaithidir — all s.y. prull. A 
superlative is nessom, s.y. ore tr^ith, from ^nedhsmmo-Sf 
cognate with TTmbr. and Osc. neatmo-f Skr. ^nah, naddha^ 
nadh. So airegdam, s.y. prull. Moaib, s.y. nemnall, is 
probably a scribal error for mdam, another superlatiye. 

Infixation of pronouns is still in full force. Thus : 

sg. 1, fo-m-geillsat, s.y. pain, and imm-om-loiscet, nom-leicid-si, 
rom-bia, atom-glaite-sa, all s.y. prull. 

8g. 2, ro-t-huc, 8.y. mumu, and inn-ot-bia, ni-t-aicelladar, nit- 
acelladar, s.y. prull. 

sg. 3, masc. d: ro>d-torimiai, B.y. letheeh; immi-d-comairc, s.y. 

prull. 
dm do-dn-gnf, 8.y. niih; do-dn-g6na, s.y. mhos 

forosnai. 
idn : ass-id-cfd, s.y. prull ; imm-id-r6t, s.y. rout. 
for-idm-bi, s.y. la%th\ ar-idm-boi, s.y. %mha% 
foromau 
tn: co-tn-gair, co-tn-aitib, no-tn-acclestar, all s.y. 

Utheoh. 
a : d-a-bir, s.y. imhas forottMi. 
da: con-da-forruimi, s.v. ntntM; cota-gair (=coth- 

da-g.), 8.T. tmhas forosnai. 
n : do-n-iced, s.y. neseoit ; amach-Q-imp[ar3ra, B.y. 
imhas foroinai ; ta-m-bert, s.y. ore, 
fern. 8 : ni-s-fuair, s.y. prull. 
pi. 3. da: ata-gladastar (=ad-da-g.), B.y. prull. 

08, 8: imm-os-coemorcuir, B.Y.prull; do-s-leicit^ ro-s- 
fuair, s.y. ore. 



OF OORMAC*S GLOSSARY. 153 

Xvifized relative : intan didiu ro-m-bo broithe, 8.y. Utheeh ; imm- 
an—derg^thar, s.v. leo9\ intan tra do-n-anic, 8.y. pruU; is do con- 
a-s^»crad, s.y. mairL 

TThe verbal forms are equally archaic. Thus : 

"V'crbal prefix infixed : f ort-ro-r-gell, 8.v, imhas ; far-ro-laig, 8.v. 
kM^€h; do-de-r-saig (a*to-di-ro-od-8ech), s.y. m«^ ^itm; do-r-onad, 
8-y* pruU\ pi. do-r-onta, s.v. mtig dms; fo-r-acbad, 8.v. mha$. In 
do-ixid-ar-scansat (from ^-fnd-ro-scansat), s.v. mug hne, as in f(h- 
ind^-^r-Ud (gl. subintravit) Wb. 8* 6, and ^tmm-or-iuit, Lism. Lives, 
222, 225, we seem to have a metathetic form of the posttonic ro. 

Orthotonic forms of compound verbs : ad-miudur, s.v. mann ; 
sd-rfmed, s.v. r%nene\ at-cess, as-bert, s.v. prtUl; as-con-grad, s.v. 
^^ ^m$\ as-soilgi, s.v. loeh\ atopair, s.v. imbasforosnai; con-acrad, 
8.V. letheeh; con-daig, s.v. mw^ ime ; do-ad-benar, s.v. imhai foromai ; 
do-^tni, S.V. samrad ; do-bert, 8.v. wrc ; do-fuisim (*to-f o-ess-sem-), 
^^- letmae; do-fuarascaib, s.v. mhos forosnai; do-g6ni, s.v. mug 
^*^ ; do-indarscansat, s.v. mug dme ; fo-r-acaib, s.v. mug 6me ; for- 
rooaither, s.v. netedit ; imme-luinethar, s.v. loes. 

EUiclitac forms of compound verbs : (con)epert, s.v. mug ime ; 
(^O&rgaib, s.v. mhasforoifiai; (hi)fuirmither, s.v. ne9cait'y (co nach) 
^''^^csca, (ama) tarmescatar, s. v. mhos fwromai ; (n{) tardad, s.v. 
^^ 9 (na) tarta, s.v. mug im9\ (ama) eillnet, s.v. r6ut (where 
^m/js^/ stands for Ssslenety cf . the orthotonic as-linaimm, Sg. 54^ 8) ; 
'^ttx^perthe, s.v. pnUl; tait, s.v. milgetan; (na) hacaldai, s.v. 
P''^^^^ In tohrethf s.v. lethech, tohert, s.v. mug eme, and ta-m-hert, 
••^- ore, an enclitic is wrongly used for an orthotonic form. 

•I^lie conjugations are still distinguished in the third sg. 
Pi'^^ent ind. act. Thus we have the following subjoined (or 
* w^TvJTinct ') forms : 

^^^'Verbs: atf6t, s.v. htheeh; imm-id-r6ty s.v. rdut; atopair, s.v. imhaa 

/oro«^4ii . con-gair, cota-gair, dicain tbid, ; di-eim, s.v. ollam ; do*beir, 

^'^ii", s.v. imhas farotnai; do-fuisim, s.v. ksmac; do-fuarascaib, 

*•▼• ^tnhai/aroinai; do-ling, b.y. prull; do-roich, s.v. pom; do-d-roig 

fc.v. rx5ii<; fergaib, s.v. imhas farosnat ; fo-cren, s.v. rdut, for-dingair, 

B.VV. if^i^ Utheehy mairenn; con-daig, s.v. mug im$\ t-ic, s.vv, 

orc^ ^wulgg, and apparently (nad) fosad-som, s.v. legam. 

^-▼erbs: do-bruchta, s.v. tipra; forosnai (from *for-od-sunna), 
con- end, 8.V. mbas faroinai ; tim-chella, s.w. hes, sceng ; f odera 
C=-f«Nl-fera) b.t. mairenn ; do-SKnga. s.v. rop. 



154 MR. STOKES— ON THE BODLEIAN FRAGMENT 

e-yerbs: do-d-gnf, s.y. niih; pass. sg. 3, do-gnftber, s.v. imlas 
forosnau 

I-Terbs: assoilgiy s.y. loeh; contuili, s.y. imhas forotnai; do- 
aitniy s.y. samrad ; docmidi, s.y. loeh ; forruimi, s.y. pndl. 

Tbe so-called consuetudinal fonns in -arm, -enn do not occur. 
Tbis points to a date before a.d. 1100. 

Absolute forms: benaid s.y. orCy socbtaid s.y. prtdlf sluinnitb 
s.y. laithf bfid s.y. langfiter^ doitbid s.y. mug iitne : witb suffixed 
pronouns: cingtb-i, s.y. pndlf^ rant-ai (=rannaitb-i), s.y. ore. 
Pass. sg. 3 6titbir> s.y. Idmand; bitbir, s.y. imbas foromai; 
letbaidir, fertbair, s.y. lethech^ promtbair, s.y. pain, maeltair, s.y. 
rane ; pL 3, deiligtir, s.y. triath. 

Subjoined (or ' conjunct') forms: pi. 3, dofocbat, immom-loiscety 
r-ecaty s.y. pndl, Do-s-leicit, s.y. ore^ sbould be do-s-16icet. 
Pass. sg. 3 doadbenar, dognftber, s.y. %mha» foromaif aracialltar, 
fossaimtber, s.y. legam, imma ndergtbar, s.y. Uos, co n-oenaigtber, 
s.y. turigin, doberar, s.y. Uthech^ imcomarcar, s.y. naime^ adfiadar, 
s.y. neseott, fotbruictber, s.y. pruH, i ngabtbar, s.y. sen; pi. 3, 
i nglanaiter, ara nglanaiter, s.y. rdut. 

Reduplicated preterites are : 

sg. 3, atcboDnairc {^derk^ Skr. dadarga, ^dBoptce), s.y. prtdl; 
do-cbuaid {^kud, Skr. eodayamt)^ s.yy. mug eme^ prull ; do-deocbaid, 
s.yy. ninus, prull, rosead ; pass. sg. 3, docbuas, s.y. ore ; ra-cuala 
(from *kuklave=z^]LT. gu^ava), s.y. letheeh; ni tbanic, s.y. mug dme, 
do-n-anic, b.y. prull, co farnaic, s.y. ore {^ank, Skr. y^af, anamga)\ ro 
gdid {^^}sx, jag&da), s.y. ore\ dorrumidir (Gotb. miian), s.v. la%th\ 
do coemnacair {^nank, Lat. nanc-ueor), s.y. ore ; immos-coemorcuir, 
s.y. jPTtt// (cf. imcbomarcair, LU. 62» 7, \/ar^=Skr. prag). 

pi. 3, con-accatar Wkas, Skr. ^cdksh for eakas\ s.y. prull; ro 
lellatar {y/li, Skr. -lilye), s.y. morann; romebdatar (leg. -memdatar, 
^matf cf. Skr. mamdtha), a mbatar, s.y. prull. Absolute form: 
batir, s.y. ore. 

T-preterites are : 

sg. 3, ro-dn-ort, s.y. lathirt (cf . ro ort * delevit,' LU. 48«, ^org) ; 
conammelt {^mel, Lat. molo), s.y. mug ime ; luid, do-luid {^lu=plu), 
s.yy. ore^ mug ime; doriucart (■»*to-ro-od-gar-t), fris-gart {Vgar, 

^ Such forms originally, perhaps, belonged to the middle voice. Thus berid 
may be from *beretS—<p4p€TcUf as no^bered is certainly s=^-^cp«ro. See H. 
CoUitz in Bezz. Beitr. xvii. 232, note, where he suggests that the absolute 
inflection is developed from the old middle, and the conjunct (•beir^hBLt./er't) 
from the active. 



OF OORMAC'S GLOSSARY. 159 

lith. giriii)f s.v. pruU; as-bert, s.v. Istheeh; to-bert, s.v. mug ime. 
ad-acht, s.y. rtnene (cf. do-sn-acht, imm-act, ^ag^ Lat. ago, Gr. df^tv). 
pL 3, bertatar, as-bertatar, b.v. prtdl {^her, Lat. forOy Gr. (f>€pu)) ; 
for-ra-achtatar, b.t. mdU (cf* do-ra-achtatar, Z^. 457, ^ak, 
Skr. flf). 

Reduplicated future: eg. 1, ni geb-sa, s.v. mug ime\ sg. 2, 
asb^ra, s.y. lethech] sg. 8, do-dn-g^na, s.v. imbas fafroinai; ass-m- 
b^ra, 8. V. Uiheeh. Secondary form, sg. 3, nach eperad, s.v. morann. 
S-future, sg. 2, ni naiss, s.v. mdl: deponential : meser, s.v. 
megamlae ; eg. 3, ara ndig, s.v. r6ut : dep. cini fiastar, s.v. pam ; pi. 
d (relative), cichsite, 8.t. mann. Secondary forms : sg. 3, co feissed, 
ara tissed, s.y. Utheeh ; co taudchissed, s.v. ore} The curious form 
itfrMiM, s.y. r6ut, seems= iuro^, the secondary redupl. s-fiit. of 
f^gim^^ plus a suffixed pron. of the 3rd plur., just as iitrthund, 
XU. 108» 19, is'^iurad plus a suffixed pron. of the 1st plur. 

Passive forms of the verb substantive are bithir, s.v. tmhas 
Jbrosnai, and -tathar, s.v. pruU. Bethir^ -hether occurs in the 
IV'urzburg codex, but I have not met tdthar in Old-Irish. 

So much from the grammarian's point of view. To the 
lexicographer all the articles now published, but especially 
ranc and rdut^ are of value. The historian of the move* 
ments of races in the British islands will find one of his 
most trustworthy documents in the article Mugh Sme, Irish 
folklore and mythology are illustrated by the articles Imbass 
farosnai, Lugnaaad, Leos, Mtliiud, Morand, Mugh ^me, Man" 
annan mac iir, NUt^ Ninus^ Ruad ro/eaaa. Sin, and Triath : 
Irish manners and customs by Laarg, Leithech, Milgetan, 
and Ollami Irish law by Matt, Mann, N6e, Noes, Nbhs^ 
Segamla and Sau : Irish romance by Nesc&it, Ore, Prull, 
Rincne, and Serb. All these matters are more fully 
noticed in the preface to Three Irish Oloasaries, and in 
the annotations to O'Donovan's translation of Cormac's 
Glossary. To that preface and those annotations the notes 
at the end of this paper may be taken as supplemental. 

W. S. 

^ See as to turad * occideret,* Thumeyseii, Sevue Celtique, vi. 96, 372, note, 
and Hogan, ibid. Tiii. 636. With Ir. orgimy cf. the O.W. orgiat (gl. caesar, 
i.e. caesor), and Ganl. OrgetO'ia, which Penaon connects with Gr. ip4x9», 
'Ep§xMs. 



156 MR. S^COKES— ^N THB BODLEIAN FRAGMENT 



» 



FRAGMENT OF CORlfAC'S GLOSSARY. 

Laud 610. 

[fo. 79*.] hunisa a thasoelad ind. 

IMbas foroenai .i. dofuarascaib Beclii[p]ret bas maitb lasin filid 
7 bas adlacc ' do dofaillsigud. Is amlaid 6n didiu dognitber 6ii 
.i. concnd infill m{r docbamu dirg muiee n6 ebon no cbaitt, 7 dabir 
iamm forlicc iarcul nacomlad, 7 dicbain' dicbetal fair, 7 ato[d]pair 
do deib idol, 7 cotagair do, 7 nifargaib iarum amabaracb, 7 dicain * 
iarum f oradabais/ 7 congair deo ^ idol cbuici ama tarme8cat[b]ar * 
acbotlutb, 7 dobeir adabois imadalecooinn ^ 7 contuili, 7 bitbir 
oca[f]borairi amacb nimprd' 7 connacb tarmesca' necb, 7 do 
adbenar ^^ do iarum anf aridmboi '^ cocend nomaitbe " no ado n6 
atH fut gardi cotmeiss^i " occond audbairt. M ideo imbas dieitur 
A. bas disia 7 bass anall imacenn. Atrarpi'^ Patraicc anisein 7 
atenm Mida, 7 fortrorgell ^' [a briathar] napa[d]mme na talman 
nacboen dodng6na, ar is diultud batbiss. [col. 2] Dicbetal docben- 
daib immorro i cortM cherd» foracbad '' son, ar issoas foWa son« 
7 nibeoen audpairt do demnaib oocu, acbt aisneis dicbendaib 
acbndme focb6tair. 

Laecb .i. a laico. Laicbes .i. laeob 7 fess : is^^ dinfess foas la laecb. 

Ligor .i. tengu. 

Lugnasa[d] .i. nasad Loga mato Etblend .i. oenaob nofertba 
laissom imtbaitti fogomair. 

Lelap .i. laulep .i. lau cecb mbec. 

LesmaQ .L lismac .i. arindi dofoisim liss dond[f]ir nd donmnai 
inti is lesmac donecbtar de. Sic lessatbair 7 lesmatbair. Les*' 
dano qtMfsi lis^' .i. immorgal^ no debuid. 

Legam .i. ligem, dindlige liges anetag. [N6] ligditb .i. aith fri 
liga. N( aria em nad fosadsom*^ fri cacb n6tacb cenibetis liga 
imbi, acbt ismenciu aracialltar ** 7 fossaimtber an^tacb Kga quam 
aliud uestimentum. 



* adlaic, Y. " d6cai, L. ' dochain, Y. * foradibois, Y. • tie, Y. ; 
dano, L. * tarmascthar, T. "^ adibois ima dilecain, T. ^ n-imparrae, T. 
* toirmescae, Y. *® doadbanar, Y. " aradmbi, Y. " ii6maide, Y. 

*' fut ngair eomesa&df Y. ** Atrorbe, Y. '* fotroirgell, Y. *• fodracbad 
son i conis cerdae, Y. " sic, Y. ; 7, L. ** Lis, L. ** lias, L. ^ imargail, Y. ; 
imorgal, L. '^ ua denand som foe, LB. ** araciallathar, Y. 



OF OORICAC'S GLOSSARY. 157 



TRANSLATION. 

easy to disclose it therein. 

Itnbat foroma * Manifestation that enlightens ' : (it) discovers 
▼hat thing soever the poet likes and which he desires to reveal. 
Thus then is that done. The poet chews a piece of the red flesh 
of a pig, or a dog, or a cat, and puts it then on a flagstone 
hehixLd the door- valve, and chants an incantation over it, and offers 
it to idol gods, and calls them to him, and leaves them not on the 
morrow, and then chants over his two palms, and calls again idol 
9^ t;o him, that his sleep may not he disturbed. Then he puts 
his tr^o palms on his two cheeks and sleeps. And men are watch- 
ing him that he may not turn over and that no one may disturb him. 
And t;lien is revealed to him that for which he was (engaged) till 
the end of a ndmad (three days and nights) or two or three for the 
long or ^Q short time (?) that he may judge himself (to be) at 
toe ofiering. And therefore it is called Imm-bas, to wit, a palm 
(om) on this side and a palm on that around his head. 

-^^t^ck banished that and the Tenm Idida ' illumination of song,' 

^d <l^clared that no one who shall do that shall belong to heaven 

^^ ®^^^h, for it is a denial of baptism. Diohetal do ehmnaih 

^^I^^Onpore incantation,' however, that was left, in right of art, 

^^ ^^ :Ss science that causes it, and no offering to devils is necessary, 

^ declaration from the ends of his bones at once. 



^'A'hero' from laieut. Laieheu 'heroine,' that is laeeh and 

***,. . ^»^eep : ' it is from the sleep which she sleeps with the hero. 

•, that is * tongue.' 

^nad * lammas-day,' that is the festival (nasad) of Lugh, son 

^ *'^^8:x:iiu, to wit, a fair that was held by him at the beginning of 
autui 



^^^^ * child,* that is lau-lsp. Lau is everything small. 
^^'^^mae 'step-son,' that is lis-mae, because dissension QU) is 
cans^i^^ to the husband or to the wife by him who is step-son to 
eithe^^ of them. Even so Us-athair 'step-father,' and les-rndthair 
*^®I^^^rnother.' Lm, then, quasi /w, that is 'contest or quarrel.' 

^^^im ' moth,' that is ligem^ from the licking with which it 

"*i** t^he raiment, or lig-dith, that is, 'sharp against colours.' 

iSol o^eause it does not settle on every cloth, though there be no 

coVoTizTs thereon, but the coloured cloth is perceived (?) and is settied 

ou oft^QQ]. than another vesture. 



158 MR. STOKEd— ^>N THE BODLBIAK FRAGMENT 

Lecconn .i. leccend .i. leth chenn. 

Lassamuin ab eo qt<oi sst lass- [col. 8] -itn .i. cech ndn. 

Lemlacht .i. lacht t^itb .i. lem oech t6ith. 

Loch d^de fordingair .i. loch .i. dub. nude dieitur assoilgi laith 
loch bronna .i. dub bronna .i. cid dorche samud diich 7 eid rundae ^ 
riam 7 iarum, docruidi aruna laith do 61. Loch d&no .i. huiH, xmde 
dieiiuT loch dub .i. huile dub. 

Laarg .i. leo ball 7 arg .i. laech. ball, dagldich insin .i. 
ahage.' 

Lam .i. loam in ' chuirp. 

Lamas .i. lamibss .i. foss Idme. 

Ldmand .i. Mm ind .i. ind na laime etithir * di. 

Lautu .i. lau ceoh mbeco .i. m6r is lugu fil forsinlaim.' 

Lie ab eo quod est lidos grAce et interpretatur lapis. 

Laith .i. d6de fordingair .i. laith gaile 7 Idith .i. med* ut 
protfdiximus etir laithi Lugbai [li sula sochar, Jl] .i. ameid Lugbai 
cherrda dorrumidir Fachtna anargat amabu 7 rl. INtan tra as 
forsail foridmbi issand sluimiit[h] ^ hoc. 

Langphetir .i. aiogliss insin. lang diiiu .i. fottae, phetir 
immorro .i. glass, langphetir di^^iu .i. glas Iota .i. itir chois 
niarthair [fo. 79^] 7 chois nairthir biid.^ Kon sfc aurchomul .i. 
aur accomol .i. itir dichoiss airthir bfd son. 

Lecht ab eo qtiod est leotus. 

Long bis formuir, ab eo qtiod est longa.' 

Luachair .i. taitnem, ab eo quod est luceo vel lux^^ quasi 
lucar. 

Lebor quasi libor .i. a libro. 

Lott qtiasi lott, ar is lott do mnai mertrichus. 

L&th quasi luth" [.i. iarsinnf luthas]. 

Loman .i. lorn fann, sech is lom is fann. 

Lathirt .i. laith ort .i. laith rodnort .i. 61 corma. 

Lugbort melius est .i. lubgort^' .i. luib gort, ut dieitur gort 
luibe. 



* nmdae, Y. ; nrndu, L. • Fof' .i. ahage, F. hat Lorg .i. Ifii ar^ .i. laich i, 
no lar rice. * HCfY.; an, L. ^ eidigther, T. ^ ise mer u lauga isin 
laim e, Y. • .i. Ifeith 7 laitk gaile .i. med, L. ' sluinid, Y. » biia, Y. 
' longao, L. *^ .i. soilisi, Y. '' L&t qxiasi lut, Y. >' Lubgart meliiM 
est .i. luigbort, .i. luib gout, ut dica^tfr, L. 



OF OORMAC'S GLOSSARY. 159 

Zsceann * cheek,' that is Ucoenn, that is ' half/ Ulhy ' head/ eenn. 
LasMmuin * flamy,' from /<M«tm ' everything bright.' 
Lemlacht * new milk/ that is smooth milk, for lem is everything 
smooth. 

Zoeh means two things, to wit, loeh 'dark/ whence is said 'ale 
opens dark wombs/ that is, though dark be every one's council, 
and though it he secret before and after, to drink ale elicits (?) ^ 
its secrets. Zoehf moreover, means 'all,' whence is said hehdub, 
that is ' all black.' 

Laarg 'fork,' that is leo 'limb' and arg 'champion,' a good 
champion's limb is that, that is, his joint {i,e, his honorific portion 
of food). 

Ldm ' hand,' that is pilot {luam) of the body. 
Ldmos ' sleeve,' that is Idm-foss^ that is the resting-place (Joss) 
of the arm. 

Zdmand ' glove,' that is Idm-ind, i.e. the end {ind) of the hand 
(ldm) is clad thereby. 

Laulu 'the little finger,' i.e. lau is everything small, for it is 
the smallest finger on the hand. 
Lie ' stone,' from the Greek \t0o9, and it is interpreted lapis. 
Laith, two things it means, i.e. laith ' a valiant hero,' and Idith 
'a measure,' as we said before. 'Between the scales of Lugba,' .... 
that is, ' in the balance of Lugba the goldsmith Fachtna adjusted 
the money for the cows.' It is when tiforsail ' mark of length ' is 
upon it that it signifies this. 

Langpheiir, ' spancel,' that is English : lang then means ' long,' 
and phetir means ' fetter.' Langphetir then means a long fetter — 
between a hind-foot and a fore-foot. I^ot so is aurehomul, i.e. aur 
' front,' aeeomol * binding,' that is, between two fore-feet. 
Leeht ' sepulchre,' from Uctus. 
Long ' ship,' which is on the sea, from hnga. 
Luachair * brightness,' from lueeo, or lux quasi luear, 
Lebor ' book,' as if libor, i.e. from liber. 

Lott ' harlot,' quasi lot * destruction,' for harlotry is destruction 
to a woman. 

Ldth 'hero,' quasi luth 'motion,' because he moves with suppleness. 
Lommann * cloak,' that is lomm-fann, because it is bare (lomm) 
and weak (Jann), 

Lathirt, 'drunkenness,' that is laith-art, i.e. laith 'liquor,' 
rodn-ort * destroyed him,' that is a drinking of ale. 

Lugbort, better Lubgort 'herb-garden,' i.e. luib 'herb,' gort 
' garden ' : as is said, gort luibe, a garden of herbs. 

^ Cf. rO'Cruidi, F^. ep. 136. 



'MO MR. STOKES — ON THE BODLEIAN FRAGMENT 

L(n .i. a Hno. L^ne a Unfa. 

Ldnamain .i. Idn-somuin insin,^ aris leth-somuin nechtar de 
secharaile.' 

Lethech .i. deide for[d]ingair .i. aimn [c^^amus] dochenelia else 
dianid ainm [lethech] .i. ara lethet 7 ara thanadetaid, ar is mor 
lethet neich de bfs' in ociano. Ainm dono do losait, arindf 
lethaidir bargin fuirri, ennal iehert Cruiteni '[file] fechtas luid 
dothig alaile 6icis 7 agilla leis .i. 6c8ini [6]8ide com^nmain 
fithidrea> Farrolaig didiu Cruit6ni fadesin himmaig. Leicis agilla 
do hoigidecht' dothig ind6icis. Tobreth' didiu tarr tuirc do 
iscabdiP 7 ro [col. 2] boi ealleic int^ices occacallaim indeicsine, 
7 occur aula tar a frithgnam. Rorathaigestar * iarum inteices 
mor-m^nmain indeicsine 7 laget afrithgnama. Intan didiu rombo 
bruithe intairr asb^ inteices fiadindecsinia .i. 'Dofotha tairr 
tein ' .i. ismfthig athicseil dintein,* 7 ba ^^ cofeiss^ son ciafrecra 
dob^ad inteicsine do, fobith rachualasom inteices ocga[a]maidim ^^ 
alanaile nairecc ningantae amal bidhe feisin aratiss^i^' 7 niro 
[creit]8on immorro inteces," [Ba aire atbert intecesj dopromad 
indeicsine .i. 'tofotha tairr tein/ et tribtM uicibiM dixtY, et non 
respondit 6i ad ullum unburn. Laissein diitu attraig inteicsine 
7 teit cosfiinmbale ^^ imbai Cruiteni, 7 atfet scela do, 7 asb^rt 
fris anaithesc roradi inteces .i. ' tofotha tairr tein.' ' Maith,' ol 
Cruiteni, 'intan a8smb6ra^^ affrisi^' asb^asa fris occa .i. 'Toi lethaig 
foin fris. [adaind indlis .i. toi lossait faon fris] .i. fon tairr, 
7 frisindle chaindiU do deicse^^ d^ inbruithe intairr.' INtan 
didiu dofessid inteicsine isintig thall [79^ 3. J dixit intecess aoetnae, 
et dtxi^ inteicsini [col. 3] 'Toe lethaig' 7rl 'Maith son,' ol 
inteces, ' ni bru ecsini rodtorinnai : ^^ isin ocus ata Cruiteni. 
Cotngair demuig ! ' Conacrad iarum. Ferthair failti fris, 7 
dob^rar biad naill hicoiri dosuidiu, 7 babecc a m^mae de ineicsine 
fobith cotnuaitib " inteces cein notnaiclestar* Cruitini. 

Leos immdergad'^ immandergthar duine iamahair no iama- 
hecndug. 



^ For insin, F. hat each dib dia chelei. ^ ' Y. adds : Aliter lanamain quasi 
lenamain, ar ni fil etarecarad doib acht ar Dia. ' nich bis de, L. * fithidire, T. 
* do hui^decbt, L. ; do aigidecht, L.B. • Dobreth, Y. ' iscabul, Y. 

^ £orataid, Y. * athiscal do teinidh, Y. ^^ isba, L. ^^ oc giunaidim, 
Y. " arahissed no araradissed, Y. ^' intei^in, Y. >* maighin, Y. 
>* assinbera, L. *• doridisi, LB. •' dia deicsin, Y. " rotonnai a 

inne, Y. ; rotoirne no rotorma, LB. ^* cotnaitib, Y. ^ cech notnaichlestar, 
L. ; ce chotnaidefitar, Y. *^ imndergad, L. 



OP C0RMAC*8 GLOSSARY. 16t 

Zln 'flax/ from Unum ; line ' shirt/ from Imia. 

Ld,noma%n 'a married couple/ i.e. Idn-somain 'full property/ 
for each without the other is only a. half property. . 

.L^tAsek^ two things it means: it is a name for a kind of fish, 

because of its breadth and its thinness, for great is the breadth 

of i^ that is in the ocean. It is also a name. for a kneading- 

troa^li, hecause a cake is spread {htkaidir) on it,, as Cruit^ne said 

once when he went to the house of a certain poet having with 

him hus gillie, a bardling he, with a teacher's pride. Cruit6ne 

himself remained outside, and sent his gillie for guesting to the 

poet* 8 house. Theii a hog*8 belly was given the bardling in a 

caldron, and meanwhile the poet began conversing with him, and 

castixi^ an eye on his diligence (in preparing the meat). Then 

the poet perceived the great pride of the bardling and the small- 

ness of his diligence. . So when the belly was boiled, the poet 

said in the bardling's presence, Dofotha tairr tetn, that is, 'it 

is tixKie to take it off the fire ; ' and [ii was in the poetic dialect 

he flpoke], in order that he might know what answer the bardling 

would give him, because the poet had heard him falsely boasting 

of certain wonderful inventions, as if it were he himself that 

had invented them. But the poet did not believe that, and 

therefore he said, to prove the bardling, Dofotha tairr tein. Thrice 

he said it, and not even a word he got in answer. With that 

^ bardling rises and goes to the stead where Cruit6ne was 

■^yiug, and tells him his tales, and repeated the words which 

^6 poet had uttered, even Dofotha tairr tein. "Good," quoth 

Cnut^ne: ''when he says them again, thou shalt say to him, Toi 

^^igfoinfris adaind ind lis, i.e. * put a kneading-trough under it, 

(i.f. Qoder the belly), and light a candle to see whether the belly 

^ hoiled.' So when the bardling (returned and) sat down in the 

^ouse there, the poet said the same, and the bardling replied, * Toi 

^^ig^ etc. ' That is good ! ' saith the poet : ' it is not a bardling's 

▼omb that hath produced it : Cruit6ne is at hand : call him from 

^thoat.' So then he was called: welcome is made to him, and 

^^er food is put for him into the caldron ; and small was the pride 

^ the bardling because the poet had mocked him until Cruitene 

^ spoken to him. 

-^'m, a blush wherewith a person is reddened after being satirized 
^' »«tUed. 

^ Traai. 1891-8-3. II 



162 MR. STOKES — ON THE BODLEIAN FRAGMENT 

L6e8 .i. suilse, ut Mt inBuil [Roscad] : grin[n]iud lois .i. 
dfbdud^ suilfii .i. caindli. Item aiged fir ixnme lois luinethar .i. 
immatimchella ' suilse. 

Luacharnn .i. qttaei lucemn .i. a lucerna. 

Modebrothy ol' Patraico, quod Scoti coitapte dicant. sfc autem 
dici debet .i. muin duiu biaut. amain diJtM is mens, anduiu is deus, 
ambrant is index. 

Marco .i. ecb. marcacli .i. eich imdai laiss, nt dieittfr buasach 
infer lasanibit b^ imdai, airmnech d&no infer techta« arbur nimdai 
7rl. Sfc airgtech/ coilctbech. 

Matbair quasi mater^ ariss^i rotmaillned and. 

Mid .i. combrecc rotmaillned and .i. med. 

Metbil quasi metil, ab eo quod est m6ta ' vel meto. 

[fo. 80^ J Muccairbi .i. mac fuirmid .i. is mac ddn dofuirem' adana. 

Malland .i. f^itb b{s tarsa mullag annas, quasi mulland. 

Mass quasi a ma[s]sa.'' 

Milliud quasi misillind .i. sillind olc* 

Miscath ' .i. mi insce. scath .i. insce. 

Milgetan .i. Mol-cbniten .i. cnit Muil dorsada Temrach. Mol 
didiu a ainm side, deg in muil noferad foma duinib ^° .i. tait 
ass, tait ind. inde dk^'^ur molach. 

Melgg as, arindi mblegair." Melgg dayso .i. hi%, unde melg-tene 
•i. tene mbdis. 

Morann .i. mordnd. bahed insin ahainm dob^rt amdtbair do, 
7 asb^ robad bibdu^' biis nach eberad" fris [anainm sinj. 
Moo muin immorro iBsed ainm dob^ aatba[i]r do .i. ba muin 
maith inmac. 7 ba bidbu bdis nacb ep^rad fris inainmsin. 
Rolellatar ^^ iarum a d4 nainm infer in dun ^' oenanma. Mac som 
Cboirpri Chind chaitt immorro. 

[80, col. 1.] Menath .i. min iith. 

Moth .i. each ferdae .i. cecb [fer-] insce, ei nomen [est] uirili 
membro.^' 

Man .i. a manu. 

^ didbdud, L. ; dibad, LB. ' immathimchela, L. ; imatiincliella, Y. 

' od, L. * airgtbech, L. ; airgtec, Y. ; airgdecb, LB. * .i. buain, Y. 
• du fmrim, Y. ' .i. on cfeir, Y. » drochaiUiMrf, Y. » Mlssach 

.i. inimsce, L. *® forena d6uub, Y. " melgair, L. ; blegar, Y. 

» bidbu, L ; bidba, Y. and LB. " nacThjitnepred, Y. " Roleltar, Y. ; 
rolennstar, LB. ^^ indun, Y. ; arson, LB. ^^ .i. ball ferrda, y. > 



OF OORMAC'S GLOSSARY. 168 

Lon 'light,' as is in the Book of Maxims, grinniud lou, i.e. 
extinction of a light, t.e, a candle. Also aigedfir imme lots luinethary 

* the face of a man which light surrynnds.' 

Zuacham * a lamp/ quasi lueem from lueema. 

Modehrothy says Patrick, which the Scoti say corruptly. It should 
be said thus : muin Duiu hraut : the mum, then, is ' my/ the Ihtiu 
is ' God/ the hraut is 'judge.' 

Mare 'horse/ mareaeh 'one who has many horses,' ut dicitur 
htauuh^ the man who has many cows, airmnech, the man who 
owns much com, etc. Even so airgtech, ' a moneyed man,' coileteeh^ 

* one having quilts.' 

Mdthair ^ mother,' quasi mater^ for this was corrupted in it. 
Mid ' mead.' Welsh was corrupted in it, to wit med. 
MtthH ' a party of reapers,' quasi mettl, from meta or tneU. 

Mueeairhe^ that is a macfuirmid (a poet of the sixth grade) 

Malland^ a vein that is across the crown of the head {mullaeh), 
quasi mulkmd. 

Masi ' a mass,' as if from masM. 
MiUiud ' destruction,' quasi mk-siUiudy i.e. evil-eyeing. 
Miseatk ^ a curse,' i.e. mi-insee ' evil-word/ seath, i.e. word. 
MiJgeiany i.e. MoUchuiten, the share of Mol, the door-keeper of 
Tara. Mol, then, was his name hecause of the talk (mo/) which 
be used to hold with the people, to wit, ' Go out ! Gk> in ! ' Hence 
is said molaeh ' hoarse.' 

lidg ' milk,' hecause it is milked {melgair). Helg, also, ' death,' 
whence nulg-tene, that is fire of death. 

Marann, that is m6r-Jind, i.e. great-&dr. That was his name 
that his mother gave him, declaring that whoever should say that 
name to him would be guilty of death. Mae-mdin was the name 
that his father gave him, i.e. the son (tnae) was a good treasure 
(m^Mi), and whoever should say that name to him would be guilty 
^f death. So his two names clave to the man instead of one 
''^oie. He was son of Coirpre Catshead. 

-^^iffuiih ' an awl,' i.e. min * small,' dith ' sharp.' 
^ ^oth, everything masculine, i.e, every masculine word, et nomen 
"^iimembro. 

* hand,' from manui. 




164 MR. STOKES— ON THE BODLEIAN FRAGMENT 

Manach .i. a monacho.' 

Mooach .i. clessach, ab eo quod 0st mon .1. cles. 

Methoss .1. a meta.' 

Molad [fo. 80, pol. 2.] [.i. mol-sod] .i. is mol ara menci 
7 is[86]od aragnaith[ch]i. 

Menmchossach .i. mtfumae chassacli .i. caingnech, [lais]. 

Muirtchend .i. ab eo quod est morticinium.' 

Muilend ,i. mol 7 ond .i. docb, aris^r? dede asmuilend immulinn/ 

Mertrech ab eo quod est meretrix, merendo pretium stupri.* 

Matt .i. mucc, unde est tsna Brethaib Nemed .i. forruachtatar 
m&ttsL mothuindi torgabail 7rl. 

Mann .i. unge .i. rmde Sencha dtxt^ : 

M6 ailib imdergad Emnse. 
Admiudnrde secht cactu ciohsite crisu. 
Secht muga moigfite morgnimu mugsaine. 
. . Secht manna oir lorloisothi * frifialgnuis ^ cona chaurathaib 
Conchobuir. 

Munnu .i. m6 finnu * .i. ainm buide Fintain enim dictiM est a. 
\mds Moedoc Fema dtxi^ occ dir Munnu mate Tulchain : 

A chel^n D6 chumachtaig ! 
a mate Tulchain, a bachlaich ! 
rue mac nannsae * diamnintir 
mathair rothuc, a Fintain ! 

Milis quasi mellis .i. isinand 7 mel. 

Mil .i. rf *° 7 fili, \mde dieitur Ni naiss or " na hargatt [fol. 80, 
col. 8.1 acht for mal. 

Mairenn .i. dede lordingair. ainm cetamtM domnai .i. morfiind, 
ainm dano do gai .i. mfrind ^' .i. droch rind .i. fodera b4s. 

Mugeme ainm incetna oirc oe tar aba inHere .i. Coirpri Muso ceta 
tucside in Ere atirib Bretan. Ar intan ropumor cumaohta nan- 
Goedel forBretnaib " rorannsat Albain eturru hiferanna, 7 rofitir 
c4oh durais diacharait " leo. 7 nibu lugu notrebtafs Goedil ^* frimuir 
anair quam in Scotia, 7 doronta anairusa ^' 7 arigduine and. Inde 
[dieitur] Dind Tradui .i. dun tredui " .i. tredue Crimthain Moir maic 

* manacbo, L. ; manco, Y. • on crich, Y. ' .i. marbadh, Y. 

« is mvileiid inmullim, L; is muilend immuilend, Y. ; asmaillem immuilend, 
LB. * .i. dligid si fiach a saothair, Y. • aithlegtba, LB. ' fialgniw©, Y. 
•» mo a finda, Y. • L. ituerts nic nansae. " Y. inserts 7 cis. " 6ir, L. 
" mirfind, L. " L. and H. insert 7. ** 7 a caniid, H. " goedelo, L. ; 
gaoidil, Y. " anairse, H. anairlisi, Y. *^ sic H. tradui, L. 



OF C011HAC*S GLOSSARY. 165 

Manack * a monk,^ from tnonaehuB. 

Hmuich * tricky,' from man ' a trick.' 

Metkost ' a goal/ from meta, 

Molad 'praise,' i.e. mol 'a mill-shaft,' from its frequency, and 
96d ' taming,' from its usualness. 

MenmehoMoehf i.e. menmae 'a mind' eastach 'disputatious' hath he. 

MmrUhmn * carrion,' from mortieinium. 

MuiUnd * a mill,' i.e. mol ' mill^shaft,' and <md ' stone,' for these 
are the two things that are together. 

Merir$eh 'harlot,' from mereMx^ .heing entitled to {merenio) 
the price of her defilement. 

Matt 'pig,' whence in the Judgments of the Notahles, mdtta 
'^pigs ' have attacked my . . . • , etc. 

Mann ' an ounce,' whence Sencha said : 

' Greatest of disgraces is the reproach of Emain. 
I adjudge therefor seven hondmaids who will emhroider (?) girdles : 
Seven slaves who will perform the great works of slavery, 
Seven ounces of refined gold for a nohle face, with Conor's 
champions.' 

Munnu, i.e. mo Finnu, ' my Finnu,* a loving name. Pinntan was 
so called. Hence said liaedoc of Ferns when satirizing Munnu, 
son of Tulchan : 

little vassal of mighty God ! 

son of Tulchdn, shepherd ! 

She bore a child troublesome to his family 

The mother who bore thee, O Finntan. 

. J/f7M ' isweet,' quasi meliis : it is the same nsmel ' honey.' 

Mdl ' king' and 'poet.' Hence is said, ' Thou ahalt not bind gold 
or silver except on a mdl,* 

Mairenn^ two things it means, first, it is a name for a woman, 
i.e. mdr-find ' great-fair :' then it is a name for a spear, i.e. mi-rind 
' evil point,' that is, it causes de^ith. 

Mugh-4me ' slave of hilt,' the name of the first lapdog which 
was in Ireland. Coirpre Muse, he first brought one into Ireland 
out of the lands of the Britons. For at that time, great was 
the power of the Gaels over Britain, and they divided Albion 
among them into estates, and each of them knew his friend* s 
abode. And the Gkiels used to dwell to the east of the sea not less 
than in Scotia. And their dwelliDgs and their royal forts were built 
therein. Hence is sai^ IHnd Dradui^ i.e. Dun Tredui, %,$, the triple 



166 MR. STOKES— ON THIS BODLEIAN FRAGMENT 

Fidaig .i. r( Heirenn 7 Alban comnir nlcht. 7 inde est Glassdinibir 
[.na nOoidel] .i. cell forbru inara hicbt. Iwed aruiBS insin irrobai 
Glass moo Caiss moicid rfgh Hirbuaithi ocomucaib for mess, 7 isb^ 
iDsin dodersaig ^ Patraicc iartain .1. se ficbit ' bliadoae iam[a]g:mn 
dofiannaib Mate Con, 7 is dindraindsin beos ata Dind Map Letani * 
bitlrib Bretan Com .i. Dun mac Liatban [ar is mac iiinf as map 
isin Bretnus]. Sfc rorann oaob ceii61 disiu, arroboi acbutrumu allsB 
anair, 7 robatar fonobumacbtu sin cociana cid iaiticbtain Patraicc. 
Disin trd didiu roboi Coirpri [fo. 80^, col. 1] Muse occ aitbigid 
sair coamuintir 7 coacairdea. Nitbanic didtu indinbuidsin oircci 
tir nilirenn, 7 ascongrad ^ laBretnu natarta ' oirci ind£il nabaiscid 
nacaratrad na oommain doGoed^^i^. 

lSb{ tra cain roboi indinbaid si laBietnu: Cecb bidbu innacbinaid 
doneocb foroesad [a cbdin].' 

(Boi) didiu oirci aimind ^ isseilb cbarat Oboirpri Muse bitiiib 
Brettan, 7 atroe ^ Coirpri buad. Dochuaid dano Coirpri corfectais 
do atbaigsidi,* 7 forricc ^^ failti act immonoirci. Scfan amrae dano la 
Coirpri Muse 7 imd^nam de or 7 do argat imabeimb, 7 bas6t amrae 
7rl. Tob^t iamm Coirpri beoil comor 7 conammelt saill fo an em.^^ 
Po/acaib inscfn arbelaib indoirci. Eogab iartim intoirci cocnam 
indeo^e comatain .i. roluited iarum inscfan ^nabud alain[d] '' 
[ijamab^racb. Dog6ni iartim Coirpre c4is moir desin, 7 ba 
bronacby 7 condaig cert nimbi coacbaraitt. '* Indraicside, dano; 
[col. 2.] icfasa incbinaid," oUsuide. "Nigebsa," ol Coi[r]pr*, 
'' acbt bi fil bi cain Bretan .i. cacb bibdu " inacbinaid." ^ larsein, 
tra, dobretb ^ intoirci innacbinaid, 7 dobretb '' ainm do .i. Mug 
beme. 

Ba bancbu, tra, intoirci, 7 batorracb intan dobretb tairis anall. 
Ailill Flann Bee, tra, ba rf Muman intansin, 7 Cormao bua Ooind 
birrig^ Temracb. Do indarscansat sidi locbetoir cuingbid [7 
cosnam] indoirci. Isbe iarwm crutb rocoraiged eturra atriur, 
aimser chinti dobeitb doncboin bitig cacbae. Doitbid iannn incui 
[7] robuc cdcb cuilen di[a]cuain, " 7 is oncboinsin oirci lELeirenny 
ut Scot! p^bibent. 

* dedersaicb, H. ; rodersaig, T. and LB. ' fithit, L. * den roind beos 
sen ata dind mapledhain, H. * isracongradh, H. * conat&rda, T. 

^ forsesed a chain no foniasnadfad ichain, T. ; rofess^ a cauain, H. ; nofuasna* 
bad cain, LB. ' annind, L. * contori, LB. » aticcsuie, H. 

atsaighid, Y. *• sic, Y.; foricc H. ; colonic, LB. We should perhaps 

read forrdnie, the perf. sg. 3 of forteivu ^^ beoil oomor ima neim 7 

conammailt sull 7 bo^aill foa, Y. *' conarho aloinn, H. ^' bidbdu, L. 

bidba, H. ** each rob ina chin, LB. " dobretha, Y.; tncad, LB. 

^B rolil, LB. ^^ Y imerU oonaprait aceli nade 7 landsat fir Eirenu etoixa. 



OF CORHAC'S GI4O8SART. 167 

{08S of Crimthaiui the G)reat, sob of Fidach^ king of Ireland and 
of Albion aa far as the Ictian Sea, and hence is Glastonbury of 
the Gaels, a church on the brink of the Ictian Sea. That is 
the residence in which dwelt Glass, scm of Cass, the King of 
Hiruath's swineherd, with his swine mast-feeding, and it is he that 
Patrick afterwards raised from the dead, six score years after he had 
been slain by the Fianns of Mac Con. And from that division also, 
in the lands of the Cornish Britons stands Dind Map Letan, that is, 
the fort of the sons of Liathan, for mae ' son,' is the same as map 
in the British. Thus every tribe divided on that side, for its 
property on the east was equal to that on the west, and they 
continued in this power long after the coming of Patrick. Hence 
Coirpre Muse was paying a visit in the east to his family and his 
friends. At that time no lapdog had come to the land of Erin, 
and the Britons proclaimed that no lapdog should be given to 
Gaels in partition, or gift, or friendship, or barter. 

Now at that time the Britons had this law : * Every criminal for 
his crime to him whose law he shall have infringed.' 

There was a beautiful lapdog in the possession of a friend of 
Coirpre Muse in the lands of the Britons, and Coirpre got it from 
him thus : Coirpre once went to his house, and was made welcome 
to everything save the lapdog. Now Coirpre had a wonderful 
dagger, with an adornment of gold and silver on its hilt. It was 
a marvellous jewel. So Coirpre put much grease upon it, and 
rubbed bacon to its haft, and left the dagger before the lapdog. 
The lapdog took to gnawing the hilt till morning, and the dagger 
was injured so that on the morrow it was not beautifuL Then 
Coirpre made great complaint thereof, and was mournful for 
it, and demandeth justice for it of his friend. * That is fair 
indeed. I will pay for the crime,' said he. ' Nought will I take,' 
says Coirpre, ' save what is prescribed in the law of the Britons, 
namely, every criminal for his crime.' So after that the lapdog 
was surrendered for its crime, and a name was given to it, even 
Mug-hne * slave of hilt.' 

Now the lapdog was a bitch, and was in pup when it was taken 
over to Ireland. Ailill Flann the Little was then King of 
Monster, and Cormac, grandson of Conn, was then in the king- 
ship of Tara. These began at once to demand and contend for 
the lapdog. This is the way in which matters were arranged 
between the three of them; that the bitch should be a certain 
time in the house of each. Then the bitch litters, and each of 
them took a pup of her litter ; and from that bitch come (all) the 
lapdogs of Ireland, as the Irish declare. 



168 MR. STOKBS— ON THE BODLEIAN ERAOMENT 

Bamarb tra intoiroi. larreib cianaib, immorro, fofdair Connlae 
iDMc Taidg ma/o C^in mate Ailella Olaim clocenn lorn iDdoirci, 
7 dobert ' hioeist dondfilid dolaid conairchetul ' dia athair .i. 
Moen mae Etnae nomen poet» illius. Tethnae iarum, int^ces tre 
tenm laido, done^^ert : * C^ tonna [tige •] hui [fo. SO**, col. 3] 
Eogain,* fth hitig hui Chaind cachthddath tobaraind.^ Basa coem 
hitig Choirpri Muse, aMug h6ine. Gend Moga heime inso/ olse, 
* incetna hoirce dobreth • inH^rind.* 

Mumu de nomine ali^tims regis, id est Eocbu Garb lilninti .i. 
Eocho Mumo .i. mo agreimm 7 agreitt 7 achumachtae oldds oachrf, 
7 isdi[a]aiimuim rohainmniged Mumu 7 Huimnig dicuntur.^ 

Mug qtMsi much, aris fo muich 7 fo thodemam fognama bfs. 

Mugsaine .1. quasi mug snime .i. snim bfss ^ for mmmain inmoga. 

Much ' tra ainm saindiles dodiaid, unde dieitur muchad. 

Mil .i. mel rotruaillned ann. 

Midach quasi medio, ab eo quad est medictM.'® 

Mer arindi iBah6eiiur bith^' isind alt imbi .1. innameracht, j 
isahoenur teit ^' quasi merulus .i. Ion, et inde merultM grotfce quod 
uolat solus, ^' 7 n(gndth en aile cid diachinel fodeisin inacoimtecbt. 

Meraoht quasi mericht .i. icht nd acht m^r .1. gnim m^r ab 
eo quod est actus.^^ 

Mairt .1. a Marie .i. o dia obatha la genti dianid ainm Mars, 
is d6 conasecrad " allathisin [fo. 81% col. 1] 7 inm{ dianid ainnt 
MdrtiuB .i. mi marta. 

Mfndecb quasi mendio .i. ab eo quod est mendicw.'* 

Mart a morte, quasi mort.^'' 

Mortlaitb .i. a mortalitate.^^ 

Manannan fuoc lir .i. cennaigi amrae roboi inninis Manand. Ish^ 
luam '* isdech rob6i formuir iniartbar domuin. Eofinnad trianem- 
gnaebt (.1. gne nime) .i. tria deicsin indaeoir,**^ inderet nombiad 
intsuitbnenn[7 in doinenn]'' 7 intan conoemchlabad*' cechtarde 
arre. inde Scoti et Brit6nes eum deum maris uocau^runt, et 
inde filium maris esse dixerunt .i. mac lir ; et de nomtne Manann 
insola ^ Manannan dicta ^ est, 

^ combert, LB. * donairoetnl, L. ; 00 n-i&i f.i.] aircbetnl, LB. * nr, H. 
^ ecgom, L. ^ oechtratha dobarind, H. * dobretha, LB. "^ dicantur, L.; 
dicunt, T. ^ L. omitt ; biss, T. ; ibs, LB. * Mug, L. ; much, T. and LB. 
^^ .1. liagb, T. ; liaigh, LB. >' arindi bis ina aonar, T. *' isin aonar imteit, Y. 
^' .i. etellaigid [a]aonar, Y. " .i. gnim, Y. ** condosecrad, Y. " bregach, Y. 
>^ morte, L. ; mc^ .i. on b&aa, Y. *** on mortlaitb, Y. *' luamaire, Y. 

^ ina, deicsin gne in nime .i. in aeoir, Y. '^ ind airet nobitb indtsoinind 7 
ind donend, Y. ^* eodaocblobadh, Y.; noeclaechl6bad, LB. ^ insole, L. 
a* dicatiM, L. 



OF OORMAC'S GLOSSARY. l69 

Now the lapdog died, and long afterwards Gonnlae, son of Tadg, 
son of Cian, son of Ailill Bare-ear, found the bare skull of the 
lapdog, and took it as a puzzle to the poet who had come with 
a eulogy to his father. Moen, son of Etna, was that poet's name. 
Then the poet solved it by tmm Idido 'illumination of song,' and 
said ' Cain tohna, etc., Thou wast dear in the house of Goirpre 
Muse, Mugh-6me.' This, says he, * is the head of Mugh-6me, 
the first lapdog that was brought into Ireland.' 

Ifumu 'Munster,' from the name of a certain king, to wit, 
Eochu Garb Mumu, . «.». Eochu Mu-mo, t .#. Greater {m6) his 
hold and his valour and his power than that of every other king, 
and from his name Mumu ' Munster,' was named, and Mumnig 
* Munster-men,' are (so) called. 

Mu^h * slave,' quasi much * mist,' for it is under the mist and 
tribulation of slavery. he is continually. 

Mugh^aine * slavery,'' quasi mugh snitne, the sadness that is on the 
mind of the slave. 

Much, a name proper for smoke, whence is said mtiehad 'to 
simother.' 

Mil ' honey ' : mel has been corrupted therein. 

JUidaeh, the name of a Tuath d6 Danann leech, quasi medie, from 
medieus, 

Mir * blackbird,' because it is alone in the wooded valley where 
it Hves, that is, in its madness (?). And it is alone it goes, qua^i 
mertUus, i.e. merle, and hence in Greek merulus, because it flies 
alone, and another bird, even of its own kind, is not' often in its 
company. 

Meraeht 'madness,' quasi mer-ieht, i.e. ieht or acht mir, i.e. 
'action of fingers,' from actus, 

Mairt ' Tuesday,' from Mars, a god of battle with the Gentiles, 
whose name was Mars. To him that day was consecrated and 
the month whose name is Martius, that is March. 

Mindeeh ' a bad man ' (?), quasi mendic from mendicui^ 

Mart ' a beef,' quasi mori, 

Mortlaith ' mortality,' from mortalitas, 

Manannan mac lir, a famous merchant who dwelt in the Isle 
of Mann. He is the pilot who was best at sea in the west of the 
world. He used to know by studying the heavens, that is by 
looking at the sky, the time that would be fine weather, and foul 
weather, and when each of them would change. Hence the Irish 
and the Britons called him a god of the sea, and hence they said 
that he was Mac Lir,. son of the Sea, and from^anannan's name 
the Isle of Mann has been called. , - . v \ J ^^ -^^^ 

/ -^-^ '■ 01' Tii^'^^ ^X 

[ - -•■ i V 4ii. :. .,',"■;' \ 

^ ,-• ^^ -: 

\ -^ >J "•• ^ -■' .. /, 



170 MR. STOKES — ON THE BODLEIAN FRAGMENT 

Nie .i. mac sethar. ut Gu Chulaind dtxt^, profetans de Cbrbti 
aduentu .i. nie duine tiofa .1. mac sethar .i. moc sethar duine tic&i, 
ipse est lesus. 

Kemnall iaaed ainm as moaib ^ de anmonnaib H^irenn ' insin 
.1. nemnaall .i. naall fer nime imbL 

KfniM .i. nin[f]os .i. nin .i. tonn [rogab foss] .i. tond [tanic 
don] fedrgi anfar. dodeochaid* iarsindaeor ^ndaforruimi^ isintlr 
[n-ucut], condeirg6ni' tiprait df. Inde dioitur Corcammad 
Kfnuiss. 

Kemeth .i. nemiath anas dfrde * doheclais : nemdith anas dirde^ 
de ocaib : nemhuath anas dfrde ' dofil^^b. 

K6it .i. dia catha lagente GoedeL l^emon uxor illius. 

Nacc ^ quasi n^. 

[Nith] .i. guin duine. 

Ner .i. tore allaid, ut est isnaib Aistib .i. fail nir n6it grife 
grddaichdae. 

Koe .i. duine, uni^ dicitur diandama noe iorHr .i. dfan dama 
duine fortir. 

Koes ^ .i. fess ' n(5nbair .i. tri rfg 7 tri epscoip 7 tri suid .i. sui 
filidtf^A^a 7 sui litre 7 sui b^lrai Fene. Bobatar huili ocdenam 
intSenchasa [Mair]. 

Kim[b] .i. broen, ab eo quod est nimbus. Inde dm^tir isnaBrethaib 
N6med .i. Oengtft foh6iblib ^^ imbais aricht roluisc leth ^^ lonimib 
nimb. 

Kaimne .i. glaine,^' no naimne, amal bid naire nobeth and« 
SenbelrsB diditf annairesin.^* Jsinann 7 asb^ha 6cin. Isgnath 
b^lrasom [col. 3.] tra [cid indiu] lablrmumain maxime, unde 
dieunt ''Infil ni bestoich duit?" "Fil naire," arinti dianimcomarcar 
•i. fil eicin. 

Kith .i. guin duine. Nie infer dodngni. Keithes dano .i. guin '* 
duine. ut est nf idnae nethis nemthigedar. 

Kescoit .T. ise senchas Goidhel .i. intan fechtae cath Muige 
Tuiredh bui Goibhne gobae isin cerdchai oc denum na narm do 
Thuathuibh nDea Domnand, 7 bui Luchtaine soer Iri *' denum 
crand isna goa, 7 bci Credne cerd fri denum semmend innangoe/* 

^ mo, Y.; mou, LB. * in domuin, Y.; in domain, LB. * tonn 

dodechaid dond fairgi aniar, Y. ^ co torchnir, Y. ' condema, Y. 

• dir, Y.; dirge, LB. ' nac, Y. and LB.; niaoc, L. * Noe, Y. 

« fis, LB. w fohfeibUb, Y.; fohuiUb, L.; fo&ibUb, LB. " Roloiaoeth 
a leth, Y. *' gl&ni, L. ; glain, Y. ; glfiine, LB. *' anairenisin, L. ; 

innairesin, Y. ^* guine, L. ; guin, i . and LB. ^^ oc, Y. and LB. 

!• iBsna goib, Y. 



OF OORMAC'S GL068ARY. 17t 

JTta a sister's son, as said Cdohalainn prophesying of Christ* s 
sdvent : ' The sister's Son of Man will come, and He is Jesus.' 

2($mnuaUf this is the noun that is greatest of the nouns 
of Erin, that is, Nem-nuM the acclamation of the men of. 
He«ven around it. 
2fmu» a place^aame, i.e. nin-foss, i.e. nin ' a wave/ which got 
J^^SB < an ahode,' Le. a sea- wave from the west which came along 
^e air until it settled in yon country and made a well of itself : 
'Whence is said ' Gorcumruaid of Ninus.' 

-^emeth 'a chapel,' 'a sanctuary,' that is nem-iath * heaven -land,' 
"^^Mit is due to the church : neim-dith ' poison-sharp,' what is due 
^ Qoldiers : neim-uath *• poison-horror,' what is due to poets. 

^Hty a god of hattle with the heathen of the Gaels. Nemon, his 
?«. 

Jfacc ' no,' quasi nee. 

-^ith ' the slaying of a human heing.* 

^er ' a wild hoar,' as is in the Metres, to wit, ' the lair of a 

r, the nest of a • • • griffin.' 
^^Vm ' a human heing.' Hence is said, '' If thou sufferest a n6e 
oo. land," that is, if thou sufferest any one on land. 

-^6es ' customary law,' i.e. ndi-fias the wisdom of nine persons, to 
^^^"t:^ three kings, three hishops and three sages, to wit, a sage of 
P^^^try and a sage of history and a sage of the language of the 
*^^nL They were all composing the Senchas Mor. 

• -^imh * a shower,' from nimbus. Hence is said in the Judgments 
^ %h6 Notahles: '^ Oengus under the sparks of the imhass which 
^^^ i&vented, half (of him) humt under heavens of showers." 

-bairns 'pure,' or nairne as if it were na%r$. Old language is 
"^^►t nairne. It is the same as if iieen < indeed,' were said. It 
^ oommon speech even to-day, in West Munster chiefly, whence 
"^^y say, *• Is there aught that is- pleasing to thee?" " Fit naire,^* 
■••ys he who is aisked, that is, " There is indeed." 

«ZVitA ' the slaying of a human heing.' iV^, the man that inflicts 
^^ N9iihe$ then, the slaying of a human heing, as is said, *' Not 
& ^^eapon of slaughter that dignifies." 

^m6it *a boil.' This is a story of the Gaels. When the 

C*^cond) Battle of Moytura was fought, Goibniu the Smith was 

^^ the forge making the weapons for the Tuatha d6a Domnann, 

A^^ Luchtaine the Wright was making shafts for the spears, and 

^^^^^ the Brazier was making the rivets of the spears. More- 



I7i MR. 6TOKE8— ON THE BODLEIAK FRAGMENT 

et dtctmt autem Scoti quod Oobne^ gobae [hjastas &ciebat fri teora 
gr^ssa^ 7 ba feth ' Ingres d6denach.* 

Dognfth dano Luchtaine soer [na ornnda] fri teora snassa, 7 ba 
feth intsnas dMenach. Sfc [faciebat] et Credna nasemmand. 
Dobidgged iarum Ooibne aaintenchair nagoa conglentais isindaor- 
sain. DoUeiced Luchtaine saer nacranna innandCaid/ 7 balor insma 
doib. Dobidged iarum Credene cerd nasemmend ab^laib' natea- 
choire, 7 balor [fo. 81^, col. 1.] insmai doib 7rl. Cene tra buf 
Goibne gobae occondfsin Ifthe bine foramnai. Adfiadarson ^ddiu 
doGobnindy 7 basaeth lais, 7 bahetaid imbi. Issed didiu dog^ni 
fris .i. bui crann innalaim intan adf^ss* do a8c61 .i. ness a 
ainm, ia imbi dognfther indaiimaisi criad. Dicain [bricbtu]^ diifm 
forsaxsrandfiiui, 7 each fer doniced dob^red fuasmaind' donorand. 
Ma adellad iartim imduine dohurgbad foc^toir cnoec ann Idn doloimm 
chru • 7 gur, 7 foloisoed induine amal tene, ^* 7 huaire ba fuath ** 
incraind dianidainm * ness ' Bobid lorsincnucc. Isaire roainpinigset 
ondanmuimsin as nesooit .i. ness [.i. cnoc] 7 scoftt .i. lind. 

Ness tra cetharda fordingair .i. ness c6tamt4t ainm indanmandaL 
7rl. Ness ainm donchrund ut pro/'diximus.' Kess dai>o ainm 
dondaumisi criad fessin," amo/ asb^t ben araile goibenn ^ dorigeni 
marbnaith doa^^ fir, dieens : 

Badirsan lim 8ella[d] [col. 2.] f/iss, 
forbir ifr[ajig derc ** aniss, 
ba binniu nochantais dord 
fri derc aniss adabolg. 

Ness dano ainm donbeimm 7 doncrecht, ut est isint[S]enchas Mar .i« 

agrannib cachtomtM, aFenib gachfortM, ammafnib each mess. 

a dirib cuirp duine, citili afuili,^^ rohordaiged ^^ ness. 

.i. ama/ mbes aurgpatu ^^ inbaill isin duiniu hifuirmither 
increcht isfai ^' dano bfth inderaic .i. u^bi gratia .i. mad indagaid 
no inetan*^ no ismech forrumither increcbt nd inaithiss'^ ismoiti 
aheraic 7 rl. amal rongab i8int[S]enchu« Mar .i. mad fohetach 
immorro bis increcht n6 indainim islugu son.^ 



* Gobniu, Y. * feith, Y. • degenach, L.; in tress gress, Y. * -diaig, L. 
^ agobaib, LB. ^ adfedes, L. ; atfes, Y. ; atoflas, LB. ^ tie^ LB, ^ fuasmad, 
Y.; fuasma ind, LB. * do lind chrCi, Y.; do lindchro, LB. ^^ tenid, LB. 
»* arba h6 fuath, Y. and LB. " fissin, L. *• gobenn, Y.; ^bann, LB. 
^* dia, Y. and LB. ^^ deiro, L. ; derc, Y. ; dercc, LB. ^* oidhat ill fuile, Y. ; 

ciadat He a fuile, LB. ^^ rohairdiged, Y. ; rohainmniged, LB* ^^ urgnacbi, 
LB. >' issai, L. ; isfai, Y. ; as foi, LB. *^ anitan, L. ; inetan, Y. 

'^ foiruimther indathais, Y. '* luga, Y. 



OF CORlCAC's GLOSSARY. 173 

OTer the Scoti say that Gt)ibniii the Smith* used to make spears 
by three processes, and the last process was the finish. Then 
Lnchtaine the Wright would make the shafts by three cuts, and 
the last cut was the finish. So also Credne made the rivets. Then 
Ooibniu used to cast the spear-heads out of the tongs, so that 
they stock in the door-post. Luchtaine the Wright used to cast 
the spear-shafts out after them, and this was enough to. set them 
in. Then Credne the Brazier would fling the rivets from the 
jaws of the tongs, and this was enough to insert them. Kow 
while Goibniu was at that business, a crime is charged, against 
his wife. That then is told to Goibniu, and it was grievous to him, 
and he grew jealous thereat. This is what he did in consequence. 
When the tidings were told him there was a pole in his haod, 
Ness is its name, and round it the furnace of clay is made, and 
he sings spells over that pole, and to every man who came to 
him, he would give a blow with .the pole. If, then, the man 
escaped, a lump full of gory liquid and matter would rise and 
bum the man like fire. Ahd because the form of the pole called 
yius used to be on the lump, therefore they called it ne9''Se6it from 
that name, to wit, ness ' a lump,' and scdit * a liquid.' 
. JVtfM, then, means four things, to wit : . 

J^e9s, first, the name of the animal (the weasel), etc. 

JV«M, a name for the pole, as we have said before. 

iVtfM, then, a name for the furnace ol clay itself, as said the 
wife of a certain smith who made an elegy for her husband, saying : 

* 'Tis sad to me to look at him : 
The red flame of his furnace grows into the wall : 
Very sweet was the bass which his two bellows 
Would chant to the hole of his furnace.' 

^e$$, then, is the name for the blow and for the wound, as is 
(said) in the Senchas Mor : 

From grains every measurement, from the F6ni every law, from 

treasures every appraisement, 
From the fines for a man*s body, though many be his wounds, the 

n€9s was graduated. 

That is, according to the dignity of the member in the person on 
whom the wound is inflicted, so then is the eric. For example, 
if the wound or the insult is inflicted on the face, or the forehead, 
or the chin, the greater is its eric, etc., as is in the Senchas Mor. 
If. however, the wound or the blemish is under the clothing the 
eric is less. 



174 HR. STOKES^ON THE BODLEIAN FRAOMFIVT 

[011am .i.] oUdam .i. oil addm .i. cethrar ar . xx . Alit^ [ollam 
.i.] ollhuam .i. am[ail]a8 nannsae saigid huame bis inaild sic 
isdoiFb^ saigid for d&a 7 ei[c]8i indollaman. Ailit^ [oUam a.] 
olldieim .i. oil anf dieim.* 

(yi .i. caoru,' unde dieitur oisc .i. 6i seise. 

Omelgg * .i. 6i melg .i. isi aimser insiii hitic ass cairach ' melgg 
[col. 3.] .1. ass, arindi melgair.' 

O'en qtkisi (m, ab eo quod est \mu». 

Oenach .1. une^ ech. 

Oeth .i. luige." 

Oar .i. gath.* 
' O'ech .i. namae. 

Orb .i. nomen oiii a quo Oirbrige.^* 

Og quasi ob, id est ouum.^^ 

Ordd .i. ab ordine. 

Olohobor .i. occobOT laiss 61. 

Ornn " .i. orguin." 

Om in bid, gr^co truaillnide insiii .i. omon isinGr^tcc .i. craor 
isinLaitin. 

Opar " .i. ab op^ratione. 

Ong fochid ^ 7 cose : inde drc^ton est eossaig domo^, aFitbil, 
cotreseat** aoing. 

Oraitt .i. 6ratio.*' 

Oslugud .i. huasleoud .i. lecud suas nacomlad. 

Ogt^ch ^^ .i. ogtbech .i. ogude atheeh diabuith fair. 

O'r quit si aur .i. ab auro. : 

Orec treith .i. nomen domoc Hg, trfath enim rex nocatur [unde 
dixit poeta :] Oenacb nuirco treith .i. biad 7 6tach logmar, cluim 
7 coilcid, [cuirm 7 c^ma **] 7 brannub 7 fithohell, hech 7 carput * 
7 mClchoin 7 6isrechta olchena. 

Orcc dano ainm do brattan. Inde dixtY cend Lomnai druith 
iamabein" de : ** Orcc brecc brondfinn bruchtas [fo. 82% col. 1.] 



1 amat/ is doilt^ naim bis fo aill do togail no saigid uaime bis fo aill, sic is 
doilig, Y. ' is oU inni ditness .1. cethrar ar fichit, no oil diemid .i. ollani 
ernes nacesta, Y. ' caora, Y. ; caera, LB. * 6inielc, Y. ; oimelc, LB. 
9 ciirich, L.; is aimser andsin tic as cairach, Y. * blegar, Y. and LB. 

^ aine, Y.; (ine, LB. ^ luidhe no eithech, Y.; luige no dthech, LB. 

» LB. adds no gairm. **> orbroige nominatur, Y. " .i. og, LB. ; .i. 

ogh, Y. ** om, Y. ^* creach no orguin, LB. " ««, L. and Y. ; 

opair, LB. ^* ong .i. foiched, Y. " cot sescait, Y. ; con secait, LB. 

1' ab oratione .i. aimaigthi, Y*. ^^ ogthach, Y. ; ochtach, LB. ^' tic, LB. 
20 branndub 7 fidhcella, eich 7 caipait, Y. '^ iama beimeim, Y. 



OP CORMAC'S GLOSSARY. 175 

Ollam 'professor/ that is oU 'great' his ddm 'retinne,' that 
is, foor-and-twenty men. Aliter ollamy i.e. oU-uam, as it is 
difficult to go to a cave {uam) which is in a cliff {aU) so it is 
hard to attain to the poetry and learning of the Ollam. Aliter 
Ollam, i.e. oll'dmrn, i.e. great [pll\ is what protects {di$im) him| 
(the train of twenty-four persons). 

(yi ' sheep.' Henoe is said diw, i.e. ' a dry ewe.' 

(yimelgg 'the first day of spring,' that is di-melgg 'ewe-milk/ 
that is the time at which the sheep's milk comes. Melgg, i.e, 
milk, because it melgor ' is milked.' 

Oen * one,' quasi ^n, from tcitiM. 

Oenach * a fair/ that is une eoh ' contention of horses.' , 
. Oeth ' an oath.' 

Oar * a voice.' . 

Oeeh * an enemy.' 

Orb, the name of the man from whom are the Orbraige. 

Og ' egg/ quasi oh, i.e. ovum* 

Ord ' rank/ from ordo. 

Olehohar, a man's name, Le. drink {ol) is a desire (aecohor) with 
lim. 

Dm * destruction.' 

Dm ' raw ' of the food. That is Greek corrupted : wfibv in the 
<}reek : eruor [leg. erudum ?] in the Latin. 

Opar ' a work/ from operatio. 

Ong 'tribulation/ and 'chastisement': hence was said, 'Chastise 
thy son, Fithel, till his tribulations ' . . . 

Orait ' prayer,' ' blessing/ from araiio. 

Oilucud ' an opening,' i.e. uas-lecud ' up-letting/ i.e. letting up 
the valve. . 

Ogtach ' ridge-pole,' i.e dg-thech ' perfect house,' i.e. the house 
{tech) is the more perfect {dgu-de) from its being thereon. 

O^r ' gold,' quasi aur from aurum. 

Ore triithf a name for a king's son, for a king is called triath, 
whence a poet said oenaeh oire treith ' a king's son's fair,' to wit, 
food and costly raiment, feathers and quilt, ale and flesh-meat, 
draught-board and chess-board, horse and chariot and grey- 
hounds, and toys besides. 

Ore also is a name for a salmon. Hence said the head of Lomna 
the Pool, after it had been cut off him : " A speckled white-bellied 



J76 MR. STOKES — ON THE BODLEIAN FRAGMENT 

demagur fo moirib nilabar nitoe tuathe toro nadric roimsi 
rorannais raind nad chert Coirpri." 

lade 6dd*u boe anisein dossom. 

Find buaBaiscne, is do robo druth Lomnae midlach. Luid didiu 
Find latbe nand forcnaird selga. Darroaraid^ Lomna hiftM.' 
Bill ben laFind, de Luignib ' di, iror * em cech roilbe 7 cech rofid 
nognathaiged Find conafeind ' . nobith ben aurdaltae arachind as 
each tfr banessom do beos. Batir banbrugid son, 7 batir maithi 
do[im]'fulang nafiann, arnolethad ananae tar[na]tire, arnflamthe^ 
nech olc friu. Docoemnacair' di(fttf fecht nann Find hiTethbae 
co[n]afeinn, 7 luid forcnaird seTga. Domiaraid tra Lomnae hifuss. 
Ambuiside occ imthecht immuig cofamaic Coirpr^ .i. fennid 
doLuignib, illiugn lamnai Find hitdidin. Rogaid didiu inben 
doLomnae adichlid, 7 bas^th laissom bratU Finn. Tie Find iarom. 
Benaid Lomnai didiu ogum hifleisc chetharchuir. - lased [col. 2.] 
bui intiside .i. 

Cuaillne fernae hi felaim* argaitt ath[aba] hifothrocht.*" Fer 
mna druithe druthlach lafeinn ^^ foircthe. hifroech ^ forhualaind 
linim luigi.*' 

Bofitir iartnn Find ascel, 7 ba dognasach donmn^. Kofitir 
didiu inben is oLomnai rofess, 7 docuas '^ huadi coCoirpr^ cotand- 
chisstf^^ coTomaxbad indruth. Dognith son dano iartnn, 7 tallad 
achenn de, 7 tamb^ Coirpr^ laiss.'^ Tic iarom Find dond huarbuid 
deed lai, cofamaic in cholaind cen chenn. " Colunn sunn cen 
chend!" arFind. "Fintae dund," olind fiann "coich incholunn." 
Pob^^ iartim Find ahordaininnabeolu, [7] dicain ^* tre thenm-laido, 
condepeirt : '* Nfconrubai . doini niconarlaig nicontopaig nais 
nicodergraig " n{corubai thorcc nicotorgro^ niconarbairt '^ alige 
Lomnai. Goknn Lomnai inso," olFinn, ''7 rohucsat *' namait 
acenn de." 

Ticsid** d{naib*^ conaib 7 dosleicit fwsinslicht.'* T6id didiu 
Fiod forsin slicht na noc, 7 rosfuair'' hifastig ocfuiniu heisc 
fmndein,** 7 bae cenn Lomnai forbir hicinn natened.** IKcet 

* Domaraid, Y. and LB. ' ibfus, Y. ' w, Y. ; laignib, L. 

* ar6n, Y. * fein, Y. * in, Y. and LB. ' tarna tirib 

cona lamad, Y. ^ deccomnacuir, Y. ^ feilm, Y. ^^ athaba 

ifothlacht iiolurg .i. forlurg, LB. " fene, Y. '* is fraocfa, Y. " limm 
Inigne, Y.; luimm Itiigne, LB. ^* teit techta, LB. >^ 00 tndchaid aide coro 
marb an druth, Y.; 7 co tall a cenn de combert lais, Y. ^^ 7 dochain, Y^ 
" ni en dergraigi, Y. ** ni con ruba tore ni confomae ni contorgrae ni 
cnrarbairt, Y. »» ronncaat, Y. «> Ticaaid .i. benaid, LB. «i Ticaat dona, Y. 
" for BUcht na nog, Y. » fosfuair, Y. »* for indinin, Y. ** for bir 
ooon teinidy Y. 



OF CORMAC'S GLOSSARY. 177 

salmon that bursts Thou hast portioned a portion that is 

not fair, Coirpre." 

Thus, then, that happened to it. Finn, grandson of Baiscne, had 
as his fool Lomna the Coward. One day Finn went on a hunting 
excursion. Lomna remained at home. Finn had a wife of the 
Luigni, for on every mountain and in every forest that Finn and 
his Fiann used to frequent, there was a certain woman awaiting 
him fi*om the neighbouring country. These were female land- 
holders, and they were good to support the Fiann, for their wealth 
spread over the territories, so that no one durst do evil to them. 
Once, then, it happened that Finn was in Teffia with his Fiann, and 
he went on a hunting excursion. Lomna stayed at home. While 
he was walking outside he found Coirpre, a champion of the 
Luigni, lying secretly with Finn's wife. Then the woman 
entreated Lomna to conceal it, but indeed it was grievous to him 
to betray Finn. Then Finn comes back. Lomna cuts an ogham 
on a four-cornered rod. This is what was therein : ' A stake of 
alder in ... of silver, deadly nightshade in brook-lime. The 
husband of a lewd woman . . . with the well-taught Fiann. 
There is heather on Ualann . . .' Then Finn understood the story, 
and he became disgusted with the woman. So the woman knew 
that it was from Lomna, and she sent to Coirpre that he might 
come and kill the Fool. That then was done, and the Fool's head 
was cut off, and Coirpre took it away with him. Finn, at close 
of day, comes back to the hunting-bothy and found the headless 
body. ' A body here without a head ! ' says Finn. ' Make known 
to us,' say the Fiann, ' whose is the body.' Then Finn put his 
thumb into his mouth, and he chanted by tefm Idtdo * illumination 

of song,' and he said : ' Not 

from Lomna's bed. This is Lomna's body,' says Finn. 'His 
enemies have taken the head from him.' 

They slip the sleuth-hounds and put them on the track. Finn 
goes on the track of the warriors and found them in an empty 
house cooking fish upon a gridiron, and Lomna's head was on 
FhU. Trans. 1891-41-S. 12 



178 MR. STOKES— ON THE BODLEIAN FRAGMENT 

[col. 3.] luoht doralad^ dindin[deoin] rantai Coirpr^ doa' trib 
nonbtiraib, 7 ni tardad dantmfr imb6ola incbinn oLkxiain.' Bagels 
lafiannu.^ Is andsein iartmi asb^ [a cend] iriu : ** Oreo brecc 
bronnfind [bmcbtof di magar fo moirib] inbratan di magor .i. 
iaaed adomnae." 

INlucbt tanaissi daiio doralad' dindinn[eo]m rantai' Coirprtf 
itnriim priori modo, ooclos ni daito, aeenn : ** Rorannais raind 
fonnail^ nathrainn rann^ natbbaig iarmuig medbae mos batoieb 
mir metnail* bit fuaitne finna^^ lib Luigni. ''Cuirid immag 
acend, secbas mi[f jocul dunn/' olCoirpri. Coclos n(, acenn dimnig. 
" Bomecbtatar cletb cbuiri rintb ruirtbecb. Age catba cet ambd 
mescbaitb. bacomme lib moage. ba condimdai biid moag tein 
doalasfaid Luigni laFind didiu. 

[Lasin dodecbaid Finn oucu, conid romarb."] 

Patraicc a patricio : patricitM autem qui ad latiM regis sedet. 

Pell .i. a peUe." 

[Pennait .i. a] pennitentia. 

Peccad .i. ape[c]cato. 

Pattu .i. poitbo .i. p6i coss : is to didiu forruimi inpbattu [fo. 
82^, col. 1.] achossa, ol ni lugu bfs finn fo a demaind oldas fuini 
an lias. 

Propost .i. prq>ost .i. pro^ositiM.^ 

Port .i. a portu.** 

Parcbe .i. a parrucbia. 

Pamn .i. nom^n do blo[a]cb mara. Nf cecb" dfhalt" trarosegar 
inne.^^ Nipmacbdad lanecb cini fiastar can doroicb bloacb dindi 
as pamn, et alia similia. 

Puincne .i. screpul mode inbicbe. isbe insin screpul Gk)idel .i. ofing. 

Pugin [selMn] imbe .i. selMn cemn».^" 

Pafn .i. bairgen, a pdne. Inde est isin Gain Ecbdacb mate 
Lucbtai .i. mo tbri finndni fomgeillsat .i. imm ailt nEocbacb 
ailcbetail gaire deloilig find forscing scailtir codipil promtbair 
pain la pugin puinceimn lasiail oernnacb ^* cennnais coimmilg 
cuicbo bitbbf ootamuicc midligen goss gessen cenosmessed connacb 
innabetbu baa. 



I rolaad, Y. « dia, Y. » diiodain, LB. * la find, Y. 

» dorala, Y. « randais, Y. ' fond naile, Y. ; fofael, LB. * raind, Y. 
» mo mir metail, Y. »<> fianna, Y. " sic, Y. » .i. 6n croicend, Y. 

1' ri.Tnech reniBuidigthe, Y. i* a porta .i. on port, Y. •* thiced, L. ; 

oecn,Y. and LB. ** ndialt, Y.; dlalt, LB. ^^ ni ceoh ndialt tra rosegar 
a inne, Y. ^' Puingind .i. selland imme .i. selland cemae, Y. ; Pinginn salund 
imbe .i. seland cemae innain .i. offaing, LB. ^* oennach, Y. ; cenach, LB. 



OF GORMAC'S GLOSSARY. 179 

'a spit hard by the fire. The first batch that was taken from the 
gridiron, Coirpre distribntes it to his thrice nine men, and not 
a morsel thereof was pnt into the month of the head. That was 
a ffeii (magical prohibition) of the Fiann's. So then spake the 
head to them : ' A speckled white-bellied are, the salmon from 
a • . . this is his . . / 

The second batch, then, that came from the gridiron Coirpre dis- 
tributes it again in the same way. Then was heard somewhat : 
the head (saying) : ' Thou hast portioned a portion . . .' ' Put 
oat the head,' says Coirpre, 'though it be an evil word for 
vs.' Then was heard somewhat : the head from outside (saying) : 
* Jtameehtaiar,* etc. 

With that Finn came up to them and slew them. 

Pdtraie 'Patrick' from PatrieiuB. Fatrieius is one who sits 
beside the king. 

PeU * a covering,' from peUis. 

Pennait ' penance,' from poeniUntta. 

Peeead ' sin,' from peeeatum, 

PaUu ' a hare,' i.e. poi-t6, i.e. pot ' foot,' silently (t6) does the 
hare put its foot down, for not less is hair under its sole than 
on it aboTC. 

Propoit * provost,' i.e. prepostf i.e. praeposttus. 

Port 'a port,' irom parttu, 

Parehe ' parish,' from parochia, 

Pam^ a name for a whale (?) of the sea. Not every syllable 
attains a meaning. Let no one wonder then though he know not 
whence hloaeh applies to a pam, et alia similia. 

Pmnens, a scruple of the notched beam. That is the scruple 
€ff the (Hels, to wit, an ofin^,^ 

Puffin [leg. Puinginn ?] 

Pain ' a cake,' from panis ; hence it is in the Oaire of Echaid 
son of Luchta, " My three," etc. 

^ The afainff .i. screpul of 0*Dayoren, p. 48. 



ISO MR. STOKES— ON THE BODLEIAN FRAGMENT 

Poincemn 6idiu cemn tomais sellae 7 med tomais indiLe .L in med 
inbech. 

[col. 2.] Fend .i. a penna. 

Pertico .i. a p^rtica .i. forrach tomais tfre. 

Pairt .i. a parte. 

Poc quasi pac .i. a pace, ar is ardhe sfda in poc. 

Praind .i. a prandio. 

PruU aidbligud 7 m6tugud,^ nt dixit ingen hid Dulsaini 
inbanl6cer< ^ fri Senchan Torpest hiBlanaind .i. 

immomloiscet ' moddno^ pmll. 

Frisgart iarum int^csine dimuintir Senchain .i. 

incherd mae hui Dulsaine 
oLia[i]cc doTaursaigi TuU. 

ISde tra bui do Senchan anfsein. 

Tochairastar ' fair techt imManaind .i. fecht aniusa do chor 
chuartaindi. coeca eices allin cenmotha* 6ic8iniu. I8[ing] mdrodbu£ 
didiu '' riam imnach naili n6ices samail inchumdaig bdi um 8ench^ 
cid cenmotha a thugain ^ suad 7rl. Arrombuf dech dano do thim- 
thaigib ' ier (sie) nGoidel issed b^rtatar nmpu indeicis chenae. 

INtan tra documlaiset forfairgi 7 dochorsatar aorlunn^^ fritir^ 
atagladastar gilldae Louise a,mal dastai " inandiaid dentir : ' Kom- 
leicidse lib,' ols^. Do^cet huile [col. 3.] ingilla. I^iipn data ^ leo 
iartim alecnd chucu arba fie " leo nibu 6n comadas inna neill.^* 
Arbadochroid ah6cosc. Intan, cetamtM, do[m]b«red nech amer iora. 
etan noth6ged athoesc digur bren [tria chluasaib] foradichulaid : a 
chongrtM chraiche ^ do daramnllach commoithan ada imdaid.^' Ata 
lanech assidcid batar " caib ahinchindi romebdatar trea chlooenn. 
Cuirrithir og luin adfsuil, duibithir 6cc [a drech :] luaithidir 
fiamuin ^ [a fegad] ; baidithir or rinn a fiacla, glassidir bun ouilinn 
ambnn : dalorgain lomchoila : daseirid'' birdai** brecdaba fou. 
Dia tallta de inceirt bui imbi nfbu decmaing di techt forimirghi 
a hoenur, manif uirmithe cloch fuirri, arimbud am£l. Dorriucart 
commor" fri Senchan 7 asbert fris : ' Beatorbachasu ** deitsiu,* 
else, 'olldas ind re foruallach'' forbaeth fil imut' 'Insetir" 

1 m^dognd, L. ; metagad, T. 'in bainlethcerd, T. ; in banlecerd, H. ; 
in bannlicerd, LB. ' imanomloiscet, L. ; imomloiscet, T. ; immomloisoet, H. 
* modeno, L. ; modanoo, H. ^ docortMtair, T. ; dei conMiar, H. ; docorustar, 
LB. • cinmothat, LB. ' Z. repeats didiu, * stuigen, LB. • H. inserts 
flathe. 1^ docorustar a I6i no urland, Y. ^^ an indtMtae, T. ; an industai, H. 
« cfein, LB. " ar fee leo. H. " dia n-elta, LB. " craicei, H. »« imdad, 
H. " sie, Y. ; badhar, L. ; b6tar, H. " luiathidir fiamnm, L. ; loaithitir 
fiamain, Y. ^' daseirr, L. ; da seirr, H. ; da sheirith, Y. ; da serid, LB. 
» birrdai, L. ; birdae, Y. " guth mor, Y. « bem torbachsa, Y. »» He, Y. 
H. and LB. ; forthuallach, L. ** sicY,; Ins^r, H. ; indester, LB. 



^- -^ OF OORMAC'S GLOSSABT. 181 

Puincem^ then, a dish for measuring a ieland,^ and a beam for 
weighing cattle, i,e. the notched-beam. 

Penn *a pen,' trompenfM. 

Feriie, from pertiea, a pole for measuring land. 

Pairi * a part,' from pars. 

P6e * a kiss,' quasi pae^ from pax, for the kiss is a sign of peace. 

Ftaind ' dinner,' from prandium. 

FruUf greatly increasing and magnifying, as said Hua Dulsaine's 
daughter, the female rhymer, to Senchdn Torp^ist in Mann, to urit : 

** My two ears bum me greatly " {prull). 

Then answered the bardling of Senchdn's family : 

*' The artist, son of Hua Dulsaine, 
From Liacc of Taursaig Tull." 

Now that happened to Senchdn thus: It came to pass that 
he went to Mann, an excursion of pleasure, to make a 
circuit therein. Fifty poets were his retinue beside bardlings. 
Scarcely any poet had worn such a dress as Senchdn had upon 
him, besides his sage's gown, etc. And what was best of the 
men of the Gaels' garments, this the other poets wore about them. 

So when they had put to sea, and set (their) steering-oar to laud, 
a foul-faced gillie caUed after them from the shore as if he were 
mad: ''Let me go with you!" says he. They all look at the 
gillie. They did not like to let him come to them, for they deemed 
that he was not a bird fit for their flock, for his aspect was hideous. 
For, first, when any one would put his finger on his forehead, a 
gush of putrid matter would come through his ears on his poU. A 
canffrua craiche (?) he had over the crown of his head to the gristle 
of his two shoulders. It seemed to every one who looked at him 
that the layers of his brain had broken through his skull. Bounder 
than a blackbird's egg were his two eyes; blacker than death 
his face ; swifter than a fox his glance ; yellower than gold the 
points of his teeth; greener than holly their base; two shins 
bare, slender; two heels spiky, black-speckled under him. If 
the rag that was round him were stript off, it would not be 
hard for it to go on a flitting alone, unless a stone were put upon 
it, because of the abundance of its' lice. He shouted mightily to 
Senchdn, and said to him, 'I should be more profitable to thee 
than the proud and wanton crew that is around thee.' 'Is it 

^ 'a ration of hoaey consiflting of four eggfulB,' aee O'Day. 118, s.y. 



18?. MR. 8TOKE8 — ON THE BODLEIAN FRAGMENT 

latsu/ olSenchan, 'tuidecht iarsind Idith issm ohoracB?' 
' Promthit^' olse. Cingthi iartim iarsiiid luid isinourach, luaithidir 
lochaid iar forga[r]muiii, combui isincuraoh. Basuail tra corroib- 
[fo. 83*, col. 1.] -thide' in curach cona Inoht arindi rombrocbsat * 
indeicis riasom ^ asmdala leith indleBtair isBalleth* naill, 7 asb^rtatar 
amo/ bid asind o6ngin : ' Totro[r]pai * peist, aSenchan ! fifd si 
dano domuinter buile acht conerUm ^ dochom tfri." 

Is disein rohainnmiged dosom Senohan Torp6ist .i. Sencban 
dororpai peist.^ 

Recat iamm iManaind. Dofocbat* acoblacb bitfr. Ambatar 
didiu bitir ocimthecbt iarsintr^g conacatar insentaiiidi mongl6itb 
moir forsin'^ camdcc, ocbudin nafemna" 7 inmniitboraid^ olcbenai. 
Airegdai gratai acossa 7 aldma, acbt natbdi^ etacb maith impi, 
7 bui anf 6tb gorta fairri ; 7 ba liacb 611 immorro^ drbasfsi iiibaiil§oerd 
ingen buiDulsaine d6 Muscrugu Lfaoc Tuill acrfcb buaFfgenti. 
Docbuaidside forcuaird H6iienit 7 Alban 7 Manann, 7 ba marb 
amdinter buile. 

[col. 2.] Bui iartim abrdtbair mao biii Dulsaine, cerd amrae 
sidbe, oca biarmoracbt focricba H^irMiit, 7 nfsfuair 7rl. 

INtan die^f tf atcbonnairc ^^ intsentainne iiinabe{c8i[u] immosooem- 
orcair^' ceptar b6. Asb^ iartiin araili dfb : 'Maitb in re^' 
immidcommairc. Sencban beices H6ir0nit buiH [inso]' 7rl. 
'INnotbia de bumeldoit? a Sencbdin,' olsi, 'anad rommatb- 
escsa ? ' " ' Bomb{a 6m/ olSencb^. 

Kipsa eola imnid odbaig " 
ceso femmuin mbolgaig mbung. 

* Gate a letb-cbomorgg ? ' ^' Socbtaid iarum Sencban 7 indeicis *^ 
buili dano. Doling lasodain ingilla remep^he arbeolu Sencbdin 
anasb^: 'TV^ acbaillecb! nabacaldai Sencbdn,*" nf oomadas deitt. 
Atomglaitese amne, ol nitaicelladar'^ nacbaile dim muintirse.' ' Cid 
dddiu ? ' ol in banleicerd, ' cate a letbrand ? ' ^ Ki anse^^ olse : 

De muin carrci mara Manann 
doronad mor saland sund. 

^ promfit, T.; promfid, H.; proimfimit fris, LB. * corroibdithe, T.; naoh 
belte, LB. ' robVoccsat, I^B. ^ roimenum. ^ <i«, H. ; isalleith, L. 

^ dotrorbai, T. ; dotrobai, H. ^ coroisium, LB. ^ T. adds no paist. 

• fonacbat, T. ^^ ianm, Y. " na femnmge, Y. " mwrthoracbt, Y. 
19 nadmbui, H. ^^ atcondairc, H. ^^ immoecoemcorcuir, L. ; immiMCSDmorcair, 
H. ^* niath ire, L. ; ni athire, H. ; matb ire, Y. " £nm aitheecsa, Y. ; frim 
aithesxa, H. ^^ adpaig, L. ; odbai^h, Y. ; adbaig, H. and LB. The ri^ht 
reading must be odbaig ^ rhyming with mbolgaig. >' lethcomarc, Y. ; 

letbcomrac, LB. ^ inddices, L. '> asbert sta, Y. ; asbertta, L. ; antMb^ftai, 
H. ^ niraga U-leth Senchain, LB. ^ olnidaculdadhar, H. 



OF COBMAC'S GLOSSARY. 183 

possible for thee/ said Senchdn, ' to come along the steering-oar 
into the boat?' 'I will try it/ says he. So he goes along the 
steering-oar into the boat, swifter than a shuttle along a weaver's 
beam, till he was in the boat. Now the boat with its crew was 
nearly swamped, because the poets pressed before him from the one 
side of the vessel to the other. And they said, as it were with one 
mouth: ''A monster hath come to thee, Sench^, and it will 
be thy whole family, provided we escape to land." 

Hence he was named Senchan Torpeist, that is, Sench^ to 
whom a monster (pdist) hath come. 

Then they reach Mann and haul their boat on land. Now 
while they were ashore, going along the beach, they saw an 
old woman, grey-haired and big, on the rock, reaping the sea- 
weed and the other sea-produce. Refined and noble were 
her feet and hands; but there was no goodly raiment around 
her, and upon her was the evil aspect of famine. Now that 
was a pity, for she was the female rhymer, the daughter* of 
Hua Dulsaine of Muskerry, of Liacc Tull, from the territory 
of Hui Fidgente. She had gone on a circuit of Erin and 
Alba and Mann, and all her company was dead. 

Now her brother, the son of Hua Dulsaine, a marvellous artist 
he, was seeking her throughout the territories of Ireland, and found 
her not. 

So when the old woman saw the poets, she asked them who 
they were. Siud a certain one of them: ''Good are those thou 
questionest. This is Senchin, the poet of the whole of Erin/' 
etc. '' Wilt thou be humble enough, Sench^," says she, '* to 
wait and give me an answer ? " "I will, in sooth/' says Senchdn. 
[Then she said :] 

** I have no knowledge of knotty^ tribulation. 
Although it is blistered sea-weed that I reap." 
''What is its corresponding half-quatrain?" Then Senchdn is 
silent and all the other poets. But then the aforesaid gillie springs 
before Senchdn and said : " Hearken, hag ! address not Senchdn : 
'tis not fitting for thee: but address me, for no other of my 
family holds speech with thee." " WeU, then," says the female 
rhymer, "what is this other half -quatrain?" "Not hard," saith 
he: 

" From the surface of the crag of the Sea of Mann 

Much salt hath been made here." 

^ i,e. dark and hard, like a knot in a tree. 



184 MR. STOKES — ON THE BODLEIAN FRAGMENT 

'' IS£[r," old, 7 mletbrann bfain daiio .i. 
immomloiscet modi nao prall. 

Cate aletbrann, adenchain, beos?" [col. 8.] ''Amin," olingillay 
''oosaigid deit acallma Senchain. nitacelladar 6m." **CeBt diifti, 
oate latsom ? oKsseqi.^ Olsessom : 

« 

inoexd moo Hui Dulsaine. 
oliac do Tairsaige Tall." 

''Ffr son," olSenchan. ** Intussu ingen Hui Dulsaine inbanl6icerd 
occa tdthar chninchid sethno ' H6ir«}in ? " 

'' ISme 6m," olsf. Fothruicther iartim laSenchdn,' 7 dob#rar 
dechelt^ amra' impe, 7 dodeochatJ laSencdn inHeinnn.' INtan, 
tra, donanic Benchan inH6irtnii^ conaccatar ingillai remeperthe,^ 
ba b6clacb comoing 6rbuidi caissidir carra mennorott. Tlacbt 
Hgdba imme : delb isairegdam ' atoess for duine fair. 

Dotboet dessiul Senchain 7 amuintere, et nusquam apparuit 
ex illo tempore. Dubium ita^n^ non est quod ille poematds erat 
sptrt^s. 

Rectaire .i. rettor [leg. rector] a rege. 

Boss .i. trede fordingair .i. ross fidbuidi 7 ross Hn 7 ross bnisci. 
sain didiu accuiss as(ro)bainmniged cacbae .i. ross fidbuide, cetamiw 
.i. roi^** OSS : ross lin dano, robds : ross huisoi dano rofoss son, [fo. 83^ 
col. 1.] oir n(bi " acbt for marb-boisci. 

Eemm ainm do fuirseoir fobitb cecb fuirseoracbta 7 cdcb 
riasta[r]da dob^ foraaigid. 

Bintaid ainm dofir aercbaid " linnas cacb naigid. 

Rout .i. reset .i. mo oldas s6t, semitta son unius animalis. Atdtt 
tra ilanmann forsnaconairib .i. B6t 7 rout 7 ramut 7 slige 7 lamrottea 
7 tuagrotea 7 botbur. 

s6t cetamtM ut pr^diximtM. 

rout .1. ddcbubat carbat 7 daoeneohda imbi.'' doronad ^ibecbraite 
mennato immedon.'* 

ramut .i. [mo] oldas rout .i. aurscur bis fordun rig. cecb 
comaitbecb asa tir dodroig ^^ dlegar do a glanad. 



1 olsi, T. and H. * cmngidh seohnon, T. ' T. adds dl. * ^tach, LB. 

* dobret dechelt n-amra, T. * docom neirenn, Y. "* tangatar docnm Derend, T. 

* nemeperthe, L. ; remepertae, T. * is aireedai, L. ; is airechdam, T. *® HCy 
LB. ; reoi, T. ; raf, L. *^ nirbi, L. '' aorcbaid, T.; aerad, LB. ; fsBSch&id, L. 
^* dacubat caipat / daonecbdae imme, Y. ; dacumat n6 dachuat carpal dooenacb 
dae imme, LB. Tbe text here is corrupt, and my Teiaion is a mere gnesa. 
^* armedon, Y. ^^ dotroich, Y. ; dor6 chuige, L. 



OF OORMAC's GLOSSARY. 185 

'' That is true," saith she, ** and tliis half-quatrain moreover : 



Mj two ears bnm me greatlj. 

What, Sench^, is its half -quatrain also ? " 

"Verily," says the gillie, ''thou art attempting to converse 
with Senchdn; he holds no speech with thee." ''What then," 
Baith she, " what is it according to thee ? " Saith the gillie : 

*' The artist, son of Hua Dulsaine, 
From liacc of Taursaig Tull." 

" That is true," saith Senchdn. ** Art thou the daughter of Hua 
pulsaine, the female rhymer, for whom there is searching through- 
out Ireland ? " 

''I am, indeed," saith she. Then she is washed by Senchdn, 
and a wonderful dress is put upon her, and she went with Senchan 
to Ireland. Now, when Senchdn arrived in Ireland, they saw 
the aforesaid gillie, that he was a young hero with golden-yellow 
hair curlier than cross-trees of small harps : royal raiment he 
wore, and his form was the noblest that hath been seen on a 
human being. 

He went right-hand-wise round Senchdn and lus people, and 
thenceforward he never appeared. It is not, therefore, doubtful 
that he was the Spirit of Poetry. 

Reehtaire * a steward,' i.e. rector^ from rex. 

RoMy three things it means : rou ' a wood,' and ro9B ' flax-seed,' 
and ross of water ('duckweed'). The cause for which each of 
them was so named is different. Ross * a wood,' in the first place, 
that is roe-OSS an abode (?) of deer ; ross * flax-seed,' then ro-dss 
* great growth ; ' ross of water, ro-foss that is * great rest,' for it 
(duckweed) is only found on stagnant water. 

Bemm, a name for a buffoon, because of every buffoonery and 
every distortion which he puts on his face. 

Rintaidf a name for a satirical (?) man who wounds every face. 

R6tU 'a road,' i.e. ro-set ' a very great path,' i.e. greater than 
a sH, the path, that, of one animal. Now there are many names for 
the ways, to wit, sit and r6utf and rdmut, and sli^e, and lam' 
r6tae, and tuag-rdtae, and hdthar. 

sHy in the first place, as we have said : 

r6tdy two lengths of chariots and two one-horse vehicles (?) on it : 
it was made for the horses of a mansion in the middle. 

rdmtUy i.e. greater than a rduty an area which is in front of 
the King's fort. Every neighbour whose land reaches it is bound 
to cleanse it. 



186 MR. 8TOKE9-— ON THE BODLEIAN FRAGMENT 

sligi daMO doscuat^ carpait sech innaile, dorronad fri himcomrac* 
dacarpat .i. carpat rig 7 carpat epwoip arandig cechtar dib sech 
araile.' 

laimroitea .i. et^ d& sligi .i. • sligi tar tuaisoexd mendato alaile 
tar deiscerd frilessu frioui doronad. 

taagroitea focren fer trebar conair do ascnam raitea n6 sleibe. 
issi iaEtfin afoohraico anamin each dfne * immidret cachala blioJain. 

bothur, talla daboin fair, ala[n]ai forfot, alaili fortarsnn, aratalla 
allaegu 7 angamnu an[a fjarrad,' armad inna ndiaid beit iurthus ^ 
inbo bias diheiss. 

Attat tri glantae docachae. Teora aimsera inglanaiter ^ .i. 
aimser [col. 2.] ecb-ruathair 7 amser chue 7 amser cboctha. Itte " 
a tri nglantae,' glanad afeda 7 ahuisci 7 acoclaid. 

IThe aicsin '^ aranglanaiter .i. ama heillnet acarptea ^^ ocdul for 
cuiy" amahuallnet ^' ecbraide 00 techt do oenuch. A choclaid 
arnahesarlaither nech fair octecht fotbressu.'^ 

B^ .i. griiCf reoi mim gr#ce gelu inttfrpr#tatur. 

Einone .i. quasi qoinqutf. Inde dieilur Fercbes mac Mosecbess 
dtxtt intan adrfmed Einn hua Baiscni cecb coiciur ahuair^' de 
filuag Luig[decb] mate N6it^' dochuincbid iiid[f]6indedo.i. Ferchess. 
Lassin didiu adacbt Fercbiss tren foacbnamai secb Find, 7 doUeici 
inslig for Luigid ^^ conid romarb, 7 asb^ occo : *' Bincne caiiincne 
ris ^ riig " .i. arbabed asb^ed ^* Fimi beos adrfmed each coiciur 
arbuair.'^ Rincne quasi quinque. 

Bobutb quasi rebutb .i. rembu[b]tbad." 

Ketgblu .i. ret gle .i. ar[a]sailBi. 

Rotb .i. a rota.** 

Kucbt .i. inar, ut Ferebertne dtxt^ .i. [ro ir dam] deeb rucbtu 
dirggu. 

Eudratb .i. roduratb. 

Euam quasi rom a Eoma. 

Eane iBaed a[B]8eiss0i cen61 namaili. Eane didiu it6 na bussin ^ 
arda. 



1 doecuet, T. doscuchad, LB. ' himcomarac, L. himcomarc, LB. 

' He, T. arandiche hi Bechindili, L. ; condechaid cachae dib sech araile, LB. 
* anam each dine no each mil, Y. ^ f or a tallut a laeig n6 angfrmna inafail, LB. 
8 inrrns, L. inrthaas, T. iurtais, LB. '' T. inaerU teorae tucaite aranglanaiter. 
8 Iteat, Y. » elanta, Y. »<> achnifl, Y. " cairpthiu, Y. « cai, Y. 
" ama hnilled, Y. " fotaressae, Y. fothrew, LB. " Intan boi 

find na baiscni oc airim each coicir amnair, Y. '* niadh, Y. ; con, LB. 

" Inidig, L. ; Ingaid, Y. " ms, Y. " arbaheth atbeired, Y. » anair, Y. 
** Robnth quasi remfuath no robudh di^fttf .i. rembnptadh bioe, Y. " .i. on 
cnairt) Y. ; a cnairt, LB. ^ hnsine, LB. huisin, Y. 



OF OORHAC'S GLOSSARY. IS*^ 

Jilig$^ then, chariots pass b j each other on it : it was made for 
the meeting of two chariots, to wit, a king's chariot and a bishop's 
chariot, so that each of them may go bj the other. 

IdnM^toBy i.e. between two highroads, i.e. a highroad to the 
iiortb of a mansion and another to the south: for garths . . . 
• • • it was made. 

^^ui^-rotae, a husbandman buys a path to get at a pasture 
or li mountain. This then is its price, • ... for every lamb 
^t passes it every second year. 

^ihar, two cows fit on it, one lengthwise, the other athwart, 
tha^ their calves and their yearlings may fit on it along with them, 
heai^e them ; for if they were behind them the following cow would 
|orie them. 

Por each of them (the ways) there are three cleansings, (and) 
"^'^^^ times at which tiiey are cleansed, to wit, time of horse-racing, 
^^ time of winter, and time of warfare. These are its three 
^^^^^^Ksings : cleansing of its wood, and of its water, and of its 



^I^ese are the causes for which they are cleansed, that their 
ch&x-xots may not be soiled in going -on the way, that horses 
^'^y not be ... in going to a fair. From weeds that no one 
I'^y^ he delayed (?) on it when going ... 

^^o * frost,' i.e. Greek, for reo (/>«7o«) in Greek is interpreted ^elui 

^^nene, quasi quinque. Hence said Ferches, son of MoSechess, 
whex^ Finn grandson of Baiscne was counting every five in turn 
of t^e host of Lugaid, the son of Mac-neit, to seek the champion 
Ferolies. With that Ferches gave . < . . past Finn and cast the 
>peaar on Lugaid and killed him, and said thereat Rinene eairincne 
rit (leg. ru9 ?) riif, for that is what Finn used to say when he was 
nun^l>enng every pentad in turn, Rinene quasi quinqw^ 

StfSuih 'a warning,' quasi rebu^, i.e. rem-hubthad 'a fore- 
threatening.' 

S^t^l^ < g star,' i.e. tH gli * a bright thing,' because of its 
l^ri^^tueas. 

•R^tA * a wheel,' from rota. 

-R^^^ht « tunic,' as Ferchertne said, ' He gave me ten red tunics.' 

'^^^^ath ' period of prescription,' that is roHirath * very great 
daratiot^> 

' a cemetery,' from Roma. 
-> this is the sixth kind of baldness. Rane^ then, here the 
*«°^P^%a are high. 



188 MR. STOKES— -ON THE BODLEIAN FRAGMENT 

Bach daiio, ishe inrout namcdli otha intedan ^ corrice amullach. 

Bomaili .i. et^ ind& b&o huile. 

Saal tria assa .i. assa firm- [col. 8.] -ullaig' maeltairside combi 
achaisi buidi inamullacb amal sail duine tria assa. 

Bude-reid ' dan^ .i. mael acbenn buili isidu combf gl61om. 

IMspeilp ^ daiio .i. bfd folt for cectar indalethcbend 7 bid inrout 
r6id otha int6dan corrice claiss ind£chulad.^ 

Ite insein se ciiiolai namdili. 

Eigan .i. a regfna, u$l apui Scotos r( rfgan huad, 

Bath ab eo quod est rata latine. 

Euam .i. luss dobeir cucht su^^r faciem combi derg, xinde dieUur 
ruamnad.^ 

Eoscad .i. ro scad .i. roinnsced de d^ochaid inord insce ' .i. scatli 
•i. insce. 

Eelic .i. a relicis saficforom. 

Bop .i. rap. rop " dii^ui nach nanmanna gonas, nt stmt uaccae. 
Bap immorro nach nanmanna dosrenga [chugae'], ut sunt sues, 
wd tanun uicissim community dicuntur. Bap diJttc ab eo quod 
est rapio. Bop Yero ab eo qitod est robusttM. 

Bibar .i. criathar. 

Biss .i. sc61, BiBsi .i. scela. Inde Coirpr^ mac Ethna dixit 
isin c^^na iii doronad. inHere .i. cendil dam rissi ro[bj8en 
Brisse. 

Buad rofessa .1. nomen donDagdu. 

Bomnae aiss .i. lias 7 buidetu. 

SanctBrigit .i. noebBrigit.^^ 

Suil quasi sol, aristrithi ata soilsi do duine. Suilsi ab eo quod 
est suil. SoltM .i. a sole.^^ 

Sobraig .i. a sobrio.^' 

Senod .i. a senodo.^' 

Salonn .i. sal ond .i. cloch sdili.^* Sal xinde est sale. 

Sanas .i. sain As. 

Segamla .i. blichtmaire. seg didiu .i. blicht. Inde dieitur isna- 
Brethaib Neimed .i. meser bu ara segamla.*' 



I indetan, T. * firmnlliig, T. ; firmhullach, LB. * Buge raid, L. ; 
Bugereth, T. ; Buide reid, LB. * Imspelip, T. ^ conice acb6I, LB. * diamuitt 
nianaidb, T. ' roindsciged J. doaeocDAidh innord nindsci, T. ; r6mdBce .i. 
dodhechaid in ord insce, LB. " m«, T. ; pop, L. * 9ie, T. ^^ tic, T. ; 
noembrint, L. ^^ .i. on gr^in, T. *' a sobrio .i. on snbacbus, T. 

>^ Byno£, T. '* cloch sail nnde dicitor saile, T. ^ T. adds .i. ar a 

mbhcbtmaire. 



OF OORMAC'S GLOSSARY. 189 

rachf also, this is the road of baldness from the forehead as far 
as the crown. 

romdik ' very great baldness,' to wit, all between the two ears. 

9dl tri (usaf ' heel through hose.' From his very crown he is 
bald, so that his yeUow caise (?) is in his crown like a man's heel 
through his hose. 

bude^ad, 'yellow-smooth,' then. Bald is the whole of his 
head in him, so that he is completely bare. 

tmspeilp, then, there is hair on each of the two half-heads, and there 
is the smooth road from the forehead to the hollow of the poll. 

Those are the six kinds of baldness. 

Eigam 'queen,' from retina: Tel apud Scotos ri 'king,' rl^an 
from it. 

Sdth * a surety,' from rata in Latin. 

Euam, a herb that puts a colour on the face so that it is red, 
whence is said ruamnad. 

RoMcad 'maxim,' i.e. ro-seady i.e. ro-inseed. It went into the 
order of words, i.e. seath ' a word.' 

JSsiic ' a burial-place,' from the rtliquiae^ ' relics,' of the saints. 

£op and rap. Hop is every animal that gores, such as kioe : 
r^p, however, is every animal that drags to it, sed tamen vicissim 
communiter dicuntur. Bap, then, from rapw, but rop from rohustus. 

Rihar ' a sieve.' 

Biu ' a tale,' rim ' tales.' Hence Coirpre, son of Ethna, said in 
the first satire that was composed in Ireland, ' Not to pay story- 
tellers, may that be Bress' fate.' 

Ruad rofusa * lord of vast knowledge,' a name for the Dagdae. 

£ow%na0 aiss, to wit, greyness and yellowness. 

Sonet Brigity i.e. Saint Brigit. 

Siitl ' eye,' from tol * the sun,' for through it is light to man. 
SoiUe light, from Siiil. Soku ' manifest,' from soL 

Sohraig from sobrtm, 

Senod ' an assembly,' from iynodus. 

Sakmn ' salt,' i.e. sal-ond, i.e. a stone of salt. Sal, whence is 
saile 'brine.' 

Smuu * a secret,' ( 'an etymologicnm '?) i.e. sain^Jus ' rare knowledge.' 

Seganda 'milkiness,' Beg then is milk. Hence is said in the Judg- 
ments of the Notables: ' Thou shalt estimate kine by their milkiness.' 



190 MR. STOKES — ON THE BODLEIAN FRAGMENT 

[8m6r6it .i.] Smerfoaitt [fo. 84% col. 1.] .i. 8iii6r .i. tone 7 
f ude ^ .L fuidell. 

Sirem .L iarnndi sires olucc coUaco in ca|ate et in toto ooipore. 

Serr .i. a serra. 

Snuad .i. folt. 

Secc 7 seccda ondi i8[8]iccas. 

Secnap ' .L second-abb ' .i. seeunduB abbati/ 

Secht qtMsi sept' ab eo quod est septim. 

[Se ab eo quod est sex. 

Spimt ab eo quod est spiritus. 

Spiracnl ab eo quod est spiraculam.] ^ 

Sponc .i. a sponcia.^ 

Sest .i. nomen domedon lie, quasi sext .i. a sexta bora. 

Semind .i. seixn coahinn. 

Screpul qtMsi scripnl, ab eo quod est scripoltM. 

Sceng .i. imda, unde est imsceng .i. both beoc immatiincella 
imdai. Inde dieitur ferr [imscing] adbul " il. 

Sorbb .i. locbt, quasi sord .i. a sordendo.* 

Slabrad quasi slab[a]r-iad .i. iadad ctfmang .i. slabar each ctonang 
7 eslabar cech fairsing. 

Samrad .i. sam isindEbru, sol isindLaitin,^® unde [dieitur] Samson 
' sol eorum.' Samrad didiu riad reithess grfan, is ann ism6 doaitni ^^ 
asuilsi 7 ahairdi. 

Sen ab eo quod est senez. Senser .i. senex 7 fer. 

Sailchoit .i. sail coit. coit .i. caill isin Combreicc. Sailchoit 
didiu ,i. caill m6r dosailchedain roboi ann.^ 

Sadb .i. so adba. 

Sine " qfMsi so^^. 

Serb .i. ingen Scethimdi in druad deChounachtaib. Ish( ros^ 
feda Atha Luain .i. Bron 7 Duba 7 Daurdibeoil^* .i. intan rodalai 
natri dalai ocAth loain fri Cormac Conloinges ^' (.i. Cond naloingse) 
mac Conchobair. 

Sin .i. each cruind, xmde s(n 'hLaie mufn '* .i. epis^iY bai imma- 
bragait [col. 2.] fri forcell firinne .i. intan" nobbed" flrindi 
bafairsing d[i]a braigit. intan " bago bacumang.^* 



1 fnait, T. ' Sechnap, L. ' secnnda, L. * abbate, L. 

* septa, L. * aiCf Y. ' spongia, Y. * adbar, Y.; &dbai, LB. 

'* a soidento .i. on tsalchar, Y. ; a soraido, LB. ^^ sam hebraioe, sol latine, Y. 
^ ' doaithne, L. ; doatne, Y. ^^ do sailchetain and prins, Y. ^' sene, L. ; 
sine, B. ^* daurdebeoil, Y. ; dur 7 dibeoil, LB. ^^ sie, Y. ; Coinloinges, L. 
10 main, Y. " tie, Y. ; antan, L. ^^ atberedh, Y. ^* cumac. LB. 



OF CX)RMAC'8 GLOSSARY. 191 

Smirdit 'embers,' that is smSr-Judit; smer 'fire/ and fuait 
* remnant/ 

Sirem * disease/ because it searches (aires) from place to place in 
the head and in the whole bodj. 

Serr * a reaping-hook/ from terra. 

8n{Md ' hair.' 

8eee * frost,' $eee * dry/ eeetaf seeda ' dried/ from stecue. 

Seendbh * a prior/ i.e. Mcund-ahh, i.e. secundus abbati. 

Secht * seven/ quasi sept, from septem. 

8i ' six/ from sex, 

Spirut * spirit/ from spiritus, 

Splractd * a breatiiing-hole/ from spiraeulum. (Genesis, vii. 22.) 

Spone * sponge/ from apongia, 

8est, a name for mid-day, quasi sext, i.e. from sexta hara, 

Semind ' a rush,' i.e. slender {sdim) to its end (tW). 

Serepul, quasi sertpul, from sortptdus. 

Seeng 'a bed,' whence is im-seing, a little bothy in which a 
bed fits. Hence is said, ' Better is a bed-booth than a . . .' 

Sorb ' a fault,' quasi sord from sordendo. 

Slahrad * a chain,' quasi slahar-iadf i.e. a narrow closing {iadad). 
Slahar is everything narrow, and esslahar everything ample. 

Samrad ' summer,' sam in the Hebrew and sol in the Latin, 
whence is said Samson ' sol eorum.' Samrad then is the course 
which the sun runs. Then («.«. in summer) its brightness shines 
most and its height. 

Sen 'old,' from senex. Senser 'ancestor,' from senex and/trr 'man.' 

Sailchdit (a place-name), i.e. sail-edit, Coit is ' a wood ' in the 
Welsh. Sailchoitf then, a great wood of willows that was there. 

Sadb ' a dwelling ' ? that is, so-adha * a good abode.' 

Sine ' a teat,' quasi snige ' a dropping.' 

Serb, the daughter of Scetheme the wizard of Connaught. She 
it is that planted the trees of Athlone, to wit. Grief and Darkness 
and Hard-Dumb, when she trysted the three trysts at Athlone 
with Cormac Conloinges (t.^. Conn of the Exile) son of Conor. 

Sin, everything round. Whence is Mac moin's sin, i.e. an epistle 
which was about his neck for declaring truth. When he used to 
pass a true judgment the sin was wide for his neck. When the 
judgment was false the stn was narrow. 



102 MR. STOKBS — ON THE BODLEIAN FRAGMENT 

Serrach .i. serr cech nuachell ^ i. cech noccla. Inde dtW^ifr serr 
[cech] ocdam. N6 serrach .i. ser-ech .i. indiaid amathar bfs. 

Scuit .i. a Scota ingen rig Foraind rig Egept[e]. 

Sath .i. biad, xknde dicitur sathech. 

Sanb .i. mac Augaine M6ir, xmde Mag Saiob dicitur. 

86gda .i. cosmail [fri seig 6] arafegi 7 aglicciM.' 

Seng cech segda isinDuil Feda Moir.' 

Subaid .i. subiate {sie : leg. a Bobrietate ?). 

San .i. Tiinget. 

Srol^ .i. soilsi, xmde est apnd Scottos [diu' srol .i.] dies soIm. 

Sopaltair .i. sepnltair .i. a sepultiira .i. relicc nanduneba .i. 
inadnaictis.^ 

Saim .i. cech corait bis et^ da duine ^ nd iter da hech. 

S6d' .i. OSS allaidy nnde eet s^dgoine' .i. fer gona« ossu 
allaid (sie : leg. alltai). 

Sanc[t] ab eo quod est sanctus. 

Sen .i. Kn ingabthar 6oin, wade sen-bretha. 

San each soer ut est isnaBrethaib I^eimed : faircditeir ^® mairo 
mathi maccaib sau socharde." 

Triath .1. rf .i. tremi etha iath.'* 

Tf .i. bratt." 

Tort .i. bairgen .i. nom^ de sono tactum, [inde dicitur] 
tort^ne .i. bairgeneni. 

Toiso .i. uoluntas hominis. each nf iarton aslaind 7 isadlaic ladoini 
is^ atbtfir istoisc dam. 

Trfath dano trede fordingair .i. trfath .i. rf, 7 triath muir, 7 
triath tore. [col. 3.] Deiligtir diiitt innar^mendaib. Triath ri 
didiu tr6ith areim. Triath muir .i. trethan areim. Triath tore 
dano treith[i] areim.'^ 

Tech ab eo quod est tectum. 

Tore quad pore .i. cendiochrtM. 

Tarb quasi taurb .i. a tauro. 

Tethru .i. rf Fomori, ut est isind Immacaldaim in da Thuaru 
.i. eter tri[u]»u Tethrach." 



1 nuallach, LB. • Y. adth 7 ara gabailchi. ' Y. prefixes Segon quasi 
Began .i. tten segdae, ab eo quod est. * sroll, Y. '^ sic, Y. * relec na 




breo ar fuit. '* Y. addt Triath .i. ri, tir-sith a taithmech. Triath muir, 

tir-nath a taithmech. Triath .i. tore, tir-sod a taithmech. ^^ trethrach, L. 



OP OORMAC'S GLOSSARY* 193 

Serrach * a foal.' Serr everything proud (?), everything vehe- 
ment (?). Hence is said: Every band of warriors is mtt^ Or 
Hrraeh i.e. Ur-eeh * heel-horse,' ue. behind his dam he is. 

Seuiit ' the Scots^' from Scota, daughter of King Pharaoh, king 
of Egypt. 

8dth ' food.' Hence is said sdthech ' satiated.' 

Sanh, son of Augaine the Great* Hence is said Magh Sainb 
•the Plain of Sanb.' 

8epda, he is like to a sep (hawk?) for his keenness and his 
acuteness. 

8en^ is everything $epda in the Book of the Great Wood* 

Subaid subiate (?). 

San, i.e. rifriget [read refrigeret ?]. 

8r6l ' sun/ whence is said by the Scots ' diu »r6l,* i.e. Sunday. 

SopaUair^ i.e. 8$pulta%r^ from stptdtura, i.e. the graveyard of the 
plagues, in which they (the heathen) used to be buried. 

Saim, every yoke that is between two persons or between two 
horses. 

Sdd 'a deer,' when is said sid^uine, a man who kills wild 
deer. 

Sand ' holy,' from ianctus. 

Shi * a net in which birds are caught,' whence sin^hretha • bird-* 
net judgments.' 

Sau, every noble, as is in the Judgments of the Notables : ' Let 
good horses be kept by the sons of a noble host.' 

Triath * a king.' Through him {tremi) are the foods {etha) of 
the lands. 

2\ ' a mantle.' 

Tort 'a cake,' a name made from the sound. Hence is said 
torUne ' a cakelet.' 

Toise, one's desire, everything that is pleasing and desirable in 
a person's eyes he says it is a toise, it is a desire, of mine. 

Driath, three things it means, to wit, triath • king,' and triath 
•sea,' and triath 'boar.' They are distinguished by their 
genitives. Triath 'king,' triith is its genitive. Triath 'sea,' 
irethan is its genitive. Triath * boar,' triithi is its genitive. 

Tech ' house,' from tectum. 

Tore, quasi pore, a mutation. 

Tarhh * a bull,' quasi taurhh, from taurus. 

Tethru, a king of the Eomorians, as is in the Dialogue of the 
two Sages : • among Tethru's mighty men.' 

FUl. Itani. 1891-2-a. 13 



\ 



194 MR. STOKBS'-ON THE BODLBIAN FRAGMENT 

Tenlach ab eo quod eit tene lige. 

Tet .i. nomen de Bono foc^tmi. 

Temair .i. Qrno rotniaillned sun .i. teomoro .i. conspicio. 
Temair dii^fti oaohloc aanad airgna decsiiiy un^^ dieitvr temair 
natuathe 7 temair intige. 

Tern cech ndoroha, xmde dmtur temen. 

Tipra quasi topra .i. iarsindi dobmchta [aiBoe as] . n6 tipra .i. 
tepersiu. 

Troech ' .i. ca^h nechtraide .i. each neottir.' 

Turigen .i. H. Torigein (.i. tuili) gein .i. gein tollin [.i. tolin] 
a hi (a8in)daicii«W ^onoenaigter fri fir na screptra. Alitor turigin 
•i. tur i[n]gena .i. tenga. Aliter turigein .i. geia tori.* amal bfs 
tuir mor ocfiilaog thighi 7 illama essi [sic ised in teg in doman 
centarach']. IShf intuir firindi recbta aicni'i. Itte na hillama 
[asin tuirid .i.'] ilchialla 7 ilchonora inbreithemnais 7rl. Aliter 
toirigein .i. rf, ut est isinDuil Koscad 7rL 

1 cech loo as mbi aury^^iaiii deicsi itir mag 7 tech, at, Y. * Troeth, Y. ; 
TrasBthf LB. ; traeth, Book of Lecan, 160^ 2. * .i. gach neuttfr .i. 

nemnechtardea ' not-either-of-them/ Y. [ = nenmechtardha (gl. neutmm), 
IJraicecht, Book of Lecan, fo. 145^ 1]. * quasi gein a tuir, Y. * sic, Y. 
^ tie, Y. Y. adds the quotation: ai tolach fri toirigin tuigethar tnile mar 
miiime * not a hill for a king who ... a great flood of spears ' [muirenn .i. 
slegh, 0*DaT. 105, mairenn, supra, p. 16]. 



OF OORMAC'S GLOSSARY. 195 

Terdaeh * hearth,' from tene ' fire ' and li^e ' bed.' 

Tet ' a string,' a name made from the sound. 

Tnmnr^ Greek was corrupted therein, i.e. teomoro {Oewpcw?), 
that is, conspicio. Temair, then, is every place from which there is 
a remarkable prospect, whence is said the temair of the country, 
and the temair of the house. 

Ttm, everything dark, whence is said temen ' dark grey.' 

Tipra 'a well,' quasi topru, because water bursts {dohruchta) 
out of it, or tipray a dropping {tepeniu), 

TVoethj everything neutral, every neuter. 

7\irigen ' a king,' turt^einf i.e. tuti-gein 'flood-mouth,' i.e. a mouth 
that fills itself, i.e. its truth fills out of nature, so that it is made 
one with the truth of the Scripture. Aliter turigin^ i.e. tur in gma 
' the tower of the mouth,' i.e, a tongue. Aliter turxgein^ i.e. gein 
turt, as there is a great tower supporting a house with many arms 
out of it, so is the house in the present world. This is the tower, 
the truth of the law of nature. These are the many arms out of 
the tower, the many meanings and many ways of judicature, etc. 
Aliter turigein * a king,' as is in the Book of Maxims, etc. 



196 MR. STOKES — ON THE BODLEIAN FRAOME?(T 



NOTES. 

Imboi foroinai or Imhass farosnaf see LIT. 55^ 14, 125^ 9. My 
translatioii of this difficult article is merelj tentative, Imlas, 
LL. 30^ 29. Imbass, H. 3. 18, pp. 70, col. 3, and 635, col. 3« 
Itnhass, gen. imhais, supra, s.v. nimb,^ from ^emhi-hat'to-, cognate 
with the JjAiin fateor. 

fifrostMi, from ^/or-od-suntia, verbal norm /urmnnud, seems cogn* 
with Goth, sunnd^ A.S. sunne. As to cama derg^ see Laws, ii. 
202, where derg-eama is contrasted with eama hruithe * boiled 
flesh.' In tenm Idida 'illumination of song,' we have a derivative 
of ten *fire,' ^tep. 

ZdiohesSf voc. pi. a Idichessa, LIT. 103^ 35. 

Zugnasadf gen. oenach Tdilten cech lugnasaid, LIT. 52* 20. As 
to Lug mac Ethlenn, see LU. 78* 18 and Eevue Celtique, xii. 127. 

Lelap. The lau 4ittle' here and s.v. lautu seems borrowed from 
an Old- Welsh *lau = Ir. lug, Gr. i-Xaxw, Skr. laghu. 

Lemlaeht. lem each maeth, H. 3. 1 8, p. 635<^. The Uith (leg. 
teith f) here cited occurs compounded in Uith-milhi ' smooth-sweet,' 
an epithet for the rivers in Fairyland, LU. 131, 1. 37, and teith-hlaithi 
'smooth-soft,' an epithet for a woman's thighs, H. 2. 16, col. 716. 

Loeh, O'Donovan renders asioilgi laith loehhronna by * prosperous 
is a king of dark secrets' — a signal instance of unlucky guess- 
work. * Aiioilgi is 3rd sg. pres. ind. act. of a verb, ^mIc, cognate 
with Milaicim 'I open,' and arosailether, Ml. 14<^ 15, inoleoth 'an 
opening,' dufuasatlce, Ml. 29^ 10, and the verbal noun tuaalucud 
'an opening': laith is a common word for ale, cognate with 
Com. lad (gl. liquor), Gr. Xaraf , and Lat. latex ; and hronna is the 
ace. pi. of hHtf a fem. nn-stem. 

Laith, from *{p)latif cogn. with vaXijy iroXcytto?, as Jlaith with 
valeo, Ldith * balance,' perh. from ^tldti, cogn. with Gr. rdXatrroy, 
Skr. tuld, Lat. tulij {t)latus. The t of the soundgroup tl seems lost 
also in lucht 'a charge '=0.^. tluith in or maur-dluithruim (gl. 
multo uecte). The grammatical term forsail is thus explained in 

^ And 80 ba rand immais, LL. 187^ 15. fomesa ceard n-eicse donessa cerd 
n-iumuifl, he who reproaches the art of imbass reproaehee the art qf poetry, H. 3. 
18, p. 62, col. 3. 



> 



OF OORMAC'S GLOSSARY. l97 

the XJraicecIit, BB. 318^ 20 : forsail in fuill^ ele doheir cumang 
fedha for in son dia fot, amatl ata sron, slog [leg. sron, slog]. So 
in BB. 328* 44, forsail, nt est sron 7 slog 7 mor. 

Zangfiter. The corresponding loanword in Welsh is lawhethyr^ 
Laws, i. 558, now llowethir in Cardiganshire, according to Rh^s, 
Wehh Fairy Tales, 36. 

Lommand, The ace. sg. tummain is in Mac Conglinne's Vision, 
LB. 213^. The adj. lomm (W. Uwtnm) here mentioned is, according 
to Strachan, from ^lup-mo', cf . O.-Slav. lupiti detrahere, Lith. lupti 
Bchalen, laupiti ranben. So Old-In dmm 'hand,' Ascoli, Gloss. 
pal.-hib. p. xl, seems from *6p'fnm, cogn. with Lat. apere^ aptus^ 
apuei, Skr. t/4p« 

Zeihech 'plaice* ('flounder*?), cogn. with Mid.-High-Oerm. 
vluoder ' flounder.' 

JUhidreaf sg. gen. of fithithairy Laws, ii. 344, 348. dat. do 
fithidir, Laws, ii. 228. 

Ldes, The stem of the verb -luinethar here cited seems in form 
identical with that of frX^pttf from irXvp-i-tif (EZ. vi. 89). Root 
{p)lu. With the tmesis, imme^ toes -luinetharf cf. as* rkna -rUndid, 
Fel. Jan. 12. 

Ifo dehroth. This is spelt mudebroih in the Bk. of Armagh, 6^ 1, 
mudebrod, ibid. 7* 2. 

Mare. In airmneehf cogn. with arha, pi. arhatm, the tn is for 
r. AirgUeh (which I have«not met elsewhere) is derived from 
argetz::zargentum. 

Mueeairhe^ see Laws, iv. 360, 1. 6. 

MiUiud, LU. 97* Cf. oc admilliud ind rfg, LU. 86*. do 
admilliud ind rig^ LU. 87^ fer tuadcdech co suil millethaig, 
LU. 97*. 

Jfilgetan is an honorific portion of an animal ; but ichat portion 
is not known. 

Ifely. The compound melgthem is exemplified in the Amra 
Conroi : flru batar fo meilgtine .i. has. 

Morann, See now as to him Irisehe Texts, 3^ serte, 1. heft, 
pp. 206, 207, 208. The Audaeht Moratnn is in the Book of 
Leinster, pp. 293, 294. With bihda hdts, cf. is bidba b^s (gl. reus 
est mortis), LB. 165« 42, 169^ 59. 

Menad, W. mynawyd, M. Bret. menaueU Etym. obscure. 

MofMchi men is cognate with Ch. Slav, manfa, maniii 'triigen, 
tauschen,' Lith. m6na% * Trug,' Lett mdnis * Gaukler,' as to which 
word see Fick in Bezz. Beitr. ii. 201. 



198 MR. STOKES — ON THE BODLEIAN FRAGMENT 

Methoss * boundary-mark,' cogn. with Skr. mit * post.' The Ir. 
m^de * neck ' (from ^meitio-) and Lat. mita are radically connected. 
As to the twelve kinds of boundary-marks, see Laws, iv. 142. 

Muilenn, The maiUem of LB., which corresponds with the 
muihnd of L. and Y., was rendered by O'Donovan * together' 
— a mere guess. 

MdU. Cogn. perhaps with Lat. mdjdlis ' barrow-hog.' Another 
form, mdta, gen. sg. mdtai .i. mucci, occurs in LTJ. 109* 30. dat. 
sg. matai, Book of Leinster, 118* 48, 50. As to the Bretha 
Kemed, cited here and s.vr. nimby segandae, tauy see Laws, i. 
112, 113, ii. 70. Nemed includes a chieftain, cleric, sage, poet, 
judge and chief artificer. See also O'Don. Supp. s.y. neimheadh. 

Mann, The Sencha here quoted was doubtless one of the Senchas 
mentioned in the Laws, i. 22. The second word of his decree, 
dilib^ is the pi. dat. of dil from ^agli-y cogn. with Goth. agU 
* disgrace ' (Strachan). Ctehsite seems a redupL «-fut. (relative form) 
from '^qik, whence also eicht, a carver or engraver, with which 
Windisch connects Pict%, Pieionea, Pietavi, 

mdly gen. mdil^ from *maglO'By cogn. with Lat. mag-nu-a. An 
Old-Celtic ' Maglus Conomagli filius ' is cited by Becker, Euhn's 
Beitr. iii. 349. Mdl i CUu M&il, LL. 21^. Mdl mac Telbaind, 
LU. 90» 32. 

Mug-itme. In the footnotes marked H. are given the more 
important various readings of the oi^py of this article in Harl. 
5280. 

As to Coirpre Muse, the ancestor of the Muscraige in Tipperary 
and Cork, see LL. 38^ 2. He fought in the battle of Cennfebrat, 
A.n. 186, according to the Four Masters. 

The statements here made as to the power of the Gaels in 
south-western Britain agree with Nennius, and have of late 
years been confirmed by the discovery of some Ogham inscriptions 
in South Wales and Devon, which were certainly the work of 
a Gaelic population. 

As to Glastonbury of the Gaels, see the notes to the Calendar 
of Oengus, Aug. 24 (Glastingbera na nGaedel i ndeisciurt 
Saxan, A.S. Glasstingabyrig). 

For the resuscitation of Glass, son of Cass, see the Tripartite 
Life of Patrick, Rolls edition, p. 122, and the Book of Armagh, 
fo. 14» 2, cited ibid. pp. 324, 325. 

As to the Qiaio Liathain, here said to have settled in Cornwall, 
see Nennius, ed. Stevenson, c, 14, the Irish Hfimnius^ edd, Todd and 



OF CORMAC'S OL06SART. 199 

Herbert, p. 52, and "Rlifs, The Arthurian Legend, 329, whence 
it would seem that the Children of Liathan were in Pembroke- 
shire ('in regione Demetorum '), the peninsnla of Gower {Ouir) 
in Somerset, and Kidwelly ( Cetgueli), in Carmarthen* 

To the British law * every criminal for his crime, etc.,' there seems 

to hare been something similar in Ireland : see a story cited from 

ZeSw Gabdla in Petrie's Tara, p. 195, where a horse is adjudged 

to lose the leg with which he had kicked a boy. The offending 

thing, not its owner, was liable (0. W. Holmes, The Common Lato^ 

P- 10). So under Solon's law : eypayfte ^e xal fiXafirf9 rerpawo^ivu 
^^^^mo¥f iv cp Koi Kvpa SoKOvra wapahovpai xeXevei xXotif rerpain^xet 

^^^fupop.^ By the Roman law of pauperies^^ the owner of the 

Ax^jTmal doing damage had either to surrender it to the injured 

P^x-ty, or to make full compensation, see Ulp. in Dig. ix. 1, 1 ; 

^^s^^tio. Inst. iv. 9. So Eiog Alfred enacts : Gif neat mon 

9^"^nindige, weorpe f neat to honda o^^e fore jTingie.' These 

*K^d other ancient laws, in that they give the owner of the 

P^<2>cant heast the option of making compensation for its crime, 

less primitive than the Celtic and Solonean rules above cited. 

it, the name in LB. of the eulogy brought by the poet to Tadg, 

^ ^n interesting word. The gen. sg. is «aM,^ ace. sg. uith^ pi. gen. 

•*^A Thumeysen, Irische Texte, third series, part i. pp. 127, 128. 

^^^06 the Old-Irish nom. sg. would be wi, which becomes at in 

'^^^Idle-Irish, as drm druid, sui sage, become drdi, »di. With this 

•*•• I venture to connect Gr. vp,po9 for v-/4i/o9 as inrep, inro for inrep, 

v*"«>. CnrtiuB' etymology of v/ii^o9 (v0-^i^o9) is, as Brugmann 

B^o^^rs (Studien, ix. 286), impossible, because the suffix fivo when 

^'^^ to a root ending in a consonant requires a connective vowel ; 

^^^ Bnigmann's own combination with Skr. tyUman 'Band, 

^^^'^eifen,' is questionable, as the v of vp,vo9 is short by nature. 

^'^^ Aesch. Ag. 990 : top i*dp€v Xvpa9 ofiw9 vfAPi/^Set, There 

■^^'iM to have been an Indo-European ^H, * to call, cry, proclaim,' 

''^^b.^nce Ir. *«-i, Gr. V'fipo9 ; but whether the Vedic ueS (uv6 . . . 

y^^^l^Tafkga bhavishyati I proclaim how it will be indeed, Bv. 10, 

^^- 7) is to be referred to it I do not venture to say. 

I ^nt. Solon, 24 (ed. BeUke). 

-I^ this for *p6p . . . cognate with Skr. pdpman^ with which Frohde and 
^^•^kernigel have connected Gr. ♦■^^ from ^ithijuul, ♦wirr^P For Lat. au 
^^^ 4 see Stolz, Lai. Or, 272. The suffix may be dne to confosion with 
'^^^^^'^ * poTerty.* 

* rfc?*^* Laws, etc., ed. Thorpe, i. 78. 
^ uady Corm. s.t. ercne. 



200 MR. 8TOKE9— ON THE BODLEIAN FRAGMENT 

Jfueh ' smoke,* W. mwff, Ann* mMx, and perhaps Qr. fffivxio from 

Mir 'blackbii^/ has been deduced from ^meisro-, cognate not 
only with W. mwydleh (from *fM%aalh(h- ?), but with Germ. metM, 
Cormac's etym. is taken from Isidore, Origg. xii. 7. 69 : Alii 
merolam aiunt uocatam, quia sola volat, quasi mera uolans. So 
also Featas (ed. Miiller, p. 124) : Merum antiqui dicebant solum ; 
undo et ayis m$rula nomen accepit, quod solivaga est et sola 
pascitur. 

Jffiii^<;A=miimeach .i. breg, 'falsehood,' 0*CL But of. the 
nom. pi. mindig a tigh .i. drochdaine no an t-oes deroil, Harl« 
5280, fo. 41\ Cognate are Lat. menda^ mendum, Skr. mrndd* 

Manannan mac lir. See also BB. 258% 11. 48-56. 

NemnuaU* This obscure word is glossed by ceol aingeal 'music 
of angels,' in the Book of Lecan vocabulary. 

N$meth. See Petrie's EoeL Arehiteetur$i 59. As to nem-uath see 
LL. 38^ 88. These absurd etymologies are thus given in H. 2. 16, 
col. 120 : Kemid intan is fri heclais .i. nem-iath .i. iath neme. 
Kerned flathae .i. neim (no nim)-aith .i. aith neime (no nime) for 
armaib inn oesa flathae. Nem^i filed .i. nim*uath [i.] uath nime 
for tengthaib na ^ed — ^the sharpness of poison being on the 
weapons of the nobles, the horror of poison being on the tongues 
of the (satirical) poets I 

N6e. The .quotation is explained in the Book of Lecan, 151^ 1 
=BB. 828i» 43. 

N6m custom^ * customary law,' whence ndtM^e, "LL. 234^, 
and the verb ro-^esegedf LTJ. 90^, nonoisigthe, LL. 106^, is 
obscure. The synonymous n6s is borrowed from W. nawt 'dis- 
position,' which is cogn. with Ir. ^ndi * custom,' and gndth * usual.' 

Ne^eoit, An abridgment of the story of the Second Battle of 
Moytura is now published, with a translation, in the Revue 
Celtique, xii. pp. 56-111. For the part corresponding with Cormao 
see ihid. pp. 92-94. 

. a Math na imehoire^ lit. < out of the lips of the tongs : ' lei from 
*gveilo-9^ cogn. perhaps with Goth. q%\an^ A.S. ewe^an. So Lat. 
labrum with Ir. labraim * I speak.' 

Ollam. Compare Laws, iv. 358. In the earliest times his retinue 
was 24 in number, BB. 333* 1. The ddm here mentioned is a fern, 
d-stem, and, aco. to Windisch, cogn. with Dor. hafio9, 

Oi, The gen. pi. of 6%ee occurs in the Saltair na Rann, 2844 : 
CO tret n-ooisce lacob. 



07 OORMAC^S GL()SSART. 201 

Oimete, So in Harl. 5280, fo. 32% samsnan co hoimelo, ho 
oimelc co beltine, ho beltme co brontrogam 'from the first day 
of winter to the first day of spring, from the first day of spring to 
the first day of summer, from the first day of summer to Lammas.' 

Oech, A.S. ^B'/da * foe,' Goth, faih Betrug, Lith. peikiHf peikti 
flnchen. For-oesad^ p. 18, 1. 14, may be the 2dy «-fut* sg. 3 of a 
cognate verb. 

Om. This seems cognate with Or. ^<9, ^A^79, tlprf and Skr. drush 

* wound,' f-nd'ti. From the same root are ort .i. bds, ort .i. orgain, 
O'Bav. 109, and air: in-air a prim-roit 'for injuring his chief 
road,' E. 3. 5, p. 39^, cited in O'Don. Supp. The W. amest 

* monomachia,' which has been compared, is borrowed from A.S« 
§omMt ' a duel.' 

OekUak. The ace. sg. occurs in LTJ. ld8^ 22 : focheird iarom in 
roth CO hochtaig ind rfgthigi then he east the wheel to the ridgepole 
ef the palace. Is it the same word as the (corrupt) 0.*lr. octijiaf 
(gl. pinus), octhgaeh as aidu alailiu (gl. habies), MS. Lat. 7260 
(Bibl. Nationale), fo. 9^ ? Of. the gloss Ailm dno .i. crann giuis 
a. ochtach, Book of Lecan, 149^ 1=BB. 325» 50. 

Ore triith^ cf. LL. 187^ 49 : i n-oenuch thuirc thr6ith^ .i. i n- 
oenuch mate ind rig .i. cluim 7 cholcid 7rl. 

Ore. The story of Lomna's head is noticed in Ehys' Hihhert 
LeetureSy pp. 98, 99. Bat surely the head's displeasure at not 
being given part of the salmon is nothing but vexation at the 
breach, in its case, of a law of hospitality which bound the Eiann, 
and the ** offering to the Celtic Dis" is mere ingenious guesswork. 
For other tales of a speaking head see LTJ. 98* 35 and Three 
FragmeiUe of Irieh Annals , pp. 44, 46. 

The indeoin which I have rendered tentatively by ' gridiron,' is 
iome kind of cooking apparatus. Hence the verbal noun indeonad 
« to broiL' 

a eenn, pp. 28, 1. 30, 30, 11. 5 and 9. Hence eenn seems to have 
been sometimes neuter. Cf. friscichset for cenna dfb, LTJ. 89* 44, 
and BB. 320^ 3 : hilugud labartha, ut est is ^ a cenn 7 is lia-te 
ind aurlabra. 

conid-ro marh. Here the d {id ?) is an infixed pers. pron. of 3rd 
pL See G.C. 332. 

PaiiUf gen. pi. pattan, LB. 183^ 37, seems a loan-word, cogn. 
with Fr, patte and pataud. See Diez, s.v. patta. So pdi * foot,' 
is a loan from O.Fr. poe (E. Meyer). 

I do thnrc Thomair 'to prince Tomair,* Book of Rights, 206, 3. 



202 MR. STOKES — ON THE BODLEIAN FRAGMENT 

Pdin. The Bretha Echach maic Luchta are mentioned in the 
Laws, i. p. 18, 1. 23. Some of the words in the obscure extract 
here quoted are thus glossed by D. Mac Firbis: form 9eing .i. 
ar mo leaba : pain .i. aran : punoem .i. tomas : huiail (leg. la siail?) 
.i. oileamhain : ^oss ^ .i. g6dh : ^eisen .i. eala. 

F6e: sain-poc 7 pocnat (gl. osoulum), Sg. 46* 2. df laim im 
£tain 7 poc di, ol Midir, LTJ. 131, 1. 24. It is borrowed from an 
oblique case of Lat. pax. 

Frtdl, Of the copy of this article in Harl. 5280, fo. 75% the 
more important various readings are given in the footnotes 
marked H. Ehys {Hihbert LeoturM\ p. 567, note 1, equates prull 
with W. prwystl 'tumultuous.' With the description of the 
monster which appeared to Senchdn cf. the description of the 
piihaeha in Hoemle's edition of the Jaina Uvdtagaddso, pp. 
65-69, translated by Morris, Philolog, 8ac, Froceedingt, April 
15, 1887. Ehys {Mibbert Zeotures, 567) says that Cormac's 
picture of the Spiritm Fo&matu is *' in fact a description of 
a corpse in an advanced stage of decomposition,'' and accounts 
for this by alleging *'the ancient notion that poetry traced its 
origin to the world of the dead, whose king was sometimes given 
the outward appearance and lividness of a corpse." But where, 
is there any evidence of the existence of this *' ancient notion " ? 
And what people had a corpse-like king of the shades? Not 
the Indians, nor the Greeks, nor the Finns. Neither Yama nor 
Hades, nor Tuoni had anything corpse-like about him. Hel, no 
doubt, is described as bid (or h&lf'bld) ; but she was a goddess. 

promthit for pramJU ex protn/a+it, as gebaity LXJ. 87» 45, for 
giba'\'it, 

luaithidir lochaid tar forgarmain ' swifter than a shuttle along 
a weaver's beam * : cf . Job vii. 6 : * My days are swifter than a 
weaver's shuttle.' The primary meaning of the word rendercd- 
' shuttle ' is ' mouse.' The Welsh term for shuttle, gtoennol gweydd, 
lit. 'weaver's swallow,' also recognizes the swiftness of this 
instrument. With for-garmatn cf. lu-garmain, Laws, i. 150, 1. 7, 
and Welsh ear/an, 

setir, 'SitoTf pass. pres. indie, sg. 3 of sitaimf a sister-form of 
fetaim 'possum,' both from an Old-Celtic *itent6 cogn. with Goth. 
svin]^8 ' strong,' w%n\jan xparety, tvin]^nan KparaiovaOai, 

lue 'steering-oar,' dat. luith^ from *lu{p)et, cogn. with Slav. 
lopata * shovel.' 

^ A loan either from A.S. got or O.N. gdi. 



OF CX>KMAO S GLOSSARY. 



203 



Another inatanoe in Irish literature of challenges to give 
corresponding couplets will be found in the Lobar Brecc, p. 85, 
lower margin, where the challenger is the Devil and the person 
challenged is Saint Golomb cille. The practice seems to exist 
in Portugal, where singing ao desqfio is a favourite amusement : 
see Latouche, IVav$U in Portugal^ p. 47. So in India : 9ama»yd 
* the giving to another person part of a stanza, and requiring him 
to complete it * (Benfey). 

odhaig^ sg. ace. fem. of odbaeh^ derived from odb (=>W. oddf 
'tuberculum'), later /a(^, with prothetic/. 

hung (»01d-Geltic ^bongd, later hongaim) is, as E. Meyer first 
saw, the act. pres. ind. sg. 1 of the verb of which hiiain (^ss=*hogni) 
p. 39, 1. 12, is the infinitive, and hoeht 'poor' an old participle 
passive. Skr. ^bhanj, hhandj'mt, * 1 break,' Lith. bangd * wave.' 

Td, cf. Td (.i. dostid) cein, LU. 86». 

The description of the Spirit of Poetry, after he had assumed 
his proper form, is thus given in Y. cols. 75-76 : 

Intan, tra, tangatar docum So when they came to Ireland 
nErend conacatar in gillae they beheld the youth aforesaid, 



remepertae, ar ba hoclaech 
ruithentae, righ[d]ae, romor, 
roiscletan, mormileta eseom, co 
muing orbuidhe orsnaith fath- 
maindigh caisidir carra menn- 
crot. Tlacht rochaom rigdae 
uimbe : milech orduide a n-im- 
dunad in tlachta sin. Sciath 
corcra cobradach cetharochair, 
Ian do gemaib carrmogail 7 liag 
logmar 7 nemthand 7 cristal 7 
sathfire, for a ch'a. Cloidhemh 
coilgdirech, co tairchetlaib [« 
trocheltaib, LL. 231» 47] oir 
deirg, for a deiscib. Cathbarr 
airg[d]idhe co coroin orduighe 
ima cend. Dealb is airechdam 
7 is aidbli luchracht [?] bui 
for duine riamh fair. 



for he was a hero radiant, royal, 
huge, broad-eyed, warrior-like, 
with golden-yellow hair of gold- 
thread, fathmainneeh (?), curlier 
than croBs-treesof small harps. A 
costly, royal robe around him: a 
golden brooch closing that robe. 
A shield purple, bossy, four- 
edged, full of gems of carbuncle 
and precious stones and pearls 
and crystals and sapphires, on 
his left. A sword- straight 
glaive, with inlayings (?) of red 
gold, on his right hand. A 
silvern helmet with a golden 
crown on his head. A shape 

the noblest and grandest 

that had ever been on man was 
on him. 



In the phrase earra menneroit E. Meyer takes earra to be for 
rra, ace. pL of eorr, the curved cross-tree of a harp, see O'Curry, 



$04 MR. STOKEg— ON THE BODLEIAN FRAGMENT 

Manner* and Cuitomi^ iii. 256. Mennerott (=mendcrott, LL. l8^ 
50) seems a compound of mmd * kid ' (cogn. with Alb. ment ' to 
suck,' as dinu * lamb ' with Lat. fi-lare^ Ghr. Oji-aarOf Strachan) 
and crott=W, crwth. O'Curry's benn ehroit (M. and C. iii. 305) 
is a mere figment. 

RoBB, The ace. pi. of rQ99 'a wood/ occurs in LB. 208* 1 : ama/ 
trascraid sloig doch^in r6id rossa do thnagaib rog^ra ' as hosts firom 
afar fell smooth woods with keenest axes.' The gen. sg. of tobb 
' flax-seed/ roi«, occurs in Laws, ii. 368. It is said in the 
ITraicecht (Bk. of Lecan, 143^ 1 = BB. 317» 43) to have been oner 
of the nine components of the Tower of Babel — ^the others being 
clay, water, oil, blood, lime, flax, shittim and bitumen. The ' rou 
of water ' is = ro« laeKan (gl. lenticula aquatica). Rev. Celt. ix. 236. 

du * growth,' is from *{p)dt'Uh^ cogn. with Or. frareofiaif 
a-ira<rro9, and Qoth./ddjan (Strachan). 

Similar explanations are given in the TJraicecht (Leb. Leo. 150* 
1 = BB. 326* 50) : ross .i. roi oss, quando .i. intan, is ros caille, 7 
rass iar lind intan is ros usee .1. rof hos mad for marbus^', no roidh 
ass mad for sruth, 7 rofhas intan is ros Ifn .i. ara luas 7 ar[a]thighe 
f hdsas. 

Remm^ Spelt reimm and glossed by fuirseoir no druth in Laws, 
iv. 354. 

Rout. The whole article is printed and translated by O'Donovan 
in his edition of the Book of Eights, Dublin, 1847, pp. Ivi-lviii. 
As to rdtU and ratnut, see Laws, i. 232, 1. 18. Erom ramut comes 
the adj. ramataeh^ Laws, iii. 112. The description of ilige 
reminds one of a/ia^iTOi 'a carriage-road.' As to the five great 
tligid of Ireland, see LL. 155i> 8. With Idmrotae cf. O'Gl. Gl. 
hmrod .i. rod laimh le rod aile. With the gen. sg. raitea cf. raiU 
.i. cuairt ingelta, O'Don. Supp. Bdthar seems borrowed from an 
Old- Welsh *hautr, V^d, whence also Skr. ji-gdtiy a-gat, Gr. ^-fiav, 
A.S. papy O.H.G. pfad. In amser chu$ (rectius cMtad) we seem to 
have a cognate of W. eawad 'imber,' Com. eouat (gl. nimbus)* 
See H. 3. 18, p. 639^ where eiuidh is glossed by gemred 'winter.' 

Rinene, This article is, to me, unintelligible. The usual 
meaning of rincne is * spear ' : do rindcne .i. do sleigh, Rawl. B. 
488, fo. 4^ 2. 

Rohuth, gen. robaid, LU. 57* 36, 87* 13, 15. pi. dat. rohthaihf 
LL. 57^ 22, seems a compound of ro-=pro and *beuto» from ^gveuto-^ 
oogn. with poFa and 7009, Fiok *• 406. So the Welsh rhyhudd 
' monitio,' comes from ro and ^peud^ *p0ufO', y/qu 'to cry,' Fick* 880# 



OF OORlCAC's OLOS8ART* 20$ 

jRuehi from *ruk'tUf cogn. with Germ. J^ck, RoeJien, A.S. rocc^ 
Low Lat roceus, whence Fr. and Eng. roehet, pronounced roeket 
in the Go. Cork. The quotation is from the Amra Gonroi (H* d< 
18, p. 49): ro ir dam .x. ruchta derga *he gave me ten red 
tunics.' 

JRitam ' cemetery,' from JRdma, See the Tripartite Life, p. 656^ 
coL 2. See also the Divina Commedia, Par. ix. 140 : xxvii. 25. 

Jtane seems horrowed from a British cognate of Lat. runeo, Persius 
4. 36, runeina. For the ahlaut a — u, see Frohde, Bezz. Beitr. 
xvii. 306. The imspeilp here mentioned is a compound of tmm- 
=?=Lat. amh', Gr. d/i0<, and 9psilp horrowed (with prothetio « and 
metathesis of /) from the gen., sg. of Lat. peplum, Speilp (gl. co- 
opertorium), Lr. Gl. 730. Cor' scail in speilp hoi imme, LB. 
160* 42. 

Rath. The Low-Latin rata occurs often in the Canones Hihem« 
Ducange explains it hy ' stipulatioi contractus ; ' hut it seems 
always to mean ' a security ' or * guarantee.' 

Euam : hence ruamadh^ O'Curry, M. and G. iii. 119. 

RoBcady pi. nom.. dotiagait asna foclaih sin . . . roscaid 7 fasaig 
7 aircheadail, L.Lec. 143»» 2==BB. 817»» 30. 

Rihar 'sieve' seems cogn. with Lith. ri-tu^ the suffix heing 
borrowed from Lat. erihrum (krei-dhro-)« So in Idichesi, supra, p. 
8, the suffix is=sLat. -ma, Gr. -unra, 

Riss, pi. n. ail rig risi rede, LL. 187* 37, rissi ruada, LL. 187^ 20. 

Sohraig. Bop sobraig, LL. 343^ 3. 

Se/nind, So semind .i. s^im co hind, LL. 186* 29. 

Sen. tenser is, ace. to Windisch, from *8eniasterO'^ O'sser 

* younger ' and Lat. magie-ter, minie-ter^ are similar formations. 

Sa%Uelmt=8uleh6idf Four Masters, A.n. 1602, p. 2312, is a loan 
from ealieoitum, the Old-British pronunciation of Lat. ealicitum. 
The Welsh eoit here cited is from *^'to-n, whence the Old-British 
eitan in Zito-eitany Luit-coit 'grey- wood,' now Litchfield. See 
Bradley, in the Academy, October 30, 1886. The Goth, haithi 
' a heath,' is cognate. 

Sine is compared by Bugge with Lith. epenye ' teat.' But ^rt- 
phne, LIT. 77* 38^^ri sine, LL. 75^ 21, seems to point to a 
primeval Celtic evenio, Cf. however the redupl. perf. eephaind 
'played (the harp),' which Strachan brings from ^ephendh^ 

* zucken,' whereto Fick ^ 149, refers Skr. pa-spande, Gr. <r0€£ai/o9« 

Serb, As to the Feda, or woods, of Athlone see the Four 
Kasters, ed. O'Donovan, a.d. 1536, p. 1435, note 0. 



206 MR. STOKES — ON OORMAC's GLOSSARY, 

Sin. Se^ now Irisch^ Texte^ 3^ eerie, 1. heft, pp. 188-198, 
where this and eleven other Irish ordeals are described. The 
' epistle ' was got from S. Paul. 

Scuitt, So in the Bk. of Lecan, 152* 2 : Nel mac Feniasa dofnc 
Scotai ingen Eoraind, conid dia hanmaim-sein dogairthear Scuitt 
* Nel son of Eenius wedded Scota Pharaoh's daughter, so that from 
her name the Scots are (so) called.' 

Sopaltair. Soppaltair la Forbraigi, Tripartite Life, p. 250. 

Saintt oogn. with Skr. sarndtitf Gr. ^/io, 6^169, ofiov, Qerm. stf- 
santtn^n. 

Sin 'net* (W. Atryn), from ^seg-no^ ^segh^ whence Gr. ^w. 
To the same root Strachan refers Ir. temmenn 'rivets/ 8.v. nescoit, 
from *8estmen, ^Begh-B-mm, 

Sid, pi. gen. re trichait sedh lurganda, Bk. of Lecan, 149^ 1 =» 
BB. 325^ 

SaUj O'Donovan Supp. takes Bder to be for Boeff and accordingly 
renders Bau by * carpenter.' 

Ti 'mantle,' Hi dubglass co lubain airccit, Egerton 1782, fo. 
37^ 2: pi. n. tii dubglasso, Ir. Texte, 3*« ser. 1. Heft, p. 239, 
1. 136. From *fc)-rMi<>-, ^v$b (Strachan). 

THath * king,' may be cogn. with Lat. Btrft-avuB * an ancestor 
in the sixth degree.' As. to the superstition here referred to, — the 
influence of a righteous king on the seasons and crops, — see the 
Odyssey, xix. 109-114; Horace, Odes, iv. 5, 11. 7, 8; Manu, ix. 
246, 247; and the Irish documents mentioned in the Tripartite 
Life, pp. clx, 507, 670. 

Tethra. The quotation from the Dialogue of the Two Sages 
may be found in LL. 187^ 37. 

Tem^^A.&. \imm. The adj. temen occurs in Bumann's poem in 
praise of the sea. Laud 610, fo. 10*, as an epithet for a wave : cwai 
dutraio tuinn temen cruaid. It also seems to have been a proper 
name, whence Temenrtgi, Book of Armagh, 15* 2. 

Iktrigen, The similarity of tur- with the first syll. of Gr. 
rvpappo9, said to be borrowed from Lydian, is probably accidental. 
The D&il RoBcad * Book of Maxims,' cited here and s.v. A$m, is said 
to have been composed by Cennfaelad, son of Ailill, after his skull 
had been split in the Battle of Moira, a.d. 637, and his ' brain 
of forgetfulness ' {inchinn dertnait) removed : see Laws, iii. 86, 88, 
550. He is also alleged to have been the author of the TJraicecht. 



207 



v.— NOTE ON THE PRONUNCIATION OP THE 
ENGLISH VOWELS IN THE SEVENTEENTH 
CENTURY. By Russell Maktineau, M.A. 

[BMd at the Soeiety't Meeting on Friday^ Nov 6M, 1891.] 

The elder Joannes Buxtorf, the great Hebraist, was bom 
at Eamen in Westfalia in 1564. He was Professor of 
Hebrew at Basle for 38 years, and died in 1629. His son 
of the same name was a worthy son of his father, and 
carried on his work. Several of the father's works, left 
unfinished at his death, were edited by the son. One of 
these appears to be the '' Epitome Grammaticae Hebraeae," 
which, probably owing to the esteem in which the younger 
Buxtorf was held in England, where he resided for some 
time, was published first in London in 1653. This was 
written in Latin ; but at the beginning, where the Hebrew 
▼owels are enumerated, their value is given by citation of 
both Oertnan and English words. These give us interesting 
information respecting the pronunciation of the English 
Yowels in the age before 1629 when the author died, or at 
least before 1653 when the book was published. It might 
indeed be objected that, Buxtorf being a foreigner, his 
estimate of the English vowels is not to be implicitly 
believed. But very fortunately for my argument this 
objection is met by the existence of an English translation 
of this little grammar made by John Davis and published 
in London 1656. The translation omits the German 
instances, but retains all the English ones — except in one 
case where Buxtorf gave two instances, and the translator 
thinks only one necessary. Thus we have native English 
evidence of the time. I shall also show (though it is not 



208 PRONUNCIATION OF THE ENGLISH VOWELS 

the subject of this paper) that the German instances afford 
interesting indications of BuxtorTs German pronunciation , 
showing him to have retained (at least in the vowels) the 
Low-German sounds of his native country of Westfalia, 
and not to have adopted the High-German of Basle, where 
he lived for the greater part of his life. 

I now give the passages (at the beginning of the grammar) 
in which Buxtorf explains the Hebrew vowels/ and the 
corresponding ones in Davis's translation, 

Buxtorf, Efitomb Grammaticab Hebraeae. London, 1653, 

Vocales, 
Longae. Breves. 

Kametz K A obscurum Pathach K A clarum, ut in 

quasi cum mixtum, ut in Fisr^^^rpater, Samlen colligerci 

Germanicis Avend^ vespera, Dach tectum, Angl. art ars. 
Samen semen, Angli salt sal, 
small ^ parvus. 

7W& K E purum et siccum, Segol K E impurum et 

Ehr honor, Lehr doctrina, dilutum, ahr^ arista, Angl. 
Angl. f€e nos» an eare of corn, bed lectus. 



> Of the consonants he gives only three English sounds: <Sajin T> 7, 8 
lenissimnm, Angl. z'; <Caph 3, 20, ch, x Oraecorum, Angl. c'; 'SchinfiT, 
300, sc Tel ax aut sch Gennan. Angl. sh.* 

> This is Frisian ; Dutch is avond. Buxtorf was bom at Kamen in Westfalia^ 
and his Gennan pronunciation here and in Note 4 is Low — not as one might 
imagine from his long residence at Basle eminently High, It is clear that he 

S renounced ftvend. ia. this word the a tends in many dialects towards o, as in. 
wise obed. 

* He means the modem aw (Sw. &) sound in aalt, tmall. It is the modem 
Jewish pronunciation of kametz, and Buxtorf*s teachers #ere Rabbis. Christian 
scholars generally treat the kametz as the ordinary long & of father. 

* Legend, ahr, properly ahre. Engl, eare is meant not as translation of ahre, 
but to show pronunciation of segol, and is retained in the Engl, translation in 
that sense, ft would appear therefore that the vowel of Engl, eare is identical 
with that of Germ, ahre ; but both these examples are puzzlmg, for they would 
seem ratiier equal to 6 Tzere, with the vowel of Ger. ehr, lehr. rerhaps 
Buxtorf had a recondite reason for selecting ahr, ear. The pronunciation of 
the segolate substantive in pauea (which, moreover, is the pronunciation adopted 
in the transliteration of proper names in the Greek and Latin versions) lengthens 

the segol (8} into kametz (a), as in ^^p, pausal ^f n "AiScX Abel. Thus there 



IN THE SEVENTEENTH CENTURY. 



26d 



Langae. 

Chirek longum *K I Ihr vos, 
Ihnen ipsis, Slider [legend. 
Glider] membra, Angl. Alive 
vivus. 

Cholem N 1« Ohr auris, 
8ohn filius, Angl. open apertus, 
ater supra. 

Schurek \ JJ Umer noster, 
Unsehuldig innocens, gesund ^ 
sanos, AngL a lute barbitum, 
testado. 



Breves, 
Chirek breve K t Irren 

« 

errare, Sinn sensus, gelitten 
passus, to live yivere. 

Kametz ehatuph S S Ort 

Tt T 

locus, Son sol, Trott torcular, 
Angl. to trot, succussare. 

Kibbutz « u vel Y, Sunde 
peccatum, verfuhren seducere, 
Angl. but sed, shut^^ clausus. 



Impropria voealis est, quae vel non semper, aut non 
omnium literarum voealis est ; estque Simplex vel Composita, 
Simplex dicitur Sheva . • . valetque E brevissimum, quale 
in berahlt,^ gestrqfft, quod rapide effertur quasi brahlt, gstrafft, 
ut e med. in Angl. beggery. Vulgo vocatur Sheva mobile. 



seems to be an affinity between tbe Hebrew and e^ so that Buxtorf might 
treat it as an umlant like the Genn. a, a. Anyhow, the Kng. bed (his 
other instance^ is uneqniTocal ; the surprising thing is that he should pnt side 
by aide ear ana bed, the vowels of which surely neyer can haye been identical. 

^ The German instances of long H seem faulty, i.e. to be short rather than 
long: unser, unsehuldig (he probably means both vowels to serve as examples), 
gesimd. Happily for us the Eng. lute is unequivocal; for even if that was 
pronounced lyute, the vowel which represents the Heb. shurek is not yu but u. 

' These identifications are interesting ; and, as with shurek, the Eng. instances 
are most obviously correct. But^ shut are clearly to be pronounc^ with the 
orifiiinal short u as in b%ill. The Cferman instances are curious : tunde^ verfuhren. 
It looks as if he pronounced the Heb. ii like Germ, ii with umlaut, though his 
Eng. instances will not lend themselves to that. But the umlaut is not written, 
nor I believe used by Buxtorf in these words. Fiihren had by no means 
generally the umlaut in the earlier times down to at least the middle of the 
lixteentn century, and in the Netherlands we hnve voeren [pron. fiiren]. 
similarly in the other instance : aunde is the Hi^h- German pronunciation ; but 
the Low is in Dutch zonde^ which implies High-Germ, sunde, as D. 5sG. u, 
e.g. ofui^s=unter, ron^fs rand, «;7roii^ = sprang.) 

^ This word berahlt is not easily exphcable, as no verb at all like rahlen is 
known. It seems to be a misprint. 



PML T^mni. 1891-8-8. 



U 



210 



PRONUNCTATION OF THB ENGLISH VOWEW. 



John Davis's Transtjition. London^ 1656. 



Long Votceh. 
Kametz K it's an obscure 

T 

A as though the vowel 
were mixt with it, as in 
English salt^ small. 

Tzere K it's a plaine, and 
single E in English thus, 
we. 

Long Chirek ^K it's the 

vowel I, in English thus, 
alive. 

Cholem X IS is the vowel 
in English open^ over. 

Schurek, it is the vowel 
XT in English thus, a lute. 



Short Vowels. 

Pathach K is a cleare and 
shrill A, in English thus, art. 



Scegol M is a dubble or 
mixed E/in English thus, 
an eare of come. 

Short Chirek K it's the 
vowel I, in English thus, to 
live. 

Kametz-chatuph M M it's 

t: T 

the vowel 0, in English to 
trot. 

Kibbutz M is either u, or 

else it is y, as in English 
but or shut. 



The simple improper vowel is Sheva, and it's cast under 
the letter in two pricks just down right thus i>t, and it hath 
the force of a most shorte e, as in English thus beggery; 
it's commonly called the movable Sheva. 



211 



VI.— THE GREEK INDIRECT NEGATIVE. By E, 

R. Wharton, M.A. 

[Read at the Society's Mnting held on Friday^ Nov. 6, 1891.] 

Thb real nature of the particle fi-q is allowed to be the greatest 
difficulty of Qreek syntax.^ I shall endeavour to show that (1) 

primarily and essentially fiTf is not a negative or prohibitive 
particle, but an interrogative ; that (2) many ;ii;-sentences which 
are at present printed as assertions might better be printed as 
questions ; ' and that (3) even in other cases the apparent negation 
contains or pre-supposes an interrogative meaning. 

(1) Greek /17 is plainly the same word as the Arian m& (Old 
Persian and Zend ma^^ Sanskrit md : ^ the so-called ' prohibitive ' 
particle). That both are identical with the Accusative of the 
Pronoun of the first person (Latin m^, Zend md, Sanskrit md 
enclitic), used, as in Vergil's ' me, me, adsum qui feci,' to call 
attention to the speaker, is an old but over-bold suggestion. 
It is possible that /tj/ and fiijv were originally byforms like uv 
and pWf i^iti and iywv, since t/ fi^ and t/ fii^p equally mean 
(like our * Why now,' introducing a sentence) * of course ' : but 
in practice fi^ was confined to questions, fi^p was not.' That 

> See Hadvig, Greek Syntax, sections 122-124, 131, 200-212; Goodwin, 
Moods and Tenses,* sec. 259-283, 289-301, 305-310, 315-316, 333, 352-354, 
365-371, 550-551, 576, 595-599, 685-688, 734, 784, 811-819, and Appendix 
II. ; Monro, Homeric Grammar,* sec. 278, 281, 303, 316, 328, 358-361 ; 
Gildersleeve, American Journal, of Philology, I. p. 45-57 ; J. Cook Wilson, 
Transactions of the Oxford Philological Society, 1890. 

* Of course our present system of punctuation in Greek, dating only from 
pofft-clasaical times, can prove nothing as to the usage of the classical period. 

* See Spieeel, Grammatik der Alteranischen Sprachen, sec. 388. 

* BdhUingk, Sanskrit- Worterbuch, s.v. ; Whitney, Sanskrit Grammar, sec. 
67^-^0; Delbriick, Altindische Syntax, sec. 177, 191, 203, 265-267; Speijer, 
Sanskrit Syntax, sec. 353, 405. 

* Before a mute or spirant fi-fiv reeularly became fi/y, e.g. in the combinations 
fUw TM, flip ydpf whicn form then became common before vowels also. See 
Bmgmann, Grundriss, I. sec. 611, and to the consonant combinations there 
mentioned as shortening the preceding vowel in Greek add Sy : KtMs for 
*«i|8-y^r must be an Ionic derivative from ic^Sos (the form in other dialects would 
be *ica8r^ff, ^m jca3of), and ^tMs stand for *t^'B'¥6t from ^^y *to rub.* 
The Greek grammarians notice the shortening of the semivowel before { in 
^o«iri| K^fv^ beside ^ytica iHipOtca. 



212 THE GREEK INDlBECT NEOATITB. 

/(i; usually begins the clause and firiv never does is no objeciioii 
to their connexion: bri is often initial in Homer, though neyer 
in Attic. 



(2) In the Homeric interrogative formula ^ firi (Od. 6. 200, 
9. 405-406), as in the Attic combinations apa fifjf fiwv fi^y ri firj^ 
the fM.ri merely adds a slight emphasis to the initial interrogative 
particle. In Attic (see Madvig 199 b) firi in direct questions 
has an indeterminate force, the answer expected may range over 
the whole field of possibilities, from * No ' through * Perhaps ' 
and * Probably • to a covert or ironical * Yes * ; it is only the 
context that can show which answer is intended. But that this 
use of fiTf in questions was Homeric as well as Attic may appear 
if we view the following combinations as interrogative : 

(a) fkii with Indicative : II. 9. 698 ;««/5' o0€X€9 XlaaeaOai ; 
' Ought you to have prayed ? (We cannot say that in such cases 
the ;tiy, for ov, goes with the Infinitive, for then we should have 
equally /i^ XP^^ XitrceffOai, firj icaXov fjv XiatreaOai.) — Eur. Med. 822 
Xef€« (Xef^ff is only a conjecture) ^e firjhiv ; * WiU you say 
anything? — Ar. Av. 195 (see Goodwin 686) /iri '710 votifia 
Ko/jLyfrorepov rJKovffd vu) ; * Did I ever hear ? * 

(/3) firi with Subjunctive, used as future: II. 16. 128 /iiy ^7 
vfja^ iXtvffi; * Shall they seize the ships?' — Eur. Or. 776 fiij 
Xapu3<Tt 0-e ; So, first in Herodotus and often in Plato (Goodwin. 
265), where the answer really expected is 'Yes,' $,g. firj (l>avXov 
rj : * May it be bad ? Quite possibly ' : cf . the use of apa for Zp 

oVf Soph. Oed. Rex 822 ap* elpvv xaxoi ; ap^ otr%« va^ di/ayyo9 ; 

—Also with Optative, a less direct form of the Subjunctive : 
Od. 7. 316 fLTf loirro (f>iXov A a iroTpl r^cvono ; * Should this be 
the will of Zeus ? ' and so /^rj f^ivotro ; * Should it be so ? God 
forbid.' 

(7) /^y with Imperative, which is really (see Rutherford, 
Babrius, p. 38) a Future with a sense of command : ' /ii/ KXevre ; 
' Will you cheat? Don't.' — The corresponding use of md in 
Sanskrit with Imperative is not found in the Yedas (the only 




iyt *But if you 
questions (Whitney 572 b). 



THE ORBBK INDIRECT NEGATIVE. 213 

instance quoted is probably corrupt, Whitney 579 c), though 
common in Classical Sanskrit ; whence we may infer (Monro 328) 
that the Imperative was originally used only in commands, and 
perhaps that its extension to prohibitions was due to the analogy 
of the Subjtinetiye. 

(S) ftrf with imperatival Infinitive: II. 2. 413 /irj irplv iw* 
^eXiov Suvat ; ' Is the sun to set before ? ' i.e. let not the sun 
set before. 

(c) f»ff used elliptically, without a Verb : Soph. Oed. Col. 

1441 €1 XP^ ^vovfUMi, — fk^ trv 7'; £K\* cfiol frtOov 'What, you 
die ? Nay, hear me ; * Ant. 577 /ti) rpifia^ ir ; * What, still 
tarrying ? * So /ii) ort (or oww^) -is the interrogative equivalent 
of ovx on (or o!jritf»), and in Bem. 18. 200 (=295) ti» owx* 

Koreirrvtrev &9^ trov ; ftrj *^ap rq9 woXetii ye the latter clause is 

as much interrogative as the former. 

In the same way we may analyse every complex sentence in 
which the second clause begins with /*i^ (usually translated ' lest ' 
or * whether,* though the latter would property be ei) into an 
affirmative clause + a question : ^ 

{a) ft^ with Indicative in oaths in Homer: H. 10. 829-330 

lirfm ¥o¥ Z€V9 . . . fi^ fjL^v Toiv t7nro^1riv i,»n^p iTro^xritnTai uXKo9 ; 

* Shall any other drive them ? I call Zeus to witness, No,' and 
so 15. 36-42. 

(0) f»^ with Subjunctive or Optative in final sentences: II. 1* 
522 awotmxe' fi^ n vofjtn^ ; 'Shall she notice anything? With- 
draw,' i,e. * Withdraw, or she will notice something.' • . 

(7) /fti} in sentences expressing apprehension, with Indicative: 
Od. 5. 300 ^tSw' firj ^ frdvra Oea mjfiep^ia ^Twev; ' Was it all 
true ? I fear it was,'* see Fasi ad loo. ; Plat. Theaet. 145 b Sim' 
fM,ij woiXi^v ^eyeu ; 'Was he joking? See to it'; Soph. Ant. 
1253 eiffofieaOa' ^rj t« . . . KaXwrret i 'Is she concealing some- 
thing ? We shall find out ' (Goodwin 369 note) ; Thuc. 3. 53 
^f^ofiovfkeOa' fi^ afMxfHniptov ^fiafnrJKa^v ; So with Subjunctive 
in the sense of a Future : II. 11. 470 htlbu)* fiii ti frd&jjatv; Plat. 
Oorg. 512 d opa' firf SKko ri , , , ij ; 

^ In Sanskrit the Verb after ma Im always enclitio (Delbriick 267) i.e. the 
^nee was a principal one, not dependent, and the m9 cannot be translated 
' lest.* In really dependent sentences nid, not mi, is used. 



214 THE GREEK INDIRECT . NEGATIVE.- 

(3) Finally, even where we cannot print interrogatively we 
may fairly see a question underlying the use of fiij. Instances 
in Homer are comparatively rare (Goodwin 315, Monro 361), 
while in Attic they are numerous and in later Greek preponderate 
(see Gildersleeve) : the Greeks increasingly loved ' dubitantius 
loqui,' to view facts as possibilities. 

A. fij before a Verb : 

(a) After a Relative (Monro 359 b) Homer always has ov, except 

m II. 2. 30], 60T6 ^€ watnc^ fiaprvpoi ov9 firj Ktjpe^ ^fiav Oavaroio 

(pepovtrat, which contains the question ' Are some of you gone ? 
Perhaps ' ; 'and so Soph. Ant. 546 fiffS* a firj *0ty€9 woiou ir€avr^9f 
' Had you a hand in it ? No : then do not claim to have had.' 
So after a ' final ' Relatival particle : iva /iij r^evrfrai is a less 
original way of saying fAtj 761/17x01, which (see 2 fi) is really 
interrogative. 

(fi) In Protasis. A hypothetical sentence may be viewed either 
affirmatively, *This is so, then that follows,' or interrogatively, 
' Is. this so ? then that follows.' To the first form belong the 
Sanskrit hypothetical sentence, which always has the direct 
negative nd, not mi (Speijei" 405 Rem. 2), and the Greek use 
of oif after cc : ^ Homer always (with one exception, Od. 9. 410, 
see Monro 359 0) has el ov with Indicative unless the natural 
order of the clauses is reversed and the apodosis put first, and 

so Soph. Aj. 1131 ei . . ovK ea>, Thuc 1. 121 fin. €t • • ovK ajrcpovffiVf 

and with Subjunctive II. 3. 288^ c< . . . ovk i&e\w<rtv^ Plat. ApoL 
25 b iav re . . ov ffnjre (Goodwin 384). We cannot explain such 
cases by saying that here the Negative forms one idea with the 
Verb, since this is equally true in all cases, $,g. Dem. 21. 205 
uv re firj (f)w^ Thuc. 3. 68 ovvre firj (j^tev^ * the negative particle 
was treated originally like the prepositions, placed immediately 
before the Verb and closely connected . with it ' (Monro 355). — 
Rut the commoner way in Greek was to regard the Protasis as 
involving a question : el ^rf n exuf ovr* iiBtofu, * Have I nothing ? 
then I give nothing.' 

(7) In Oratio Obliqua after on. or *09 the regular Negative is 
ov : Soph. Ant. 685 o7rw9 <rv fi'q XC7C19 6p0w9 raBe is a kind of 
auacoluthon for an interrogative fi^ \er^et9 op0w9 raSe, Rut in later 
Greek (see Gildersleeve) on firj became common, reported state- 



^ ffl in itself no more neceflsitates /t^ than its compound hc'ei does, which 
is always followed by 06. 



THE GREEK INDIRECT NEGATIVE. 215 

ments were regarded as inTolying questions. — ^Before Infinitive 
also o^ is used (Goodwin 685), except where the narration is but 

half oblique, after Verbs like ' hope, promise, swear, agree ' : 

11. 9. 132 ofiovfiai fill wore t^9 €vv^9 iwifirffievai (' Shall I ever do 

it? I will swear an oath'), Thuo. 2. 101 dTi<rroutn€9 ainby fiij 
lyfe^y ('Would he come? They doubted'), and so 1. 139 
wpovXtffov , , fiL^ Ap yif^veaOai voKefwvy as if it Were ' They agreed 
not to fight.' 

(^) ov fiff rf dvffrai .{oT yei^aerat) plainly, see Goodwin p. 391 
note 1, contains the question ntj f^etnjTai (or rfevrjirerai) ; ' Shall it 
be ? No,' and so=' There is no question of its being.' 

(c) £irr€ ov with Infinitive (instances in Goodwin 598) contains 
an assertion, Soph. £1. 780 iv^ ovre pvkto9 vwvov • • . ifik 
irre*ya^€tvj * I never slept : such was the consequence of my alarm,' 
while the more usual ware fi^ contains a question, ' Did I ever 
sleep?' 

(^) Before a Participle : ov wiarevtov involves the assertion ' He 
disbelieves,' fiij Triarevwv (' if he disbelieves ') the question ' Does 
he believe?' So Soph. Oed. Bex 397 6 fi^iSev elBwv OlBtjrovf 
(' Do I know anything ? Perhaps'), 289 /i^ wapwv Oavfid^erai 
(' Is he here? It is strange if he is not,' while ov wapwp would 
involve a definite ov Trapeari). So Thuc. 2. 40 fin. elBwt ovk . . . 
aroSwirwv^^* He will not pay, and he knows it,' while 2. 17 
wpoijiei ftij • » , KaTOiKi<rOri<r6fi€vov=* Would it be occupied ? The 

oracle knew,' 1. 76 la/nev firj &v tjatrov vfui^ \v7nfpov9 r^epo/iepov9= 

* Would yon have been gentler ? We know all about thaV 

B. ft^ before a Noun: Ar, Eccl. 115 heivov , , if firj 'fkireipla 
involves the question 'Is there experience?', Thuc. 1. 137 fin. 
T^v r&v f^eipupwp , • . ov hiaKvai¥ contains the assertion ta^ r^eipvpa* 
ov hiekvaav, So ThuC. 1. 22 to firj fwOwhev ainwv presupposes 
'Is there a mythical element in it? Perhaps not.' In Soph. Aj. 

1231 or ovSev &v rov fitjhev avreirrtfi vwep we have both the 

assertion ' Yon were worth nothing ' and the question ' Wat he 
worth anything ? ' — But in Thuc. 1. 118/17 taxet^ can only be 
a Byzantine gloss (see Eutherford, Thuc. lY. p. xxxiv) which 
has taken the place of the original PpaZeU, 



216 THE ORBBK INDIRECT NBOATIYB. 

The particle md is used (a) in Old PereiaOi the Oathas, and 
Sanskrit, with the * Infunettve,' i.e. Aor. or Imperf., almost always 
nnaugmented and so having no connotation of time ; ' (fi) in Old 
Persian, the younger Avesta, and Classical Sanskrit, with Optative 
(or 'Potential'); and in the two former languages also with 
8uhfuneti99 (in Old Persian only after mdtya\ while in Yedio 
both Mpods almost always take nd, not ma) (7) in Classical 
Sanskrit alone, see above, 2 7, with Imperative] (^) in Sanskrit 
epics with t\Uur$, In Sanskrit (see note 7), and therefore doubt- 
less in the Iranian languages too, md-clauses were always regarded 
as principal, not dependent ; and I would explain them in Just 
the same way as Greek ytii/-clauses, sec. 2, %,$. as really questions. 
The Injunctive then will be a ' timeless ' Indicative, the Subjunc* 
tive a direct (the Optative a less direct) Future, the Imperative 
a Future of command. 

* The InjunotLve In Sanskrit often takes nd instead of mi^ D^dbriick 203. 



217 




f 



^ . ■; i. ■- 1 



Vir.—THE COMPENSATORY LENGTHENING OF 
TOWELS IN IRISH. By J. Strachan, M.A., 
Professor of Greek and Comparative Philology, Owens 
Ck)llege, Manchester.^ 

Though the lengthening of vowels in Irish by way of 
coixipensation for a lost consonant has often been noted in 
isolated cases, the subject is one that has never been syste- 
matically treated as a whole. The unsatisfactory state of 
the question was forced upon the writer's notice by a remark 
of Dr. Richard Schmidt in ' Vol. !• of Indogermanuche 
F^i^^^'^cAungen, to the effect that for dSr *tear'=:*flfacrw, O.W. 
d(^r, one might have expected *ddr. That d^r stands for 
*daer^ there can be no doubt. But should dacr* have 
given *ddr ? The investigation of this question led on to 
the consideration of other similar problems, until at last 
it seemed desirable to bring together and discuss as many 
instances as possible of such compensatory lengthening with 
* view to determining the laws by which it is governed, 
^or the sake of completeness there have also been included 
^^ this paper those combinations of which the laws are 
^ready known. It is not to be expected that all the details 
^' this difficult subject h^ve been finally settled : a further 
analysig of words that have hitherto defied analysis will 
^oiibtless fill up some gaps and correct some of the following 
^^tenxents. But it is hoped that something may have 
°^^ done to advance the question, or, at least, to clear 
^^o Way for further research. 

^^ is necessary at the outset to call attention to the 

P^uliar difficulty of the subject, which is that in Irish 

^^ there is of ten nothing to indicate that a consonant 

•'•^^ irriter's best thanks are due to Dr. Wliitley Stokes for much friendly 
and ior information freely given from his stores of Keltic lore. 



218 COMPENSATORY LBKGtHENING IN IRISH. 

has been lost, and, even when that is certain, there is still 
often an embarrassing variety of possibilities. Sometimes 
help is to be got from the British dialects : thus ^n * bird ' 
might, so far as Irish is concerned, go back to ^egno- or 
*ecnO' or *etn(h; O.W. ein decides in favour of the last. 
When this help fails, the only course left is to look around 
for cognate words in the other Indo-Germanic languages. 
Such evidence is naturally not so conclusive; a wider 
knowledge might reveal a more satisfactory etymology; 
a word put under one heading might have to be transferred 
to another. 

I. Sound Groups ending in n. 

1. den > in.^ 

I4n * sorrow,' Gael. Udn 'wound, grief, vexation,' Mid. 
Ir. Unaim *I wound '=*/?/flk?ndjid : Lith. pidkti * strike/ 

^ Stokee [KuhnU Zeitschrift xxix. 375) has suggested that pretonic gn, dn, 
bn (so also pretonic kn^ tn^ pn)^ became in Keltic e \ee)y /, j9, as in Teutonic kk, 
ttf pp. There is a considerable amount of eyidence in fayour of this: cttoee 
<hill' may be most simply explained from ^enoenot Germ. nMeken (Kluge, 
£t!/m, Wb. S.T.) ; hoc * tender *= *6Atf^iM$«, Skr. bhugnd$ *bent*; aiee 
*bond'=V<*<^^«» Skr. pi^a- * fetter/ Gr. irt(yrvfu (Kuhn «. SchleieherU 
Beitragt^ viii. 332) ; Uee * flat stone/ W. lUch * lapis, tabula saxea, * Uyeh * what 
is flat'»V|<^> ^Pb^'f Gr, w\d^, Lett, pli^ku 'become flat' {KSB, viii. 
317); leeo *maziUa*«i«/uM-, Old Pruss. laygtum, Ch. Slav. lie$ *Yultus' 
(KSB, yiii. 439); boce 'he-goat,' W. hweh, £ng. buek^Hhugndt, Zd. buza 
* he-goat' (cf, however, Skr. bkukka)y meniee 'often,' W, mgnyeh^menogni-, 
■Goth, m^nagt; so Johansson {KZ., xzx. 426) would derive eaee 'excrement' 
from eaqtt'f Skr. gakndt : here, however, kk appears outside Keltic, Gr. ledKiai, 
Lat. caceare; brece 'variegated, speckled,' W. brgehss*mrign6' *mr<fn6' a 
participial formation parallel to mrktS', in W, braith, Ir. mrechtrad ; if muee 
' pig,' W, moeh is to be connected with Gr. /uficrfip etc., Skr. munrdti * lets 
loose.' (Stokes KSB, viii. 316, Brugmann, Orundriu I. 327), it might be 
derived from *fnukn& (original decL ^muknif ^muknuitf etc., cf . Thumeysen, KZ. 
xxviii. 149, Stokee, KZ. xxviii. 291, J. Schmidt, Bluralbildungm der Indo' 
germanischm Neutral 64 sq). In view of a forthcoming paper of Mr. Stokes it 
is unnecessary to enter further into the subject here. There is one point, 
however, to which I should like to call attention — the two-fold treatment of 
Old Ir. ee in the modem dialects. It sometimes appears as «, boe, ' he-goat,' 
mue 'pig,' sometimes as g, beag 'small,' Old Ir. beee, W. byehan. Probably 
under certain conditions of sandhi the double consonant was reduced. There 
was probably also a still earlier interchange oi ee^ e m Keltic, cf . W. clweh 
' crag ' a ^eluceo' by W. dtcg ' crag,' Ir. dock ' stone ' s ^elueO"^ ^cluea, nnles^ 



PROF. STRACHAN. 219 

Qr. 7rX»7<r<r<o=±*7rXaiftai, y/plak. By the side of a tenuis 
a media often appears at the end of a root,^ and thus we 
find also pldg-y^ which in Lat. plangere, Goth, faifldk 
^lament' shows the same development of meaning as in Un 
* sorrow.' 

l^ne 'camisia, shirt, shift '=^n-y Lat. lacema, lacinia? 

m^n i. Ml ' mouth ' *=*i?k3kj/i- : Ags. niaga, Eng. ma/r, 
N.H.G. fnagen^=i*fnay6n. At first sight the meanings lie 
far apart, but it seems possible to reconcile them. Min 
means not simply ' mouth/ but ' open mouth, rictus, hiatus,' 
as is evident from the derivative m^naigim^ Ml. 71<^ 4 
menaigtey gl. inhiare, i.e. 'qui inhiant,' mdnogud 'hiatus,' 
Sg. 8^ 11, 40^ 8; cf. also Gael, meunan 'gape, yawn,' The 
transition of meaning would then be from 'gape' to 
' throat,' and from ' throat ' to ' stomach.' Cf . Lith. 
gomurgs * palate, throat,' Lett, gdmurs ' windpipe,' Gr. 
'Xaaicta, X'if^^*^ further the change of meaning in Gr. 
arofJLaxo^.^ Formally ^makn- would stand in the same 
relation to */nak6n, as Skr. ahna- to dhan ' day.' ^ With m^n 
has been compared W. min 'edge, lip.'^ Thurneysen re- 
marks, "die worter konnen nur verbunden werden, wenn 
fiir das brittische ein stamm auf i i^megni-) angesetzt wird, 
in welchem das e durch das folgende % umgelautet wurde 
wie im cymr. llUh aus lat. lectio." But short i in a final 
syllable does not produce umlaut in Welsh*: nith 'niece' 

.here ^elueo' comes from ^eluc', ^elucco' from ^eluen'. Where eg etc. remained 
before n and disappeared only at a later period, we should then have to assume 
that the accent fell on a preceding syllable. 

^ Cf. Brugmann Orundrist I. 190 sq., Z4S; Orieeh. Ghram,^ 61 with the works 
referred to there. 

» Pick I*. 486. 

* Une represents the development of an n stem, Uteemm the mixture of an 
n and an r stem, cf. Johansson, Beitrage z. Grieeh. Spraehkunde 110. 

« Cf . Stokes, Metrical Lith QIomm 84. 

* Persson, Zur Lehre v. d. Wurzelertceiterung 139. 

* If old Slay. ieladukU < stomach,* is connected with zelad& 'acorn,' one 
might conjecture that ieladukU meant first * Adam's apple,' then * throat,' then 
' stomach.' 

"^ Cf. Brugmann, Morphologuehe TTntertuehungen II. 166, sqq. 

* Thurneysen, KeitoSomanisches 69 ; Stokes, Metr. Ir, 01, I.e. 

* Gramtnatiea Celtiea* 175 ; Windisch, JTZ. xxvii. 167. 



220 OOMPENSATORT LENGTHENING IN IRISH. 

(i=*nepti) may be directly compared with Skr. napfu 
Perhaps a stem ^megni- would give the desired result, only 
then the British stem would be different from the Irish, 
which points to -no* or -tt^- ; we should have in this case t6 
postulate a pre-Keltic megh- with ablaut mogh- in Teutonic. 
On the assumption that mSn comes from ^macti' the W. 
form admits of a simple explanation. To '^macn- there 
might be an ablaut form ^m^cn- and *m^no* or *micHd 
could in W. give mi/t. There seems to be no doubt that, 
under some circumstances at all events, c was lost before n 
in the British dialects, that is to say en had become gn by 
the time that g was lost before n. Only thus can be ex- 
plained W. eroen * skin ' by Bret, croc'henn, Tr. croccenn ; 
croenz^'^erogn't ^eroen* with a weak form of the suflix. 
W. dwyn, Bret, doen *to carry,' have been explained as 
standing for ducn-^ and whether Zimmer^ is right in his 
account of due or not, dwyn can hardly be explained 
otherwise. See below ir^/i, tdn. Where c appears as ^, 
as in W. mign 'bog'' it is easy to suppose that a vowel 
has been lost between e and n. It seems simplest then to 
refer wAi, magen^ min to mdk-^ mik; though the possibility 
of megh', mogh* is not altogether excluded. 

8chi 'shyness, fright ' = *««wnD- : Lith. 9zdktif 'jump, 
spring,' etc.* From this can hardly be separated acuchim 
*I depart '=*«feicid. Zimmer* postulates seee^y but the 
cognate languages show an ablaut <fdtA'-, sUdk-^ and the 
Irish forms may be equally well explained from acac- ; for 
8euehimzz*8eacid cf. eeehuin^^*e€eane. The perf. rosedich 
z=.*prO'8edee is not in itself absolutely decisive, as such 
perfects also come from undoubted e roots, as rogdd, \liJiedh. 
With the use of the perf. roscdieh in the sense of ' it is past, 



' Rhys, Bevue Oeltiguef vi. 17 ; Emault, Hetumnaire etymolojii^ue du brtUm 
tnouen 276. 

^ KZ. XXX. ISS sq. 

* Bezunbergert BeitrSgey xtu. 303. 

* ^J?. xvii. 303 ; Fick, I*. 41. 

* Keltitehe Studien II. 80, 97. [If, howeyer, M.H.G. tehthen 'run, hasten,* 
is to be compared, we must assume a *^ sKeky cf. Franck, Btpmologiteh 
Wordtnboek d. Nederlandtche Taaly 286 ; tdn would then«*MM;ffO-.] 



PROF, 8TRACHAN. 221 

was post/^ c(. Itith, prassdkti, in expressions like sdu metas 
greitai praszdko, ' this year has passed by quickly.' ^ 

br^ 'rotten, stinking/ W. braen 'putidus, tabiduB,'= 
*mraenq', cognate with Ir. mraich, braich 'malt' (=*f/ir«ci-), 
Lat marceo, marddus. For a trace of the primary meaning 
in. the Latin words Osthoff^ refers to corpus occisi marcescit, 
Pliny, H.N. z. §134. He points out also the analogous 
development of meaning in Ags, meaU 'malt/ Ags. melton 
'dissolve, nielt/ O.N. maltr ' rotten, corrupted, become sour/ 
O.H.G. malz 'melting away, soft, flabby.'* In *mracn(h ra 
would represent r ; cf . fraig ' wall,' Gr. iFipya>, flaith 
' sovereign ty,'=*t?Ai/w, y/uel-^ Lat. uelie.^ 

blin ' inguen ' may=*mi<8^/id, Gr. fjuiXatco^.^ 

Where the following syllable originally contained a slender 
vowel ^ i appears as e6i, g. k6in^=i*lacni {Lives of Saints 
from the Book of Lismore^ 3206) ; scedin {do chor scioin 
' to cause terror,' LL. 302» 24) ; bledin ace. of bUn (Wind. 
Wb.). This ed may make its way further by analogy, e.g. 
Gael. nom. ledn after gen. ledin. 

On the other hand den > dn. The long vowel resists the 
change of quality. 

cdinim ' ' I revile, satirize 'r^*cdcnio : O.H.G. huohdn. 

1 Windiflch, Worterbueh 763. 

' Kunchat, LittauUek-deutsehei Worterhueh 437. 

' Moiphologuehe Untermckungen Y. 105 sq. His further comparison of Lat. 
frace$ is tempting, but the change of initial mr to /r in Latin cannot be regarded 
as established. 

* Osthoff, Morph. JJnterfueh Y. iv. sq. Flaith g^oes better with Lat. valeo 
(Fifk. I.^ 541) ; another Keltic cognate of vaUo seems to be W. gweU * better/ 
ef. Oscan valaemom * optimum.* 

^ Formally Ir. bUn is very like W. bUun 'cusniSy summitas, pars interior* : the 
gender is different. Are the words the same, Ir. bUn representing some highly 
specialised meaning ? 

* As R. Schmidt points out (IndogermantMehe Fortehungen I. 65) this infection 
took pUce only when the following slender Towel was actually lost. In the 
modem language a levelling has set in. According to 0' Donovan* s Ir. Oram. 
S5 ei is now the usual form of the gen. If the Irish Tales edited by Mr. 
Douglas Hyde represent the popular usa^ in this respect, they only confirm 
O* Donovan s observation. In no single instance have I observed meot in the 
same paradigm. The levelling is commonly at the expense of the eSi forms, 
beui — bM^ few — fH/r^ muin^ul — muiniH,. It may, however, also take place in 
the opposite direction, as in deor * tear ' after gen. deoir. In Gaelic, so far as it is 
possible to judge from grammars and printed books, this levelling has not gone on 
to nearly the same extent. 

^ At the same time it is also possible to connect edinim with Ooth. haum 
'mean, humble,* haut\jan * rsatttyow * ; Lett, kauns 'disgrace,* on the assumption 



222 COMPBNSATQRT LENGTHENING IN IRISH. 

crdin * bow * z=*crdcntx, properly 'grunter'; Lith. ktokfi, 
'grunt/ Lett, krakt, 'snort, rattle/ Lat. crocio, 

2. agn > dn, 

din * driving ' {ac din liathrdiU ' driving a ball/ LU. BO** 6) 
=z*agni', agim 'I drive/ So the compounds immdin to 
imm-agim ' I drive about/ and tdin ' cattlespoil/ to do-agim. 

grain 'disgust, loathing '=*^ra^rii- W. *graen 'lament* 
abilis, luetuoBus ' (Davies). 

stdn 'tin/ W. pstaen, borrowed from Lat. stagnum.^ 
Guterbock^ remarks: "die lange des vocals in 8idn=z 
stannum ist vielleicht aufzufassen wie die in den seite 17 
erorterten wortem wie bare, spiritdldle,'* etc. (where the 
mark of length is put over vowels naturally short, but long 
by position). But W. ystaen points conclusively to deriva- 
vation from a form atagnum\ cf. Ital. stagno, Span, estaho^ 
Fr. ^tain? 

'dn=^ -ngnoSf Broccdn=iBro€cagno8, etc.* 

fan ' sloping ' (etir r^id y amreid etir fdn 7 ardd ' both 

smooth and unsmooth, both slope and height/ Ml. 140^ 2 ; 

glenta 7 fdnta ' glens and slopes/ LL. 101* 45 ; baralham 

a tilcha cofailet ina fdntaib ' we have cast down their hills 

of an ablaut koijtn', kauH-, with loss of u after the long vowel, Schulze, JlZ, 

xxvii. 420 sqq. ; J. Schmidt Plur. d. Idg, Neutra^ 407 ; Kretschmer KZ. xxxi. 

451 sqq. ; Streitberg, IF. I. 278. The effect of the loss of w in a final syllable 

on the accentuation of the preceding long vowel it disputed. Streitberg, 

IF, I. 270, supposes that on the loss of t or « the preceding long vowel took 

the circumflex, k&Vf Skr. gdm^*j,o^m. On the other hand Hirt IF, I. 22n. 

thinks that i and u did not have this effect, cf . nom. dual, 6r. kyp^^ Lith. butu 

=011, and explains ^¥ by a rather complicated process of analogy. It may be 

worth while pointing out that, in Lithuanian, where in the interior of a word u 

has been lost after i? (=0), the accent is the acute (gestossener accent) — kSpiuy 

kispti * to heap' : kaHpti, *to heap,* kaupaa *heap* ; «/a>, szlffti 'to wipe,* 

8zl&ta * besom': pret. tzlaviaii^ 6r. KK6(t»; dSbti * to hollow out': dubei 

'become hollow,* gusti 'bewail': gauati^ *wail,' gaudiu^ 'pitiful'; dffzis* 

'breach* : daSiti {daidti Leskien), 'strike hard.* On the other hand, so far 

as I have observed in such cases appears with the circumflex (geschliffener 

accent) kopti 'heap,* by kufpti\ glddinu 'make smooth,* glodita n. gladu 

' smooth,* by glaudiu^ glaudu ' smooth.' Does this indicate that the loes 

took place at different times? Perhaps some Lithuanian scholar will give an 

explanation. 
I Ebel, K8B. II. 163. 

* Latexnxscke Lehnworter im Irischen 41 n. 

* Schrader, Spraehvergleiehung und Urgeschichte^^ 316. 

* Cf. Holder, AUkeltitcher ^aehtchatz 60, 164. 



PROF. 8TRACHAN. 223 

80 that they are in their valleys/ LL. 95^ 11) may stand 
for *€agnO' or *vdcnO' : ^udg, udk, Fick I.* 123 ; W. gtcaen^ 
* plain, meadow/ pL gweunydd points to *cdgnO' or *«!cno-.^ 

3. atn > dn. 

idnaise * secundos,' imthanu * alternation ' {inUhanu aidche 
'noctis vicissitodo/ Ml. 21® 3; inna imthanad gl. proprias 
Tices, Ml. 93<' 7; innimthdnad gl. talionem, Sg. ISl*" 6) 
z=*i(hatn; *ambi'to»atn' : , V at *go/ Skr. a^,. with which 
Schulze ^ has connected Goth, afin * year.' From the same 
root may be derived Ir. amm * time' =^*atsnien. 

an 'drinking vessel' has been connected by Bezzenberger ' 
with Gr. iraravrj, Lat. patina {dn=^*patnd). Against this 
is the Old Keltic anax ; patenam et urceum qui anax dicitur, 
Greg. Tur. Mir. 2, 8 (quoted by Holder, Alt. Kelt. Spr. 
137), as ^ is not lost in the Gaulish dialects. Stokes^ has 
already connected the word with Skr. pdnam. 

4. apn. 

There is no clear instance of lengthening here, dna 
* wealth ' has been connected with Skr. dpnaSy Lat. ops. o is 
also found in Irish in aomnie 'rich,' domrne * poor'=*«w-<>p- 
fm<h, etc. : thus it is improbable that we should also find a. 
On the supposition of an ablaut 6, o, dnae may be explained 
as^^*dpn' *opn- (d^p'jnaios, Stokes*), 

5 ec9n. 

tr^n 'strong,' compar. treasa W. trech=:*trexio8^ is commonly 
derived from *trecma8. This is not free from difficulties. It 
is impossible to separate from trdn W. tren, 'impetuous, 
strenuous,' and the Welsh word indicates that can was treated 
like «ii, i.e. became n without lengthening the preceding 
vowel; compare the parallel change oi csm like sm > mm, 
and csl like si > II (see below). It is difficult to get any 

1 For ae > eu when a syllable is added cf. aeth ' ivit,' euthom * ivi/ haer 
* oertus,' hsurwyd * certa scientia/ Oram, Celt, 103, cf. aLso Nettlau, Beitr, z, 
Cifmr. Gramm. 61. Bret, ffueun * marsh,* seems to point to a form *vacn' 
or •r^ff-. 

» KZ, xxTiii. 164. 

* In Stokes* AUkeltUeher Spraehtehatz 27. Through the kindness of Mr. 
Stokes I haye been able to see the first few sheets of this work. 

* K8B. Tiii. 334. Perhaps a British reflex of <in is to be found in Com. 
m^urbor * patena/ Is. *curbir a corruption of Lat. eiborium ? 

» AUkett. Spr, 14. 



224 COM PENS ATOKY LBKOTHrENIKO IX IRISH. 

certain instances of cm. Irish /on ^ elk 'might be derived- 
from *locsnO'' and connected with Bussian ioai 'elk,' which 
like ^locsno* points to a base hle.^ A different form of base 
is found in Ags. eolh, O.N. elgr, O.H.G. elaho. We may 
postulate as the original base eloU", whence by different 
weakenings e//c; loK-, Keltic iocano- might be regarded 
as an expansion of an 8 stem, cf. Lat. alnua 'aldertree'=2 
*al8iio8, Lith. elksznis, Ags. alor, O.H.G. elira, Ch. Slav. 
j'eUcha. men ' farina/ might be derived from mecsn-, an un- 
nasalized form of metik^ 'grind,' Ch. Slav, mqka *meal.' 
It might, however=*m^- and be connected with Lith. 
fninii 'tread,' Ch. Slav, meti ' e\lfi€iv/ Gr. fidrvfu {BB. 
xvii. 205), though the meaning does not come so close. If 
^trecsnos became in Keltic *treno8, Ir. Mn might be ex- 
plained as follows. At one time there stood side by side 
pos. *treno8, comp. *trec8t08, whence C8 made its way into the 
positive — *trec8no8, *trec8ia8; *trec8no8 then by a later law 
became irSn. There is an analogy to this in the treatment 
of the prep. e88=:ex, e.g. inirt * weak'=*ec«ii^r^w, where ecs 
was retained or restored from cases where it regularly 
remained, as before a vowel e.g. eaomain 'fearless,* Gaul. 
exobnus. cal becomes //, hence *ecs'ididy ' I escape,' should 
have given ^elldifi^ ^ellaim. We find instead 4iaiin. *ecs 
(or ess) was restored (or retained) as in the previous 
case and essldlp became by a later law *Slatm. Similarly 
ecsm in imdim 'am unable '=:*^c9m^(/d, enclitic to asmidim 
(W. meddu cf. Gr. fieSioov, fiiSovre:). The meaning of W. 
tren points to derivation from ^thregh, Gr. rpix'Oif, Gaul. 
vertragus, whence may also come Ir. tress ' battle '=*^rec«-, 
further trice 'quick, nimble *:=.t^ghni-. 

6. encn > en. 

lenu (ace. pi.) ' meadows,' Uana ' meadow, swampy place ' 
O'Reilly, leana * meadow,' Mac Alpine, Manx lheanee=^*leticn' : 
Lett, lekns, lekna 'depression, wet meadow' (=:*feriAn-). 

^ Ab lo in. this word fpes throughout the Slavonic languages it seems 
impossible with Job. Schmidt, Vocali»mu9 146, to explain it as coming frcnn ol. 
At the same time one must reckon with the possibility that ^ is a Tariation of 
the stem elen- 'deer.' 

'Tick I*. 611. 



PROF. STRACHAN. 225 

There is also a form lian which can hardly be connected 
with these words ; it may stand for lein<h and be compared 
with W. lltcyn ' lucus, nemus, saltus/ Gr. Xet/iic&i/.^ 

7. egn > en, 

fSn, wagon =*e>e^no- : ablaut to ON. mgn, y/iy^hy W. cywain 
*vehere.'* 

"g^n * I knew ' (etirg^nsa adgdnaa). The Idg. form of this 
was *^egnd(u), Skr. jajh&.u. With u infection ^§e§n6 would 
give in Irish *gSun. The isolated form could hardly maintain 
itself against the mass of the perfects originally ending in 
a, so we find not g^un but g^n. 

g^nar ' was bom '^gegn-, ^/gen^ 

rSn *8pan'=:*rc^wo-.* The quantity of the e in the 
modem rSise is remarkable, as *rec8id should hare given 
\ei8e. The long i may have been taken over from ren. 
This is more probable than to refer r^n, riise {zzz^prendano-, 
*prendsid) to ^{8)prend, Lith. 9prMi *to measure a span.' 

8^ *net'=*«e^«o- ^segh, Stokes/ who also compares W. 
/ncf/ngn, or hoengn, 'a hair of the tail of a horse, etc., gin, 
springe.' We have seen above that egn in W. becomes 
ain ; hence if 8Sn and hwgn are to be connected, we should 
Iiave to assume an ablaut ^segno-, ^aogno^. It is very doubtful, 
liowever, if the words have anything to do with one another. 
8^n is found only in the sense of * net for catching deer or 
birds'^; on the other hand, for hoenyn Da vies gives only 
'Pilus ex Cauda equina vel bovina, etc., pilus majusculus, 
seta ' : it is only in a secondary sense that hoenyn comes to 

1 Gramm. Celt^ 96 ; Cmiiiis, Orieeh, Etym,^ 366. 

* Ebel, KSB, II. 177 ; Stokes, Metr, It, 01, 68. 

s Why Zimmer, Kelt, Stud, II. 127, should say, <' WindiscVs ansicht <dass 
dieses e erst auf speciall irischem boden ein^treten ist,' entbehrt jedes beweises/* 
is not clear. The eyidence is clearly agtonst Zimmer*s view that the 9 is here 




haTe to be regarded as an analogical formation after other e perfects, and the 
existence of such e perfects in Idg. is yery doubtful, cf . Wiedemann, Dot Htauische 
J^rdteritum 106 sq. and the works quoted there. On the other hand there is 
nothing against the origin of gett' from ffegn- in Irish ; in the middle the weak 
form '^tt' is regular. 

• Stokes, Metr, Ir, 01, 90. 

• Academy y Dec. 12, 1891. 

• Cf . Stokes, Metr, Ir, 01. 90. 

Phil. Tnni. 1891-S-S. 15 



226 COMPENSATOBT LENGTHBNINO IN IRISH. 

mean * gin/ because made of hair. Is hoen to be compared 
with Lat. saeta ? For the vooalism might be compared coed 
'wood/ Lat bucetum. Can Irish sSn be a borrowing 
of some kind from Lat. sagena, for instance, through Ags. 
tagne? 

8. etn > en. 

4n 'bird/ O.W. etn^: G. Meyer* compares Alb. ipen^ 
ipeae 'bird.' 

Before slender Towels eon, egn appear as iui, eni, eai ; etn 
as eui, eai. The following are the instances in the Old Ir. 
Glosses :—^rftitVi Sg. 96» 4, Ml. 46<> 23, triuin Ml. 30» 11, 
triuin Ml. 30» 10, 36^ 1, treiin Wb. 27» 7: etargiuin^ Ml. 
42« 10, ingiuin 69* 15, athgeuin, ingeuin 52, etargeiuin Sg. 
197^ 10, adgSuin Wb. 12« 13, under the influence of -g^, 
ginammar, etc. etirgein Ml. 24* 19 : euin Ml. 127<' 12, eiuin 
Sg. 93» 2. 

9. ^/t. 

I have no clear instance of this. Stokes^ connects ten 
^&re'=:*tepnos with Zd. tqfnanhf in which case jt? would have 
disappeared without affecting the preceding vowel. This, 
howeyer, is not certain, as there is also the possibility that 
teh^=:*tep8no8, as tens * heat'=*^^«^w-.* So ttmme ' heat,' may 
:=z*tep8tnid ; it might al90=*^6pmid. 

10. ebn. 

Stokes^ derives Ir. indedm 'anvil,' Corn, ennian, Bret 
anneffn^ from ^ande-bni-s, ^ben& ' I strike,' ^-^hen. But dan 
'lamb,' W. oen=*o'^n(h (see below) proves that jn in the 
interior of a word in Keltic did not become bn, as b was 
not lost before n ; cf . Gaul. Exobnus, Ir. essomun ' fearless,' 
W. ehofyn. The Irish word might be derived without 
difficulty from ande-gni-, but this could not give Breton 
anneffn. Could this have come from a parallel ^ande-j^heni-^ 

> Ebel, K8B, IT. 130. 

> Albanesisohea Worterbuch, 413, Alb. Stud. iii. 40 n. 

> GSum might =*feffHai and be compared directly with Skr. jt^ne. The 
ending -at in the 3 sg. perf. mid. is established by Goth iddja * he went,* Ags. 
dyde, cf. SB. xvii. 238. 

* KZ. xxix. 380. 

6 E. Schmidt, Idff. Forseh. I. 73. 

• Altkelt. Sprach. 15. 



PROF. 8TRACHAN. 227 

onde-heni' ? It may be doubted whether, under any con- 
ditions, jn became in Keltic bn, Brugmann,^ indeed, 
•••^0368 this change in *mnd=*jnd«, gen. of hen *wife'= 
*S^»t<i. But h of *bnd8y whence fwwd, may very well have 
coino from cases with *6ew- ; a declension * Jend, ^gnds wotdd 
uKT^ haj no chance of surviving. 
H . emn. 

A^ccording to Windisch,* OsthoflF,' Wiedemann* *mSnar 

*I thought/ is developed regularly from *memnar. This seems 

very doubtful. I am unable however as yet to prove that 

^ any of the cases where mn come together in inlaut they 

^*^ not originally separated by a vowel.* We have an 

l^^^tance in auslaut if Ir. slemon 'smooth,' W. Ili/fn=i*8limn0' 

w to be connected with Ags. sHm, Gr. Xet/iof .• It is safer 

^ look upon mSnar as due to the analogy of genar. Points 

^^ contact are found in the inflection of the present, cf. 

Q^^nedar * nascitur,' dodmainetar ' putant hoc' 

12. ocn > dn, fian. 

^^itan, crdn 'red, orange '=*(?ro(?no-, Gr. tcpoKo^ (Stokes). 

^^6in *bog,'=*mocnt-: Ch. Slav. tnoktU *wet,'moca 'bog.' 
^ and r stems are often found side by side.^ 

^»», dm * loan ' perhaps = ^pocni-, Lat. paciscar. Cf. 
i^f^laim below. 

13, ogn > dn, Uanfi 
^6n * sorrow ': W. brwyn. 

2 ^nnu/riMl. 380-1. 
, ^Z. niii. 246. 
^ -^r Ge9ehiehte det PerfeeU 10. 
^ -2>a« Litauueh* Prateritum 107. 

. "^ possible instanoe is dotnna, rigdomna *■ tlie material for a king, a crown 
^^^> W. defnyd * element, matter,' which may come from domn-, Gr. 94ftm, 

* )fi^ WuruUrwHterung, 110. 
g^^^- Pedersen, KZ. xxxii. 240 sqq. ; Johansson, Beitrag$ zur grieehisehem 

Sr*^*****^ ^ ^%* ^* *'^*'^ * P^^ ^^ ' ^^y ^ derived from tnaett', in 
*^^ to fffom-. If Thnmeysen {Paul u. Braune** Beitrdge xiii. 436) is right in 
''^^^^i&g 6 of m6r to the inflnence of the preceding m, m6in might come from the 
••^lorm. 

— . ^ every instance ua may be assomed to have passed through the stage o. 

^ ^ precise conditions under which 5 became ua are not very clear. If, as 

*^J*8yaaim, Grundritt, I. 67, suggests, o passed into ua beK»re a following 

^^^^ vovel, then the regular representation has been very much interfered with 

wy ^ttlooy. Why, for instance, should *clopni' have given eluainy when in 

^'^^^ aU the cases the following vowel was slender ? There is an obvious 



228 COMPENSATORY LENGTHENING IN IRISH. 

8rdn 'nose/ W. ffroen^ Bret. /roan:=^^8prognd.^ As we 
saw that e was lost in W. before n, there is the possibility 
of *brocnO' (or *brucnO') ^aprocnd. No certain etymologies 
have been proposed. 

uan 'Iamb/ W. oen (pi. un/n:=i*ognl):=*ognO'f *og^n<h : Lat. 
agnus, Gr. a/ivo^. The vocalism is strange here, as other Idg. 
languages shew a. Can o be due to the f< sound following 
the g? u seems to change a to o in tnog 'slave/ cf. mam 
'servitus/ Goth, magus. The alternative is to assume an 
ablaut d d (d in Gh. Slay, agnid, jagnlcl ' afivo^ '), for which 
see Bartholomae BB. xvii. 121 sq. 

bUain 'reaping' {cofinbuanaigit ' vindemiant/ Ml. 102* 12) = 
^bogni' : an unnasalized form of the root appears in bocht 'reap- 
ing/ O'Gl. also Broccan's Hymn 1. 29, lathe buana di mad 
bocht 'on the day of reaping to her a good harvest.'^ Brug- 
mann,^ however, and R. Schmidt^ derive biiain from *bongni' 
— which seems also possible. 

Hiane 'green,' may stand for *vognio- : Gr. vypo^^ O.N. v4>kf>a 
' wet.* uane and vypo^ (Ir. fir, see below) would be another 
instance of n and r stem side by side, ^vognio- should have 
become yuaine. The loss of / may either be explained as 
in errach, espartain,^ or uaine may be regarded due to con- 
tamination of mg-, and ug-. That *ugntO' should have 
become itaine is highly improbable, as the change of ugn to 
dn can be explained only through an intermediate ogn^ 
where u has become o because of a following broad vowel. 

14. opn > dn, dan. 



difference of treatment of e and o in cases where they arise by compensatory 
lengthening. Unlike ^=^, this e does not (except dialectically) become la, while 
becomes iia like o from ou. This would seem to indicate that o> Ha was later 
than e > M'f that when e arose b^ compensatory lengthening, the change of ^ to 
la had already taken place, while o fell together with o^ou and ^ared its 
fortunes. The subject requires further investigation. It may be noted that ua 
is very much more frequent than 5. 

1 Ebel, KSB. II. 8? ; Stokes, KSB. viii. 336 ; otherwise Windiach 
XSB. viii. 431. 

* Cf. Stokes, Metr. Ir, GL 43. 
» Gnmdrissy I. 382. 

* Id(^. Forteh. I. 77. 

» Stokes, KSB. riii. 344 ; Zimmer, Zeitsehrift fur deutsches AUerthum 
xxxii. 279 sq. 



PROF. STRACHAN. 



229 



dan 'harbour* (Irish), 'sea' (Gaelic) =*«)/)n- : Ags. 
hmfene ' haven/ Ags. heqf, O.N", haf ' sea.' So already Kluge. 
Wb. 8.V. hqfen. 

eltkaitt ' meadow '=*cfopMi-: Lith. BzJdpti 'become wet,* 
szlapiaa 'wet,' azlapimd 'a wet spot,' kKiva^ vorepov miK&he^^ 
tcXerro^ vorepovj Hesyoh. cl&ain and /cXhro^ may be added 
to the examples of parallel n and 8 stems given by 
Pederssen, KZ. xxxii. 252, Johansson Beitr. s. Or. Sprach. 
21, 28, etc. 

cUiain ' deceit ' =*cfopm- : KKhma (Stokes). For cluain, 
'meadow,' we must postulate ^Ulep, as Lith. %zldpti shows, 
for cliiain 'deceit,' ^klep, Gr. /cXeTrrca, Goth, hii/a, Pruss. 
auklipts ' hidden.' 

15. ten + broad vowel > ^ecn, in. 

L^n, a proper name^=Gtiul. Licnos. 

The Gaulish -icnos in Druticnon and the like has been 
compared with the Ir. diminutive ending -in^ It is 
impossible to connect in directly with -tcnos, as the modem 
form 'in indicates the loss of a slender vowel after n. 
Moreover in, though it is found in Middle Irish and is very 
common in Modem Irish, is not found in the Old Irish 
glosses. The most common diminutive suffix in Old Irish 
is -dn, and next in frequency come -tat, ^nat: ^ine is found 
in some eight or nine words, there are a couple of instances 
of ^/}, and Ane is found once in glainine 'maxilla.' This 
'ine might be directly compared with 'icnioa in Gaulish 
Te^^igniua^ "icnoa would in Irish give 'in and may account 
for some of the diminutives in -in, but it will explain neither 
'ine, nor -in. It would be possible to explain these latter 
as arising from an admixture of -^=-tona«, and "ine^-icnios, 
bat it is very likely that these diminutives have more than 
one source. Some diminutives in -^, ^ine may very well 
come from stems in -ec-, ag. larine 'equula'=*ferec-nii5 from 
lair 'mare' stem Idrec*. There is the strongest reason for 



1 Stokes, Liifet of the Sainti xxxi. 

s Stokes, K8B, ui. 71. ; £ev. Celt. tu. 107. 

* JC8£. iii. 208. 



230 COMPENSATORY LENGTHKNIKG IN IRISH. 

suspecting the working of analogy in the suffix -in^ which, 
as we have seen, does not occur in the oldest language, 
and is found with increasing frequency in the later language. 

16. f^n+broad Towel > *egn, in. 
dogina ' he will do ' ^=z*tO'gigndL 

nitg&nsa * I will not slay thee,' LTJ. 68^ 34 : gSn=*gigndm. 

Brdn^ a man's name =*Brigno8f Gaul. Arebrignoa (Stokes). 

8^n * blessing/ from Lat. signum^ if it did not rather come 
from segnum. Before a slender vowel ign > iuin in Brtidn 
gen. of Brin. 

17. urn + broad vowel > *oc/t, dn, fian. 

The instances here are somewhat uncertain, as dn might 
be also derived from eucn^ auen, with a strong form of the 
root. This remark also applies to most of the other cases 
in which d may be derived from u with compensatory 
lengthening. 

bf^n 'fragment,' may=*Mrficno- or *bhr<mcnO' and be 
connected with Lett, brukt 'crumble.' But it may equally 
well stand for *bhrout8nO' *bhroud'8no*^ Ags. brh)%tan ' break/ 
O.H.G. brdama. 

tAn 'podex'=*^M^«d, or ^tiicnd: Teut. */>euha-' * thigh.' ' 
From idn cannot be separated W. tin podex; tin may 
be derived from *tUcnd. That Ir. tdn should also go 
back to *tucnd is highly improbable, as ft is not liable to 
umlaut (cf. ur below) : there is nothing very strange in 
the fact that the two branches of Keltic should shew two 
different grades of vocalism. *tucnd points to an n stem 



* If these diminntiTes in -m be ancient, they might come from ^imis {-it weak 
form of -tot) if we are right in supposing (p. 36), that short t did not change 
preceding t to iui. With regard to the rdation of W. -y» to Ir. -^ Thumeysen 
Jiev. Celt. vii. 325 compares W. di/ni/n *mannikin' with Ir. duin^, a being lost 
without a trace in unaccented syllables. W. -yn as well as Ir. -en might then s 
'icnoB. . Thus W. -yn proves nothing as to the antiquity of Ir. -tn. 

' Thumeysen (KZ. xxxi. 77) has made it probable that the vowel of the 
reduplicated syllable was t not e. Comparing eeehna^^eicanat with ^gena^ 
*gi'ffn'dtf we see that in the one case there is a strong^, in the other a weak form 
of the root. The supposition lies nigh at hand that these strong and weak forma 
were originally combmed in the same paradigm, the strong forms appearing in 
the sinff . act., the weak in the dual and plural, and that the historic pAradigms are 
due to Teyelling in one direction or the other. 

» Cf. Fick III.» 136. 



PROF. STRACHAN. 231 

^leuken: For n stems from o stems of. Bartholomae £ezz» 
Beitr. zv. 25 sqq. Of. also Zd. ndohhana ' nose ' bj ndanha. 
luan, ion ' light/ ma j stand for *leuc8n0' or *lauc8no^f Zd. 
raoxina ' shining/ Lat. luna.^ It might, however, equally 
well=*/Mc»o- ; of. Old Sax. logna 'flame.' The meaning 
' moon ' seems to have been taken over from Lat. luna^ 
whioh resembled in form the native word, or we maj saj 
that the Irish borrowed luna and adapted it in form to the 
similar liian. 

18. ugn. 

diian *poem*=Vti^«4 or *dougn&i Gr. revx^f'Vi Goth, dugan} 
ciianine 'pugil/ from Lat. pugnus, or perhaps rather j9o^nti8. 
Of ucn, ti^n+ slender vowel, I have found no instances; 
we might expect -um. iuiine was discussed above ; (tain 
* leisure,' may be derived from *eucni'' : Gr. evKtiKo^ * free from 
care, at ease,' Skr. dkas, 'comfort, dwelling-place,' rather 
than from *ficnt- [ucu' may, however, appear in uim * time, 
opportunity ']. 

19. fi^n+broad vowel=*o^n, d», 6fln(P). 

friian 'lasting ' may perhaps stand for ^bhu-tna-f and be com- 
pared with Lith. bittinoB^ 'essential, lasting' (=*Mfi^nfio-), 
with which has been compared Lat. 'bundu8=*bhutn(h^ I 
know of no other instance of suffix -tno^ in Keltic. 

20. udn' > *odn^ dn, fiait. 

9mudinim * think '=*«mfiin- or ^smoudn- : Goth, gamaudjan 
'remember,' Oh. Slav. mysA- 'thought.' If uan herezitM/n, 
it must have arisen before a broad vowel. 

21. vpn > *opn, 6n^ uan. 

citan 'host' {Druim Criaieh c4te c4t ciuin 'Drum Oree 
meeting-place of a hundred hosts,' LL. 151* \.)=:*oupn' or 
*eaupn^ : Lith. kupd ' a heap, a multitude, an assembly, e.g. 
of men,' Germ, haufe, 

8uan 'sleep,' W. hun^^aupnos: Gr. vttpo^, Oh. Slav, s&nii, 
Alb. gume 'sleep.' The gen. sAain can hardly be the 



^ Brognuum, Orundrits II. 132. 

* Tharneysen quoted by Osthoff, Paul u, £raime*t Beitrige, ziii. 421. 

* Cf. Brugnuum, Grundriu II. 162. 



232 COMPENSATORY LENGTHBNINO IN IRISH. 

4 

regular deyelopment of ^supni : M comes rather from cases 
containing a broad vowel. For suggestions as to the 
phonetic changes of pn see Brugmann, Orundriss I. 272. 
According to Stokes' law ^sdpnoa must have been accented 
like Gr. vrrvo^. 

22. m. 

Thumeysen ^ suggests doubtfully that W. gum ' toga,' Ir. 
fkan 'lacema'=*fdno- come from *vomO'. But that is im- 
possible; 8 disappears before n without lengthening the 
preceding voweL* *rdno- may be derived rather from 
^vo^ouno'^ Lat. ind-t40, ex'Uo^ sub-U'euh, Lith. auti 'to put 
on shoes, stockings, and the like.' 



II. Sound Groups Ending in m. 

1. dcm. 

One would expect by analogy of acn, acr, im. 1 have no 
instance. 

2. agm > dm. 

gldm * outor J* zz*glagmd: Germ, klagen^ ^glagh, 

mam ' servitude '=*mn^mtf-: mog 'slave/ Goth, magus. 

tldm * handful of wool '=*^/a^w- : Germ../fo(jA:e, Gr. Xa;^i/o9,' 
with which Stokes^ has already compared thcht * garment.' 

Hence Mod. Ir. brdim, Gael, braim (g. brama^ pi. bramati' 
nan), W. Com. bram, Bret, bramm * crepitus ventris,' to O. Ir. 
braigim * pedo/ cannot =*6ra^»i^n. Nor is it likely, as Thur- 
neysen ^ doubtfully suggests, that it should have come from 
*brangmen ; cdim * step * =z*cengmenf and similar cases are 
against this, bram is probably to be explained €LB=i*brag»men, 
cf. bamm * morsel '=*6o^«m^,* semmand 'rivets' (^segh, 
Stokes) from a stem *8egsmenf W. drem 'look, aspect ' = 



^ KeltO'Soimmisehst 64. 

* Stokes, JTZ. XXX. 669. 

' Siitterlin, Bezz. Beitr. xm, 164 sq. 
« Metr. Ir. Gl, 99. 

* KehO'Somanisehei 98. 

* Not *boffmen as R. Schmidt, Idg. Faneh, I. 30, foggests ; that miist haye 
giyen *b6im or *6ttaim. 



PB07. STRACHAN. 233 

*drec8md *dri€8md, Ir. drech * aspect, ooniitena.nce* y/derX, Qael. 
dream 'wisp of hay or straw/ dreamag * handful of corn' : Gr. 
Spdy/ia. 8 is often present before certain suffixes in other 
Idg. languages, ag. 'Smo-, -anch 'Sla-^^ and its presence must 
be also recognized in Keltic, e.g. amm ' time '=*a/«m^n 
(above), from * heavy,' W. tncm^HrudsmO' : Goth. f«- 
thriutan 'oppress.' 

For 6% in Mod. Ir. brdim^ cf. dirdhearc 'glorious '=0. Ir. 
airderCy and in the Wb. glosses bdill limbs, sg. ball is 
frequent.* 

3. eem > im. 

rSim 'shout' {doberi rim curad asa bragit 'he gave a 
hero's shout from his throat' LTJ. 76« 10, reim curadh i. 
geim euradh 'a hero's shout,' 0'Clery)=reciw- : Ch. Slav. 
reka 'speak,' lith. rekii^ 'roar, cry'; for the long vowel 
in relcH, see Bechtel Hauptprobleme der Idg. Sprachwksen* 
%ehaft 162. * Examples of words with the meaning of cry, 
roar, and the like applied to articulate speech will be 
found in Persson.* To this belongs W. rhegen 'quail'; 
for the origin of the name cf. Diez. Etym, Wb, s.v. quaglia. 
In Old Ir. a similar name is found for the peacock, giaachtach 
from gessim * cry.' 

4. egm > im. 

Biimeth ' oSspringj * =i^Begm€to-^ 

drimire 'ladder,' according to Stokes* stands for *dregm-, 
l>Qt it seems rather to come from drengm^f cf. dringim ' step^ 
advance,' and so to fall under the following head. 

5. engm > emm, im (in auslaut). 
ciimm 'step,' W. cam=z*kngmen. 
liimm 'leap,' W. llam=^*lngmen. 

giim * shout' :=z*geng men : cf. Lith. iwSngiu ' neigh.' If this 



* Oithoff, Fortehwtgm I. 190 aqq. ; Bnigmaim, Orundrist II. 133, 140, 163, 
1««, 187, 196, 196, etc. 

' 'AocQtdiiig to ThmneyBeii, PtuU u, Braunet Beitrige, xiii. 436, o is here due 
to tliepreoediiig UluiL 

* Irunthnoeiterumg 244. 

* Stokes, XiTM oftht Saints 399. 

* Stokes, Lin^Utie Valus of the Irish Ammls 371. [He now {BB. xnu. 62) 
men it to *drsHj^mm}. 



234 COMPENSATORY LENOTHBNINO IN IRISH. 

comparison is to be maintained, we must assume a double 
Idg. form gfteng, geng, u having been lost in certain positions 
in the parent language.^ 

r^imm, rSim 'cursus/ serving as the infinitive of rethin 
* I run/ The double m of rSimm cannot be explained eithei 
from *reimen or from *reidfnen,^ as R. Schmidt^ suggests 
Either we may suppose that ^rSimzn^retmen became riimni 
under the influence of c^imm, griimm^ or we may place i1 
with W. rhamu * soar/ Idg. ^renghO *run, spring.'* 

6. efidm > emm, im, 

grStmm * progressus '^=*grendnien : ingrennim ' persequor.' 
Gael, teum * bite/ W. tarn, Bret, tamm ' piece, bite/ Com. 
tam^ pi. tgrnmyn^^ndmen : Or. rivSo), Lat. tondeo. 

Brugmann ^ doubts if endm became regularly emm. Foi 
griimm he suggests the analogy of cSimm^ which, from the 
similarity of meanin^r, is quite possible. But if teum is 
rightly deriyed from *tendmen no such analogical explanation 
is possible. We must then with Schmidt, IF. I. 77, look 
upon endm as becoming regularly emm, and there is nothing 
in the way of this. 

7. enm. 

beim 'a blow/ has been explained as standing for *ben' 
^^*h y/jjien. But this explanation is very doubtful for 
the Irish word (according to R. Schmidt® one might have 
expected ^bemm'^), and it is impossible for Bret, boem, 

1 Cf. Brugmaim, Grundrm 11.476, 802; Solmsen, KZ. xxxii. 277 sq. If it was 
in endisia tmit u was lost, it must be remembered that in certain positions the 
Ide. Terb was enclitic. 

' It might however come from ^reidtmen, 

' Idg, Forseh, I. 77. 

* Fickl*. 118. 

» GrundriM I. 382. 

^ Idg, Forscl^. I. 77. His own explanation leaves Bret boem ont of account, 
and does not satisfactorily account for Ir. beimm, for how could beimm be in- 
fluenced by words of so totally different meaning as eetmm and griimm f On 
the other nand, as Schmidt suggests, beimm 'journey,* '\l%em, may have been 
influenced by these words ; it may however = ^^gtutm-imm, 

^ Anmaimm dat. of ainm 'name* ma^i—* anmembi as well as ^anmenmi, 
A difiiculty arise in connexion with the form ainm, U ainm^*anmen *aimm 
might have been expected ; if it comes from ^an'^meny we should rather have 
expected *ainmh. Perhaps the difficulty may be overcome by the hypothesis 
that two forms of the stem *anmen and an" men were origmaUy in use in 
different cases, and that they acted and reacted on one another. (In Mod. Ir., 
Gael. meamnas:0, Ir. menme, such an nm has become in inlaut mn,) A. 



PROF. STRACHAN. 235 

Com. bamm, pi. bommen 'blow.' All these forms maj 
be explained from *ben8men, which in both branches 
of the Kelts would become first *bismen. ^BHmen 
would give by assimilation in Irish *b€mmen, biim. In the 
Brythonio dialects,^ too, bismen would become *bimfnen^ 
whence regularly Mid. Bret, boem. Mod. Bret, bomm (cf. Mid. 
Bret, toem *hot' (W. twym) probably =*^^/)tf«iiio-, Mod. tomm). 
In older Cornish e appears as ti», 01, in the later texts as 0, 
Mm ' hot/ later tommys * heated/ Oramm. Celt? 97. Now 
bomm is found only in texts in which appears for i. It 
is also possible to derive beimm, *bemmen from *bei8men : Ch. 
Slav, biti 'strike/ with which Thumeysen^ connects Ir. 
be9iim ' I strike.' 

8. ogm > dm, Ham. 

fduximm, f&aim 'sound, noise '=*fY>^m^y ct. /ogur 'sound, 
noise.' The double m here is not due to phonetic develop- 
ment, but comes by analogy from other cases where it arose 
regularly, e.g. daimm ' sewing ' (to iuigim ' I sew ') =:*eugsmen. 
In such cases mm became in auslaut m, so that here /uaim 
and iiaim would coincide, and this coincidence would lead to 
the introduction of mm into fiaim. 

idm before a broad vowel seems to become *edm im in 
fO'dima 'he will suffer '=VW/W(I^ But as fod%dmat^*d^da^ 
mdnio is also found, it is probable that in fodSma we have 
an analogical formation, for, though in all likelihood in 
these reduplicated futures strong and weak forms of the 

difficulty of a somewhat siinilar nature is met with in W, ewrw *beer,' 
Oanl. Kov^iu (Ir. emrm) compared with garm 'outcry' (Ir. gairm) where 
the m is preeerred. The difficulty might be solved by postulating in the 
latter case a stem ^gartmen cf. Lith. gar §09 * noise,' Lat garrio^^garno^ Alb. 
gtrias * invite.' With a syllabic division *gar | tmen this would have given 
*gormmeH, and mm did not become / {e%pr%p=^ewruf cf. Com. corofj. How 
i% the I of icovpfu to be explained P Can it be that we have here a nom. -me {n), 
as in Ch. Slav, im^ * name,' Schmidt. Pluralbiidungen 90 P frop/m might be 
looked upon as a Graedaed form (or ar^p/m : Kovpfu s= Skr. nima : nima ?). 

* It is not certain to what extent these changes are to be referred to proto- 
Keltic. I know of nothing to shew that the change of ^bentmen to ^betmen and 
of busmen to bemmen did not take place before the separation of the Keltic 
dialects. 

s Rheinitehet Muuum zliii. 351 ; KZ. xzxi. 88. 



236 COMPENSATORY LENOTHBNINO IN IRISH. 

stem were at one time found in the same paradigm, it is 
not very likely that both survived in this solitary instance, 
particularly when we bear in mind the tendency of the i 
future to encroach. Stokes suggests that d may have been 
lost before m, before the time when intervocalic m was 
aspirated. He refers to JrSm * root ' (Mod. Ir. freumh) = 
*vridmdf W. gureidd: Lat. radix^ uamonn *skin '=*orfw-: 
Lith. iida ' skin '= (ablaut 6: d). But this is not yery 
certain. 



Sound Groups ending in r. 

1. acr > ir, 

d^r * tear/ O.W. dacr=^daeru : Gr. Sd/cpv. 

Sr .t. uasal, * lofty/ 0'Davoren=*flcro5 : Gaul, axpotalvs, 
Gr. atcpo^.^ 

m4r *&uger *=z*macro8: Gr. fmtcpo^,'^ ficuceBvo^^ fiotep&va; 
rov o^vv, ^EpvOpaXoi Hesych.* 

Mod. Ir. c^ir *gum/ Manx keeyr, Gael, edir might be 
explained from an ablaut ^cdcH', *cdcn-. We may compare, 
perhaps, W. cig * flesh * (Ir. cich * teat ' ^) =*ciC'. The ablaut 
series would then be kdk-^ kik-^ kok- (=Ir. cdc-). The W. 
name for 'gum' is cig y dannedd, lit. ^zahnfieisch* One might 
conjecture that the Irish word was onginally used with 
some such qualification, c^ir inna n-dit or the like ; cf . fedil na 
fiacal (gl. gingiva), Stokes, Ir. 01. 150. 

Before a slender vowel edii medir, gen. sg. and nom. pi. 
of mSr (Gael. gen. sg. medir), Gael. Mod. Ir. de&ir gen. sg. of 
deur 'tear' (with change of declension, Mid. Ir. gen. dire^). 
For the vocalism of cHr see p. 36. 



1 Stokes, AUkeU Spr, 5. 

* Bezz, Beitr, xvii. 299. 

' Brugmann quoted by R. Schmidt, Idff. Forteh. 1, 63. 

« Stokes, Ir, Gl, p. 150. 

> The nom. pi. na tUfr iB fomid Ml. 23* 18, indicatiiig a nent. M-stem ; dh 
might here =*datiru like Ved. purd (Schmidt, Plur, 42). There are howe?er 
other poeaibilities, cf. Brugmaui, Orundriss II. 626. 



k 



PROF. STRACHAN. 237 

2. agr > dr. 

dr '8trage8'=*a^ra«, "W. aer 'preBKum/ O. Bret, air, Gaul. 
Veragri, Suagros, Gr. Srfpa} 

sdv' ^exceedingly' (as a prefix), W. haeru 'affirmare/* 

sdr 'insult/ can hardly be separated from W. aarMu 
'contumelia afficere/ Corresponding to Ir. d one would 
expect W. ae ; e must have been lost in the pretonic syllable, 
though I have no other instances of this. For examples of 
ae > ain other positions, cf. Nettlau, Beitr. z. Cymr. Gramm. 
61 sq. 

ndr * modest *=*na^ro- (or ^ndgrth) : *nd^ho *to be sober,' 
Gr. vifi^? 

3. tr. 

t is not lost before r ; an anaptyctic vowel has sprung up 
between t and r, and intervocalic t then becomes th, e.g. 
tarathar^horex^'i terebra, reperpov, criathar ' sieYe' z=^*creilro», 
hriathar *word'=*4m^r(2,* mothar *dark '=*»iii^ro-: Arm. 
mut* " tenebre, nebbia," mthr " oscuro," * riathor * torrens ' 
=z*reitr(h : Skr. ri ' let flow, run,' Lat, riuos. 

4. adr > dr. 

Only in composition drim ' number '=*a(ifrim(2 to rim 
* nnmber.' 

6. egr > ir. 

fir * grass,' W. gwair, Com. gtPf/r=i*v€grO' : yjveg *tobe 
moist ' Fick P. 545 P 

Before a slender vowel /^iWr Sg. 68* 10, /«<*> Ml. 90*» 8. 
Mid. Ir. /i^oir, Windisch. Wh. (by analogy dat. feor, LU. 
74»> 38). 

^^/*a^, girait 'champion' is probably cognate with the 
shorter greit 'champion.' Qreii may stand for *gredni', 
^ghredh-ni' {^/ghredh Fick I*. 418); girait would be a 
parallel form with reduplication. In FSlire, Sept. 27, 
a passage to which Mr. Stokes has called my attention, 

^ Rhys, Sev» Celt. 11. ; Stokes, Spraehtehatz 7. 

* Rhys, Lectures', 395; Welsh has also taeru^^tO'tagr^, an additional 
example of the original coincidence between the Welsh and the Irish accent. 

» Fick I*. 499. 

* Bnigmann, Grundrus I. 470. 

* Bugge, KZ, xxzii. 19. 



238 COMPBNSATaRT LBNOTHENINO IN IRISH. 

it is used with fern, di, which shows that, as might have 
been conjectured, it was originally an abstract fern. noun. 

6. tcr+ broad vowel > *e(?r, dr. 
{arro)cMr ' redemi ' =*-cii?ra.^ 

9tnMU 'burning coals, sparks, embers' mB,y^=*smicronti' 
and be compared with Lat. micare^ V(8)miA;-. Otherwise 
Stokes, Metr. Ir. OL 97, but the length of the e is established 
by the modern language, smeurdid ' charcoal,' Foley. 

Before a slender vowel tfii, doraeAiuir, * redemit,' Wb. 2* 9, 
duarchiuir, Ml. 73^ 6. 

7. t^r+ broad vowel > *egr, ir. 

{fris)gira *he will answer' =*^^r(I^, (ar)gerat^ ML 112^ 8= 
*gigrdnto. It might be expected that i. in such a case would 
remain. We have perhaps an example in dir, 'proprius, 
conveniens, iustus'=*fl^*ro- or ^digro- : ^deijc^ dei^, Seucirufu, 
SUff, Lat. dignus, Goth, ieihan, taikns} 

8. ibr seems to become *ehr^ Sr in bSra 'he will bring '= 
^bibrdt. But b is not lost before r ; as in the case of tr an 
anaptyctic vowel is develc^ed between b and r and b becomes 
a spirant, e.g. dobur 'water,' W, dwfr=*dubro-, Gaul. Vem<h 
dubrum,^ gabor * goat,' W. gafr^gabro^^ Gaul. Oabrthsentuni, 
Qabro-magus, It might be said, perhaps, that this may 
hold good only in auslaut, while in the middle of a word no 
such vowel sprang up, and in this position b was lost 
before r. But in abra 'eyebrow,' Gael, abhra^ Mod. Ir. 
fabhra : Gr. i^pv^, br is treated in precisely the same way as 
in dobur, gabor. It follows, then, that b4ra is an analogical 
formation after ^gira and the like. The € future has a 
tendency to spread, of. 8c4ra to acaraitn^ limaither to Idmaim 
(LU. 63» 15), gH to gataim, aderad to adraim (ESB. vii. 
23). The same yiew must be taken of miraid 'he will 
remain' to maraim; there is no evidence that m was lost 
before r. 

1 Cf. R. Schmidt, Idp. Forteh. I. 63 note. 

' Cf . Kluge in Panl's Grundrist der Oermaniaehen PhiMogie I. 326 ; Bnig- 
mann, Grundriss II. 136. [I had orerlooked W. dir * certas, necessarius/ dir yw 
* oportet, necesse est/ which shows that e cannot have heen last ; the words may be 
reterredto *dero8y *dheroBy V ^^i cf. Skr. dhdman *law, order,' Gr. Biius.'\ 

' Gliick, KeUisehe Namen 35. 



PROF. 8TRACHAN. 239 

9. ogr > dr, 6ar. 

itar, fiat • cold/ "W. ow, goer^^^ogro-^ (to Ch, Slav. o^»l 
* fire/ of. tinY frigus, or Gr. irdxifV wa7€T09, ablaut a o P), 
Com. oir. For W. goer, cf. ^orcW ' hammer/ Ir. ord ; for the 
explanation of the prothetic g see Nettlau, Hev. Celt xi. 77. 

10. tt{T+ broad vowel > *oer, Or, iiar. 

citar 'crooked/ cf. corthdn (corthdnech Ascoli), Sg. 66^ 7= 
^cucro* (or *ceucr<h) : yjkeuk, Skr. A^tica^i * bend/ Ch. Slav. 
kuko-ndsH * crooknosed/ etc.^ 

d in ucTy ugr might be expected to remain. Of iker j 
have no very certain instance; fir, 'bad/ might=i*/?fiATO- : 
Gr. ix<PT€vtci]^, TrevKeBavo^f cf. iceueS^ : Lith. kenkti ' hurt * ; it 
may, however, be explained otherwise'; bdrach 'din,' bdrethar 
' clamat,' might stand for *i&cr- : Gr. fiv/crrj^, Lat. bucifM ; 
they might equally well stand for *6dr- : Gr. fiva^, Arm. 
ftw, ftw-^c ' owl. '* 6r ' fresh ' {xcrin' withered ' *), W. ir 
'viridis, floridus, novus, recens, succalentus,' has been well 
compared by Siegfried * with Gr. vypo^. The Keltic forms 
go back to *ugrO' ; Welsh proves conclusively d. 



lY. Sound Groups ending in /• 

1. acl > el. 

die ' servus ' is separated by Stokes from cile ' comrade ' 
(see below) and compared with Lat. cacula ' soldier's servant.' 

mSlacht ' disgrace ' =*mac^^e<- : Lat. macula, though this 
is not altogether certain. 



1 Stokes, Zivet of the SainU 402. 
» Fick I*. 380. 

* Stokes, Metr. Ir. GL 103. 

* CI. Persaon, WurzelenceiUrtinff 12 sq. 

' A good instance is quoted by Windisch {Worierbueh 866) from Stokes, 
7%ree Middle Irish Sotnilie* 26 : teeh do denum iein uair ein a leih ur ocut 
aroli erin^ *'to build a house in that hour, the half thereof fresh and the other 
withered.*' Cf. also Stewart's Oatlie Foemeb^^ Cia lum crann bheil an eoiil f 
ar Fionn. Adha, ar an inghean^ iodhon iir ague erion. How many trees are in 
a wood ? said Fionn. Two, said the maideo, to wit green and dry. 

* Quoted by Stokes, KSJB. yiu. 322. 



240 COMFENSATORT LENGTHENING IN IRISH. 

cil i. Ml, * mouth,' O'Clery. If this be a genuine word 
rightly explained one might compare Lith. kaklas 'neck.' 
But no weight can be laid upon this instance until the 
word is better established. 

2. agl > dl. 

ail ' disgrace '=*a^/i- Goth, agh *aUr)(pbs* Cf. also W. 
*aefe, grefyn, dolorus, trest, trwm, LI Braw aele, * Terror 
miaerandus' (Da vies). 

ail * pleasant '=*|>fl^/t-, Goth. /fl^r« *&ttbig,' /ahe\>8 *joy' 
y/pdH, pd^, Fick I.* 77.* 

dl ' brood, oflbpring/ W, ael * litter, brood,' Bret, eal * foal ' 
perhaps=*/?fl^fo- : hsLt. propago. 

gabdl^ inf. of gabim * I take '=*^flf6fl^/i-, "W. caffaeL 

mdl * chief,' "W. Mael'=-*maglo', Gaul. Maglm. 

idl ' adze,' may, perhaps, stand for ^to^aghh : Goth, aqizi, 
Engl. axe. If this were so, *'aglo- Tout, akea- would be an 
instance of / and « stem side by side, cf. Lat. oculus, Ch. 
Slav, oko 'eye.' The difficulty is to see what the particle 
to- has to do here. Stokes^ now derives tdl from *taxlO' 
(=Idg. *ta/c8lO'?): Ch. Slav, tesla *axe.' But A»/ seems in 
Irish to become *««/, //, cf. tiall * pride '=*aMg«/(I,* toll, W. 
twll * hole '=^^tuc8lO', Slav, f&k- 'pierce.' 

We have seen that acn, acr, acl became respectively en, er, 
el, but agn, agr, agl became respectively dn, dr, dl. What 
was the reason of this difference of treatment? The most 
probable answer to this is that the loss of c and of g do 
not belong to the same period, that c persisted longer than 
g, and, when it also went, the preceding vowel was treated 
differently. We have a parallel to this in Welsh, where 
*dacru becomes dagr (O.W. dacr), but *agrd becomea aer. 
In Irish itself there is something similar in the different 



^ d* Arbois de Jubainyille, Zes Noma gaulois ehe» Cesar 26 y sees this word in the 
Gaulish name Catusualis ; if that were so, the etymology would have to be given 
up. But Stokes and Rhys {Classical RevietOy April, 1892, p. 166) compare 
Catusualis better with W. ehwalu * strew, spread,* =' battle- scatterer.* 

' Zimmer 
tiesaim * I raise, 
analogical f onnation 

» KZ. xxxu. 219. 

4 Bnigmann, Grundriss II. 194. 




FROF. STRACHAN. 241 

treatment of the particles aith" and ad^zr.ate^^ ad- in com- 
position under the accent. Where the a of ate becomes e, 
the a o{ ad appears as a, compare ni ^il ' he does not die ' 
=.*dtebalet with dpelugud 'flattery '=0^;?^, Scid *he relates '= 
atC' (with accent on the verb aithchtuUd) with acct * he sees ' 
=ade'.^ Where this rule seems to be yiolated, there is 
confusion of the two particles.^ Now, as in ate^ the t is 
followed by a yowel, while in ad- the d is final, it is 
easy to suppose that the t persisted longer. The different 
treatment of the vowel in this case would then be a 
parallel^ to the different treatment of the vowel in the 
combinations acn, agn, etc. 

3. atl > dL 

anal 'breath/ W. anadl =• *anatld, ^andtlci: Skr. dniti 
* blows.' 

ddl ' meeting/ W. dadl^^datla. 

In sal 'heel/ W. sawdl, Bret. seul—^Mtld (Stokes)^ & is 
naturally long. 

4. adl > dl. 

In the compound dlaind ' beautiful '=a(^^i;i£l, laind.%. solas 
no iaitnemach 0' Dav. 102.^ 

5. eel > el. 



1 From cases like uwe 'water': 6r. ^Ittpy 08or, etc., mete 'drunk*: Gr. 

;;/0v, etc., tew 'dry': Lat. 9icem(^*9itqo») has been deduced the law that 
dg. tk became in Keltic »k (cf. Brugmann GrundrUs 378) » If that be so, then 
in aeci ad must have been restored from other words where it remained, and then 
ode became ace in accordance with a later law. There is something strange in 
the sound change here postulated, and the same law has been called in question 
for Teutonic by Kluge {Paulas Grundrise I. 327), who points out that in the 
instances quotea ak may e<^ually well go back to Idg. t-ek, tk could be explained 
in the same way in Keltic; ueee ^^uts-eid, *ud-8'kia (from stem *ude9' Gr. 
tSos, Skr. dtaa' 'spring' with which Stokes compares Ir. m ^ water,' (=*«/m-) 
in 09' bretha 'water judgments'), meec^^metS'eoet etc.; with eeee 'dry,' W. 
hy»p cf. Zend AiSAtm, Gr. I<rx,y6s fZimmer, KZ. rxiy. 212, Bartholomae KZ. 
zxrx. 525). A good instance in which ee in inlaut may be explained from tk» 
is rtAcee ' disgrace ' (Gael, ruieean ' a red pimple') —^rut^kia^ rudh'kid '^reudh 
' to be red,' cf. Ir. ruidmeh ' blush,' Gaul. Seno-rueeua (d'Arbois de Jubainvflle, 
name. gaul. 69) : so euie 'secret,' a *{;t«^* : Gr. Ktitdw (Stokes) : broee W. broeh 
* badger ' =*broU!0', * hhrodh-ko' : Skr. bradhnd ' lightred ' Ch. Slav, bronii 
' white ' i^bhrotko" : *bhrodhnO'= Lat. eatcut : canut). 

* Zimmer, Kelt. Stud. II. 70. 

* Unless e for a is due to the lost slender yowel. 

« Cf. K8B. ym. 428, BB. xi. 128; in mil the loss of t might be explained 
from dissimilation. 

* Stokes, Bezz. Beitr. xyi. 60. 

FliU. Tram. 1891-9M, 16 



242. COMPENSATORY LENGTHENING IN IRISH. 

Lat. manUe, Skr. manyd 'neck^ muscle of the neck/ Eng. 
mane. 

Stokes,' after Rhys connects cedl 'music ' with W.pib 'fistula 
tibia,' OiW. piipaur (MS. piapaur) *tibicen/ and derives 
it from a form *cvecvlO', But cvecvlo* would not explain 
the Yocalism ot pib, piipaur, which are rather loan-words 
from Lat. pipa. Apart from the Welsh words the hypothe- 
tical cvecviO' has little probability. It is very likely that 
some consonant has been lost between e and o, I had 
thought of deriving cedl from ^tcepohm, Skr. ^np * curse/ 
supposing that the meaning of * curse' developed itself from 
metrical formulae of imprecation. But that is not very 

certain; . . ; 

6. egl, > eL 

Ir. die 'comrade/ has been brought together with W. 
eilyd ' comrade/ in the phrase y gilyd * the other ' (lit ' his 
comrade ')=Ir. achSiie. cele and cilyd might come from 
*ceglid8. For i in cilyd cf. llith from Lat. lectio, nUh * niece/ 
from *neptu 

r4il 'clear' {r(laim * manifesto ') =*r^^/i-, Lith. regiti 'I 
see/ regimas ' visible.' For the vocalism of rSil see p. 36. 

sedl ' sail/ W. hwyl is puzzling. The words are 
commonly derived from *8eghlO': Teut. *8eglO' 'sail.' We 
should expecty however, in Irish *«^/, in W. *haiL One 
might at a pinch explain sedl as a new formation from the 
gen. siiiil {*segli) after the analogy of ciiiil: cedl, and the 
like ; but, though such formations are found in the modern 
dialects, I know of none in Old Irish. W. hwyl is equally 
perplexing. Breton and Cornish have for 'sail' a word 
borrowed from Lat. velum, Bret, goel, ' veil de f emme et de 
navire,' Corn, guil (Voc.). Did Welsh also once have a word 
*gwyl similarly borrowed from velum which affected the 
native word? Perhaps further investigation may bring some 



> On the long tow«1 is an unaccented syllable see Thnrneysen Se9. Cdt. Tii. 
325. Gaelic shows secondary shortening, muinealf eineal, gabhail, 
• EZ. xxTiii. 67 n. 



PEOF. 8TRACHAN. 243 

light. Ir. sedlf W. hteyl, 'course/ seem but a particular 
Implication of the above words. 

7. etl > el. ... 

Ml * mouth ^z^^gvetlo-: Goth. qMan 'say, speak ' (Stokes). 

cenSl * race/ W. cenedlz=i*cenetlon. 
. sc^l * story/ W. chwedl=:*8cpetion : Gr. iwerroD, 

Before slender vowels ^ bniiil Wb. 12* 12, M»7 74 9 : 
9ciuil Wb. 17^ 6. From cen^l the forms are numerous 
and varied— ceni?^tV Sg. 28» 6, 28^ 4, 32^ 3, 61» 24;. 
(ywiwtV Sg. 3P 16, 62» 3, 75» 6, 152» 3, 203» 19, Ml. 83>» 
5; cemM Wb. 5« 3; c<?w^itit7 Sg. 30^ 8, 31»» 12, 61* Ig; 
ceneiuil Sg. 40*, 4, 6, 18; ceniuil Sg. 32* 9, 40* 11, 61* 2; 
dochetiiuil Sg. 64* 6 ; ceineuil Sg. 4l» 3; ceneuil 33* 5, 75* 7, 
211* 16, Ml. 66* 1; docheneuil Ml. 103« 13; aocheneuil 101* 
19 ; cene&il Wb. 1* 12, 6* 6, 17^ 16. . Thus Wb., as might be 
expected, comes nearest to the later language where edi gains 
the upper hand. Most of these variations must be regarded 
as purely orthographical ; the presence or absence of the long 
mark is more or less a matter pf chance. But t6t, edi clearly 
represent two entirely distinct sets of sounds (the modern 
Gaelic pronunciation shows that u and o in these com- 
binations are long). Either the difference was dialectical 
or fttt, ^Oi arose from the loss of di^Cerent consonants ajnd 
were afterwards used promiscuously, because in n^ost cases 
the words would fall together ; why in this case edi, which 
is rare in Old Irish, should have gained the upper hand, 
it is not easy to see. That the difference is not purely 
chronological is diown by the persistence of iui in later 
times. 

epL 

No clear instance of this, fedil 'flesh' is probably to be 



* /orcainteeilsi VTb. 23<* 2, At nephceneil Wb. 6* 14, are accnsatiTes after the 
fashion of » stems, and might be explained like similar cases on p. 36. dochenktil 
gl. degener, Sg. 64* 6, u it stands for •*emetli8y would be an exception to ths 
role liud down there. But it might come from •^cerutlis (cf. Brugmann Grundr, 
ll. 1. 116). The aoc. to -*eenetli9 would be -^eenettim^ whence -*oeHeUim 
(Brugmann Orundr. II. 689), whence -*eeniil the influence of which may be 
seen in the forms quoted above. Socheneuil Ml. \Q\^ 19, docheneuil Ml. 103<: 
13, are datiyes after the analogy of i stems and consequently regular. 



244 COMPENSATORT LBNOTHEKINO IK IRI89* 

connected with 8kr. vapd 'fat/ but it may be referred to 
^tepoli'. In favour of this ia/eulae Ml. 97* 10,/euldae TO'* 8, 
ST'*?. 

8. ebl> il. 

aidhhheul (le sgread aidhbheul, Hyde, Leabhar sgeulatgh" 
ertchta 66^ ageulta aidhbheula ib. 7b), aidhbhiil * huge, yas*', 
enormous/ O.R. may come from ^atebeblos, a reduplicated 
form cognate with adbul * great, vast/ which may itself be 
connected with ^bhel 'swell/ (Persson, Wurzelerweiterung 
26). 

n^/ 'cloud/ W. n%wl:=>nebl(h. 

Before slender vowel niuil Ml. 120* 11. The effect of the 
labial is seen in the ace. pi. niulu Wb. 25'' 23; contrast 
bedlu 6« 11, beulu Ml. ISP 6, Incant. Sang. 

9. »c/+ broad vowel > *ecl, el. 
cSla ' will conceal *=z*cicldt. 

10. t^/+ slender vowel > tfiiV. 

^ififV ' adhaesit ' {rogiuil Ml. 98^* S)^*g%gk, pres. glenim. 

11. tftZ+hroad vowel=^J/, iU 
at'b^la * he wUl die '=^*bibldL 

12. iml. 

There is no proof that m was lost before /, so it is saf estr 
to look upon fn^la fut. of melim ' I grind,' as an analogicaL 
formation. 

13. ocJ > d/, ual. 

dital 'lock of hair'=rf()cfo-: Goth, tagl 'hair,* A.S. tmgl^ 
which has been compared with Skr. daf& ' fringe.' 

tin-dlaim *1 collect,' doinola adplicat: bl^^^^pod-^ Skr. 
piga 'fetter/ paciacor^ Germ, fiigen. Ablaut pa/c-, pofc-, 
pdH. 

14. ogl > d/, Hal, 

bHial *wsXeT*=bhoglO', cf. Germ, bach, "Eng. beck. 
fUal ' urine *=*ro^fo- : O.N. vfpkva * wet.' ^ 

e&ailean, ' curl ' : k6j(\jO(:' * a shellfish with a spiral shell " 
(Stokes). With kox^^, however, Franck Ndl. Wb. 262 
compares Dutch gagel, Ags. geagl ' tandvleesch, gehemmelte ' 

1 Stokes, MetT, It, 01, 72. 



PROF. 8TRACHAN. 245 

zrri^ghoghUh. If that be so, Stokes' etymology would liave 
to be given up. 

15. otl > dL 

If 61 * drink/ is to be connected with ^pd * drink/ it might 
tome from a stem *potkh.^ From 61 * drink ' can hardly be 
separated 61 * undare/ imr6l * abundantia/ for6il^ id., der6il 
* iuops/ etc.^ The root pdi, p6 shews similar meanings in 
other Idg. languages, Skr. pi^ pdyate^ * swell, be exuberant, 
be full/ Gr. wicop * fat.' 

16. odl> 61, (Ml? 

italach * burden '=jwm//!3co-? Cf. O.H.G. fasza 'bundle, 
load/ also Ir. osaar 'burden/ (W«-=*j»{>fo- *pod8', 

17. tM?/+ broad vowel > *ocl, d/, dal. 

cdala *I heard '=:*CMc/(>ra, cdalae *he heard '=*CMcfof». 

18. udl+hroaA vowel=*orf/, 61, Ml? 

buaiHm 'strike' may^^budl- or boudl-, y/bhud: Ags. bedtan^ 
Ung. beat. If ial here^zudl, it must have arisen in the 
£rst place before a broad vowel. 

19. u6/+ broad vowel > obi, 6l, Hal? 

giala * shoulder ^zn^gubl- (or gaubl-- ?) Slav, g&b' * bend ' in, 
^nqti * to bend,' G. Kv<f>6<;. Cf. O.N. bak, Ags. b(BC ' back '= 
'*bhogom, 'supple, flexible,' Germ, buckel 'back, belly,' ^bhug 
•bend.'» 



V. Sound Groups ending in i. 

1. ane, enc, nc > ic,^ 

These groups have in Irish fallen indistinguishably 
together. 

br^c * lie* =^*bhrancd or ^bhrencd: Skr. bhramga 'falling, 
loss.' 



1 TJnlefls we amune a radical yariatioii po^ poi, pou, like yAJ, phdi, gh&u, 
Penson, Wurzelerw. 117, 138, ttd, «»«, »t&u^ ib, 141, etc. 
> Aacoli, Lexicon FoUbo- Hibemieum cxr. 
* Perason, WurseUmmterung 190. 
« Cf. Bnigmaim, Orundriu I. 180, 203 ; R. Schmidt, Idg. For$ch. I. 66 aq. 



346 COMPENSATORY LBNGTHSMTKO IN IRISH. 

ffSc^ * branch *= *ceinc«2, W. cang {{.), pi. cangaui Oh; 
Slav, sqkii * surculus/ Skr. ganku- ' stake, trunk.' 

^cath * hook ' =^*ancatO' : Skr. aiiArd * hook/ Gr. ayicav, Syieoi:. 

^n ' necessity *=*anc^fi, W. a/?^^n : Qr. avdy/crf. 

Sc * death '=*jiA:w-, W. angeu, Bret. an(^ti : Skr. waf * perish/ 
Gr. y€KV<;, 

Micim 'forsake/ W. tranc^ trang * finis, obitus/ trengi 
^6bi^e, mori.* 

2. one > do. 

cdic ^&Ye*z=:*conce, *k^enk^e, Idg. *penge, W. pump. The 
^ for e must be ascribed to the preceding u sound, as 
in cuity *part,' W. peth 'thing, part '=*AJ?e2(/»-, • coirs 
• kettle '=*/:*mo-, W. pair: Skr. cdru, O.N. hverr? On 
the other hand e is unaffected in da// 'understanding/ 
W. pwt/ll z=.*k^ei8ld, * da, 'who,' W. pwg^*k9ei, dan 
'long'=*^f^fW, cf. Skr. drda 'long,* Goth, hveiia^ 
'while,' cend 'head,' W. pen=:*k^€ndO'\ cech 'everyone' 
=:*k!fekftO'f cethir ' four,' O.W. petgtcar^^k^etueres : Gr. 
Tiaa'ep€<;. In these oases (except in cethir) e appears as 
when the vowel in the following syllable is clear, as e 
when the following vowel is dark. Is this mere chance, or 
did the tf sound disappear in Irish before a dark vowel in 
the following syllable without affecting the e P cetheorn (em, 
of cethir :=*k^ete8ore8 might be explained as due to the analogy 
of the masculine. In cethir the loss of u may be explained 
by dissimilation due to the following u : in that case *k^enk^e 
for *penk^e must be supposed to have arisen within Keltio 
itself^ after this dissimilation had taken place. If this 
dissimilation is to be placed in proto-Eeltic times one 
would have expected in W. *cetguar ; petguar may be 
ascribed to the influence of the feminine. 

* For a suggestion as to the explanation of initial ^ for see Betz, Beitr, xiy. 
313. 

' Thomeysen, Kelto-Romanisehei 71. 
« Windisch, KSB, viii. 44. 
, ^ Brugmann, Orundriu II. 1. 194. 

* Cf. Osthoff, Morph. UnUrfueh, iv. 152. 
« R. Schmidt, Idg, Forseh. I. 73. . 

^ Brugmann Grundrisa I. 170 has suggested an historical connection between 
Keltio *An^>»/U^ and Lat ^iw^w> Goth.^ni/: . > 



TROF. 8TRACHAN. 



S47 



3. ine} une. Of compensatory lengthening of i or u 
with loss of n I have no instance. In the case of u there 
is an example of apparently different treatment in sluccim 
*1 swallow/ Mod. Ir. Bluigim^ O.Bret rokincas, W. Ikmc, 
where nc seems to be assimilated to ce. Of. also derucc 
* acorn* =^*derun'eO' (?), derun' weak form of stem derven-t 
cf. W. derwen *oak/ Lith. dertdnis 'made of pine wood/ 
dendngas 'resinous.' Stokes derives ticeim 'come' from 
to-enk. If ice here comes from enk (or nk)^ did it come 
through inc f Whether this lack of examples of length- * 
ening of % and u is due to chance, or to something in 
the nature of the sounds themselves I am not in a position 
to say; at all events a prolonged search has brought to 
light no instances. There is the same absence of examples 
of lengthening of t^ u in int (before slender vowel)^ unU 



VI. Sound. Groups ending in L 

1. ant, ent, nt > et} 

bSt ' hurt ' =.^-^henti' or -^-^hnti- ^j^hen. ' 

cH * hundred/ W. cantz^*kmt6m : Lith. szimtas. 

' O ' 

' c^t' 'first/ eetne 'first/ W. cj/ntaf, Gaul. Cintus^ Qintugnatus, 
=*c»w/o- : Ch. Slav, clna* 'begin/ 



1 Bmgmaim once Morphol. Uniertueh. III. 154, dsrired Meim ' I leavB ' 
from *lineim with a tranaition to the 3rd conj. That is highly improbable. 
Supposing the verb to have belonged to the let conj. e for t could have arisen; 
only where a broad Towel foUowM, and if the inflexion according to the 3rd 
conjugation is older than the umlaut, it could never have arisen at ail. He now 
(Grundriss I. 327) suggests UikfUOf but there is no evidence that U%kUid could 
give leicim. It seems to me that Old Ir. leieim, Mod. Uiff can be explained only 
from ^UinhtiOf a mixture of leiq- and linq-. How could have made its way 
into leinq^ is easily intelligible if, as may be easily supposed, there was by the 
present *linqd (or *linqio) a i\ii. ^leiqso, Aor. *(tf)&iy*J», etc. 

' Cf. Brugmann, Grundritt I. 203 ; R. Schmidt, Idff, Forseh, I. 64 sqq. 

' Schrader, Sprtichvergleiehung und Urgeieh^chte^ 537. With Cintut compare 
the Ir. proper name Cet^ npwroT. 



248 COMPENSATORY LENGTHSNIKO IN lEISH. 

cSial * song *=^*€anthn, canim * I sing * : Lat. cano. 

d^t * tooth/ W. dantz=L*dnt' : Goth. tun/>us, 

it * jealousy *:=ian(0', W. add-iant 'longing/ Gaul. Ao 
iunneni, Jantumarua, Skr. yatna" * effort.' ^ 

itim * I clothe/ itach * dress ' may be compared with 
ent, int 'weave/ Gr. arrofiaiy Skr. dtka" 'cloak/* 

itaim ' find ' : Goth, finpauy Eng. yJmf.» 

mHt size, W. mam^=:*mn(t-. 

' o 

»4t ' way/ W. ht/nt=*8entO' : Goth. %inp8.^ 
' aSt^ ' likeness '=*wm^- or *«w/-, cf. amal 'as*: Lat. «] 

O ' 

Gr. &fJL€L 

set ' treasure 'znsent- (stem uncertain cf. BB. xi. 99) : 
sdnf' 'being/ Gr. eU (Stokes).* 

tSt ' fides/ W. tant=*tntU'. 

/^7' ' way '=*^em^- or *init' : rifivto. 

trit ' flocky herd/ (g. ^re(5»^) = ^trento" : Lat. /t< 
with which Bugge {KZ. xxxii. 67), has comp 
Arm. tarm ' flock of birds.' 

B. Schmidt has acutely perceived that et^=:nt is clc 
distinguished from etz^ant, ent in the t infection; here 
nt gives ^/, St=^ant, ent gives ^litt^, ^/^, e.g. cSt 'hund 
g. c^*Y (=*ce>i^l) but, it 'jealousy,' g. ^wiV {^*yantt) 
32^ 10), ^ot7 (Ml. 32^ 9) sit ' treasure/ siuit, sioit. 

2. (m^ > dt, 

airchdt ' injury ^ z=,^{p)ari'konti'? 

moit ' oath *=z^monti'fi 

The etymology of ^^ ' sod,' is obscure ; doH ' hand, w 
gen. pi. inna n-doat 'lacertorum' (Aug. Gr. 92), ma; 



1 Stokes, CMdeliea^ 85. 

* Cf. Cr. Meyer. ^Ift. iS^Mf. III. 24. It is equally possible (Stokes, Spra 
32) to compare ^tim with Lat. pamtMy etc. 

^ Stokes, 5^. xi, 140 ; R. Schmidt, IF, I. 64. 

* Can the gen. i^ta etc. be due to association with imtheehta with 
it is often found joined, as in the phrase t oend %4ta oeu* inUheehU, e.g 
2530 13 P 

> Stokes, K8B, viii. 328. 

* If this comparison is right, it furnishes an additional proof that iia 
form of this part, was ^aenU (cf. /<fy. Fwraeh, I. 23) for tiuit comes from 
not from *»^tu 

^ Stokes, Bets, Beitr, xrii. 137. 
^ Stokes, Breton GlotMet 17. 



PROP. STRACHAN* 249 

^rived from ^daventi'^ ^dheva-, *move violently/^ cf. 
fcapjTti^ 'wrist* to KOfyrroKifLo^t O.H.G. hwerban^ 'turn':* 
^^^'^w/- to ddaff ddt, as *ipvenko8^ young man, to ode, 6c. 

3. int+hrosLd vowel > *ent > iL 

y<^taim, Bitaim *I can '=*««/i/di6: Goth, mnpa 'strong/ 
^^fy'an * tcpareiv.' * sStaim appears in setar LIT. 68^ 2. 

4. anci, enct > ict, icM. 

^^chtaige (read tichtaide) ' frozen * (Atkinson, Passions, etc. 

*'^)> Gael. Uuchd * congeal *=*^flwc^- y/tank (or tenk?), cf. 

^^^icthea ' concretionis/ Ml. 145* 3 = *con-^anc«/oti« (or 

^e- P) (also Ml. 44* 10, 62<» 4, 84^ 7) : Zd. tancista ' very 

^^•»»i,' Lith. towAt^ ' thick,' Arm. t'anlr ' thick/ * 

^aeht is given by O'Clery with the meaning of 

^^ under.' There is also a word dinge with the same 

I'^^^ning. As it is difficult to separate these two words, 

^ very likely that deacht should be written diachi^=> 

^^^<:U: dinge=*dingiO' or *dingid. The etymology is un- 

^*^in ; one might perhaps compare dingim ' I thrust, 

*"^3,' Ch. Slav, dcigii * strength ' in nedqgH * sickness.' 



^cht * murder '=em'^- to ic * death.' 
^^ehte * fitting '=*^nc^io. to *tenqd Fick I*. 440. 
^r^cht * song,' cf. Zd. drenj * repeat, utter.' 
^^cht 'power,' has been compared by Stokes^ with Skr. 
i^'^di. « power,' y/Rak-. If Fick I.* 41 is right in referring 
^ ^his root Lith. szvdnkus ' becoming,' we have a nasalised 
^^*^3i also in Lith. 

. ampt, empt > and, enct > icht perhaps in — 
ht^ gL buris 8g. 127^ 1 (1. dcht). Mod. Ir. eeucht, g. 
^^^<^Jita, Manx keeaght ' plough '=*«inc^w-, ^kamptU'i Gr. 
*^j^^-3rn», KaiAirM aporpa. 



\ ^ickj I*. 76, 466. 

C nrtiiM, Gr, Et.^ 626 ; Schrader, KZ, xxx. 473. If the connexion between 
''"P'vA and hwerban is to be maintained, KopvSs most stand to kwerban in the 
JIJ^^ relation as laiirpot to Qtoi^i.^fhwapjan * ff$€pp6imi^* lith. kvaptu * smoke,' 
"•,-^«chtel, Hauptproblemt 366, Wiedemann IF, I. 266. 

^ Stokes, Bodleian Fra^nmtt of C<>rma^% Glottary 64. 

, '^jS'Binst this Bugge KZ. xxxii. 68 sq. 

, v, ^^kes compares more probably Goth. Hdha * plough ' ; *eanetu* : leaned' 
"•»ush '^Mha: Uih. izakd^ « branch.' 




5 7 ■ r T yj 'Tn r- — m v^ 



vr\\ 



250 OOMPENSATORT LEKGTHENIKO IN IRISH. 

dricht ^^paTt*^*drempt' a nasalised form of y/dvep in Gr, 

crMit 'wound,' may perhaps be referred to *krempfdf 
nasalised form of V kerp^ in Skr. krpdna 'sword/ Lith. kerpu 
'cut' 

6. Of ond I have no example; the etymology of tdcht 
'piece/ tdchtad 'hewing to pieces' (LL. 101^ 8) is not clear.^ 

(6. ec9t . . 

appears in auslaut as i, dM, fof^^ gi from ^dkcat^ *foretsfy 
gesfit} So ec8 in 6i 'six'=*«wfA;«. But the lengthening 
here is not compensatory: rather it seems to have been a 
law that accented monosyllables ending in a vowel were 
lengthened/ 



VTI. Sound Groups ending in a. 

1. ana > ^«.* 

^si ' reins/ stem ^ansi- : Lat. ansa, Lith. qad ' handle, 

knot/« 

fisyfiadg 'beard '=f>an«-: O.Pruss. tcanso, Ch. Slav, com* 
•beard.' 7 

geh ^swBLn*=^*gan8i' : Germ, ^ana, Lat. hanser. 

2. ena > ia, 

driaaacht 'a rattling or creaking sound '=Vr«n«-ac^t/- : 
Skr. dhran {dhranati gabde), not found in literature, Goth, 
drungue <l>$6yyo<:,rOJN. drynr ' roaring.' • . 

» Stokes, K8B, vii. 67. 

' Stokes suggests *»U)nk-y *9tunk- cognate with N.H.6. ttueke, stoek^ etc. 

' Bezz, Beitr, xiv. 313, n. 

* Thumeysen, £Z. xxxi. 91. Cf. also m^ * I, me'sldg. *tne, Brugmaim, 
Grundriss II. 811. How is Gael, mi to be explained P Does it=Idg. tne ? 

* If £tu8t the name of a Gaulish deity, ^Teut. *ansu' *god,' as is highly 
probable (Fick. III.^ 18), then an9 became it in the Gaulish branch of Keltic too. 
The change ittu > is may very well have taken, place before tiie separation of the 
Kelts. 

* Stokes, Linguittie value of Irish Annals 8. 
"^ Stokes, BezB, Beitr. ix. 89. 

* Persson, WuneUrweiterungf 73; Fick, I*. 76. • 



PHOF. STRACHAV. 251 

gr^ssaehi * inciting, urging on * ^^grem- : gria * fire/ grisaim 
•incite.' Ablaut grem-, Ir. gr^88acht, Bret, ^roea *heat of 
the sun/ Skr. ghrdnsa^ 'sun's heat': ^rni-, Ir. gris^ griaaim} 

b^im=i*ben8men. See above. 

It has been held that ens in Irish became 28,^ but the 
above instances, if rightly explained, prove the contrary. 
Zimmer lays stress on mi * month/ which he seems to derive 
through *men8 from miM. But mi, g. mka, cannot be separated 
from Brythonic mis, in W., Bret, and Com., and in Bret, after 
the analogy of groez we must have had for *m^«-, *moeZy 
or *moaz. These facts indicate that ^mitts^ became in ur- 
Kelt. *min8' (perhaps mis-) ; if a long vowel + sonant + 
consonant was shortened in Keltic as in other Europeaif 
languages, this shortening must have come after the change 
of ^ to i. The borrowed words cia, piaaire from ceMus^ pensum 
prove nothing ; i for e is found in other cases where there 
can be no question of n : aita * silk,' Low Lat. aita^ airic, Lat. 
airicua. I for Lat. i appears also in Teutonic loan words, 
O.H.G. aida=iaetaf O.H,G, chrida=zcrita, OJS..Q. pina:=pina, 
Goth. (ikeit=iacetum, Ags. pialic^=^penaum.* These instances 
are probably to be explained from the closer sound of Lat. 
e, that it was something between Kelt, e and Keltic 2 and 
was expressed now by the one, now by the other, i also 
appears in Tout as in Goth, m^aa- O.H.G. miaa ; this i must 
have been a closer sound than i which becomes d.^ 

3. ent (ntt) > enas > m, ea, 
. b^ * custom '=*6^rtM-,* Gaul, biaatia,^ ^bhendh* hind* : 

Alb. beae * belief, agreement/ ^ 

.... 

• - • 

» Bezz, Beitr, xiv. 314. 

. » Zimmer, JCZ. iix. 210, 
» Cf. Kloffe, PauV $^QrundHM L ' 

^ In speakiiig of the change of Idg. i to West Germ, d, Eluge {PauVf 

OrundrtMs I. 363), remarks: 'dabei ist zu beachten, das kein S eines Lat. 

lehnworta {aeStum rhnui fnima catMa monSta u. b. w.), den wandel yon S ia ^ 

durchtnacht ; offenbar deckten sich lat S und idg.-germ. S nicht.* 

• JBezx, Beitr, xiv. 312 sqq. 

* Bestia dicitur de beasu, hoc eet more feritatis, Yirg. gramm. quoted 
by Holder, Alt, Kelt, Spr, 409. From this can hardly be separated 
Bret, boaz * custom.' But W. moes is difficult. Confusion of h and m is found 
elsewhere, because b and m fell together in certain positions, e.p, ben, men 
* wagon * ; but e should in W. haye giyen wtf. 

' Meyer, Alban, WTf, 33. 



252 COMPEKSATOBT LENGTHENING IN IRISH. 



c(%sa%ni * B\xSeT ^ =::centi' or cents* ; Lith. kencxiu, ke^ 

* suffer/ 

gUse ' brightness '=*^/en^-^-, of. Germ, glanz and oth^ 
words referred to by Kluge, s.v. 

griis 'attack' = ^re»/^t-, grend-ti-, in^rewmVn 'persequor 
Ch. Slay, gredq * come.* 

ISss Might/ l^asboire * light *=z*plenUo- : Lat. $pkndeo.^ 

aSia * music/ =*«e«^W-, aennim ' I play.* 

In such instances Gaelic has also by infection edi^ e.g. ce 
g. cediB^ km Might/ g. Iedi8^*plensiy gleus * order, condition 
g. gledia. I have no similar examples from the older Iri( 
unless in Oormac's Glossary s.v. grinniud leos i. dibad soil 

* extinction of light/ we should read ledis, as the gen. 
required; from ledis would come nom. kAa, On the oth* 
hand I have no evidence that ^ in a similar positic 
was not treated in the same way in Irish. The instano 
gSUf griiSf s^s, are not conclusive^ since it is by i 
meaus certain that a short i could effect a preceding 
in this way. In all' the cases collected in this paper the i 
long. On the other hand i remains in c^r ' gum/ r< 
'clear/ where the following syllable once contained 
Whether the etymology proposed for rSil is right or wron 
fortunately does not matter ; the verb rilaim points clear] 
to compensatory lengthening; if ^ in r4il had come fro: 
ei we should have expected *rialaim. 

4. anc8, enc8 > iss, is. 

gisca 'branch * = *cane8€aiO' to gSe 'branch '^z^cancdf ^ 
cang (see above). 

&ea * moon* ^encseaio; cf. Sig 'moon,* 0'Clery=*^nci 
The words are probably to be connected with Skr. pdj) 
'light* Gr. ^efyo^. As to end- the e may be explains 
in one of two ways ; either y/peng had a by- form pen 
or *«nct- stands by Stokes* law for *engni' ; the lattt 
explanation is the more probable, encs- in ^sca would con 
from a stem ^pengea- with the weak form of the suffix. 



» BB, xiv. 313. 

* For the doubtful ioeherUuil see p. 27. 



PROF, STBACHAN. 253 

Gael. eeu8 ' ham * g. cedis (also nom. eedt after the gen.) = 
^cencso' : Lith. kenkld * hough.' 

^ 'footstep* = •em^i-, cf. eng * footstep.' ^encBi- is 
probably developed from a neut. stem ^enges", 

gimm * I cry '=-*gencsid, cf. g^m. 

do grSa ' continuo/ grissach ' continuus*=*(grretic«-, ^grengs- : 
Lith. grikiif 'come back,' atgrqzaSf 'repetitioOi strophe/ 
O.N. kringr ' round,' N.H.G. hring} 

li% * bladder ' perhaps = ^lencao^ : Lith. lengvua 'light.' For 
the meaning cf. etromain, 

5. ent8 > €8. 

8^8', fut. stem of 8ennim {z=*8vendd) ' drive '=*«wen^-.* 
ingl^t gl. rimare Ml. 140® 7, fut. to inglennimf 3 pi. 
inglennat^ gl. vestigaiit. 

6. in8 > 18. 

gri8 *&re*:=^*gr%n80', gf^mo* (see above). From this it 
appears that t before n8 is not subject to umlaut. 
i8 * below ' W. M=t>i«- : Lat. infra for *in8rd.^ 

7. on8 > d8, 

/O'lds' (=-feik»-), fut. stem to folangim 'I endure.' 
fri8ta88am 1 pi. fut. iofn8toing^ Z. £. 1005. 

8. un8 <U8? 

I have no certain instance of this in Irish. W. cm * kiss/ 
may be explained as=*ct<;}«-, cf. /n/i/ioi = /n;-i/6-cr-<k> ^ to 
i'/cva-aa. It would be possible to explain in this way 
adgmim 'I choose'; gu8=gun8' a present stem with nasal 
infix from ^gevfi^ Gr. 7ew», Eng. choo8e. The i/, however, 
might be explained otherwise. 

9. r« 

Zimmer^ has asserted that in Irish ar8 became 8r, ers, ir ; of 
the former change he produces no instances except futures 
like adger which may be better explained otherwise (see 
above, p. 4). For er« > ir he cites tir 'land' Vter8: Lat. 



1 Leskien, Ablaut im litauischm 66 (328) ; Fick, 11^ 352. 

* Brugmann, Grundriu II. 180. 
» Thurneysen KZ. xxx. 491. 

^ Johansson, De dmvativit verbis etmtraetii Hnffuae grateae 109. 

* KZ. XXX. 211. 



^25i COMPENSATORY LENOTHENING IN IRISH. 

terra, eir * comb ' Vkera^ : Skr. karsh ' plough.* But there is no 
proof that t of tir came from irs; ^ is also found in Osc. 
teerum * land/ In tir^*(iro8, Lat. terraz:z*Ur8a we may 
have an ablaut ii. dr may be better derived from ^dsra, 
Ch. Slav, cesati ' to comb/ ceslH 'comb.' In Besot. BeUr. xiv» 
315 I have collected a number of cases (which might 
be easily increasedj in which r« becomes rr without affecting 
a preceding vowel. Until Zimmer has brought forward more 
certain instances, and has shown reason for the double 
treatment of ra^ it is impossible to accept his law. 



255 



INDEX OF KELTIC WORDS. 



Old KtUie. 

Adiantunneni, 32. 
anax, 7. 
ArebrigQOs, 14. 
AXPOTALVS, 19. 
bessns, 35. 
Broccagnoe, 6. 
Catusualifl, 24. 
Cintugnatus, 31. 
Ciatus, 31. 
Efloa, 34. 
Exobnus, 8. 
Gabromagus, 22. 
Gabroeentum, 22. 
Jantumarus, 32. 

KOVpfUf 19. 

Licnoe, 13. 
Maglus, 24. 
Senoroccus, 25. 
Suagroe, 21. 
Te'Soignioe, 13. 
Veragri, 21. 
Yernodubrum, 22. 
TertraguB, 8. 

IrUh. 

abra, 22. 
adbul, 28. 
aderad, 22. 
adgCisim, 37. 
aicc, 2. 

aidbbheul, 28. 
6il dUgraeej 24. 
ijlpleasantf 24. 
fiin, 6. 
ainm, 18. 
airch6t, 32. 
&1, 24. 
frlaind, 25. 
amm, 7, 17. 
&n, 7. 
&na, 7. 
an&l, 25. 
ir, 21. 
ar^erat, 21. 
&niny 21. 
arrocb^, 22. 
atb^la, 28. 
becc, 2. 

b6imm bloWy 18. 
\ii^mm journey f 18. 



b6l 27. 
b^ra, 22. 
b^, 35. 
b6t, 31. 
bl6n, 5. 
boc, 2. 
bocc, 2. 
bocht, 12. 
bomm, 16. 
brec, 29. 
brecc, 2. 
br^n, 5. 
Br6n, 14. 
briathar, 21. 
brocc, 25. 
Broco§ui, 6. 
br6im, 16. 
br6n, 11. 
brCian, 14. 
baailim, 29. 
b(iain, 12. 
baal, 28. 
bCian, 15. 
bCiracb, 23. 
bCiretbar, 23. 
cacc, 2. 
c&inim, 5. 
cech, 30. 
cechna, 14. 
c6cht plough^ 33. 
dkehi power f 33. 
c^imm, 17. 
c6ir, 20. 
c^I, 24. 
cela, 28. 
o^le iervMf 23. 
c61e eomradgf 26. 
cend, 30. 
cend, 27. 
ce6l, 26. 
c^ssaim, 26. 
Ca, 31. 
cHy 31. 
c6t, 31. 
c6tal, 32. 
cethir, 30. 
cetne, 31. 
cia, 30. 
ciall, 30. 
cian, 30. 
cich, 20. 
cir, 38. 



256 



INDEX OF KELTIC WORDS. 



cis, 35. 
cloch, 2. 

cltLain mMuUnc, 13, 
ciCiain deceit, 13. 
cnoco, 2. 
c6ic, 30. 
coire, 30. 
cr&in, 6. 
cr^ht, 34. 
ciiathar, 21. 
croccenn, 4. 
cr6n, 11. 
crQan, 11. 
c(iailean, 28. 
caala, 29. 
c6an harbour, 13* 
G(ian hoetf 16. 
cCiaii^ne, 16. 
cQar, 23. 
cuic, 26. 
cult, 30. 
d&l, 26. 
deacht, 33. 
de6r, 6. 
d^r, 20. 
der6il, 29 
denicc, 31. 
det, 32. 
dinge, 33. 
dir, 22. 
die, 34. 
dobur, 22. 
doff6aa, 14. 
d6it, 32. 
domme. 7. 
dorachitiir, 22. 
di^ht »ong, 33. 
drecht part, 34. 
dr^mire, 17. 
dr^ssacht, 34. 
dtial, 28. 
ddaa, 16. 
duin^y 14. 
^, 30. 
6cath, 30. 
^n, 30. 
^ht, 33. 
6ig, 36. 
eiBf 37. 
^laim, 8. 
6mdiiii, 8. 
ea, 10. 
^nirt, 8. 
6r, 20. 
^sca, 36. 
6si, 34. 
6t, 32. 
4taim, 32. 
aim, 32. 
f&n, 6. 



Un, 9. 
fe6il, 27. 
f6r, 21. 
f^, 34. 
f^taim, 33. 
fod^ma, 19. 
fodidmat, 19, 

iOT^y 34. 

fordil, 29 
fr^m, 20. 
trii'g^n, 22, 
fCiaimin, 19* 
faal, 28. 
faan, 16. 
idai, 23. 
gab&l, 24. 
gabor, 22. 
gairm, 19, 
gkc, 30. 
g6im, 17. 
g^is, 34. 
-g^, 9. 
g^nar, 9. 
g^sca, 36. 
g^rait, 21 
gSssachtach, 19. 
g^ssim, 37. 
g#t, 22. 
-g^uin, 10. 
giuil,28. 
gl&m, 16. 
gl^, 36. 
gT&in, 6. 
gr^imm, 18. 
greis, 36. 
greit, 21 
gr^a, 37. 
gressacht, 36. 
grfs, 36, 37. 
giisaim, 36. 
^Ciala, 29. 
imthanu, 7. 
inde6iii, 10. 
ingleis, 37. 
is, 37. 
larvae, 13. 
Uana, 8. 
lecc, 2. 
leco, 2. 
l^icim, 31. 
leimm, 17. 
l^maither, 22. 
L^n, 13. 
len, 2 
lenaim, 2. 
l^ne, 3. 
le^, 36. 
Ids, 37. 
1^, 36. 
lian, 9. 



INDEX OF KELTIC WORDS. 257 



s^taim, 33. 
sine, 35. 
sita, 3d. 
slemon, 11. 
sluccim, 31. 
8mer6it, 22. 
flmtiainini, 15. 
Boinme, 7. 
8r6n, 12. 
st&n, 6. 
sQan, 15. 
t&l, 24. 
t&naise, 7. 
tarathar, 21. 
techtaide, 33. 
t^chte, 33. 
ten, 10. 
teas, 10. 
ikijidet, 32. 
UX way, 32. 
ticcim, 31. 
tics&I, 24. 
timme, 10. 
tmdlaim, 28. 
tlr, 37. 
tl&m, 16. 
t6cht, 34. 
toll, 24. 
t6n, 14. 
tr^cim, 30. 
tr6n, 7. 
tress, 8. 
tr6t, 32. 
trice, 8. 
trom, 17. 
(iaimm, 19. 
(iain /oan, 11. 
(iain 2mur«, 15. 
tialach, 29. 
(iaU, 24. 
(iamon, 20. 
tian, 12. 
(iane, 12. 
tiar, 23. 
in bad, 23. 
ftr,/rMA, 23. 
usee, 25. 

Oaelie, 
abhra, 22. 
braim, 16. 
c^, 20. 
ceufl, 37. 
cineal, 26. 
dream, 17. 
dreamag, 17. 
gabhail, 26. 
ffleufl, 36. 
fedn, 2. 
leiu, 36. 

11. 1891-S-3. 17 



258 



INDEX OF KELTIC WORDS. 



meanma, 19. 
meunan, 3. 
mi, 34. 
muineal, 26. 
rnioean, 25. 
tenchd, 33. 
teum, 18. 



keeagbt, 33 
keeyr, 20. 
Iheanee, ft. 



tddiant, 82 
ael, 24. 
aele, 24. 
ter, 21. 
anadl, 25. 
angen, 30. 
angeQ, 30. 
blaen, 5. 
braen, 5. 
brocb, 25. 
brwyii, II. 
brych, 2. 
bwch, 2. 
bychan, 2. 
caifael, 24. 
cam, 17. 
cang, 30. 
canty 31. 
cenedl, 27. 
chwed). 27. 
cig, 20. 
cilyd, 26. 
clwch, 2. 
clwg, 2. 
croen, 4. 
cus, 37. 
cwrw, 19. 
cyntaf, 31. 
cywain, 9. 
dacr, 19. 
dadl, 25. 
dant, 32. 
defnyd, II. 
derwen, 31. 
dir, 22. 
drem, 16. 
dwfr, 22. 
dwyn, 4. 
dynyn, 14. 
etn, 10. 
ffroen, 12. 
gafr, 22. 
garm, 19. 



jKJffMT. 



Wthk, 



goer, 28. 
graen, 6. 
gwaen, 7. 
gwair, 21. 
gwell, 5. 
gwn, 16. 
gwreidd, 20. 
baeru, 21. 
boenyn, 9. 
bui, 15. 
bwyl, 26. 
bynt, 32. 
byap, 25. 
ir, 23. 
is, 37. 
Ham, 17. 
Uech, 2. 
Uwnc, 31. 
Uwyn, 9. 
Uych, 2. 
Uyfn. U. 
Mael, 24. 
maini, 32. 
mawn, 11. 
min, 3. 
mis, 35. 
mocb, 2. 
moes, 35. 
mynwel, 26. 
mynycb, 2. 
nitb, 3. 
niwl, 28. 
oen, 12. 
oer, 23. 
pair, 30. 
pen, 30. 
petgnar, 30. 
peth, 30. 
pib, 26. 
piipam:, 26. 
pump, 30. 
pwy, 30. 
pwyll, 30. 
rhamii, 18. 
rbegen, 17. 
sawdl, 25 
sarbao, 21. 
taeru, 21. 
tam, 18. 
tant, 32. 
tin, 14. 
tranc, 30. 
trecb, 7. 
tren, 7> 
trengi, 30. 
trwm, 17. 
twll, 24. 
twym, 19. 
y gilyd, 26. 
ystaen, 6. 



INDEX OF KELTIC WORDS 



239 



bomm, 19. 
brun, 16. 
coruf, 19. 
engurbor, 7. 
enniaii, 10. 
goU, 26. 

grwF. 21. 
mis, 35. 
€01, 23. 
tarn, 18. 
toim, 19. 
tommys, 19. 

air, 21. 
ancon, 30. 



Comiih, 



Brtion, 



anneffn, 10. 
boaz, 35. 
boem, 19. 
bramm, 16. 
cro^chenn, 4. 
doen, 4. 
eal, 24. 
goel, 26. 
groez, 35. 
gueun, 7. 
mis, 35. 
rolnncas, 31. 
seul, 25. 
tamm, 18. 
toem, 19. 
tomm, 19. 



[The reason for the past discontinuance of our yearly Dic- 
tionary Reports was merely want of funds. As the Dictionary 
work suffered in some degree from this stopping of news, and 
of recognition of the labours of our helpers — ^whose collection of 
the uses of words constitutes the chief value of the Dictionary — 
the Council hare resumed the publication of the Eeports at the 
earliest possible time, almost before the Society's funds justify 
the step. 

The issue of the present Part of Transactions with the Dic- 
tionary Reports has been delayed, first by Mr. Bradley's regrettable 
illness, and then by the pressure of Dr. Murray's other work. 

My daily help to the Dictionary is (as it has been for many 
years) cutting out slips from the newspapers, periodicals, and 
proofs or revises of Early English Texts as they pass through 
the press, with an occasional hunt for an early instance for Mr. 
Bradley. Looking back to 1858, when Herbert Coleridge and I 
started the Dictionary work, I feel satisfied that, however faultful 
the result may be, the Society's JVew JEnglish Dictionary is the 
best existing Dictionary of any modem language, and is a 
credit to our Victorian time. — E. J. Euenivall, 18th Jan. 1893.] 



261 



APPEKDIX. 

I. Eepobt ok the Pboobess of Vol. III. of the Societt's 
DioTiONABY. Br Henbt Beadlet, M.A., President (Editob 
of Vol. III). 

{Bead at the Meeting of the Society, Fehmwry 19M, 1892]. 

''The Society is already aware that in August, 1891, the 
Clarendon Pre^s issued the First Part {E — Every, 344 pages) 
of the portion of the Dictionary entrusted to my editorship. 
Of the Second Part there are now in type 115 pages, extending 
from Everybody to Extempor%%$ ) 64 pages, ending with the word 
ExhihitioHy having been finally passed for press. 

"The reviews that have hitherto appeared of Part I. have 
been, so far as I know, without exception, highly favourable in 
general tone. I have to thank my reviewers for many criticisms 
on points of detail that will be useful for the improvement of 
the forthcoming portions of the work. Even when I am unable 
to agree with my critics, their remarks have not unfrequently 
been of service in emphasizing the necessity for exhibiting, as 
fully as the limits of space permit, the evidence on which the 
conclusions adopted are based. In one or two instances, in which 
pertinent criticism was afforded by competent specialists, I am 
happy to say that I have been able to obtain from my reviewer 
a promise of his assistance in dealing with matters belonging 
to his particular branch of knowledge. 

"The words beginning with E, which occupy the whole of 
the portion of the third volume now in type, and will for some 
time longer continue to engage the attention of my assistants and 
myself, are remarkable for the large preponderance of foreign 
derivatives, the native element being very scantily represented. 
Technical words belonging to science and philosophy are ex- 
ceptionally abundant, and their treatment has often required 
a good deal of research, and recourse to the advice of specialists. 
Although the E words, as a rule, do not form a very interesting 
portion of the English vocabulary, either with regard to etymology 
or to sense-history, there are still many instances in which our 
material has disclosed facts not generally known. The articles 
in Part I. which contain matter specially worthy of attention 
are enumerated in the prefatory note. Of the points of interest 



262 DICTIONARY REPORTS, I. — MR. BRADLEY. 

occurring in the subsequent portion already in type, a few may 
be here mentioned. Under the word JExeite, I have adduced 
evidence to show that the word (which came into English from 
Dutch) is ultimately connected etymologically with the Latin 
eenmSt not, as hitherto supposed, with Agsiu, The word SJvil, 
in modem use expressing only positive badness, retained until 
the 16th century its original wider sense, and could still be 
U3t*d as expressive of mere depreciation, as in the invitation, 
'come and take an evil dinner with me.' The etymological 
equivalents Evict and Evince were formerly used indiscriminately. 
Their primary meaning is *to gain by a victory,* or *to expel 
by a victory.* In early use it was possible to speak of * evicting ' 
or * evincing ' a country by force of arms ; but the prevailing use 
referred to a victory in argument or in a legal contest, and 
eventually the two words became differentiated, evince meaning to 
establish a conclusion by argument, and evict to gain possession of 
a holding or to turn out its possessor as a result of a legal contest. 
The now familiar word exiel^ it is curious to find, cannot be 
traced earlier than the age of Shakspere, though existence 
meaning * reality,' goes back to Chaucer. The words containing 
the Latin prefixes e or ex and extra afford many illustrations of 
tlie tendency of words that originally expressed mere facts, to 
acquire an emotional sense ; thus enormoits, exorbitant^ extraordinary y 
extravagant^ all originally meant simply ' out of the common rule 
or course * ; but in modem use they express the speaker's feeling 
of astonishment, admiration, or disgust. Another instance of 
this emotional connotation of words is exquisite, which originally 
meant only 'carefully sought out,' * well-chosen,' or in an un- 
favourable sense * abstmse, out of the way.' One application 
of this word has often been misunderstood. The old medical 
term 'exquisite fever,' 'exquisite erysipelas,' did not mean an 
acute or specially severe form of the disease. The Latin exquieitus 
was a translation of the Greek aKpiprjvy accurately determined, 
so that an exquisite erysipelas meant an erysipelas accurately 
so-called, the typical form of the disease, or the genuine d^ase, 
as opposed to a spurious one. 

" The treatment of pronunciation in this portion of the Dic- 
tionary has presented some special difficulties. This is due chiefly 
to the large number of classical derivatives which, although they 
may be more or less frequent in literature, are so seldom pro- 
nounced that there exists no orthoepical usage with regard to 



BICIIONART REPORTS, 1. — MR. BRADLSt. 263 

them. In some cases it has been necessary to choose between four 
or five different pronunciations, each of which is supported by the 
authority of some dictionary of repute, and by an analogy which 
would be sufficient to decide the question if it stood alone. In 
dealing with such words I have been guided by the consideration 
that with reference to them the normal relation of written to spoken 
language is reversed : the combination of written letters being the 
real word, and the corresponding combination of sounds merely its 
symbol. In words of this class, therefore, the best pronunciation 
is that which most effectually and promptly suggests to the mind 
the written form of the word. 

'' The constant recurrence of difficulties of the kind just referred 
to has strongly impressed me with the conviction that the objec- 
tions to the adoption of phonetic spelling are far more serious than 
the advocates of reform are accustomed to admit. I will even 
venture to say that the opponents of reform have seldom, if ever, 
done full justice to the strength of their own case. I quite admit 
that the want of correspondence between our spelling and our pro- 
nunciation is a great evil. Whether it can be remedied I am not 
sure; but it is quite certain that no reform can be satisfactory which 
is based upon an imperfect recognition of the force of the arguments 
on the conservative side. I therefore think it is worth while to 
attempt to call the attention of the Society to certain considerations 
which, as it seems to me, are of essential importance and have been 
very generally overlooked. 

"In the first place, the objection felt to all orthographical 
change as such is very far from being a mere matter of sentiment. 
It might be so described if the function of writing were merely 
to represent speech; or, in other words, if whenever we looked 
at a printed page we were obliged to think of the sound of the 
words before we could apprehend the author's meaning. This, 
however, is very far from being the case. By long habit certain 
groups of printed letters have acquired for us an ideographic value, 
so that to every educated adult the act of reading is partially, 
what to a deaf mute it is altogether, a direct translation of 
printed symbols into thoughts without any intermediate transla- 
tion into sound. This is, to some extent, the case with regard 
even to common colloquial words. If I receive a letter beginning 
^My deer Friend,' my mental eye sees a picture of a homed 
ftnimal quite as soon as my mental ear hears the ambiguous suc- 
cession of sounds which the spelling represents. With regard 



264 DICTIONARY REPORTBy I. — MR* BRADLEY. 

to purely literary wordfl the case is, of oonrae, much stronger; 
their written form often conveys to our minds their meaning 
without calling up any 'sound-picture' at all, even as an ac- 
companiment. In consequence of this ideographic function of 
written words, most practised readers are able to take in at a glance 
the drift of a whole octavo page. The practical value of this 
accomplishment is obviously enormous. Introduce a new system 
of spelling, and you render it necessary for us at first to translate 
every word into sound before we can understand it; and the 
amount of time that must be spent on reading, when mere 
information is the object, will be multiplied tenfold. Of course 
habit would gradually give to the new spelling the quality of 
ideographical expressiveness; but I do not think that we who 
have reached middle age could hope ever to acquire that familiarity 
with it that would enable us to read as rapidly as we now read 
books in the traditional orthography. Supposing, therefore, that 
a really practicable scheme of phonetic spelling-reform were before 
us for adoption, the question to be answered would not be merely 
whether we were prepared to sacrifice for the good of posterity 
certain aesthetic prejudices and likings, but whether we were 
prepared to undergo a considerable loss of working power for the 
rest of our natural lives. 

'' In the second place, it seems to me that one of the current 
specific arguments against spelling reform has been treated with 
undue scorn by some of our most eminent philologists. I refer 
to the plea that * phonetic spelling would obscure the etymology 
of words.' The common reply has been that our traditional 
spelling is often positively misleading as a guide to etymology; 
and that etymology is of no practical importance to the ordinary 
reader, while for those whom it doei concern such evidence as 
the old spelling supplies would still be available, so long as the 
books written in it remained in existence. This answer would 
be quite conclusive if the language consisted only of colloquial 
words, and of words compounded of or derived from these. So 
far as colloquial words are concerned, the correct meaning is 
that which is prescribed by usage; there is no appeal to any 
higher court. If the customary meaning differs from the etymo- 
logical meaning, it is bad English to follow the etymology. 
In this class of words, therefore, etymology is never a help to 
correctness of speech, and is sometimes an actual hindrance ; so 
that even if phonetic spelling did conceal the derivation of the 



DICTIONARY REPORTS, I. — MR. BRADLEY. 266 

^ords no one would be the worse for it. The words peculiar 

^ literary use, however, are on a very different footing. The 

Dt^AJority of our most famous English writers, from More and 

Spenser to the present day, have been 'classically' educated, 

^^^ have been accustomed to presume on classical education in 

^eir hearers. Most of them have now and then invented words 

^^ Greek or Latin derivation — often without knowing that the 

^^oi-cis were not already English ; and nearly all have habitually 

^®e<i words in senses or shades of meaning which are not based 

on any existing English usage, but which are understood as a 

n*a.t;t^r of course by readers who know Latin and Greek. A large 

prx>portion of our classically derived words can be understood 

^^^tli precision and used with unfailing correctness only by those 

^Ho are acquainted with their etymology. This curious dependence 

0^ English literature on a foreign culture is perhaps a lamentable 

W'^3.1i:iie88 ; but that it exists as a fact cannot reasonably be denied. 

-^-^ a. consequence, the English vocabulary includes an enormous 

'^^^*^l:>er of words of which the written form has an ideographio 

^^^'•^o, due not to its being familiar to the eye as occurring in 

^'^^lish books, but to the fact that it reminds us of the spelling of 

c^«"ta.in words in a foreign language. These words were originally 

loi'tned as sequences of alphabetic letters, not as sequences of 

**^n<ig. The rarely heard and often uncertain pronunciation 

®^ ^ word of this kind is merely symbolical of its written form, 

"^^ intelligible only as it suggests this to our recollection. If, 

f^ Sometimes happens, we mentaUy give a wrong graphical 

"*terpretation to the sounds we hear, we are for a moment 

P^^^led to think what the word can mean. If phonetic spelling 

^^^^ adopted, we should in reading often find it necessary first 

lender the written word into so and, and then to render the 

^^d into the old spelling, in order to apprehend the meaning 

^ the light of the etymology. It seems to me that to write 

^'^ of this class in phonetic spelling would be just as useless 

^ luischievous as it would be to alter the pronunciation of 

*-ioq^al words to make it conform to the spelling. To do 

^^^t of these things would really be disfiguring an original to 

^^© it accord with an imperfect copy of itself — much as if 

'^e one were to alter a text in the Hebrew Bible because the 

^^ of the English version was different. On the whole, I 

^'^t to say that I see no practical way of very greatly 

^^tening the difficulties occasioned to children and foreigners 



266 PICTIONARY REPORTS, I. — MR. BRADLEY. 

by the anomalies of Eng^lish spelling, though many small changes 
in the phonetic direction may with advantage be gradually intro- 
duced in the orthography of such words as are really addressed 
primarily to the ear. But it is highly important that the process 
should be gradual, and the reformers ought not to begin by 
abolishing the useful distinctions that have become established 
in the spelling of such pairs of homophones as whoUy hole. 
The number of such cases is not large enough to add very much 
to the learner's difficulties. And there is no abstract reason 
why written English should be condemned to share all the im- 
perfections of spoken English ; we all recognize the advantage of 
writing proper names with initial capitals, though that contrivance 
is purely ideographic. 

''It is now my pleasant duty to express my thanks to those 
persons who have in various ways assisted in the preparation of 
the work. Amongst these the first place by right belongs to the 
readers who have furnished the quotations ; but to them I can only 
refer in general, their individual contributions to the Dictionary as 
a whole being from time to time acknowledged by Dr. Murray. 
With regard to the volunteer sub- editing, only one name has to be 
mentioned, the whole of this work for the letter E having been 
performed by the late Mr. P. W. Jacob, whose scholarly and 
painstaking labours are referred to in the prefatory note to the 
first part of Vol. III. Dr. Murray continues to give constant and 
most important help, every page from the beginning having had 
the benefit of his careful criticism. The proofs have been 
regularly read by Mr. Fitzedward Hall, D.C.L., who has furnished 
an abundance of supplementary quotations which have in an 
extraordinary degree enhanced the value of the Dictionary as a 
record of the history of words; also by Mr. H. Hucks Gibbs, M.P., 
the Rev. J. T. Fowler, M.A., Durham, and Mr. W. H. Stevenson, 
who have all from time to time contributed suggestions of great 
value. Most important aid lias been rendered on questions of 
Teutonic philology by Prof. Sievers, of Halle, and Prof. Napier, of 
Oxford, and on questions of Bomanic philology by Pruf. Paul 
Meyer. 

''My especial thanks are due to the constant and multifarious 
assistance rendered by our Hon. Secretary, Dr. Fumivall. The 
persons to whom I am indebted for information on particular 
points are very numerous. It is to be feared that the following 
list is far from complete; and I must ask the forgiveness of 



DICTIONARY REPORTS, I. — MR. BRADLEY, 267 

any occasional helpers whose names through infulyertence been 
omitted : — 

"Mr. F. Adams; the Rev. J. C. Atkinson. D.C.L. ; Mr. A. 
Beazeley, C.E. ; the Kev. T. E. Bridgett; the Rev. W. Bright, 
D.D., Canon of Christ Church ; Dr. Robert Brown ; Mr. A. H. 
Bullen, M.A. ; Mr. Ingram By water, M.A., Oxford ; Mr. J. S. 
Cotton, M.A., Editor of The Academy ; Mr. P. A. Daniel ; Mr. 
Benjamin Dawson ; Mr. Leon Delbos ; Mr. C. E. Doble, M. A., 
Oxford ; Mr. Austin Dobson ; Mr. W. T. Thiselton-Dyer, Royal 
Oardens, Kew ; the late Dr. A. J. Ellis ; the Rev. Canon D. 
Silvan Evans ; Dr. Fennell, Cambridge ; Dr. Robert von Eleisch- 
hacker ; Dr. S. Rawson Gardiner ; Dr. R. Gamett, British Museum ; 
Mr. Israel Gollancz, M.A., Cambridge ; Mr. G. Heppel ; Dr. Carl 
Horstmann ; Mr. Henry Jenner, British Museum ; Mr. Henry Jones 
(Cavendish) ; Mr. W. F. Kirby, Nat. Hist. Dept., British Museum; 
Prof. E. Ray Lankester ; the late Mr. James Lecky; the late 
Rev. Dr. Littledale : Mr. J. A. Fuller Maitland ; Prof. Maitland, 
Cambridge; Mr. Julian Marshall; Mr. Russell Martineau, M.A., 
British Museum ; Mr. F. D. Matthew ; Prof. Alfred Newton, 
Cambridge ; Mr. Edward Peacock ; Prof. Karl Pearson, University 
College, London ; Mr. T. G. Pinches, British Museum ; Mr. A. 
"W. Pollard, M.A., British Museum; Sir Frederick Pollock; Mr. 
F. York Powell, M.A., Oxford; Mr. R. B. Pressor; Mr. P. Le 
Page Renouf, British Museum ; Prof. Rhys, Oxford ; Dr. Ch. 
Rieu, Biitish Museum; Mr. J. S. Shedlock; the Rev. Prof. Skeat, 
Cambridge ; Mr. John Slater, F.R I.B.A. ; Dr. Oskar Sommer ; 
Mr. W. Barclay Squire, Biitish Museum ; Mr. Whitley Stokes, 
D.C.L. ; Mr. W. Sykes, M.R.C.S., Mexborough ; Miss Edith 
Thompson ; Dr. R. F. Weymouth. 

**I have further to acknowledge the services of my assistants 
at Oxford, Mr. G F. H. Sykes, B.A., and Messrs. Lewis and 
Bryan; also the very important aid rendered by Mr. A. Erlebach, 
B.A., in the reading of the first proofs. In conclusion, I desire 
to offer my thanks to the Trustees of the British Museum for 
having accorded me exceptional advantages for working in the 
library, and to the officers of that institution for the courteous 
readiness which they have shown in every possible way to 
facilitate my researches." 



268 



II. Eepobt on the Phoghess op the Philological Societt's 
New English Dictionaht. By Dr. J. A. H. Mu&rat. 

{Bead at the Meeting of the Society on Friday y March, Uh, 1892.) 

It is with much regret that I send this statement to be read 
instead of being personally present at the meeting. It is a great 
disappointment and loss to me not to meet my fellow-members and 
to shake hands with the many Dictionary workers who usually 
come to meet me on these occasions. It will, I think, be expedient 
in future to hold the Dictionary Evening at another time of the 
year, when there will be less likelihood of interruption from in- 
disposition or severe weather, or as in the present instance from 
both combined. I had a serious attack of pneumonia in January, 
from which I have been for several weeks so far recovered as to be 
able to do regular work, but not sufficiently recovered to endure 
exposure to cold or wet. Until Tuesday, I thought it possible that 
I might venture to oome to London, but the untoward change in 
the weather has completely defeated my expectations. 

In reporting on the progress of the Dictionary during the past 
year, it is unnecessary to dwell upon the fact that during the year 
two parts were completed and issued to the public, one of Vol. III. 
letter E, by Mr. Bradley, and a part of Vol. II. letter C, under 
my immediate editorship. As to the parts in progress, the 
"Monthly Statement" of the Press which I have just received, 
shows that of the C part, 153 pages are in type, and 112 
printed off; of the E — ^F part 115f pages are in type, and 64 
printed off. Turning to the question of what proportion of 
C is now either in type or in the printers' hands, that is down 
to Corw, I find that when we started, the material for C 
filled 320 pigeon-holes, 6 inches wide; the portion still to do 
occupies 40 pigeon-holes, or makes \ of the letter. I may pause to 
point out that 320 pigeon holes contain a pile of slips 160 feet 
high, and that we still have 20 feet thick of this to work through. 
When I turn to Wehster^i IHetionaryy and compare the pages done 
with those still to do, the proportion remaining seems larger: 
C has in Webster 149 pages, of which we have got through 119, 
leaving 30 still to do. My recent illness has, I fear, lost me some 
10 days; but my assistants have worked splendidly in the mean- 
while, and I do not think that the total back-set is very serious. 



DIGTIONART RBPORTS, II. — DR. MURRAY. 269 

I do fear, however, that I have paid less attention to cor- 
respondents, since I have had to devote all my available strength 
to actual work at the Dictionary. I hope that if some communi- 
cations have not been acknowledged at all, and others only after 
delay, that the writers will not feel hurt. The answering of letters 
is a very serious part of my duty; after delegating as much of it as 
is possible to assistants, that which I absolutely must do in my 
own handwriting, takes always one, more usually two, sometimes 
three hours of the freshest part of my working day. This being 
so, it will readily be believed that letters which take trouble and 
thought to answer have often to wait for many days till I have the 
time and thought to give, and not unfrequently are never answered 
at all, because the time never comes, or comes only too late. 

I have had the benefit during the year of the valued assistance 
of Mr. John Mitchell, Mr. Walter Worrall, B.A., Mr. Arthur 
Maling, M.A., Mr. G. Balk, all of whom have now been in the 
Scriptorium for some years, and are capable co-workers, also of Mr. 
. Sweatman, as a junior assistant; of these gentlemen I desire to 
speak with much appreciation. Much of the accuracy of the 
Dictionary in the thousand and many points, in which error is not 
only possible, but hardly to be avoided, comes from the special 
attention of these assistants. I cannot attempt myself to keep in 
mind the hundred thousand dates of authors and editions, with 
which the dictionary deals, and for these, and countless other 
matters, have to depend upon the special attention of one or other 
of the staff. 

But if credit is due to these paid assistants, much more are credit 
due and grateful thanks to be offered to the many unpaid volunteer 
workers, who have so nobly helped us from the beginning, and are 
helping us still. It is a real source of trouble to me that I cannot 
bring more prominently before the public, and especially before 
that peculiarly opaque section of the public called reviewers, the 
important work done by these unpaid workers, and the honour and 
credit which are their due. I carefully g^ve the names of all 
Sub-editors in every prefatory note; I mention particularly the 
sections which they have handled, and — the Be viewers never 
mention them, except in the rare case in which a local newspaper 
is *' pleased to observe that the name of our respected townsman 
Mr. A. B. is mentioned among those who have helped in the 
arrangement of material." But then local papers very rarely 
** review" the Dictionary, and so "respected townsmen" may be 



270 BICnONART REPORTS, II. — DR. MURRAY. 

doing silently the most assiduous and self-sacrificing work, wl 
their fellow-townsmen know nothing of it. I hope that we may 
able to devise some means of remedying this evil, and of bringi^ 
the public, especially the Reviewer, to render honour to who^ 
honour is due. My own wonder is that in such circumstances, ^ 
many of our unpaid workers have stuck so devotedly to the work::: 
I know that the want of recognition has chilled and disappointed 
some, who have consequently turned their energies to fields i ^ 
which they could do something for their own fame, instead cm 
merely contributing to swell the fame of another, and all the mor* 
do I admire and honour those who work on notwithstanding. Buj 
I trust the Society will assist me in an effort to let those who valu» 
the Dictionary know to whom they are indebted for very much o: 
its workmanship. 



I have to give the following Statement of help received from Suh^ 
editors and others from Jan, 1891 to Jan. 1892. 

The following portions of C have been re-subedited, and th^ 
new material incorporated in preparation for treatment in tli»^ 
Scriptorium-^ 

Mr. G. L. Apperson, 11, Park Road, Wimbledon. 

Consub- to Contem-, Cop-, Counteract to Counterworking': 
now engaged on Crot- to Crows-. 

Rev. C. B. Mount, M.A.., 14, Norham Road, Oxford. 

Contenance to Conti, Corr-, Covenant ; Cro to Crony ; engage 
on Cm to Crup. 

Mr. John Peto, Ravenswood, AUeyn Park, London, S.E. 

Conto to Contra, Core to Corpus, Cow to Coz-; engaged ^ 
Cur to Curi-. 

Rev. W. B. R. Wilson, M.A., Devonside House, Dollar, N.B. 

Const-, Contre to Conu, Coq to Conn, Coup to Couw-, Crad " 
Crez- ; engaged on Cu to Cui. 

Mr. W. N. Woods, B.A., 58, Elgin Road, Addiscombe. 
Com-, Cost to Coty- ; Cou to Coum, engaged on Cri-. 

Miss Edith Thompson, Brooks Lodge, Reigate. 
Conventicle. 



DICTIONARY REPORTS, II. — DR. MURRAY. 271 

I am especially indebted to Mr. Apperson, Mr. Peto, and Mr. 
Woods/ for the thoroughly workmanlike manner in which they 
return the material to be taken in hand by the Scriptorium staff, 
and to the Rev. G. B. Mount, the earliest and most devoted of 
our workers, for unnumbered preparatory investigations of difficult 
and doubtful points in the Bodleian Library and elsewhere, 
whereby he has often cleared up difficulties before I actually 
reached them. 

The following Sub- editors, who are engaged on other parts of 
the alphabet, have sent in work since Jan. 1891. 

Mr. "W. J. Anderson, Bank House, Markinch, Fife, N.B. 
Mor- to Mors- ; engaged on Mort to Mostly. 

Mr. Jamas Bartlett, M.A., Cloverlea, Bramley, Guildford, 
Get to Goz- ; engaged on Gr-. 
Mr. Bartlett is one of the most valued of the recent additions 
to our staff of volunteers ; he does work excellent both in quality 
and amount. 

Rev. W. H. Becket, The Manse, Stebbing, Chelmsford. 
Wea to Weaz- ; engaged on Web to Wei-. 

Rev. G. B. R. Bousfield, B.A., 248, Portsdown Road, W. 

Fissel to Fizzle, Foin to Font, Wi to Withy; engaged on 
Witi to Wom-. 

Mr. E. L. Brandreth, 32, Elvaston Place, Queen's Gate. 
Hat to Haz ; engaged on Kno to Knutty. 
Of Mr. Brandreth's work I need hardly speak; no member 
of the Philological Society has done so much, and I often feel 
impatient to be at letter H in order to follow his steps. 

Miss J. E. A. Brown, Further Barton, Cirencester. 

Incem- to Inch-, Indispen- to Indue-; engaged on Indue 
to Inexo-. 
One of the greatest friends of the Dictionary, to whom its 
Editor owes a debt of deepest gratitude for help in difficulties. 

Mr. J. Brown, M.A., Kendal Grammar School. 
Mah- to Mainy- ; engaged on Mair to Mak, 

1 Alas ! since this Report was written, death has carried off Mr. Peto and Mr. 
Woods, and pressure of other work has ohli^red Mr. Apperson to discontinue his 
valued assistance. — J. A. H.M., December, 1892. 



272 PICTIONARY RBFORl'S, II. — DR. MURRAY. 

Eev. A. P. Fayers, Eawdon Manne, Yeadon, Leeds. 
Nid- to Niggot. 

Mr. R. J. Lloyd, 46, Chatham Street, LiverpooL 
Hip- ; has to end of Hi-. 

Key. Dr. Rupert Morris, D.D., Eaton, Eccleston, Chester. 
Engaged on Intra to Inutterable. 

Bey. J. Small peice, M.A., St. Bees. 

Myrobolan to Myry ; has to end of M. 
An old and faithful friend, who never fails to send his half 
yearly tale of sub-edited work. 

Mr. John Dormer, Mortimer Crescent, Kilbum, N.W., has in- 
corporated new material for Dia to Dirty, and made out LisU 
of Special Wanfs, He has done the same for Cu to Cz, and 
re-subedited Cut. Mr. Dormer has given invaluable help 
during the last few years, enriching the Dictionary with 
scientific and technical quotations from books not before 
read, and contributing in every way that seemed most likely 
to be useful. 

Mrs. L. J. Walkey, 24, Milverton Crescent, Leamington, has 
some new material for Dis to Dy to arrange alphabetically 
aud chronologically. This lady has for many years devoted 
much of her time to this needful work. 

The present Beport will, I hope, be printed ; I think it therefore 
useful to present, in a tabulated form, a statement of the 

Present Position of the whole Work with reference to Suh-editingt 
and the names of those who have done the work. The year 
given shows approximately up to what date new material 
was incorporated, and gives an idea as to what will have to 
be added before we take it up finally in the Scriptorium. 

D. D to Decayer Mr. Elworthy (1882) 

Dece to Defecation Mr. Elworthy (1884) 

Deliver to Dh. Miss Brown ( 1 883) 

Dia to Dialysis Bev. W. E. Smith 

Diam to Dietist Mr. Jacob 

Diffail to Dirvy Mr. W. W. Tyndale 

(Dia to Dirty — ^the new material up to date has been in- 
corporated by Mr. J. Dormer) 
Dis to Jiz Mr. Jacob (1883) 



DICTIONARY REPORTS, II. — ^DR. MURRAY. 



273 



-P- F to Fiz 


Rev. G. B. R. Bousfield | 


; 1884-90) 


Fla to Floun 


Mr. J. Peto ( 


; 1886-8) 


^- Gem to Groundsel 


Rev. G. B. R. Bousfield ( 


;i880- ) 


Group to Gz 


Rev. T. D. Morris ( 


;i885) 


Ga- to Go- 


Mr. J. Bartlett ( 


;i888-91) 


^- Ha to Harmlessness 


Mr. G. A. Schrumpf \ 


;i882) 


Harm to Haz 


Mr. Brandreth 1 


;i890) 


He- 


Mr. Brandreth 1 


;i886-9) 


Hi to Hipwort 


Mr. Lloyd ( 


;i884) 


Ho to Homunculus 


Mr. Brandreth 1 


; 1883-5) 


Hoo to Honis 


Mr. Peto 1 


[1885) 


Hosan to Hwata 


Mr. Woods 1 


;i885) 


r ^^ 


Mr. Peto ( 


[1885) 


*" la to Inch- 


Miss Brown { 


; 1887-90) 


Incoacted to Incomp- 


Mr. T. Wilson ( 


;i887) 


Inconceal to Indiscr- 


Rev. E. H. Sugden ( 


;i885) 


Indisp to Induc- 


Miss Brown i 


;i891) 


Inva to Invent 


Miss L. Gardner i 


;i889) 


Invest to Iz 


Rev. R. Morris ( 


[1889) 


^* J- to Juxtaposition 


Rev. W. Gregor { 


;i885) 


^- Ka to Ky 


Mr. H. H. Gibbs 




^' La to Lusus 


Mr. Hulme 1 


;i883-5) 


Lu to Lyz 


Mr. E. Warner ( 


;i882) 


*• Ma to Maz- 


Mr. J. Brown 1 


[1884-91) 


Ma to Miz 


Rev. T. Sheppard ( 


1883-4) 


Mo to Mond 


Rev. S. W. Lawley ( 


;i884) 


Mone to Mostly 


Mr. W. J. Anderson ( 


; 1888-91) 


Mu to Myry 
^- Na to Naz- 


Rev. J. Smallpeice { 


;i884) 


Rev. A. P. Payers ( 


;i882) 


Ne to Nez- 


Mr. Hailstone ( 


;i883) 


Ni to Niche 


Mr. Bumby ( 


;i884) 


Nicher to Niggot 


Rev. A. P. Payers ( 


;i888) 


No to Nony 


Mrs. Pope ( 


;i883) 


Nu to Nz 


Mr. R. F. Green ( 


;i883) 


^- to Oky 


Mrs. Stuart (Miss Haig) ( 


^1886) 


Go to Opentide 
Pa to Paz- 


Rev. W . J. Lowenberg ( 


[1883) 


Miss Brown ( 


;i882) 


Peas to Pelys 


Mr. J. Britten ( 


;i882) 


Pem to Perem- 


Mr. R. McLintock ( 


; 1885-92) 


Personality to Poz 


Mr. W. J. Anderson 1 


; 1882-4) 


Pra to Pz 


Mr. Jacob i 


[1885) 


YhiL Traiu. 1891-»-8. 


18 





274 



DICTIONARY REPORl'8, II. — DR. MURRAY. 



a 


• 


Mr. Jacob 1 


;i884) 


K. 


Eec to Rigour 


Mr. Jacob 1 


;i885) 


s. 


SatoSy 


Mr. Jacob 1 


;i881-2 


T. 


Tal to Tiling 


Rev. W. B. R. WUson I 


[1881-2 




Till to Tmesis 


Mr. T. Wilson ( 


;i882) 




To to Toz 


Rev. W. B. R. Wilson ( 


[1883) 




Tra to Tralucent 


Mr. A. Sweeting i 


[1883) 




Tre to Trilogy 


Rev. W . B. R. Wilson ( 


; 1888-9 




Tua to Tz 


Mr. A. Lyall i 


[1881) 


TJ. 


XJa to XJz 


Rev. T. Sheppard 1 


[1881-2 


V. 


Va to to Vyse 


Rev. T. Sheppard ( 


[1881) 


W. 


Wa to Weaz 


Rev. W. H. Becket ( 


[1883-8 




Wi to Withy 


Rev. G. B. R. Bousfield ( 


1890-1 


T.&Z 


r 
t. 


Rev. J. Smallpeice ( 


;i882) 



There are still parts of the material which have not been dec 
with by a sub-editor — at least since I undertook the work ; th( 
are the following : — 



Material not Suh-edited. 

D. Defect to Delitescent (in the hands of Mr. Elworthy). 
F. Fo to Fyz (not touched after work of old sub-editor). 
H. Hir to Hiz (in the hands of Mr. Lloyd). Hon-. 
I. Indue to Inutterable (Miss Brown working at the one er 

and Rev. R. Morris at the other). 
K. (not since originally done by Mr. H. H. Gibbs. M 
Brandreth has begun re-editing — has in hand Kno t 
Knutty). 
M. (end of Mo and My. Messrs. Anderson and Smallpei( 

are working at respectively). 
N. Nigh to Niobium-. 

Nip to Nizam (in hands of Mr. R. F. Green). 

Noodle to Nozzle. 
0. Ole to Om (Mrs. Stuart has). 

On to Onyx. 

Opera to Ouster (in the hands of the Rev. W. J. Lowenberg 

Out to Oz-. 
P. Pea to Pear. 

Per to Person. 
R. Ra to Rebutter. 

Rile to Ry. 



DICTIONARY REPORTS, II. — DR. MURRAY. 275 

T. Ta to Taky. 

Tram to Traz. 

Trim to Tsetse. 
"W. "Web to "Wh- (beginning in Mr. Becket's hands). 

Witless to Wy- (beginning in Mr. Bousfield's hands). 

Residing, 

A good deal of "Reading^' has been done during the year, 
and much is stiU going on; it being still often found that some 
works have been unduly neglected or read only for particular 
letters, a most pernicious and deceptive practice, since in beginniDg 
a new letter, it takes a long time before we notice that a particular 
author or book has dropped out; and when such books are read 
again entirely, we get a multitude of duplicates for the letter 
or letters already read. The publications of the Early English 
Text, Scottish Text Society, and other prints of MSS. or 
early printed books, also require to be kept up with. I am 
sorry to say contemporary literature is not ; we have 
next to nothing of Andrew Lang, Austin Dobson, Qeorge 
Meredith, R. L. Stevenson, and a host of contemporary writers ; 
and I should be devoutfuUy thankful if Dr. Fumivall, or any 
other competent person, would draw up from Mudie's Catalogue 
or other source, a list of works published since 1875, which 
ought to be read in order to give us a representation of the 
literature of the last quarter of the 19th century. I cannot 
do it ; I have not a moment to spare for it ; but I will check 
it when done, and mark off books that have been read, and some 
means may be found of superintending the Reading. 

"We have kept up our plan of preparing Lists of Special Wants ^ 
for the words in front of us, though I am sorry to say that 
the number of readers who make good use of them is not so 
great as it ought to be. I enclose some copies here, and ask 
any present who will seriously use them to take a set, but not 
to take them needlessly. 

During the past year not less than 30,000 quotations were 
received. Of these, Mr. John Dormer, of Mortimer Crescent, 
Kilburn, contributed more than 5,000, including many important 
desiderata and quotations from scientific works of last century, 
with early occurrences of technical terras. This is besides the 
assistance given by him in arranging material of C and D, and 



276 DICTIONARY REPORTS, II. — DR. MURRAY. 

in preparing therefrom Lists of Wants for printing. I regret 
that lately the state of his health has interfered with his 
splendid work for the Dictionary. 

Dr. W. C. Minor, of Broadmoor, Crowthome, has sent about 
3,000 quotations, all for words in C and E, such as we are 
ready to deal with, and including many from the E.E.T.S. ed. 
of Lanfranc's Cirurgte, one of the most important works for 
Dictionary purposes ever yet published by the Society, also from 
Hall's and Grafton's Chronicles, and numerous rare books of 
travel of the 17th and 1 8th centuries, giving early instances of 
foreign words since naturalized. 

Mr. E. Peacock, of Bottesford Manor, Brigg, has sent about 
4,000 quotations, including 750 from Dryden's plays (formerly 
badly read) and 300 from Ford. 

Dr. Brushfield, Budleigh Salterton, has sent about 3,000, 
including about 1800 from Mad. D*Arblay*s Diary and Letters. 

Rev. J. T. Fowler, Bp. Hatfield's Hall, Durham, about 1500, 
including a complete reading of the Castle Howard Life of 
St. Cutbert, edited by him for the Surtees. Society, and of the 
Bury Wills. 

Mr. A. Beazeley, Thornton Heath, Croydon, has sent about 1500 
quotations from Foote's plays, previously unread for the Dictionary, 
and many Miscellanea and Desiderata. 

Rev. J. W. Hooper, Gateshead Fell Rectory, has sent about 1600, 
among them many for modem colloquialisms. 

Mr. T. Henderson, Bedford County School, has sent about 
1300, among them many desiderata. 

Miss H. M. Poynter, Park Town, Oxford, who is a new reader, 
has sent about 1250 from books specially selected by me. 

Prof. J. M. Dixon, Imperial University, Tokyo, Japan, has 
sent about 700 quotations as supplement to those in his Dietionafy 
of Idiomatic English Phrases, for all of which he has besides sent 
the full references not given in the book. 

Rev. Henry Ellershaw, Mexborough Vicarage (a new reader), 
sent about 460 from R. N. Hypnerotomachia. 

Among our American readers Mr. W. Boyd, of Cambridge, 
Mass., has sent about 2000 ; the Rev. B. Talbot, of Columbus, 
Ohio, 700 ; Mr. G. M. Philips, of West Chester, Pa., 600 ; 
Mr. Albert Matthews, Boston, Mass., (a new reader) has sent 
about 900 quotations; and Mr. W. P. Garrison (editor of the 
Kcw York Nation) has been a constant contributor. 



DIC3TI0NART REPORTS, II. — DR. MURRAY. 277 

Of our Dutch readers Mr. Caland has sent ahout 800 of modern 
colloquialisms, while smaller contributions were received from 
Dr. Logeman and Mr. Stoffel. 

Valuable quotations, both general, and of desiderata, have been 
received fix>m Mrs. Grey, Gledhow Gardens ; Miss E. and Miss E. P. 
Thompson, Reigate ; Miss Geraldine Gt)S8elin, Miss C. Pemberton, 
the Kev. Cecil Deedes, Brighton; Rev. W. D. Sweeting, Maxey 
Vicarage ; Rev. "W. Lees, Reigate ; Rev. W. C. Boulter, Malvern ; 
Rev. W. B. R Wilson, Dollar; Dr. Prior, York Terrace; Mr. 
G. Gray, Mr. Hellier R. H. Gosselin, Mr. John Randall, Mr. 
M. L. Rouse, Mr. F. Hall, Llandudno ; Mr. Ralph Harvey, Cork 
Grammar School ; Mr. J. Hooper, Norwich ; Mr. J. Whitwell, 
Kendal. [See also Addendum, p. 287.] 

Special assistance in quotations for desiderata and for a number 
of historical words including convent, corporation^ conventicle, 
covenant, erosSy etc., has been given by Miss Edith Thompson, 
Technical words have been taken under his special protection by 
Mr. R. B. Prosser, who has not only looked to the accuracy of 
the definitions, but got us the earliest quotation from the Library 
of the Patent Office for very many. Mr. R. Oliver Heslop, of 
Corbridge-on-Tyoe, has given most important assistance on several 
coal-mining terms ; valuable contributions for the Desiderata have 
been sent by Mr. A. Wallis, F.R.SX., Exeter; Mr. J. R. Gillespie, 
Stratford Grove, Newcastle-on-Tyne ; Mr. J. Dixon, Harrowlands, 
Dorking; Mr. W. Jones, Gloucester; Mr. W. Johnson, Lavender 
Road, Battersea. 

Time would fail to tell of the splendid assistance rendered to 
the Dictionary by Dr. Fitzedward Hall, who devotes nearly his 
whole day to reading the proofs of both volumes, and to supple- 
menting, correcting, and increasing the quotations from his own 
exhaustless stores. When the Dictionary is finished, no one man 
will have contributed to its illustrative wealth so much as 
Fitzedward Hall. Those who know his books know the 
enormous wealth of quotation which he brings to bear upon 
every point of English literary usage ; but my admiration is 
if possible increased when I see how he can cap and put the 
cope-stone on the collections of our 1500 readers. 

Li this department I have also to mention the constant solid 
assistance of Mr. H. H. Gibbs, whose accession to the legislative 
benches at St. Stephens has not in the least diminished his 
interest in the Dictionary. If I were not afraid of its getting 



278 DICTIONARY REPORTS, II. — DR, MURRAY. 

abroad in the City, I should express my belief that he thinki 
the Dictionary a more important constituency than the City o 
London. 

The ttev. J. T. Fowler, Vice-principal of Bp. Hatfield's Hall 
Durham, one of my earliest and most valued Dictionary friende 
to whom I am indebted for great personal favours, has recentl; 
undertaken to read a set of the first proofs, and his keen aii< 
experienced eye has already hit upon aud remedied man; 
deficiencies. If it is remembered that reading a proof mean 
giving at least eight hours a week to the service of th 
Dictionary, the credit due for such a service will be more full; 
appreciated. 

Dr. W. Sykes, formerly of Mexboro', now of Qosport, has fo 
several years given us indispensible aid with the medical terms 
the history of which really requires the care of a specialist 
any one who looks at the completeness with which these term 
have in the last two or three parts been treated will see ho>i 
much Dr. Sykos has contributed to the work. It is therefor 
with nothing short of dismay that I have learned from hin 
that his other duties will no longer enable him to continue hi 
generous work. His case is one in which I specially wish tha 
some means may be found of fittingly drawing attention to hi 
share in the work. 

The proofs have also been read and annotated by a forme 
member of the Scriptorium Staff, the Rev. J. B. Johnston, o 
Falkirk, whose ** Place Names of Scotland," just published, i 
the best book on local names known to me in the language 
and well deserves the attention of all interested in the subject 
Monsieur F. J. Amours, of Glasgow, also supplies with the earlies 
known examples of words from French, and reads the Proof 
with a special view to the treatment of these words. Professor E 
Sievers, now of Leipzig, and M. Paul Meyer, Member of th 
Institute, Paris, have continued their valued assistance in word 
of difficult etymology, Teutonic and Homanic. Many other foreigi 
scholars, particularly Senor Don Bufino J. Cuervo, of Paris ; Prof 
F. Neumann, Heidelberg; Prof. F. Kluge, Jena; Dr. J. W 
Muller, Leyden ; Prof. A. Noreen and Dr. Axel Erdmann, XJpsalfl 
have helped me with special words. [See Addendum, p. 287.] 



279 



III. — CoNTEIBXrnONS TO THE HiSTOBICAL ETYHOLOOr OP EnOLISH 

Words. By Dr. J. A. H. MuRRAr. 

{Read as a Supplement to the Report on the progreee of the Dictionary, 

March 6th, 1892.) 

CoKT, CONEY. There are namerous ME. forms: a 12-1 6th c. 
eunin, htnyne, conning, eunyng, eunig. /? 14 c. OQward eony, 
eunnyy cunnie, etc. The origin is L. cunicultu, which regularly 
gave It. eoniglioy Pr. and OF. eonil : of the later there was a variant, 
eonin, (French has other examples of this oscillation of / and n.) 
OF. eonin, gave the Eng. eunin, conning forms, now obsolete, except 
as surviving in some proper names as Coningshy, Cunnington, etc. 
OF. conil had in pi. conils, eonniz, with I suppressed as in other 
'I words; the pi. eoni% (found in Anglo-Fr. in Bretton 1292, as 
cony 8, coniyi, with the variants coninn, conyns), gave the Eng. plural 
eanya^ conies, whence was deduced the singular eony. The rabbit 
is evidently of late introduction into Britain and Northern Europe. 
It never had a native name in any Celtic or Teutonic land, the 
Teutonic names cited by Prof. .Skeat are all from Old French eonin. 
There is no mention of it in English before the Norman Conquest ; 
and it is notable that the fur seems to have been known before the 
live animal. The Moral Ode^ a. 1200, has 1. 361, ' Ne seal ]7er beo 
fore ne grei ne cunig [v.r. cunin, konyng] ne ermine.* 

In the ME. period the rabbit was a domestic or protected animal 
kept in eony -boroughs, eunny-herries or eonyheareSy eonyng-earths or 
eony-garths, eonyngers or eony-gress, cony-greens, cony-greaves, cony* 
graves. It has since escaped to the woods and sand-hills, as it has 
more recently done in Australia. 

The historical pronunciation is cunny, as in honey and money; but 
the word is now known chiefly as a Bible-word, where it is a mis- 
translation of the Heb. name of the Syrian hyrax; and as the 
domestic 'cunny' had come to have some familiar and even in- 
decorous senses, readers of the Psalms have preferred to read cdny» 
Walker in 1790 knew only the cunny pronunciation. Smart (1836) 
says 'it is familiarly pronounced cunny, ^ but cdny is 'proper for 
solemn reading.* 



v^\0^^ 



10 



lil^ 



^O^^^ 



AJ^ 



tpae 



\otvg 






dA^Y 



,ia^o^ 



ited^^ 



t\xe 



0^- 0^' TJ:i^r\^ ^^^'^Z'l^- ^'^'T.^Zm C"'^' 



f . OB. '^o '^'^okoptio^' * eft t"b« 



cuoco") 









ajv 



^°Co-- ^ 



^ete 



Aets, 



It 






o^et 



M 



e6ct*« 



oitlae 






"Bn 



itis^ 









It 

id 



lii 

B 

a 
C 

6 












^«U ox 



JoU, *" 









\)e 



*j »f^ i»»* """ '- ■•" " 



ot It*^* * 



OS 






05 "^ -, aiwo* "" , \n^ft v» •- ^.^le *>"'' •» «» earl? - ^ 



India 
ltUaar»» 



S- ^-rs 



Sou^ 



tJxetiv 



potters *- ^^^ cooUe« \ _^^^ tJxe 



etro 



,ueo>^* 



etico 



^)X9%^ 



ENGLISH ETYMOLOGIES— DR. MURRAY. 281 

that the word is native there, just as many people think that 
P&rliament' is a native English word, because parliamentary 
^^f^titutions in the modem sense are of English origin. 

OooM, CULM, soot, grim, coal-dust, small coal, brittle, inferior 

^^thracite coal, is not treated by Prof. Skeat. Culm goes back 

^ a M.E. colm whence colmy, sooty, grimy, hicolmm to begrime. 

-^ t;hink eolm must be radically related to col coal, with an m 

^ffl^, but want of OE. or cognate examples make the early 

-'^story uncertain. Coam is a northern variant of eulm^ just as 

-^^^tnfr in Northumberland is locally called Boomer, and hulk, 

f^^^«*iffer, coulter appear in So. as hook, shooder, eooter. But there 

^ ^Ibo a 16th c. spelling coame, which may be reduced from 

^***^*», as in holm pronounced hoam ; or it may correspond to 

^•-^^orse kdm grime, filth of dirt; and a sense of eoome used 

y lutler 1609, suggests Ger. kahm mould, the white film on 

^^'^^ented liquors, and so related to keem of cider. 

^^^^^MB (badly spelt comh) a measure of four bushels, has no 

*^^^^^ible connexion with F. eomhle or L. eumulm. English shows 

*^^t appears to be this word at three distinct periods and in 

!^^*^« different uses. 1 . OE. eumh * a vessel, a cup,' occurring in 

^**^h Cartul. Saxon. No. 273 of date 791-6 *cumb fulne li^es 

^^, and cumb fulne Welisces alo]?.' 

^» ME. and early mod. oomhe, eomh, a brewing tub or vat, of 

^ *^<2h we have examples from a, 1400 to 1688. * A comb, or a 

' ^er's comb, or yelling comb or tub' is that vessel into the 

^oh the wort is put to work with the yeast. 

^« Coomh, eomhe, eonih, eoom,' etc., the corn-measure, found 

^*^ "tie Bury Wills 1418, and abolished by statute about 1883. 

^liese all agree in the sense of 'hollow or deep vessel,' and 

^"^>eapond to older LG. kumh, mod. LS. kumm a vessel, in 

p^^^*^ou8 dialects, a round deep vessel, basin, cistern, trough, etc. 

.^'' wAao LG. kumm, kump a measure of com or fruit, of which 

^ Bremen Worterbuch says * kumm oder besser kump, tiefe 



T^^Bsel.' We have evidently a Germanic type ^kumho^ *kummO', 

^^^ a by-form *kumpO', the same as is established for clam, clamp, 

^doMB (badly comhe, comh), a hollow on a hill-side or slope. 

. ^^^Xtrs as cumh in OE. charters as early as 770, and plentifully 

^^ roper names dA Bataneumh, Brancescumh, Eaetcumh, Sealteumh, 



• ^'^^-eeleumh, etc. A multitude of these names survive in modern 
L, and the element is usually written comh, comhe, in accordance 
the ordinary scribal rule of avoiding the combination of u 






282 CONTRIBUTIONS TO HISTORICAL 

with m, fi, V, which gives modern tomSj son, homyy etc^ 
is interesting to note that as a separate word eumh, eotf^ 
unknown from the time of the "Wessex Charters down to ^ 
when it appears in the hotanist Lyte, a Devonshire 
*Foxeglove groweth in darke shadowie valleys or coombes.* 
was still to Ray in 1674 a South Country Dialect word, 
it is not uncommon in 1 8th c. writers, natives of the southed" 
counties, and in the 19th c., with its love of nature and summ< 
holiday haunts, it has like ekine^ glen, eleve, elattsr, iurn^ corr 
heck, claehan, Jjord,, aiguille, col, and the like, become familial 
to everybody. There is no reason to think that it ever die 
out in Wessex or the Chalk country ; but it is very reraarkable^^ 
that it never emerges in literature between King Alfred and 1578. ^ 

It is generally recognized as one of the few British words 
taken up at the English Conquest of Britain; and is supposed 
to be identical with the eum- found in proper names in 
Cumberland and Strathclyde as in Cumwhitton, Cumdivock, Cum- 
longan, Cumloden, The modem Welsh eupm, occurs in com« 
pounds as -cwm, 'fftPtn, and in syntactic combination as 
in Cwm Idwal, Cwm BoehltJoyd\ it represents a proto-Celtic 
kumhos. It was natural that in coming from a flat land which 
had no hills or valleys of its own, the Old English should adopt the 
native names diin and eumh from the Bretons; but in the case 
of the latter, the fact that they had a word cumh of their own 
meaning hollow vessel, basin, might well be a helping cause. 
I think it quite probable that if King iEifred or any of his 
men had been asked what -cumh meant in JFideumh, Sealtcumby 
etc., they, ignorant of Welsh, would probably have explained 
it quite satisfactorily to themselves as a natural hollow like a 
* cumb ' or basin. It is also to be noted that eomhe occurs in 
French in the sense of * petite vallee, pli de terrain, lieu has 
entoure de collines' Littr6 (who heis examples from the 12th c). 
This word would be known to the Normans, and may actually 
be the source of some of the names in -combe, or at least of the 
spelling -combe. It is interesting and somewhat remarkable that 
the word has thus had a meaning in the languages of Celts, 
Saxons, and Normans. The origin of F. combe is disputed : some 
think it Celtic, but others look for it elsewhere. 

Coop, a basket, etc., cannot, of course, be identified with OE. 
eype, which actually persists in its proper form kipe, kype, a 
wide-spread word for a bushel- basket, a potato-basketi and a 



ENGLISH BTTMOLOOIES — DR. MURRAY. 283 

'^^clcerwork basket or cage used in catching fish or eels ; but 

It tncLy go back to a collateral form unrecorded in OE., whence 

-^-I^u. ei^pe, Du. kuip cask. But if so, this makes impossible 

tae derivation of the latter from L. eupa cask. On account of 

^^^ ambiquity of u in southern ME. it is difficult to tell whether 

ought to read cupe in Floriz, the Castel of Loue, and Trevisa, 

^oop, or kifpe. In any case cooper is not a derivative of the 

"^^S- word coop; coops were never casks in England, and coopers 

®^^r made wickerwork. Cooper was in the 15th c. eouper, 

*j:^er; the spelling cooper is of the 16th c. and merely phonetic; 

««op, cooper^ as in stoopf droop, and elsewhere before Pf long 

-^Qs persisted and not become ow in modem English, and this 

***^^^xjetic persistence of the sound has been marked by substituting 

^ modern symbol oo for OE. ti, ME. op, ow. Cooper was 

*^*'**^t>ably introduced ready made with the wine trade or some 

-|^ *^^i department of commerce from the Low Countries or the 

J. ^i*3e; cf. MD., and 15th c. Niederrheinish cUper, etc. The 

*^*^^ Lat. was cuparius ; but there is no evidence of the Eng. 

« being taken directly from this. 

^^ooPER is also a name given to the floating grog-ships which 

to do so much mischief among the Deep Sea Fishermen 

'the North Sea. But this is more correctly written and 

^T^^^ounced Coper, as it appears in E. Mather's Nor^ard of the 

3j^^^«", and in the publications of the Mission to the Deep Sea 

^*^«^ cries, which by its beneficent work has extirpated the coper. 

*^^ latter is simply the Dutch, Flemish, and Low German k6per, 

^^^f^^r, buyer, dealer, and according to information which has been 

^-^^cted for me at Grimsby, Yarmouth, and other places, from 



'^^ of the old fishermen, it arose some forty years ago, at a time 
p^ ^^n the fishing-fleets fished in and near the Dutch coast off 
^^^**^perdown. They were then visited by Dutch and Flemish 
^ ^t;«, which brought them fresh provisions, tobacco, etc., and took 
w ^'Jci. them their inferior fish. There then arose a trade in un- 
^^^^ded tobacco and spirits, and at length when the Ashing fleets 
^^t; far out to the Dogger Bank, and the mid sea, larger vessels 
*^**~ fitted out to follow them as floating grog-shops, still pre- 



'ixig the Dutch name of coper, though their main object was no 
^Rer to buy fish, but to sell vile spirits, bad tobacco, obscene 
*^*^^tographs, and other demoralizing trash. 

. ^ooT. The derivation of this was by Mr. "Wedgwood sought 
^ ^lie Welsh word etota short, docked; but this is inconsistent 



284 CONTRIBTTTIONS TO HISTORICAL 

with the history of the word in English, in which eoot corned 
down directly from ME. edte. The word is common Low German ^ 
Du. koet of the 16th c. points to a MDa. edU, identical with th^^ 
ME. o6te. The name appears to have heen given vaguely to 
various swimming and diving hirds ; especially (1) the Guillemot, 
called also in Du. see-koet, and (2) the Bald Coot, in Dutch 
meer-koet. Abundant quotations in both senses occur in Eng. : 
'a balled cote' is in Walter de Biblesworth, and Lydgate has 
'and yet he was as balde as is a cote,' while Tindale cites 'as 
bare as Job, and as bald as a coot.' 

CoPEB and Coupbb, in Horse-coper, etc., are often spoken 'of as 
dialectal variants. They are more than this. Cope came from 
Flemish in the 15th c. Lydgate in Zondon Lyckpenny says : 
'* Fleminges began on me for to cry, ' Master, what will you copen 
or by?'" and Hey wood has among his Proverbs and Epigrams 
''the Ducheman saieth 'segging is good cope.'" But eoup is 
the Norse kaupa, and is older in England than eope. The native 
English equivalent of both is cheap, cheapen ; and it is interesting 
that in coup^ cope, cheap, we have the original Germanic diphthong 
au, and its two derivatives LG. 6 and Eng. ia. Another cognate 
is the Sc. cofi, bought, where the consonant / seems to be that of 
HGer. gekauft. 

Copperas, I have discussed fully in the Academy, showing that 
the alleged derivation cupri rota rose of copper, conjectured by 
Diez to be a rendering of Gr. x°'^'^^^^^^ copper-flower, is only 
a fanciful alteration of cuperosa, or cuprosa, coppery, occurring 
in aqua cuprosa the equivalent of Ger. kupferwasser, Du. koperwater, 
mames of copperas, originally of its solution, flowing naturally 
from some mines, and used according to olden notions to transmute 
iron into copper by leaving iron in the stream until copper was 
deposited in its place. 

Coppice. The OF. form was copetz, colpeH, the regular repre- 
sentative of a L. oolpdttcium, that which has the characteristic of 
being cut. Cf. the L. adjectives in "icius, like advent ictus, 
Colpare to cut with a blow, was from colpus, earlier colapus, 
originally eolophus a blow, a cuff. The med. L. oopecia some- 
times given as the source of coppice is only the French or 
Eng. word with its ending latinized. 

Copt, an Egyptian Christian, is a comparatively modem word, 
app. not known in Western Europe much, if at all, before 1600. 

It represents the Arabic collective LlJ ^ quit, qybt, kfii quft^ 



ENGLISH STTMOLOOIES — DR. MURRAY. 285 

qyft * the Copts,' with adjective derivatiye quhte^ q^ft*t Coptic, most 
prob. representing the Coptic VTIIT'IOC, fCTFlX^JOC guptios, 
kuptaios,=Gr. 'Airfwmo9 Egyptian. Some have thought the 
name referred to the ancient city of Coptos in Upper Egypt, 
and it is probable that a belief in this among scholars made 
Copt and Coptic the settled form. The earlier Engl, forms were 
Cophte, Cophtie, The fact that Arabic has no g nor p explains 
how guptios^ ggptios, necessarily became ku/ti or kubti, kyfti or 
hyhti. 

Corbel has been badly treated by etymologers. The dictum of 
Skinner that it is from F. corheille, basket, has been repeated 
ever since, apparently without looking to see what French 
etymologists have themselves to say about it. A corbel has 
nothing whatever to do with a baiket in etymology, sense, 
appearance, or fancy ; it is the end of a beam or stone 
built into a wall, and projecting more than its own depth, 
forming a bracket for the support of some structure resting upon 
it. In many an old castle or ruined mansion the rows of stone 
corbels that supported the floors, can be seen still projecting 
from the walls. The word is in ME. in 15th c, and was in 
OF. corbel, mod. F. corbeau, the primary sense being * raven ' 
L. type corvellus dim. of corvus. F. corbel, corbeau, has been 
and is applied to many beaked projections, as may be seen in 
the new Dictionnaire GMral of Hatzfeld, Darmstetter, and 
Thomas (a splendid and cheap work which everybody ought to 
have) ; and the architectural sense was probably given, because 
the corbel being usually slanted away underneath, was, viewed 
in profile, a beak-like projection. In modem times, the word 
was merely technical, until caught up by Sir Walter Scott, 
who had always a good eye for a word that looked mysterious 
and sounded well. His ' corbels carved grotesque and grim,' 
have taken hold of the popular fancy, and a hundred writers 
since have adorned their pages with * grotesque corbels ' and 
* grim corbels * with the very slightest notion of what they were 
writing about ; generally any grotesque figure on an abbey wall 
was to them a ' corbel ' ; but a corbel is not an ornament, and 
omamen^tion forms no part of it, though it may be carved 
into something either florid or grotesque, just as a finial or a 
door-knocker, or a lectern may be. One thing that helped to 
confuse Englishmen was that the F. corbeiUe (L. corbieula) is 



286 CONTRIBUTIONS TO HISTORICAL 

actually applied to a basket in architectare, i,e. the baskets on 
the heads of Caryatides, or the *bell' of the Corinthian capital 
considered as a basket out of which Acanthus leaves and flowers 
are projecting; but this, of course, has no connexion with corbel 
in meaning, form, or derivation. 

CoHDUBOY is a trade word of English concoction, introduced 
between 1776 and 1787, as a name of corded fustain. It is 
impossible to say whether it was named after the inventor — 
Corderoy is an English surname, and corderoy is actually the 
earliest spelling of the fabric — or whether the inventor meant 
to simulate a French word, and thought of cards du rot, king's 
cord. But no such name has ever been known in French ; 
indeed in a French work of 1807, Voyage dans les D^partemenU 
du Mtdif by Millin de Grandmaison (for a knowledge of which 
I am indebted to our esteemed Dictionary helper, Mr. Prosser), 
among the manufactures of Sens are mentioned ' 6toffes de coton, 
filatures, futaines, kings cordes^ mollerons,' etc., which shows 
first that corde du roi was not the French name ; secondly 
that the interpretation king'^s cord was so current, that the French 
manufacturers borrowed it in their imitations of the English 
fabric. The word duroy as the name of a coarse woollen fabric, 
manufactured with serges and druggets in the West of England 
in Defoe's time, has evidently no connexion. A sample of the 
historical falsehoods, uttered in the name of etymology is the 
following, from the Evening Standard of 28th August, 1884, 
' Corduroy is the coming material . . the new corde du roy will 
be a dainty silken fabric, as indeed it was in the beginning.' 
This is a sheer invention, delivered as a bit of history. 

CoBE is a difficult word; the one thing certain about it is, 
that it is not as Skinner taught, L. cor, or F. eoeur heart. 
The primary meanings are the hard unbumt centre of charcoal 
or of a lime-shell, the hard centre of a boil or tumour, and 
the hard, scaly, or uneatable part of a fruit. For these 
the earlier name was colk, which now survives in the unbumt 
coke of lime or coal. In all, the sense is that of something hard, 
objectionable, or unusable. So with the early figurative senses 
of * something that sticks in one's throat,' * something we cannot 
swallow or that will not go down with one,' and of a grudge 
that remains in the heart even after a quarrel has been composed. 
It was only late in the 16th c. that etymologizing writers began 
to think of some connexion with heart, and to use * core * in senses 



ENGLISH BTYMOLOGIES — DR. MURRAY. 287 

itt '^W'liich * heart ' had previously been used. Shakspere's ' I will 
'^ewLT him in my heart's core, I, in my heart of heart,' from 
''^Hicli * heart's core ' has (only since Eeat's time) become a 
Id^li c. stock phrase, was, I have no doubt whatever, a pun, 
' liostrt's core ' suggesting the L. cor, and so ' heart of heart.' 

f^ADDEKDiTM TO Be. Mferay's Bictionahy Bepoet, pp. 268-278. 

A^s my Annual Heports for 1889-91 were not printed in the 

^9^^nsacti<mSf I desire here to repeat the special acknowledgments 
'^tioire made of the work of Mr. Halkett Lord, of Hawthomden, 
Scotxih Plains, New Jersey, U.S., who during those years was our 
important contributor, sending us more than 4000 quotations 
specially -chosen, and in many cases rare books, which it 
^^'O^ld have been difficult for us to get at, except in the public 
'^^^'^^es. — J. A. H. MuEEAY.] 




Tuxasubek's Cash Account, 189L 



QO 



C 



«« 



A O 


CO o o 


•«<«C< 


CO C) ^ 


00 04 

00 


OC4 o 

^4 



C9 



d 






c 



<« 



eg 



.2 



«D K« 



"* b- 



CO o 



o 



GO 



^ ^' 



■■si 

< 

• GQ 
eo I v 



P^ 












I 



o 



— o 

s 



«.2; 



pq 



§ 









'^2 
§ S 

GO'S 

c 



So s 

1^ 



a =3 



3 



5^ _ 

^'9 03 



^'* 03 p3 U ^ 

«-*« S o S 



I 

no 

P 



.a 






•2 

eo 



^ »o 



00 



o 
eo 



CO 
eo 



00 d 



00 



eo 

CM 



o 



H 

• MA 

»P4 

o 



9 
U 

5 
e 

o 



CO 

O 
ft 

M 



o o o o 

00 O C4 C4 






<« 



00 o ^ 



c^ 



P4 
GO 



CO 



O) 



2P 



GO ^"^ 

p . pPh 

08 O rt 



O 
P 

OS 
1^ P 

is 

p-C 

p p 
Ha2 



o 
p 

OS 

s 

CO 

p 

o 

•43 

P 

C 

•*A 

p 
o 

O 

ce 
u 



— • ^ 



OQ 

.s 

P 

eo 



to « 

•«* to 

CO 



M P 

HP^ 

o 

s - 
i 

H 

CO 2 

o 






o 

s 

P4 

o 
o 

e ^ 

^^ 

V o 

.fa-* 

•- 2 

« o 



a 
-i 

o 

•*» 

a 

p 

o 

2 



a 


C9 


• r^ 


a> 


f. 


oo 


5 


»-t 


M 


« 


O 


a> 


9 


tA 


► 




« 


Pt 


^ 


04 



^ 



PHILOLOGICAL SOCIETY. 

{MEETING AT UNIVERSITY OOLLBQB. LONDON, W.O.) 

189a. 



COUNCIL, 1892-93. 

Prttidint. 
PROF. A. S. NAPIER, M.A., Ph.D. 

Fiee^Presidents, 

WHITLEY STOKES, D.C.L., LL.D. 

THE REV. RICHARD MORRIS, LL.D., M.A. 

HENRY SWEET, M.A., Ph.D.. LL.D. 

JAMES A. H. MURRAY, LL.D., M.A. 

THE REV. PROF. W. W. SKEAT, Litt.D., M.A., LL.D. 

THE REV. PROF. A. H. 8AYCE, D.C.L., LL.D. 

HENRY BRADLEY, M.A. 

Ordinary Members of CotmeU. 



E. L. BRANDRETH, ESQ. 
PROF. T. DE LACOUPERIE. 

F. T. ELWORTHY, ESQ. 
TALFOURD ELY, M.A. 

C. A.M. FENNELL,A.M., Lrrr.D. 
H. HUCKS GIBBS, M.A. 
I. GOLLANCZ, M.A- 
A. S. HARVEY, M.A. 
T. HENDERSON, M.A. 
J. POWER HICKS, M.A. 



PROF. W. P. KER, M.A. 
R. MARTINEAU, M.A. 
REV. J. B. MAYOR, M.A. 
W. R. MORFILL, M.A. 
J. PEILE, M.A., Litt.D. 
THEO. G. PINCHES, ESQ. 
PROF. J. P. POSTGATE, M.A. 
PROF. C. RIEU, Ph.D. 
W. H. STEVENSON, ESQ. 
MAJOR R. C. TEMPLE. 



Treasurer, 
BENJAMIN DAWSON, B.A., The Mount, Hampetead, London, N.W. 

Hon, Secretary, 
P. J. FURNIVALL, M.A., Ph.D., 3, St. George's Square, Primrose Hill, N.W. 

Bankers, 
MS88B8. BARCLAY, RANSOM, & Co., 1, Pall Mall East, S.W. 



Emtbakcb Fbb £1 Is, ; Subscription, £1 Is. a Year (due every Ist ol January), 

OB £10 lOs. FOB LIPB. 



Publishers of the Transactions, 
KEGAN PAUL, TRENCH, TRUBNER & Co., Limitbd, London. 



II 



I 



MEMBERS OF THE PHILOLOGICAL SOCIETY, If 

{Corrected to January, 1893.) 



HONORARY MEMBERS. 



Professor Henri Gaidoz. Ecole des Hautes Etudes, 22, 

Servandoni, Paris. Editor of the *'Iteime Celtiqt^,*' 
Professor Kern. Leiden. 
Professor F. A. March. Lafayette College, Easton, Pa., TJ.J 

Author of "-4 Comparative Grammar of Anglo^Saxofi 
Professor Matzner. Berlin. 

Author of the " Engliache Grammatik.'' 
Professor Paul Meyer. Ecole des Chartes, Paris. 

Editor of Flamenca, etc. 
Professor W.D wight Whitney. Yale Coll., Newhaven, TJ.J 

Author of ** Language and the Study of Language^\ e 
Professor Windisch, Ph.D., Editor of Irische Texte, etc. 
Professor J. Zupitza, Ph.D , Editor of Beowulf, etc. 



ORDINARY MEMBERS. 

* 0OIlPOITin>KB« FOm LIFK. 

1853. Dr. Altschijl. 9, Old Bond Street, W. 
1886. J. Amours, Esq. 2, Clifton Place, Glasgow. 
1879. ♦J. B. Andrews, Esq. Le Pigaut^, Menton, Alpes 

1886. R. N. Bain, Esq. British Museum, W.C. 

1 883. Alfred D. G. Barriball, Esq. Dunheved, Blenks 

Road, Bolingbroke Grove, Wandsworth, S.W. 
1881. rrhe Rev. A. L. Becker. " Appin," Ma 

Road, Beckenham, Kent. 
1870. Alexander Graham Bell, Esq. Scott Circle, We 

ington, U.S.A. 
1891. J. J. Beuzemaker, Esq. 87, Southampton R 

London, W.C. 
1856. J. P. Bidlake, Esq. 339, Essex Road, N. 



I 



t 



Members of the Philological Society, Jan. 1893. iii 

€9. ♦Demetrius Bikelas, Esq. 4, Eue do Babylone, 
Paris. 

5. Henry Bradley, Esq. 6, Worcester Gardens, Clap- 
ham Common, S.W. {Joint-Editor of the Society's 
Dictionary). 

2. E. L. Brandreth, Esq. 32, Elvaston Place, S.W. 
J. S. Bribrlt, Esq. 42, New North Eoad, Hudders- 
field. 

0. S. Illingworth Butler, Esq. Licensed Yictuallers' 
School, Lambeth, jS.E. 

^ ^^ ^^^0. CAMBRnxsB Philological Society. 

"^^^^^0. *Canterbury College, New Zealand. (Care of 
-. E. Stanford, 55, Charing Cross, W.C.) 

^^^^A Robert L. Cassie, Esq. 27, Colebrooke Row, 

London, N. 
5. F. Chance, Esq. Burleigh House, Sydenham Hill, 

r-^^^^6. William M. Christie, Esq. Safed, Palestine. 

^^^^^7. Miss Caroline Churchill. Trenant, Wilbury Road, 

- Hove, Brighton. 

}: ^"^S. ♦Hyde Clarke, Esq. 32, St. George's Square, S.W. 

^^^C. Fred. H. M. Corbet, Esq. Royal Colonial Institute, 

- Northumberland Avenue, W.C. 

^^7. Miss Louisa B. Courtenay. 34, Brompton Square, 

^SS. G. MiLNER-GiBSON-CuLLUM, Esq. Hardwick House, 
near Bury St. Edmunds. 

"^V. Benjamin Dawson, Esq. The Mount, Hampstead, 
1 s> N.W. (Treasurer). 

^^BO. ♦The Rev. Prof. A. J. D. D'Orsey. 4, Trafalgar 

Terrace, Coatham, Redcar. 

^^^. R. T. Elliott, Esq. Trinity College, Melbourne. 

^-^Sa. ♦The Rev. M. James Elliott. (London Missionary 
- Society, 14, Blomfield Street, E.C.) 

^^76. Fred. T. Elworthy, Esq. Foxdown, Wellington, 
Somersetshire* 

^865. ♦Talfourd Ely, Esq. 73, Parliament Hill Road, 
Hampstead, N.W. 



IV Members of the Philological Society^ Jan. 1893. 

1842. The Rev. WiUiam Farrer. Oakleigh, Ark\ 

Bead, Hampstead, N. W. 
1875. Dr. G. A. M. Fbnnell. Barton Cottage, 

bridge. 
1877. *Donald W. Ferguson, Esq. (Messrs. Kegan 

Trench, Triibner & Co., Paternoster House, CI 

Cross Road, London, W.C.) 
1888. T. o Flannaoile, Esq. Pendennis House, E 

Road, Upton, E. 
1872. Gaston PhiKp FoX, Esq. 34, Do Vere Ga 

Kensington, W. 
1842. Danby P. Fry, Esq. 138, Haverstock 

N.W. 
1847. ♦Dr. F. J. FuRNivALL. 3, St. George's S 

Primrose Hill, N.W. {Hon. Sec.) 



1859. ♦H. Hucks Gibbs, Esq. St. Dunstan's, Rx 

Park, N.W. 
1879. Dr. J. Hall Gladstone. 17, Pembridge S 

Bayswater, W. 
1892. I. GoLLANcz, Esq., 154, Houndsditch, E.C. 
1862. Dr. Clair J. Grece. Red Hill, Surrey. 
1869. The Rev. Walter Gregor. Pitsligo Manse, I 

burgh, Aberdeenshire. 



1868. Prof. John W. Hales. 1, Oppidans Road, Pr 

Hill, N.W. 
1862. *Sir Reginald Hanson. 4, Bryanston Square, "^ 

1879. *Prof. J. M. Hart. Cornell University, Ithaca 

York, U.S.A. 
1889. Alex. S. Harvey, Esq. 16, Hanover Terrace. 

broke Square, W. 
1892 Prof. Frank Heath, 29, Douglas Road, Canoi 

N. 

1880. *Prof. H. R. Helwich. 29, Neugasse, Oberdd 

Vienna. 

1881. T. Henderson, Esq. Bedford County School, Be 
1849. The Right Rev. Lord A. C. Hervey, Lord 1 

of Bath and Wells. The Palace, Wells, Son 
shire. 
1868. J. N. Hetherinoton, Esq. 4, Lansdowne 
Netting Hill, W. 



Members of the Philological Society, Jan. 1893, v 

Jt854. *John Power Hicks, Esq. Clifton Lodge, Blomfield 

Road, Maida HiU, W. 
1864. ♦Shadworth H. Hodgson, Esq. 45, Conduit Street, 

Regent Street, W. 
1875. C. R. Hodgson, Esq. 42, Queen Square, W.C. 

185-. Martin H. Irving, Esq. (care of Messrs. Robert- 
son & Co., Warwick Square, E.G.) 



1875. H. Jefferson, Esq. 234, Evering Road, Clapton, 

N.E. 
1878. C. S. Jerram, Esq. 134, Walton Street, Oxford. 
1888. P. De Lacy Johnstone, Esq. 1, Park Crescent, 

Oxford. 



1892. ♦Prof. William Paton Ker. 95, Gower Street, W.C. 
1882. R. N. Kerr, Esq. King Street Institution, Dundee. 

1886. Prof. Terrien de Lacouferie. 54, Bishop's Terrace, 

Fulham, S.W. 
1890. Monsieur Raoul de La Grasserie. Rue Bourbon, 4, 

Rennes, France. 
1869. •The Hon. and Rev. Stephen Willoughby Lawley. 

Spurfield, Exminster, Exeter. 
1890. Prof. Jas. Alex. Liebmann. Rondebosch 7, Cap© 

Town, South Africa. 
1862. ♦D. Logan, Esq. 
1884. The Rev. Richard Lovett. 42, Sisters Avenue, 

Clapham Common, S.W. 
1842. *Dr. E. L. Lushington. Park House, Maidstone. 



1883. *The Rev. A. MacDiarmid. The Manse, Qrantown- 

on-Skey, Scotland. 
1890. W. Stuart Macgowan, Esq. 1, Montague Lawn, 

Cheltenham. 
1892 George MacLean, Esq. 27, Montague Street, Russell 

Square, W.C. 
1886. W. C. G. Macpherson, Esq. Howrah, E.I.R., 

Ben^l, India. 
1867. Prof Kussell Martineaij. 5, Eldon Road, Hamp- 

stead, N.W. 



VI Members of the Philological Society^ Jan, 1893» 

1842. C. P. Mason, Esq. Dukesell, Christchurch Boad, 
Streatham HiU, S.W. 

1873. The Eev. J. B. Mayor. Queensgate House, King- 

ston Hill, S.W. 
1884. ♦F. D. MocATTA, Esq. 9, Connaught Place, W. 
1854. ♦Lord Robert Montagu. 41, Queen's Gate, S.W. 

1874. ♦W. R. MoEFiLL, Esq. 4, Clarendon Villas, Park 

Town, Oxford. 
1862. The Rev. Dr. R. Morris. Harold Wood, Essex. 
1871. Sir Charles Murray. The Grange, Old Windsor. 
1868. Dr. James A. H. Murray. Sunnyside, Banbury 

Road, Oxford. {Joint^Editor of the Societt/'s 

Dictionary,) 



1886. Prof. A. S. Napier {President). Headington Hill, 

Oxford. 
1892. H. A. Nesbitt, Esq. 7, Newburgh Road, Acton, W. 



1881. T. L. Kington Oliphant, Esq. Charkfield, Gbsk, 

Auchterarder. 
1874. Owens College, Manchester, (care of Cornish^ 

33, Piccadilly, Manchester.) 



1873. Prof. Arthur J. Patterson. IX. Lonyay-utca 11,. 

Budapest, Hungary. 
1892. Arthur Paul, Esq. 4, Berkeley Road, Crouch End,. 

1866. Dr. J. Peile. Master, Christ's College, Cambridge. 
1889. Miss C. Pemberton. Jagerhaus, Meran, Tirol, 

Austria. 
1886. Theo. G. Pinches, Esq. 62, Newman Street, W. 
1880. *Prof. J. P. Postdate. Trinity College, Cambridge. 

1888. Prof. R. Quattrocchi. 147, Via Arpi, Foggia. 



1884. The Rev. J. Richardson. (London Missionary 

Society, 14, Blomfield Street, E.C.) 
1882. *William Ridgeway, Esq. Gonville and Caius 

College, Cambridge. 
1869. Prof. Charles Rieu. British Museum, W.C. 



Members of the Philological Society, Jan. 1893. vii 

®S9. M. L. RousB, Esq. 3, The Manor Way, Black- 
heath Park, S.E. 



S. •The Rev. A. H. Sayce. Queen's College, Oxford. 
^2. John Sephton, Esq. 90, Huskisson Street, Liverpool. 
1 ^.^M. J. G. E. SiBBALD, Esq. The Admiralty, Whitehall, 

-f ^ S.W. 

-■-^^3. ♦The Rev. Prof. Skeat. 2, Salisbury Villas, Cam- 

bridge. 
^^^©. ♦Eustace S. Smith, Esq. Bonner Road, Victoria 

Park, E. 
^^^3.. ♦T. B. Sprague, Esq. 26, St. Andrew's Square, 

Edinburgh. 
^ SS©. The Rev. W. G. Spurrell. St. David's, Pembroke- 
^ shire. 

}^ ^QS. J. Herbert Stamp, Esq. Trinity College, Cambridge. 
*^^S6. J. H. Staples, Esq. Bruachdryne, Braemar, N.B. 
-"^^"^S. The Rev. Dr. Thomas Stenhouse. Stocksfield-on- 
^ Tyne, Northumberland. 

^^93. W. H. Stevenson, Esq. 7, Tooting Bee Road, 

- London, S.W. 

^»S8. Dr. Whitley Stokes. 15, GrenviUe Place, Cromwell 

- Road, South Kensington, S.W. 

|^S7. Prof. Strachan. W<x)dbank, Marple, Cheshire. 
^^82^ •Mrs. A. Stuart, jun. 19, Regent Terrace, Edin- 
1 3^ burgh. 

^^69. •Dr. Henry Sweet. Cambray, South Park, Reigate. 




Major R. C. Temple. (H. S. King & Co., Com- 

^^*Sl. Henry Walter Thomson, Esq. Hazelbank, Syden- 

, ham Hill, S.E. 

^6e, Samuel Timmins, Esq. Spring Hill, Arley, 

1 S) Coventry. 

^9X. Toronto Public Library. (C. D. Cazenove, 26, 

, Henrietta Street, Covent Garden, W.C.) 

^^87. Edw. Tregear, Esq. Government Buildings, 

1 c> Wellington, New Zealand. 

^^86. Trinity College Library, Cambridge, (care of 
Deighton, Bell, & Co., Cambridge.) 

*^^7l. Dr. E. B. Tylor. The Museum House, Oxford. 



VIII Members of the Philological Society, Jan. 189? 

1892. University of Minnesota. Minneapolis, 
sota, U.S.A. 



1864. •£. ViLES, Esq. 16, Wetherby Gardens, Sout 
sington, S. W. 

1873. M. J. Walhouse, Esq. 28, Hamilton Terrace 
1880. Eichard Ware, Esq. Heath Street, Ham 

N.W. 
1851. •Dr. R. F. Weymouth. Collaton House, Brer 

Essex. 
1863. Henry B. Wheatley, Esq. 2, Oppidans 

Primrose Hill, N.W. 
1882. *Thomas Wilson, Esq. Rivers Lodge, Harj 

St .n.lD&.Ti8 Herts 

1870. Nicholas Wilcox Wyer, Esq. 3, Matford Terr 
Leonard's, Exeter. 



Bankers : Messrs. Barclay, Ransom, and C!o., 1, Pa 
East, S. W. 



Publishers of the Transactions : Messrs. Keg an Paul, T 
TrObner & Co., Limited, Paternoster House, C 
Cross Road, London. 



8TEPHEK AU8TI^ AKD dOMS, PUMTKRS, HKRTPORD. 



I 



289 









V :. - . r> 






V :; ^_ ,;.^ .' I' _^ 



VIII.— ETYMOLOGIES. By J. Strachan. 

1. Ir. tallaim 'fit in, find room ' agrees well in meaning 
with Lith. tilpti * fit in, find room/ as a couple of examples will 
show. Compare, for instance, meit gamlias hi tallait trichait ^ 
n- gamna^ " as large as a winter fold into which go thirty 
yearlings" (Feast of Bricciu, §91), with suvdre i itdrta, 
aU dd i id tvdrta netilpOy " he drove them (the swine) together 
into the pen, but they could not get into the pen " (Leskien 
und Brugmann, Litauische Volkslieder und Marchen, p. 200). 
One might render i td tvdrta netilpo in Middle Irish by ni 
rothalhat isin chrd, Kurschat, in his Lithuanian dictionary, 
furnishes a couple of additional examples — toj haznycziqf teipa 
iukstantia zmoniu ** in this church a thousand men find room," 
toj pUczkoj stdpa netelpa ** into this bottle a atopa (a certain 
measure) does not go." Compare with these Irish examples 
like a toill ind ina seasam ni toillet ina suide ** the number 
which fit in it standing, would not (lit. do not) if sitting," 
dotallfasu indi ^* thou wouldst fit in it," and the like (Atkinson, 
Passions and Homilies, 910 ; Windisch, Worterbuch, 807). 
Formally tallaim might be connected with tilpti by assuming 
that the Irish verb started from a present in -nd-; tallaim 
might come from *talpndmi—ldg. *t9lpnami or the like. In- 
stances like dundalla, Milan Glosses, 3I<^ 17, where tallaim is 
treated as though it were a compound^ ^o*a//am do not prove 
that tallaim is in its origin a compound verb. Other verbs 
beginning with ty which are certainly simple verbs, are found 
treated in the same way. Thus tongim, W. tyngu * swear ' is 
treated as though it were for do-fongimf inti do-d-fongad =iB 
qui id iuravit, Ml. 36*. W. tgngu shows conclusively that 

ait 
* The ace. triehait-n must be a blunder for nom. trichaf MS. xxx. So also 
mHt Hat for meit Uit, 

Phil. Trans. 1891-2^. 19 



290 ETYMOLOGIES — BY J. STRACHAN. 

tongim cannot hare come from to-fongiM. In verbs compounded 
with to'fo'^f t6- alternates with do-fo-^ and it is to the analogy 
of such compounds that we owe dodfongad,^ 

2. Ir. gemel * fetter' <*ytf»iA)- : Old Slav, iima * press/ Gr. r^imo 

3. Ir. r6% * planities ' < ^rovesid : Lat. rtts, Zend ravanh. Different 
from this is Ir. roe ' battle '=W. rhae^ which is probably cognate 
with immidrdi i. rohriss * broke,' Leabhar na h-Uidhri, 85^ 20, 
and further with roen ' victory,' roenaim * break, defeat.' Cf . 
epeiTru)^ Old Norse rifa * break,' rifna * rumpi ' ? 

4. Lat. harena, Sabine fasenaj * sand ' is commonly connected 
with Skr. hhdsman 'ashes' and its cognates, cf. Ascoli, KZ. 17 
347, Fick I.« 160, Bersu, Die Gutturalen 131, Johansson KZ. 
80 432. The difficulty in this etymology is the representation 
of initial hh by h in Latin. There are certainly instances where 
h varies with f^bh, dh, haba, faba, hordm^ fordtu^ hebris, febris, 
etc. But Yon Planta, Grammatik der oskisch-umbrischen Dialekte 
463 is doubtless right in regarding this change as dialectical ; 
these cases could then be quoted in support of the above 
etymology of harena only if we suppose that harena came from 
this dialectical area, and of this there is no further evidence. 
Of the above change in common Latin words I know no certain 
instance. The connexion of horreum with far (Ascoli, KZ. 17 
338) is exceedingly improbable; the vocalism is difficult as well 
as the initial consonant. JBerba has been compared with Gr. 
(pepfiofiaif (popfii^, but the connexion is by no means certain (von 
Sabler, KZ. 31 278, compares herba for *gherdhd with Kpi9fj 
which again is improbable). There remains harenUf and for it 
another etymology is possible.' It may come from *gha8ind and 
be connected with Irish ganem * aaad* K^ghasnimd or *ghasnemd. 
The former is the more probable. The root is in all probability 
ghaB^ from this might come ^ghami-, whence, with secondary 
suffix md as in Lat. laeruma, would come *ghasnimd» As to 



^ doiagat LIT. 60* 30, 66^ 11, might at first sight seem to he a similar 
analogical hreaking up of the simple verh Uagaim * I go,* hut Zimmer KZ. 
30 25 is right in regarding doiagat as standing for dothlagat. Another 
old instance of th unexpressed hetween vowels is adroneettar gl. sustinoi Wh. 4^ 
35. This is evidently for adron^thestar^A compound of the same verh as is 
seen in amethim gl. sustineo, expecto, Ml. 46^ 20, 46<^ 14, 60^ 8, 9, etc. 
2 Stokes, Spracbschatz 111, adds O.H.G. unqtutnilo 'racemus.* 
' Dr. Whitley Stokes informs me that he had previously arrived at the same 
conclusion. 



ETYMOLOGIES — BY J. 8TRACHAN. 291 

*ghami' it might come from an n-stem *ffhasen-j cf. Lat. eollis 

< *colni8 by koXwv in KoXtvvo^^^ and to *gha9en' there may 
have been a parallel -m- stem *gha8eS' whence might be derived 
harma < ^yhaaesnd. Further cognates are uncertain. Do these 
words go with Skr. gh<u ' verzehren, essen/ as "^009, yfrafifio^ 
with Skr. hha% ? 

4. Ir. dalta * fosterling ' < *daltaios, or the like. Dal- may come 
from dhal- dhl- and be connected with Lat. J^lare, Qr. Oi^aaro, 
^dhe, whence, with strong grade of root, Irish dinu * agna.' 

5. Ir. truit^ druit * starling ' < *trozdi€ : Lith. strazdas * thrush,' 
Lat. turdus for *tor%do8. The words in the Brythonic group of Celtic 
dialects, Welsh drudwy^ Com. troden, Bret tretf dret, dred cannot 
have been regularly developed from *tro%dU — zd in this group of 
dialects gives th; if there is any connexion it must be one of 
borrowing, and that seems not unlikely for the Cornish and 
Breton words; the relation of the Welsh word to the others 
is not clear to me. As to truit, druitf the sinking of the 
tenuis to the media finds a parallel in Irish gee * branch ' 
compared with Welsh cang. Under what conditions the 
weakening took place it is not easy to determine. The tenuis 
would become a media after a preceding closely connected word 
originally ending in a nasal ; thus * inna n- c^e in the gen. pi. 
would have given inna gie^ and the media may have spread from 
such cases. Quite distinct from this is the sinking of a tenuis 
to a media in pretonic words as gaeh * every ' from each, do * thine * 
from to^*tovo or * love, cf. Skr. tdva, Lat. tutM, 

6. W. troeth ^wash, lye, urine,' <*troet&: Gr. rdpyavov * vinegar' 
(with different grade of vowel rapf^'^^trg-), with which Prellwitz, 
Etym. Wb. d. Griech. Spr. compares O.N. ^rekkr, O.H.G. dreeh, 
N.H.G. dreek. Another form trtoyth is given, which would come 
from some form like *troctiy wy for oe being due to the nature 
of the following vowel, cf . wyn * lambs ' < *ognl with oen * lamb ' 

< *ogno8. 

7. W. gwyw 'withered, faded' < ^vi-vo-s : Lat. vietus * withered,' 
Lith. v^stu * wither.' The suffix -wo- may have a participial force 
as in Ir. marb *dead' < *marvos; other examples are given by 
Johansson, EZ. 30, 443 note. 

8. Lat. ligula, G. Meyer, Indo-Germanische Forschungen II. 
368, rejects the derivation of Ugula from ^leigk * lick,' and 

^ Cf. Johansson BB. xriii. 13. 



292 ETYMOLOGIES — BY J. STRACHAN. 

postulates as the original Latin form *lu^ula. His reasoi 
for rejecting the derivation from leigh is the Old Slav, lidia 
* spoon ' ; according to him liffula and lUiiea are diminutives o 
a *kipa 'spoon.' In this he has overlooked some Celtic word 
which, from the close connexion between Latin and Celtic, hav 
more weight in determining the original form of the Latin word 
and which Stokes, Kuhn u. Schleichers Beitrage VIII. 328 
has already brought into connexion with ligula, Ir. Ikag * ladle, 
W. llwg, Com. lOf Bret, ba * spoon,' point to a Celtic wor 
*leigd * spoon,' of which, with a different grade of vowel Lat 
ligula is evidently a diminutive. The connexion of ligula wit! 
^leigh may then be maintained. With regard to Slav, lidic 
1 do not venture to express any opinion. 

9. Ir. fdar * I found,' frith * was found.' These words hav 
been discussed in Curtius' Grundziige,* 742, with no satisf actor; 
result. It is remarked there : ** Zu fiiar gehort als praet. pa« 
frith fofrith inventum est. Daraus lasst sich im giinstigsten fall 
eine irische wurzel far, d. i. var, erschliessen, mit fo (sub) zusam 
mengesetzte fo-far^ im perfektum zu fiiar verschmolzen ; das prael 
pass, frith ohne die praposition, fofrUh mit derselben, enthalt di 
geschwachte wurzelform, deren ri einem skr. r entsprechen wiirdt 
obwohl man hierbei nach dem muster von breth (vgl. skr. bhrta] 
praet. pass, von herim ich trage, eigentlich *freth erwarten sollte.' 
It seems to me that fiiar and frith might formally be explained a 
follows. We may postulate a root uere- with another form url- 
compare the variation in e/>€a>, /nfO^vai. The strong stem of th 
perfect from this root would be ueuor-, the weak ueur-. Thus w 
may suppose that there was at one time an Irish paradigm 1 sg 
*vevora, 3 sg. ^vevore, 3 pi. ^ceurontor. These forms would becom 
in the first instance *vovora, *0ovore, *vourontor. The last woul 
give regularly the historical fuaratar * they found.* Can *vovof 
have given regularly ^f^r^ fiiar'^ This sound-change seems t 
me to be established by the contraction of the combination < 
particles to-vor* to titar- as in tiiar-aschat * proferunt * < *to-fot 
ex-gahantOf to-vo^ to tua- in tualang * aptus, peritus,' < *to-fi 
lang-. We may then look upon fiiar as coming from *vovort 
fuair from *vovoref without calling in the analogy of the plun 
or dwelling on the fact that the 3 sg. fuair might also represei 
the middle form *ueytrai, A parallel to fuair * he found ' is mc 
with in forkair * he caused ' > *fo-ro-fiia%r pret. of fo-feraim. A 
for frithy unless we are to believe that it differs in formation froi 



ETYMOLOGIES— BY J. STRACHAN'. 293 

other preterites passive, it must be supposed to have come from 
*vritO'. In that case the i cannot be short, as ^vrito- would have 
given *freih. It is true that for the most part the ♦ of this 
word has no mark of lengthening over it, but the long mark is 
occasionally found, and the length of the vowel is sufficiently 
guaranteed by the absence of inflection. In some other words 
ending in th the long mark is found very sporadically. If then 
the i of frith is long, the form may be explained very simply : 
it comes from *vrito8, which stands in the same relation to ^vere- 
as Gr. /JiyTO» to ipeo) or rprfro^ to repe-rpov. That, as Mr. Stokes 
long ago suggested, fdar is etymologically connected with ^vploKw 
is very probable, but the formation of the Greek word is not 
clear. From a disyllabic root euer- (for examples of such roots 
see Persson, Zur Lehre von der Wurzelerweiterung 227 sq.) might 
come on the one hand uer-^ on the other eur-^ but how then is 
the rough breathing to be explained ? Can analogy of some 
kind have been at work ? That must remain a mere guess until 
some word is pointed out near enough in meaning to €v/><Wa> 
to have influenced its form. There are other possibilities, but 
all too uncertain to be mentioned. "Whether there is further 
connexion with Skr. vdrate * shuts in,' Old Bulg. vrHi * claudere,' 
Lith. suverti ' close,' must be left an open question. For the develop- 
ment of meaning might be compared perhaps Gr. jrepifidWecOai 

* to bring into one's power.' 

10. Ir. bras, W., Com., Bret, bras 'great, big^ may come 
from *bratto8f and this from ^mrattos, ^mr^dh-tos: Gr. pXivOpo^ 
for *Ppa}$p6^^ Skr. murdhan 'head,' Old Bulg. bribdo ' height,' etc., 
cf. Johansson, KZ. xxx. 449. There is also an Irish word bres 

* great,' which might be derived from *brittoa, *mrdh-tos, 

11. Ir. medar * mirth ' < ^medro^, medraim * disturb, confuse ' 
{cid notmedrasu 'what disturbs thee,' LL. 57^ 27, rotmedair 'has 
confused thee,' LL. 308» 36) < *medraiji: Skr. mad 'gliicklich 
aein, sich berauschen,' cf. Fick, Wb.* 105. 

12. Ir. tailm 'sling/ Bret, tdlm 'sling' (in dial, of Yannes 
' coup de tonerre '), may stand for ^taiksmi-, cognate perhaps 
with Old Bulg. tlUka ' schlage ' ^telk-, though the meanings 
differ considerably. Closer in meaning to the Slavonic words 
are "W. talch * grist, or coarse meal,' Corn, taleh ' bran,' W. 
ialeh ' broken, bruised,' cf. Euss. tolokno * gedorrtes hafermehl.' 
Talch might come from talko- : it might also by Stokes' law come 
from talkkoj talknd-, -nO' part, from telk-, and so^Euss. tolokno. 



294 ETVMOLOOIES — BY J. STRACHAW. 

13. It. toue 'wish' < ^to-visei', *tO'V$n$k%' : Germ. «no 
< *wunik6t Skr. vdiichd ' wish.' ^ As Irish en may come li 
n, to'venski' may contain the same grade of vowel as *wuih 
uensk' is a -sko- formation from uen-, Skr. vdnati * desire,' G< 
vunan ^sich &euen/ etc., which appears perhaps in Celtic 
"W. gwmu ^ to smile, to look pleasantly.' 

14. Ir. ddssaim in pass, with prep, imm ^to rage,' Wind 
Wb. 407, LL. 69» 8, 266^ 8, 268» 11, ddsacht ' insai 
ddsaehtaeh 'insanus.' Ddssaim may come from ^dhudstdjfi, 
we may compare Lith. dreati ^breathe/ dvat^ ^breath, gh< 
dusmas 'anger,' Ags. dvas^ 'hebes, fatuus,' Dutch dwaas 'thoric 
Persson, Zur Lehre v. d. Wurzelerweiterung 81 sq. The Irish i 
ambiguous, but if the word is rightly derived from dhuis 
is simplest to suppose that it is based on a stem *dhu6i 
Cf. similar Lat. formations, Brugmann, Grundriss II. 11. 26. 

15. Ix.fuinim ' set' of the sun might be derived from *vo-n 
Gr. viofiai ; fuinid in grian literally ' die sonne geht unt 
The explanation of the cognate fuin * sunset ' is not quite cL 
I have examples only of the ace. /uin,^ A nom. fuin mi 
come from ^vo-nea-s a radical noun, which may have fa 
together with i- stems, and so have acc./ain* (for *fuine=''^von€s^ 
It is, however, possible that it may be a late deverbative h 
fuinim^ of the same kind as Lat. pugna from pugnare, 

16. Ir. fws« * blow, wound,' < *nec8o- or *necsa : Old Bi 
niza * infigo,' pronoziti * perfodere,' Gr. vvffffto. The Irish w 
may be based on a neut. stem neglies-. From ness seems to ec 
the Yerh fo-nessaim, LL. 74^ ^Ofosnessa sleig eulind ina bond trai 
* he drives a spear of holly into the sole of his foot' (the LU. vera 
73^ 15 has forneBsa), Another form of the root appears in 
c*7xo», cf. Prellwitz Et. Wb. s.v. The two forms negh, engh n 



^ Mr. Stokes informs me that he had previously compared with vSn, 
icunschf Ir. dufaacaiget, Ml. 33^ 3, toi»e, Toite may then, as he suggests, y 
■well come from *to-von»ki', 

^ With Ags. dtpae»f Dutch dwaas Ir. ddsaaim goes very well in meani 
Ml. 60^ 2, in a gloss on non nisi mentis iueumpotein dieit ptovidaitiee negmtiif 
The gloss runs neeh tastkibi eiail 7 immandaister ishe asber nad Jil dU 
re{medeieseu) da diadulaib^ '*he who has no understanding and is out of 
senses {mefttis ineofnpos), he it is who says that there is no law of providence 
God to his creatures." 

* Windisch, Wb. s.v. fuin, star eo fuin * westward to sunset,* LL. 10 
t=Book of Ballymote 33» 44. 

♦ If we could assume a radical nent. noun *vo-nes fuin would be the regi 
ace, but I can find no instance of such a formation. 



ETYMOLOGIES — BY J. STRACHAN. 295 

best be explained from a disyllabic ene^h which became differen- 
tiated into n#yA, en^h, under different accentual conditions. This 
form of the root is also found in Irish in composition.^ The oldest 
instance is Wb. 4^ 13 adcomcisset ilhiim firm 'they struck many 
blows against him': adeumci88et^*aith-com'anff'i88et. For the 
phonetic change cf. ni ehumcaim^ 'I am not able '^^ni eom^angaim. 
The corresponding 1 sg. ataeometu (^msaith-da-comangm) * I struck 
them' is found LU. 114^ 11 ataeomeu88a com Idu 'I struck them 
(the doors) with my heel.' Cf. also LL. 107* 14 atacomaihg na 
teora formaela Mide * who smote {i.e. cut off the tops of) the three 
Bald Hills of Meath.' The enclitic form of aith-comaing is iemaing^ 
iemoing * he struck, cut/ Windisch Wb. 5 1 7-5 1 8. The simple comang 
is found LL. 90* 21 cumahg8a in m-birsa iri8%ndara n-ai dih 8ium 
* I drove this spear through one of the two/ the 3 sg. cumaing in 
the same passage. As to the Irish vocalism I have with great 
hesitation written ang: a^ however it is to be explained, is un- 
doubtedly found in Celtic in the e-o series. But it is possible that 
the Irish forms are based on a perfect — *onga, *onge with regular 
ablaut. Such perfect forms are cumang^ eumaing, which might come 
from *comongaf *comonge, but also from *comanga^ *comange, 

17. "W. breuan * carrion-crow.' As "W. hreuan * handmill/ 
Ir. hrd comes from *^dro»-, *jrdro»-, Skr. grdvan^ ' stone for 
pressing the Soma/ so breuan 'carrion-crow' may be derived 
from *brdvon'f ^^rd-von- : Gr. fiifipwaKU) * eat.' "We have a 
cognate formation in W. breuad 'a grave- worm, a worm that 
eats the bodies of the dead ' < *jro-wo^. 

18. Ir. dabach 'cask' < *dabakd, *dhabhakd: Gr. Ta0o9, rd</>po9, 
$aimt)j and cognates. 

19. Ir. derg *red.' So far as I know no cognates have yet 
been discovered for this word. May it not be possibly connected 
with Ags. deore^ Eng. * dark ' < *dhergo8 ? For the two meanings 
cf . Skr. raj * colour, be red/ raktas * coloured, red,' rajas * dunst- 
kreis, nebel, dunkel, Gr. epefio^f Goth, riqis ' darkness.' 

20. Ir. *di'nessiin^ • despicio ' : Gr. ouoaaofiai^ ouotrro^y etc., 
Zd. nod 'schmahen, cf. Osthoff, Perfekt 394 n., Fersson, Zur 

^ Cf. ZimmeTf Zeitschrift fiir Deatsches Alterthum xxxii. 253. 
^ Graphic for eumgaim which is also foand, e representing the explosive g, 
* Old-Irish examples are codunetaa gl. ut dispiciat Ml. 36* 12, andarune»U8 
gl. spemens f^when I despised them) d6« 1, donesbe gl. dispicies 112<: 3, 
9derne*9a gl. aonec dispiciat 129* 14. A compound with com is also found : 
cannesaat * they condemn * Wb. l^ 7, eia eonnenfea tuicsiu di gl. quia accusabit 
adversus electos Dei ? 4i> 15. 



296 ETYMOLOGIES — BY J. 8TRACHAN. 

Lehre v. d. WurzelerweiteniDg 198. The formation of nes^ 
is uncertain ; it might come from *tieiteid or ^nettiid^ or ''^netm 
or *nstsiid. If this connexion is to be maintained, the Ir 
'word points to an Idg. root *«^rf, to which Gr. oj/o^- woi 
stand in ablaut. 

21. Ir. ipv. tair^ 'offer* points to a compound to-arg-, Tl 
arg- is found with com- LL. 65* 48, eoropi in laech eonat 
comrae dar cend in ehoicid ali "so that he is the -warrior w 
shall offer battle on behalf of the whole proTince," 65^ 1, core 
conairr seta ** so that he shall offer treasures " : eonairr < *eo\ 
arcsetf subj. of s- aorist. We may perhaps compare Gr. opeytv a 
its cognates so that the primary meaning would be * to stretch ou 
whence comes easily the meaning of * offer,' cf. oTriroriponTi var 
Zcvv KuBof ^p^^'O *^^ ^^® like. Root oreg-^ reg^ org- (^op^yvu 
Persson, Zur Lehre v. d. Wurzelerweiterung 225. The form re 
is also found in Irish, Ascoli, Lexicon Palseo-Hibemicum, cxcvii. 



297 



IX.— ON THE ASSIMILA^TION OF PRETONIC N IN 
CELTIC SUFFIXES. By Whitley Stokes, D.C.L. 

[Head 6th February, 1893.] 

"Febneb scheint n als Anlaut hochbetonter n-Suffixe an die Nach- 
kommen idg. wurzelschlieBsendenYerschlusslaute assimiliert worden 
zvL sein. So lasst sich die urgerm. Konsonantendehnung erklaren 
z.B. ahd. Uechon * lecken/ urgerm. likko, aus vorgerm. ^li^h-nd-f 
vgl. griech. Xixvevoy, w. leijh,^^ — Brugmann, Grundriss I. § 214. 

" huy dn, gn, vor dem Hauptton wurden im Urgerm. zu hh, dd, gg^ 
daraus nach § 533, pp^ it, kk, die weiter ebenso behandelt wurden, 
^6 die aus idg. pn, in, k^n, qn, und aus idg. hhn, dhn, ^hn, 
:^hn, entstandenen pp, tt, kk (§ 530, 538).— Grundriss I. § 534.^ 
2See also ihid. II. p. 397, § 541. 

Though well aware of the danger of assuming that a phonetic 
law, good for one family of languages, is good for another, it seemed 
to me, when I read these passages, that they gave the key to the 
etymology of six groups of Celtic words, which have not, so far as 
I know, hitherto been explained. I mentioned this key, briefly, 
in Kuhn's Zeitschrift, xxix. 375, and now after five years' con- 
sideration, I proceed to submit to Celtologues the words in question 
and their respective etymologies. 



« 



I, gg from -gnu, -gni, -gnd, 

1. Ir. actt8f ocuis (now agw) *and'; Old- Celtic *akkusfi from 
agg^tt (the provection being due to the accent) ; pre- Celtic 



^ Compare Paul-Braune*8 Beitrage, vii. 133'; Osthoff, ibid, vii. 297 et aeq. ; 
Kluge, ibid, ix. 149 ; Kauffmann, tbid, xii. 504. 






298 MR. STOKES— ON THE ASSIMILATION OF 

*aghnust%t cognate with Lat. anguntu9 ; Greek tlx^^f^t (Schmidt^ 
VocaliBmus, I. 31), a7x« ; Skr. amhiiM, So Ir. ae *and' (Bodleian 
Annals of Innisf alien, fo. 35^ 2, etc.) ; W, aeh * near,' point to 
an Old-Celtic *akko', ^aggd-^ pre-Celtic ^agh-nd- . 

In the Welsh ae *and/ agos * near '= Corn, hag, ogo% ; Bret, hag 
* and,' hogo% ' presque, & pen pr^s,' the urkelt. kk seems to have 
heen simplified. The Ir. prep, ucy oo *juxta, prope, apud'=» 
W. wngt umgc * prope,' seem to come from an Old-Celtic ^onko-, 
cognate with Skr. dndmga ; Gr. ffve^Ke, 

2. Ir. aice, aie * hand,' * chain * ; Old-Celtic *akk%, *aggi ; pre- 
Celtic *pagni (the loss of p is regular), cognate with Lat. pa-n-go 
(Grundriss, § 632), eom-pdge^ ; Gr. wrj^'w/au. From aice comes 
Ir. aicds * structure,' * building.' 

This etymology of aiec is doubtful, as it might equally well 
be derired from a pre-Celtic *paJtni'f cognate with the Skr. 
pagdgati * binds ' ; Lat. pax, pacisear ; Gr. waffffoXoi (from 
*7raicia\o9) ; Qoth.,fahan, 

3. Ir. hacc (now hac) * crozier, hook ' ; Welsh hdeh * hamus, 
uncus * ; Old-Celtic *hakko', *hngg6' ; pre-Celtic Hlutg-nd, cognate 
with A. 8. hac ; Eng. hack ; Skr. ^h?uij * sich wenden.' For the 
connexion of ideas, compare N.H.G. Rilcken, cognate with Skr. 
^krune * to be crooked, to curve.' 

4. Ir. hocc (now hog) * soft ' and boec * bow ' (in OJr.fid-hoce * arcus 
ligneus'); Old-Celtic Hukko-, Huggd-; pre-Celtic Hhug-nd, cognate 
with A.S. boga; Eng. how; N.H.G. Bogen, hiegen, hiegaam; Gr. 
<p€v<ytM)y 0v77aj/a>, Lat. fugio ; Skr. hhuf, part. perf. pass, bhugnd. 
The mod. Ir. bogha is borrowed from the A.S. 

6. Ir. boec (now boc) * he-goat ' ; W. bwch ; Old-Celtic Hukko-s, 
*bugg6-8 ; pre-Celtic ^bhug-nd-s, cognate with Zend bu%a * Bock ' ; 
A.S. bucca; O.N. bukkr; O.H.G. boe. 

It is possible that this Ir. boec may have been borrowed from 
one of the Teutonic words just cited. 

6. Gaul, hrdca * breeches'; Old-Celtic *brdkkd, *bragga, pre- 
Celtic *bhrdg-ndy derived from *bhrdg * rump,' cognate with Lat. 
frdgro. For the connexion of ideas compare Lith. bulis * buttock ' 
and Skr. buU * vulva,' cognate with Ir. bolad (Old-Celtic Hulato-B) 
* smell'; Skr. puta m. du. * the buttocks' B.R., cognate with 
piky, pugatS * to stink ' ; Lat. podex, cognate with pedo ; and Gr. 
Xo5ai/o9, Zend zadhahh, cognate with x^?*^ ^^^ ^^^* ^^' 

From the Gaulish brdca (where the Old-Celtic kk is simplified) 
are borrowed, on the one hand, Bomanic words like Ital. braea; 



PRETONIC N IN CELTIC SUFFIXES, 299 

Fr. hrate, and, on the other, Teutonic words, such as O.H.G. 
pruoh ; O.N. brdkr ; A.S. hrdc, pi. brec ; Eng. breech. 
The Br. hragou seems to come from a Low-Lat. hrdea, 

7. Ir. brecc(jiow breac) 'speckled* ; W. brpeh ; Old- Celtic *mrA^©-*, 
*mrgg6'8 ; pre-Celtic *mfg-n6'S, cognate with Lith. mar gas. Another 
participle from the same root is O.Ir. mrecht * variussW. brith 
* motley, pied ' ; Old-Celtic *mrki0'8 ; pre-Celtic *mrg't0'8. 

8. Med. Lat. (Gaulish ?) cloaca * bell,' W. clock, F., Old-Celtic 
*klokkd, from an oxyton *kloggd, pre-Celtic *klognd, cognate with 
Gr. K\a^uj (perf. ic6ic\a77a), from *ic\a7«tu, and icXa^^iy ; Lat. 
ela-n-go; Lith. klageti *to laugh'; O.N. hlakka *to scream* The 
Irish cognate is cloce (now c/'^^), a maso. o-stem, deducible from 
a pre-Celtic ^klog-nd-s. 

This etymology is doubtful, as Med. Lat. clocca, W, clock may 
come from *klukkdf kluk-na ; and Ir. cloec from *klukko-s, *kluk-n6'8, 
cognate with Bulg. klilcam * stosse ' ; Servian kueati * klopfen.' 

9. Ir. cnocc (now cnoc) M. *hill'; O.Br, cnock (gl. tumulus); 
Old-Celtic *knokkO'Sf *knogg6'8 ; pre-Celtic ^knog-nd-a. Cognate 
with O.N. knakke ' Hinterhaupt ' ; A.S. knecca; Eng. neck; N.H.G. 
Nacken, see Kluge s.v. In W. enwCy pi. cnyciau * gibbus, tuber,' 
we have an unexplained preservation of the Old -Celtic kk. 

10. W. (?ry<?A 'curled'; Bret, creek \ Old-Celtic *krekko-8, *kregg6'8; 
pre-Celtic *kregk-n6'8. Cognate with A.S. kring ; O.N. hringa ; 
O.Slav, kragii * circle,' kragl^ * round.' 

11. Ir. iic {itg, 0*C1.) *moon'; Old-Celtic *enkki', *efiggi' ; 
pre-Celtic *peng'ni-f cognate with Gr. 067709 from *8phengo8y 
Mod.Gr. (f>€yr^api * moon.' The common word for • moon,' hcaCy is 
to Sic as geecae * branch * is to gSc, 

12. Ir. fecc (now feac) * spade ' ; Old-Celtic ^vekka, *vegga ; pre- 
Celtic *cegk-na; cognate with Lat. vanga^ v^meTf Gr. o0i//9, O.N. 
vangsni, O.H.G. waganso ^ploughshare ' ; Fruss. wagnia * coulter.' 

13. Ir. glicc (now glic) * cunning ' ; Old -Celtic ^glkki-s, *9l99^'^\ 
pre-Celtic *gklgk-ni'Sf cognate with Gr. KoKxalvtDy KaXx<>^^ J Goth. 
glaggvd, glaggvuha. 

The Scottish gleg seems borrowed from Ir. glicc, 

14. W. kwck F. *a sow' ; Com. kock\ Old-Celtic *8ukkd, *suggd; 
pre-Celtic ^sugk-na, cognate with A.S. mgu *sow' and Dutch 

There is a modem Irish suige ' a call to pigs,' and O'Reilly has 
also 8uig * a pig ' ; but this may have been borrowed from A.S. 
mgu. Why W. kwck is not kock I do not understand. 



300 MR. STOKES — ON THE ASSIMILATION OF 

15. Ir. laee (now la^) 'weak'; Old-Celtic *lakkO'Sf lag^ 
pre-Celtic ^hg-nd-Sy formally=the oxyton Gr. X1X71/09, and coj 
also with Xar^dyv^ Lat. languor, Una (from *l$ngna\ etc. Ii 
ll€iee * lazus, remissus/ we have another instance of the pre& 
tion of Old-Celtic kk, 

16. Ir. leee, leae 'an act or deed which hinds the pe 
indissolubly,' O'Don. Supp. ; Old-Celtic *l%kka (-ib-?), * 
{-gd- ?) ; pre-Celtic ^lig-na (-»<5- ?), cognate with Lat. h 
ligtUa, 

17. It. "liee in dlie {^ad-lice) and ad-laic (=aith4ice) * pleai 
W. lip in eyffelyh * consimilis'; Old-Celtic *liqqi', */t55i- ; pre-( 
*/iS-ni. Cognate with Goth, ga-leiks, leikan ; Lith. lygus ; 
linga-m (J. Schmidt, Yocalismus, i. 89). 

18. Ir. loco (now log) * a hollow ' ; Old-Celtic *lukko'f *lugg6- ; 
Celtic ^lug-nd-. Cognate with Gr. Xv^i^uy^ Lith. lugnas * bieg 

19. Ir. meniec (now minic) 'frequent, often* ; "W. mynyeh; 
Celtic *menekki-j *meneggi' ; pre-Celtic *menegh'ni'. Cognate 
Goth, manags ; O.Slav. tnHnogH (Eluge s.v. manch). 

20. W. rhoch F. * a grunt, a groan * ; Br. roe* ha * ronfler ' ; 
Celtic *rokkd, *rogga ; pre-Celtic ^rogh-na. Cognate with Gr. pi* 

po*^\afffi69, 

21. Ir. sluecim {now Blugaim) 'I swallow;' Old -Celtic *sl 
*8luggo; pre-Celtic ^slug-no. Cognate with Gr. Xv^w ( 
\v7ia>), Xvf^r^dvw and N.H.G. aehluckm. 

The nasal in the corresponding British verbs — ^W. lit 
* deglutire,' 'gurgitare'; O.Bret, ro-limcas (gl. guturicauit) — s 
due to a contamination of the regular *lu€h with the equivalc 
the Ir. longud * to eat * (W. llewa * edere,' * manducare *), or o 
Goth. fra-Blindan ' verschlingen.' 

22. Ir. trice (now trie) 'swift'; Old- Celtic *trkki'S, *tr^ 
pre-Celtic *tfgh'ni-s. Cognate with Gr. rpextf' The Ga 
ovep-rpaf^oi' iroBwK€i9 kvuc9, Ir. traig * foot,' and Goth, ^rc 
exhibit a different grade of vowel. 

Other such words are perhaps Ir. /race ' hand,' fraee ' ^ 
fraie ' shield ' ; "W. gwyeh * brave ' ; Br. sac^ha ' s*arreter,' 
point couler,' which last may come from an Old-Celtic verb ^at 
*staggo; pre-Celtic stag-no— L&t. stagno; and Br. atue^hen 'si 
which may be cognate with Skr. tunj 'to push,' as £ng. 
is with Goth, skiuhan ' to shove.' 



PRETONIC Jf IN CELTIC SUFFIXES. 301 

II. dd from -dnd, -dniy -dnd, -dno, 

23. Ir. cit * sheep/ (whence eiten, eetnait * lamb ') ; Old-Celtic 
*ketti'y '"'keddi; pre-Celtic *ked-ni, cognate with ur-Germanic 
hadna *goat'; M.H.G. hatele, with which Ir. eadla *goat* seems 
identical. 

24. Ir. euit (now cuid) * portion, *='W. peth ; Com. peth ; Br. pez ; 
Old-Celtic *qetti, *qeddi'; pre-Celtic ^qed-nk, cognate with Lith. 
kedHi bersten ; Slav, eesti * a part ' (Bezzenberger). 

The Er. pieee^ Ital. pevMy rest immediately on a low-Latin petia^ 
doubtless a Gaulish derivative of qetU-, 

The Gaelic pet^ gen. p$iUy * a piece of land/ common in the 
topography of Scotland, and pit (in terefitf Uth-fit) * a portion 
of food,' are borrowed from some British dialect, Pictish perhaps, 
from which the Icelanders seem to have got their petti, 

25. Ir. fait * cold * ; Old-Celtic *votlij *voddi ; pre-Celtic *t?o^-»i, 
cognate with Ch.Slav. voda ' water,' Goth. vatOf Eng. wat&r, 

mm 

wet. In Lith. vandii and Lat. unda we also find a nasal, which, 
though inserted in the root, may have been originally suffixal. 

26. Ir. gataim (now gadaim) *I take away, steal'; Old-Celtic 
^gattdf ^gaddd ; pre-Celtic ^ghadh-noj cognate with Skr y/gadh (I.E. 
ghadh) ; Vedic gadhia * was zu erbeuten ist ' (Grassmann) ; Gr. 
iciff<T09 (from *x€^<09) ; and Lat. hedera. 

The I.E. root ghady whence Gr. x'^^^^^'^i ^*^' pre-hendoy Goth. 
hi-gitan, is represented in Celtic by the W. genni * contineri, com- 
prehendi, capi.' 

27. Ir. *gett (now gead) *rump,' whence geaddn * buttock,' O'Don. 
Supp. ; Old-Celtic ^getto-, ^geddd- ; pre-Celtic *ghed-n6j cognate 
with Gr. x^?^9 t^^X^^^f X'*^"*'^* > Skr. had * cacare ' ; Zend 
tadhahh * podex.* 

The honorific portion of food called in Old- Irish mil-getan, 
seems to belong here. 

28. Ir. lutu (now Ididin) 'little-finger.' Founded on a base 
*Mtt ex *lMdy *lM-n, Cognate with A.S. Igtel, O.H.G. luzil. 

29. Ir. *rataim (now radaim) * I give,' do-rata ' det,' do-ratus 
* dedi ' ; Old-Celtic *ratt6y raddd ; pre-Celtic ^radk-no, cognate with 
Skr. yjradhy randh * iiberliefem, in die gewalt geben ' (see Schmidt, 
Yocalismus, I. 36). From the unnasalised form of the root comes 
added to the Middle- Welsh dg-rodes, where the ending of the 
8-pret. is the perf. rod. This rod (= O.Welsh *raud) is to the 
pre-Celtic radh-no as Gr. ep-pw<^a is to pjrf-^-fu. 



802 MB. STOKES — ON THE ASSIMILATION OF 

Windisch, Worterb. 499, brings doubtfully Ir. rat in do-ra^ 
etc., from *ro'dad; Skr. daddmi. But in Old-Celtic the v"^ 
corresponding with daddmi would probably have been dO, d0^ 
or didOy and ro-dedQ, ro-didd would have become in Old-Irish, ^ 
ro-dad, but ro-deod, ro-diud, 

30. Ir. sldet (now slaod) *a slide,' LL. 391» 7; Old-Celtic ^slaitf^ 
*slaidd6' ; pre-Celtic ^slaidh-nd-. Cognate are Lith. alidiis * glatt^ 
Lett, alaids 'sloping' (Grundriss, §634); A.S. slidan; N.H.^ 
sehlitten ; Skr. sridh ^ to stumble/ ' to mafce a false step.' 

31. Ir. tot « a wave,' O'CL; Old-Celt. *tutta, *tuddd; pre-Celt^ 
*tud-na. Cognate are Skr. tudami * 1 strike '; Lat. tttdea^ taditard 
Goth, staula; M.H.G. stutze 'stosse,' groundform stud-n (Grundris 
§ 605). 

In Ir. and Welsh tonn * wave ' ; Old-Celtic ^tundd, the root : 
nasalised, as in Lat. tu-n-do, Per-tu-n-da ; Skr. tundate. 

Prof. Bezzenberger prefers to connect tot and tonn with A.S 
jfedtan * heulen,' O.N. ]fj6ta * ertonen ' ; O.H.G. diofutn * tosei 
rauschen, schwellen,' loazzer diezo ^ wasserschnelle.' 

32. Ir. trott (now trod) *a quarrel,' pi. n. trott; Old-Celti 
*truttO'8f *tritdd6-8 ; pre-Celtic ^trud-no-s. Cognate are Lat. trudc 
trudia; Goth. tiS']^riutan * to molest,' \rut»fill * leprous'; A.fc: 
\re6tan (Eng. threaten) ; O.Slav. trudU * trouble.' 

Other Celtic derivatives from yJtruA are Ir. troacaim *I fast 
{*trud'8kd), trosc *a leperj'^romm 'heavy,' * oppressive' {*trud'Smo-s) 
W. trwm 'gravis, tristis, maestus '=Com. trom, Bret, troum. 

To these words we may perhaps add Ir. preit 'champion, 
^rut 'curds,' lott 'whore,' lott 'destruction,' and alatt (=»W 
llath) now slat, ' rod.' 



III. hb from 'hni, 'hnd, 'hn{i, 

33. O.Ir. *bappf *bopp ' bunch,' ' tassel ' ; Mid.Ir. papp, popp 
LU. 97» 3, pi. ace. pupuy LB. 127» ; Highland Gaelic baby M. gen 
baba\ Old-Celtic *bappUf *babb^f *boppU't *bobbii'; pre-Celti< 
^bhabh-nu, *A^oJA-«A-, cognate with Lat. faba, haba, Pruss. babo ; 
O.Slav, bobikf and perhaps Gr. Tro-fi-fpo^f ir€'jLL'(/>i9 from *(l>0'fi'fPo^, 
*(/>€' fi'(pi9. The initial p in Mid.Ir. popp may be due to assimilation, 
as in Mid.Ir. prapad 'the twinkling of an eye '= O.Ir. br(ifad\ 
c6ie 'five,' Old-Celtic *qenqi, pre-Celtic *penqi', and Eng. Bob a 
pet-form of Robert, 



I 



PRETONIC N IN CELTIC 8UFFIXK8. 303 

The Eng. hob 'banch, knob, plummet,' seems either borrowed 
from, or cognate with, O.Ir. hopp, 

34. W. cloff * lame, limping * ; Old-Celtic ^ktoppo-a^ *klohh6-» ; 
pre-Celtic ^kloh^nd-Ut cognate with Gr. KXa^fiof and Lith. klumhas 

* lame, limping.* 

The low-Latin oloppus 'x*^^**'*' Bucange (whence Fr. eloup^ 
clopiner) seems to rest on a Gaulish ^khppo-s, 

35. rap ^ every animal that drags to it, ut sunt sues,' Cormac's 
Glossary ; Old-Celtic *rappo-f *rabb6' ; pre-Celtic rab-ndy cognate 
with rapen ; N.H.G. raffen *to snatch.' Or with Old-Bulg. rahiti 
' secare, rapere * ? 

The above etymology is doubtful, as rap might equally well 
be referred to a pre-Celtic ^rap-nd (see infra, Nos. 56-63), and be 
then connected with Lat. rapio, 

36. Ir. 8eip * hand,' dat. sg. ina dap-scip 'in his good hand,' 
LB. 240»; Old-Celtic ^skeppi-, *8kebbi' ; pre-Celtic *skeb'ni. 
Cognate are Goth, ga-skapjan, £ng. ihape, N.H.G. sehaffen. 

In Ir. eeaptha * shapen,' 'brought into form,' O'Don. Supp., we 
have a part. pret. pass, from a sister-root keb. Hence also O'Brien's 
eib * a hand.' 

Other such words are possibly Ir. eepoo * quire-song,' lajfp 

* slime ' (now labdn^ lathe) ; W. lleihio (Br. lippat) * to lick ' ; Ir. 
opunn (now ohann) ' sudden ' (cf. Gr. uxf>uu)si) ; tappj iopp, Mod. Ir. 
tap, tapaidh * quick, active,' and timpdn * a standing-stone,' formed 
a stem ^Umppo-^ ^embbd-f from *tembh-n6 : cf. Skr. stambha * post,* 
' pillar ' ; Zend gtembana ; Lith. stambas, 

I have thus, I hope, made it probable, if I have not proved, 
that in Old-Celtic, as in primeval Germanic, the initial n of the 
accented n- suffixes was sometimes assimilated to a preceding y, 
dy or h. I will now mention some instances where the same 
n appears to have been assimilated to a preceding k, q, t, or p. 

rV. kk from -kna, -knd-, -knon-j ^-knii-, -qnd-y -qnit, 

37. W. achy F. ' generation,' achen ' lineage ' ; O.W. achmonou 
(gl. inguinibus) ; Ir. aieme * genus,' aicned * natura.' All from 
an Old-Celtic ^akkd, pre-Celtic ^ak-nuy cognate with Skr. anka, 

* flank, lap,' and dknay vyaktMy sam-akna * gebogen ' (Fick', I. 6). 

38. Ir. brocc (now brog) • grief ' ; Old-Celtic *brukko' {^brukka ?) ; 
pre-Celtic ^bruk-ndy *grukn6, cognate with Gr. ppvKw *I bite*; 
Lat. brocchuB, 



304 MIL STOKES — ON THE ASSIMILATION OF 

Ir. hrdn * grief *«=W. hrwyn M. 'a pricking/ 'smarting/ m 
be cognate and come from a paroxyton Old-Celtic ^hrkkno-. 

39. Ir. eace (now eae) ' dung '; W. each M. * fimus '; Com. caught 
Br. ea^ch * merda ' ; Old-Celtic *kakkO' ; pre-Celtic ^kak-nd- o ^ 
*kaq'n6'. Cognate with Skr. gaka and gakrt, gen. gaknds (oxyton!) 
Gr. KaKKff ; Lat. eaeare ; Lith. szikti, 

Ir. eeehair ^ slough ' may be cognate. 

40. Ir. eroean (now crogdn) 'pot'; W. <;r0(;^», founded on an Old- 
Celtic *krokko', pre-Celtic ^krok-nd-y cognate with Gr. xptaaaof* 
from *Kpu)Kio9, 

41. *grdcc (now grdg) *the croaking of a raven,' grdg infra 
No. 65, where it rhymes with ndt ; Old-Celt. *grdkko ; pre-Celtic 
^grdk-wi. Cognate with Lat. grdculus. 

42. leec (now leae) 'flagstone;* W. Il4eh 'tabula saxea'; Old- 
Celtic *lkkd ; pre-Celtic ^plkna. Cognate with Lst.planea, placenta ; 
Gr. wXa^f ir\aK0V9f w\dKivo9 ; N.H.G. flach. 

43. Ir. lecco (now leaea) 'cheek,' gen. leccon. Founded on an 
Old -Celtic ^lekko- from *lek'n6'. Cognate with Pruss. laggnan for 
*lagknan ; O.Slav, lice wpoffwirou (Windisch K.B. viii. 439). 

44. It. mace {now mac) ^Q(m^\ 01d-Celt.*/iiflj'j'o-«(ogmicgen.in«^yi); 
pre-Celtic maq-nd-s. Cognate with Lith. tndku, moke ti '^ Lett, mdzu^ 
mdzit 'konnen' ; as Goth, magus 'boy' (Ir. mug) is cognate with magan. 

The Old- Welsh map points to an Old-Celtic maqo-Sf or else the 
qq has been simplified. 

45. Ir. mucc (now muc) ' pig ' ; W. moch-yn^ pi. moch ; Old-Celtic 
*mukku' ; pre-Celtic ^muk-nu, cognate with Lat, mucuSf e-mungere ; 

46. Ir. sice (now sioee) ' frost ' ; Old- Celtic *sikkUf *siqqu ; pre- 
Celtic ^siq-nii, cognate with Lat. stccare, Skr. sikatd, Zend 
hikush (Bartholomae, K.Z. 29, 525). 

47. O.Bret. *teeham * fugio ' (M.Bret, infin. techet) ; Old-Celtic 
*tekk6 {*teqq6 ?) ; pre-Celtic ^tek-no {Heq-no ?), cognate with Lith. 
tekinas ' laufend, schnell.' So the Ir. techim ' I flee ' is cognate 
with Lith. teku^ tekdti 'to run, to flow,' and Zend toe 'to run, 
hasten, flow.' 

48. Ir. *traccaim ' I press,' in the compound du-thraie • desires,' 
i.e. 'presses after something'; Old-Celtic *trakk6 ; pre-Celtio 
^trak-no, cognate with A.8. ^ringan; O.H.G. dringan; Lith. 
trknkti 'stossen.' 

To this belongs "W. trweh 1. 'fractus,' 'mancus,' 2. 'scissura,* 
' incisio,' if it be not borrowed from Lat. truncus. 



PRETONIC -V IN CELTIC SUFFIXES. 305 

The followiDg words may possibly be explained in like manner : 
Ir. hiccim *I bellow' (W. heichio *mugire'), now beicimy It. cocea 
'boat' (W. cwch\ Ir. eicee * flesh/ Ir. ice * healing' (W. iach\ now 
iotf, and W. taleh * grist,' from *talkko'j *talk-n6-^ which Strachan 
has connected with liuss. tohkno * dried oatmeal.' 



V. tt from tndy -tnd, -tnu, 

49. Ir. hratt (now brat) * mantle ' ; O.Welsh msLj^'hrethinnou 
(gl. cunis) ; Old-Celtic *bratto-8 ; pre-Celtic ^hrath-nd-s, ^grath- 
mJ-*, cognate with Ir. Mit * a strip of cloth,' which Khys brings 
from *^rfln/t =Skr. granthi * a tie, a knot.' 

60. Ir. glota 'belly,' gen.glotan: glotain * bosom,' O'Br. Formed 
on a base gluH-, glut-n, cognate with Gr. 7X0VT05, N.H.G. Kloas^ 
Klotz. 

51. Ir. lat 'foot'; Old-Celtic natta (^lafto-?) ; pre-Celtic 
*plai'na {^plat-nd- ?), cognate with Lat. planta ' sole of the foot ' ; 
Or. irXaim ; Lith. plattis. 

62. Ir. liUiu * porridge,' gen. litten ; Welsh llith M. Derived 
from an Old-Celtic *UtO' ; pre-Celtic *pU-n6', cognate with Lat. 
puh (I.E. plti') and Gr. ttoXto? (I.E. plto-), 

53. Ir. matan 'battle' (Rev. Celt. xiii. 472), Mid. Ir. madan, 
BB. 49^ 18. Derived from an Old-Celtic *mattO' ; pre-Celtic 
*math-n6'f cognate with Gr. fi60o9 and Skr. ^math {mdntkatt\ 
mathnati) ' to crush, hurt, kill.' 

From the same root come the Ir. verb memaid * fregit,' and the 
nouns maided and maidm ' breach.' 

54. Ir. milt * short * ; Old-Celtic *miittu' ; pre-Celtic ^mUt-nu, 
cognate with Lat. mfdiltu, 

55. Ir. ndt 'buttock,' ndd .i. ton, O'Cl., gen. pi. asheir [^in]Jiach 
gohlom grdg \ a[c'] creim ndt ndmat anoeht * the bare-beaked raven 
says grdc, gnawing foemen*s buttocks to-night,' H. 3. 18, p. 73, 
col. 1 ; Old-Celtic *ndtto- ; pre-Celtic *n6tn6f cognate with Gr. 
vuno9f vunop ; Lat. n&tes, 

VI. pp from 'pndf 'pni. 

56. Ir. eapp ' chariot,' * hearse,' * bier,' dat. sg. ructhar do ben i 
capp .i. i carr, ' may thy wife be carried in a hearse ! ' LU. 6^ 30 ; 
Old-Celtic *kappO' ; pre-Celtic ^kap-nd. Cognate with Thess. 
Kairdinff chariot,' perhaps also with Ka7rdv€V9. 

PhU. Trans. 1891^2-3. 20 



306 MR. STOKES— ON THE ASSIMILATION OF 

Gr. Kami * crib ' and Lat. eapulus ' coffin ' may also be cognate. 

57. Ir. eepp * garden,' whence eeppdn and the modem e«apai 
* a piece of ground laid out for tillage/ O'Don. Supp. Ok 
Celtic *keppO' ; pre-Celtic *kep'n6' . Cognate with Gr. K^mo 
Dor. Kairo9 and N.H.G. Mube, though the vocalism does n< 
agree well. 

58. W. (Ta/* *firmu8'; Old-Celt, ^krappo-s; pre-Celtic ^krap-nd- 
Cognate with O.Slav. krepH 'fortis'; O.N. hrafa * ertragen,' kraj 
hraptr *robur* (Ebel, Kuhn's Beitr. ii. 174; Schmidt, Yocalismu 
ii. 72). 

59. Ir. crip ' swift ' ; Old-Celtic *krppi' ; pre-Celtic ^krp-ni 
cognate with the words cited under 58 ; crib * swiftness,' O'Br. 

The Welsh cryf *fortis,' 'gravis*; Corn, crif (gl. fortis); Ere 
creff 'fermus, tenax,' may belong to the same root and descei 
firom a pre-Celtic *krpmo-. For the connexion of the ideas ' swifi 
and 'strong,' cf. N.H.G. schnell^O,^, sn/allr. The Irish cogna' 
may be crimthann ' fox,' also a proper name. 

60. Ir. gopp * mouth, beak, snout ' ; Old- Celtic ^goppos ; pr 
Celtic ^gop-nds: cognate with Zend zafan * mouth' (Bezzenberger 
Now gob, whence, I suppose, the English slang-word gob, Tl 
Fr. gober, gobe* mouche, point to a Gaulish *gopO'. 

61. Ir. *ripaim *I tear,' *rend*; Mid. and Mod. Ir. ribair 
reubaim; Old-Celtic *reipp6; pre-Celtic ^reip-nd. Belated Irij 
words are rep in the compound rSp-gaeth ' a rending wind 
LL. 83^ 15, and the verbal noun repud, LB. 150^ Cognates a 
Gr. i'peiTTw; Lat. ripa; O.N. ri/a *to break,* rifna *rumpi.' 

The Ir. roe * battle '=W. rhae may have lost a p in. inlaut ai 
belong to the same root, with a different grade of vowel. 

62. Ir. ropp ' every animal that gores, ut sunt uaccae,' Cormac 
Glossary; Old-Celtic *rupp0'8; pre-Celtic Vw/>-nd-«, cognate wil 
Lat. rU'tn-po ; Goth, bi-raubdn ; A.S. redfian, reofan ; O.N. rjiifa, 

63. timpdn * a stringed musical instrument.' Formed on a ste 
*teinppO' ' a chord ' ; pre-Celtic ^temp-nd: Cognate with Lit) 
tempiHy timpti * ausdehnen,' temptyva * Sehne.' 

I am bound to admit that none of these etymologies are absolute 
convincing, with, perhaps, the exception of those numbered 4 ai 
15, where the ideal pre-Celtic forms are actually established I 
the oxyton Skr. bhugnd-s and Gr. \a7i/o9. But one of the tests < 
the truth of a theory is the number of phenomena which it explain! 
and tried by this test, I submit that the theory now brought forwai 
is at least deserving of careful consideration. 



PRETONIC y IN CELTIC SUFFIXES. 307 

I confess, also, that I cannot explain why, in modem Irish, we 
have agtUf hog 'soft/ clog, Sig, lag, log^ slugaim^ beside haCy hoe 
' he-goat/ hreacy cnoe, feaCy glioy minic, trie, and hrog, orogduy grdg, 
beside hiicimy eaCy ioc, leaCy leaeay maCy muOy stoc : why we have cuidy 
gadaim^ gead, luidin, radaimy slaody troidy beside hraiy glotaitiy slat: 
and, lastly, why we have hahy cihy erihy goh, ohann, reuhaim, beside 
eeapaehy tap, tapaidh. 

I have, in conclusion, to state that a first draught of this paper 
appeared in Indogermanisehe Fonehungen II. 167-173, and that 
the etymologies numbered respectively 2, 5, 7, 22, 43, 46, and 
49 are due to Professor Strachan. He also pointed out to me that 
the Irish hoee * soft ' (No. 4) might be deduced from *hhugn6- : that 
gataim (No. 26) might be connected with the Skr. y/gadh: that 
hulis and huli were cognate with Ir. holad (No^ 6) ; and that l^tu 
(No. 28) might be cognate with A.S. Igtely O.H.G. luzil, 

Nerviy near Genoa, 

ZOth January y 1893. 



308 



X.— OLD-IRISH GLOSSES ON THE BUCOLICS. 
From a MS. in the Laurentian Library, By Whitley 
Stokes, D.C.L. 

Geo. Thilo (Rheinisches Museum, Neue Folge, XIV. 132-133) 
was the first to call attention to the Old-Irish glosses, which are 
found in the abridgments of Fhilargyrus' scholia on the Bucolics 
contained in two tenth-century codices, one in the Laurentian 
libraiy, marked pint. XLV. cod. 14, ff. 1-22, the other in the 
Bibliotheque Rationale, MS. lat. 7960, ff. 1-17. Thilo printed 
six of these glosses, which were reprinted by Zimmer (Gloss, hib. 
supplementum, p. 5), with the addition of one from the latter 
codex contributed by Herm. Kagen. To these seven, I added, 
in the Academy for Jan. 17, 1891, pp. 64-65, thirty from the 
Paris MS. I now present a complete list of those in the Lauren- 
tian codex. Among them I hare inserted the end of a scholium 
on Eel. III. 90, which tends to show the nationality of the maker 
of the abridgment in which it occurs. The Irish glosses in the 
Paris MS. will be published in the Revue Celtique for July, 1893. 
Both MSS. are in continental handwritings, and the scribes 
were, naturally, ignorant of Irish. Hence sundry mistakes in 
copying the Irish glosses, such as bisecting and trisecting words, 
transposing letters, and sometimes putting c for t; e, o or t for e ; 
h for h'j s for / and / for s or p; m for in ; t for rf, d for 5, and 
even w for il and ic for a. Five of these glosses — viz. (16), (25), 
(47), (65), (104) — are so corrupt that I am unable to explain 
them satisfactorily. But, on the whole, they have been handed 
down in a comprehensible form, and apart from the lexicographical 
value of them all, the gen. sg. mti^^, the nom. pi. dercey the 
accusatives pi. hledmil, englemen^ dareha, yrode, the verbs eit, 
farmutnetharf tuerechaf dodihelf banraffy folloinc, fedid, tmmanaisCf 
adrethf sibrase and adcicJduSy and the participles clithi, nephglidi 
will be welcome to the student of Old- Irish grammar. 

^ Reprinted, with many corrections and additions, from Euhn's Zeitschrift fiir 
Tergl. Sprachforschung, xxxiii. 62-80. 



MR. STOKES : OLD-IRISH GLOSSES ON THE BUCJOLICS. 309 

(1) Fo. 2f7. 10 Da .i. cit (Eel. I. 19). 

(2) „ 3r. 14 Raucee .i. brongidi (Eel. I. 58). 

(3) „ Sv, 9 Dumosa .i. drisidi (Eel. I. 77). 

(4) „ „ 39 Cicadis.^ cauig (Eel. II. 13). 

(5) „ 4v. 8 Uiolas .i. scotha 1. uaccinia (Eel. II. 47). 

(6) ,, ,,11 Anethi. prqpir Iom (Eel. II. 48). 

(7) „ 6v, 10 De Meuio aero nihil rep^i ut Adamnanus' ait 

(Eel. III. 90). 

(8) „ „ 13 Fraga .i. subi (Eel. III. 92). 

(9) „ „ 22 in eruo .i. tincur (Eel. III. 100). 

(10) „ „ 23 fascinat .i. farmuinethar (Eel. III. 103). 

(11) „ 7r. 27 flauescet bUefithir (Eel. IV. 28). 

(12) „ „ 28 Arista broth (Eel. IV. 28). 

(13) f, 8r. 8 mentiri .i. tuereeha fusca enim luna mentitur 

alium colorem (Eel. IV. 42). 

(14) „ 8r. 1 labruseas .i. feadinne (Eel. V. 7). 

(15) „ „ 34 thiasos .i. elasa (Eel. V. 30). 

(16) ,, », 39 auene .i. mailan uilchi uelcu infee uel zezaniaB 

(Eel. V. 37). 

(17) „ „ 40 uioUa .i. seoth uel fobuirge (Eel. V. 38). 

(18) ,, 9r. 1 phallioras (sie!) .i. geelesB 1. aiten (Eel. V. 39). 

(19) „ „ 3 Tumulum .i. fert (Eel. V. 42). 

(20) „ „ 4 superattite (sie !) carm^ .i. sit seriptum sup^ 

tumulo .i. membrsB sup^rseripte (Eel. V. 42), 

(21) „ „ 6 In gramine .i. israth (Eel. V. 46). 

(22) „ „ 14 Intonsi .i. nephglidi 1. intaeti (Eel. V. 63). 

(23) „ „ 20 neetar eetgrinnce' (Eel. V. 71). 

(24) „ „ 28 thimo .i. ^ropir fedo (Eel. V. 77). 

(25) ,f ,f 30 damnabts .i. fisi lasu .i. res faeies^ ut uotam 

tuam multi adorarent (Eel. V. 80). 

(26) „ „ 34 cieuta .i. buinne (Eel. V. 85). 

(27) „ M 35 Pedum .i. bron braehia .i. baeulum ineuruu;?^ 

quo pedes ouium inpediuntur (Eel. V. 88). 

(28) „ „ 37 nodis" obid (Eel. V. 90). 

(29) „ 9». 18 serta .i. eoerea (Eel. VI. 16). 

(30) „ „ 19 ansa .i. dom (Eel. VI. 17). 



1 MS. cicades. 

' Adananus, Thilo, perperam. AdannanuB, P. 

' MS. nectacsB grinnae. 

* MS. facier. 

* MS. nobis. 



310 MR. STOKES : OLD-IRISH GLOSSES ON THE BCOOLI< 



(31) Fo 
(32) 
(33) 
(34) 

(35) 
(36) 
(37) 
(38) 

(39) 
(40) 

(41) 

(42) 
(43) 
(44) 
(45) 
(46) 
(47) 
(48) 
(49) 
(50) 
(51) 
(52) 
(63) 
(54) 
(55) 
(56) 
(67) 
(58) 
(59) 
(60) 
(61) 
(62) 



9t>. 22 
„ 23 

lOr. 4 



99 



n 

99 
99 
99 
99 
99 



99 
9> 
99 
99 
99 
99 



32 



9, 86 
lOr. 7 

,9 7 

99 13 

99 16 



99 19 

.9 35 

99 87 

llr. 3 
4 



5 

8 

23 

28 

38 

„ 39 

llr. 2 

Up. 2 



99 

9 
10 
13 
14 

99 



99 99 

12r. 6 
8 



99 



moris .i. merih (Eel. VI. 22). 
frontew .i. grode (Eel. VI. 22). 
timpora. a aru (Eel. VI. 22). 
nerea .i. bled * mil .i. animalia maris * (£ 

36). 
comua .i. hmn (Eel. VI. 51). 
stabula .i. lesib 1. gelbin (Eel. VI. 60). 
alnos .i. fema (Eel. VI. 63). 
p^rmessi .i. propir fluminis boe[o]tiffi (£ 

64). 
apio .i. luib B6rb (Eel. VI. 68). 
omos' .i. ligna .i. darehachis .i. ealami 

VI. 71-72). 
inguina ' .i. noroefi loei in qtio canes scill< 

bant uel melen* (Eel. VI. 76). 
arguta dresach taeh (Eel. VII. 1). 
arcades .i, sulbari 1. fissidi (Eel. VII. 4). 
deerrauerat .i. todidel (Eel. VII. 7). 
8alu[u]8 .i. slan (Eel. VII. 9). 
examina .i. saithi (Eel. VII. 13). 
seria .i. samre (Eel. VII. 17). 
setosi .i. simch (Eel. VII. 29). 
suras • .i. gairri (Eel. VII. 32). 
rusco .i. ait tun 1. ruse (Eel. VII. 42). 
muscosi .i. coennich (Eel. VII. 45). 
turgent .i. astaid (Eel. VII. 48). 
tede .i. caindla (Eel. VII. 49). 
fuligine .i. osuidi (Eel. VII. 50). 
pampiniis .i. channachdi (Eel. VII. 58). 
populus .i. pro[pir] fedo (Eel. VII. 61). 
fraxinus .i. umnus (Eel. VII. 65). 
pinus .i. oet gag (Eel. VII. 65). 
babies .i. octb gaebasardua. Lailu (Eel. V] 
populus .i. fit (Eel. VII. 66). 
sine .i. leic (Eel. VIII. 12). 
age .i. fer airli .i. cito ueni (Eel. VIII. 1' 



^ MS. animali amaris. 
' MS. omus. 
'MS. inguma. 

* MS. uelme lendulicias, where dalicias is the beginning of a ( 
Dulichias (Eel. VI. 76;. 
^ MS. surras. 



KR. stokes: OLD-IRISH GLOSSES OX THE BUCOLICS. 311 





( 
( 

(s 



:> 



Fo. 12r. 34 
„ 12r. 4 



(SO) 

CO) 

(7X) 

(7-3) 

(r-») 

(T'O) 

(7'S) 

(7-9) 

(»i) 

(852^ 

(»s^ 
(»V\ 
(»^\ 
(8^^ 
(»0< 



» 
t> 

>> 
yi 

>> 
It 
11 
»f 
>> 

t» 
99 
ff 

t9 
>f 
9* 

»» 

99 
f» 
99 



13r 



13(7 



14r. 



99 

99 

23 
37 

39 
5 

99 

6 
19 
23 
31 
12 

99 

14 
15 

18 
24 
25 
28 
3 

8 
13 
18 
20 
20 
28 
29 
34 
37 
38 
40 



cotibfw .i. lecib«« fEcI. VIII. 43). 

electra .i. orget (Eel. VIII. 54). 

ululae .i. coinnil (Eel. VIII. 55). 

cignis .i. elu (Eel. VIII. 55). 

ui[t]ta • snathse (Eel. VIII. 64). 

licia .i. englem qitasi ligia 'per que ligantur 

stamina (Eel. VIII. 74). 
necte .i. immoaaisc .i. cunliga (Eel. VIII. 77). 
bocola .i. bonat (Eel. VIII. 86). 
sera etmaill (Eel. VIII. 88). 
exuias .i. inda fodh (Eel. VIII. 91). 
corripuit .i. adreth (EeL VIII. 105). 
Hylax> .1. conbochuil (Eel. VIII. 108). 
fors .i. toe eth (Eel. IX. 5). 
examinas athi (Eel. IX. 30). 
taxos .i. far .i. arbor (Eel. IX. 30). 
anser .i. gigren (Eel. IX. 36). 
inter argutos olores .i. iter nelu. luin een ehu 

(Eel. IX. 36). 
populus .i. pro[pir] fedo (Eel. IX. 41). 
aprieis .i. elithi (Eel. IX. 49). 
omffta fert .i. folio, inel.* fedid (Eel. IX. 51). 
equor .i. muir (Eel. IX. 57). 
simae ' .i. milberaeh .i. uirgiilta .i. inpr^ naribfM 

(Eel. X. 7). 
pinifer. fin tad birtihd (Eel. X. 14). 
Bubulei .i. mueibi (Eel. X. 19). 
ferulas .i. flesea (Eel. X. 25). 
ebuli .i. pro[pir] fedo (Eel. X. 27). 
bacis .i. eariaib (Eel. X. 27). 
uinitor^ .i. finbondid (Eel. X. 36). 
serta .i. eoberta (Eel. X. 41). 
a aueh (Eel. X. 48). 

mo^tflabor .i. sibrase .i. Bcribam (Eel. X. 51). 
malle ' .i. ma aeubrimse (Eel. X. 53). 
uenabor .i. adcichlus (Eel. X. 56). 



^ ^S. hilM. 
^^^ In the MS. the e joins the barred /, so that at first sight we seem to have 
^^ ^ith barred d. 
Ms. semsB. 
^ Ms. niniator. 
Ms. malles. 



(97) 


tf ft 


(98) 


l> »} 


(99) 


„ 16r. 


(100) 


n i> 



312 MR. STOKES : OLD-IRISH GLOSSES ON THE BUCOLK 

(96) Fo. 14r. 5 cornu .i. ondidbuoc (Ed. X. 59). 

6 spicula .i. fogau 1. gaau (Eel. X. 60). 

10 Hber .i. insnob (Eel. X. 67). 
6 palus .i. cethor (Eel. I. 49). 

11 susurro .i. susurratio 1. amal quod de 

naseitur (Eel. I. 56). 

(101) „ 16v. 8 uac[c]in[i]a .i. uiole purporesB 1. subi 1 

derce ruich (Eel. II. 18). 

(102) „ „ 26 Calathis .i. octh alca ib (Eel. II. 46). 

(103) ,, ,1 29 pruna .i. airni draigin (Eel. II. 53). 

(104) ,, 17r. 25 transuersa tuentibus hircis .i. maiccini 

uerecunde, bircus eni/n lasciau/n anii 
petulcuni)' animal et feruens Bem'per i 
turn. Cuius oculi ob libidinem in transi 
aspiciunt (Eel. III. 8). 

(105) „ „ 40 Stipula .i. cuislen (Eel. III. 27). 

(106) „ 17i;. 3 corymbos .i. brutus (Eel. III. 39). 

(107) „ ,,26 malo me petit .i. cabanrag (Eel. III. 64] 

(108) ,, ISr. 11 fraga .i. subi . quidam tam^n dinmt pomt 

Urmm nata (Eel. 111. 92). 

(109) „ „ 21 in eruo .i. fond orheman (Eel. III. 100). 

(110) ff ISr. 23 cum bacebare .i. bin uel gentts herbe et 

ioeondi (Eel. IV. 19). 
(HI) „ 19r. 17 quid?* .i. eiriee (Eel. V. 9). 

(112) „ „ 18 si .i. adas (Eel. V. 9). 

(113) ,, \9v. 1 auenae' .i. mail molcbi uel cuintbe < 

genus zizaniae^ (Eel. V. 37). 

(114) „ „ „ uiola .i. fobuirge (Eel. V. 38). 

(115) „ „ 2 palliorus .i. gle elge (Eel. V. 39). 

(116) „ „ 9 ealathis .i. catlialcaib (Eel. V. 71). 

(117) „ 20v. 23 baebare .i. boethin (Eel. VII. 27). 

(118) ,f ,f 26 sinum laetis .i. genus uasis .i. bomilge 

VII. 33). 

(119) „ „ 35 ruseo ' .i. aittiun (Eel. VII. 42). 

(120) „ 21r. 17 Licia .i. englemen (Eel. VIII. 74). 

(121) ,, ,y 27 Hylax* .i. horese milebu uel conbocail 

VIII. 107). 



* MS. pecul cwn, * MS. zezame. 

'MS. quod. • MS. ruscor. 

s MS. anime. < MS. hilas. 



r 



( 



MR. STOKES : OLD-IRISH GLOSSES ON THE BUCOLICS. 313 

C ' 33) Fo. 22r. 28 minio .i. uafordin (Eel. X. 27). 
(^^^) „ 22f?. 1 spicula .i. fogu (Eel. X. 60). 
(^^^) ,, „ 4 liber .i. sinob (Eel. X. 67). 



COMMENTABT. 

L. =the Ijaurentian codex. 
P. =the Paris codex. 

cit (gl. da), imper. sg. 2 of a verb cognate with est 
^permission,' which Thumeysen (KZ. XXXII, 671) connects 
"with Lat. eido from *kezdd, Not in P. 
hrongidi (gl. raueae), pi. n. f. of hrongide eogn. with Gr. 
x,^^ ^pa^X^^ * hoarseness,* fipayx^^ 'hoarse.' 

-^ drisidi (gl. dumosa), dat. sg. f. of drtsids, derived from drtsa 

<gl. vepres) Sg. 47* 8, cognate with Welsh drysu * briars, 

brambles,' M. Bret, dresen ''epine." This or a like gloss is 

'^he source of the '^f^risidis,*' which Thilo and Hagen have 

^^iserted in their text of Servius, III. 17, 1. 1, believing, 

apparently, that it is corrupt Greek. 

eauig (gl. cicadis), a scribal error for eailig^ pi. n. of eaiUch= 

^^elsh ceiliog rhedyn ** cicada," ad v. ** gall us filicis." Com. 

helioc * cock ' (Gaul. CalidcoB ?). Cognate with Gr. jcaXco;, 

t. ealars, OHG. haldn ** berufen." 
^cotha (gl. uiolas), pi. ace. of seoth (gl. uiolla) No. (17). 
ence teotae (gl. uiolarium) Sg. 35^ 1. But seoth ordinarily 
^^means a flower of any colour {dere a seotha ** red its flowers," 
^^fe[. 3. 3. p. 59^). Connexion with Lat. seato^ seateo seems 
^^:K)8sible. Or should we compare NHG. schosz, schdszling 
^^nd Fr. ieot * tree-stump,' y/tkut? 

^^jropir loss. (gl. anethi), ** the proper name of a plant." So 
"^ve have propir fedo (24), (56), (80), (88), propir fluminis 
^38). Propir^ pL propiri^ Sg. 32^ 14, is borrowed from 
^be Lat. proprium (nomen), and lo%a is the gen. sg. of lus^ 
^VTelsh lly% ** herba," pi. Uysiau, Com. losow, losowys^ sg. 
, ^sowen, Mer. 1483, Br. lousum, 

^ ^damnanus a Latinisation of the Ir. Adamndn^ a dimin. of 

-^dam, here apparently denoting some Irish grammarian. 

Other such diminutives (which have been connected with 

Xat. nanus) are Flaithndn, Gartndn, Lachindn, Ziathndn, 

iuhndn, Zomndn, Lesendn. 



c 



c^:> 



c^:> 



314 MR. stokes: old-irish glosses on the buoolics. 

(8) 8uhi (gl. fraga), pi. ace. stthi (gl. uacciDia) (101), pi. nom« 
of *sube=:9ubha "berry," O'R. or suib O'Br. Welsh sj^ 
"fraga," sg. ayfien: Bret, sivi ''fraise." 

(9) tineur (gl. pingui ... in ervo). I cannot explain this 
gloss confidently. It seems the dat. sg. of tincar^ LU. 99^ 
30 (a tincor do lind 7 do bfud), which appears to mean 
** a supply," " an abundant provision." gen. sg. tincair 
O'Curry M. and C. III. 499, 1. 17. A homonym is 
in LL. 109* 5 : ni roacht Gull tascud no tineur dind lot 
forsa mbdi. Compare gloss (112). 

(10) farmuinethar (gl. fascinat), better formuinethar, pi. 3 for" 
doh-moinetar (gl. aemulantur vos) Wb. 19<* 27. Welsh 
gorfynu * to rival, to envy.' The gloss adligon which comes 
next in L. 7' 6, is adligar in P. and stands for adligoriee 
=allegorice. 

(11) hUefithir (gl. flauescet), a scribal error for hldfithir sg. 3 
^-fut. of a deponent verb cognate with Ir. ^/a»Lat. flavm, 

(12) broth (gl. arista). O'R. has a gloss broth .i. arbhar " com " : 
also broth ** straw." broth " a straw," 0*Br. Cognate with 
Lat. frutex ? The gloss on Argo, Eel. iv. 34, viz. manaUy 
which comes next in L. 8^ 2, is inonaui in P. and stands for 
nomen nauia. 

(13) tucrecha (gl. mentiri), should be duerseha, pros. ind. sg. 
3 of the verb whence the pi. 3 ducrechat (gl. moliuntur) 
Ml. 80^ 6 (gl. demolentibus), Ml. 85^ 3, duerechubta (gl. 
demolliar) Ml. 126<^ 23, ama/ bid an-duroehneh (gl. quasi 
. . . commentatus) Ml. 68® 11. 

(14) feadinne (gl. labruscas) should apparently be fiad-fini "wild 
vines," a compound oi fiad ** wild" = Welsh gtoydd^ and /i»«, 
borrowed, according to Cormac, from Lat. vinea. 

(15) clasa (gl. thiasos *' Bacchic dances"), elatua P. This seems 
for classa pi. ace. of cUmb (pi. n. classa buana binde, Pel. 
prol. 181 ; clasa aingel oc claiscetul, LB. 111^ 2) borrowed 
from, or cognate with, Lat. claasis, 

(16) mailaXL uilchi uel cuinfea (gl. auene). This gloss is, I fear, 
hopelessly corrupt. In (113) it appears as mail molehi uel 
euintbecha. In the Paris MS. 7^ 8 it is mailam uilhi Leui 
infee, ibid. 15^ 26, it is mailchi molehi 1. cuintbe cha. Per- 
haps the mailan may he^mdeldn **beare, a kind of coarse 
barley," (Vision of MaeConglinne, ed. K. Meyer, p. 186,) 
which may be cognate with Welsh meillion " trefoil, clover." 



f 



MR. stokes: OLD-IRISH GLOSSES ON THE BUCOLICS. 315 

(17) scoth uel fobuirge (gl. uiolla). As to %eoik v. supra (5). 
JFijhuirgey which reoccurs infra Ko. (J 14), is either a sister- 
form of, or a scribal error for, 8ohairge==»ohatrche ** hypericum 
quadrangulum,'' Windisch, Worterb. 8.v. sobrach. Have we 
tbe same word in the imperfect gloss . . . tannica fohirge, 
Zimmer, Gloss. Hib. supplementum, p. 4 ? 
^*^) geelca uel aitan (gl. phalliorus, %,$, paliurus) « gehelcee 
1. arten, P. Here geelca is^gle elge infra (115). It probably 
stands (as Prof. Strachan suggests) for *gel-s€6^ a compound 
of g$l "white" and sei 'Hhom.'' aiten, better aitUn^ dat. 
aittun^ gl. rusco (50), aittiun, gl. rusco (119), is^ Welsh 
eithinf Com. eythinm (gl. ramnus), 0. Bret, ethin (gl. rusco). 
It occurs in the compound aitensliab ** a furzy mountain/' 
H. 3. 3, p. 63^ 8. aiUnn-aliah, LL. 210* last line. 



"^^-^ fwt (gl. tumulum) i^^fert .i. adnacul, Conn. Tr. p. 79: see 
also fert ** grab," Windisch, Worterb. p. 544. pi. n. ferta^ 
.^ LL. 44^ 29. 

^"^ 3 fnemhr(E (gl. tumulo) = tnsamhra, O'Cl., who explains the 

word as meaning "a shrine or case in which are relics." 

See also Windisch, Worterb. s.v. memra, and add Less an m^mra 

"the Fort of the Shrine," Four Masters, a.d. 1143. Borrowed 

.^ :from Lat. memoria^ the grave or shrine of a martyr or saint. 

3 i9r(Uk (gl. in gramine). Here t is^Lat. m, Gr. eV, and 
'Srath, gen. srathay is the dat. sg. of 9rath " the bottom of 
m valley," " fields on the banks of a river," see Windisch 
«.v. The Welsh cognate is gstrad, Old Welsh istrat ** the 
:^at land bordering on a slow stream," Eev. Celt. II. 190. 
^n the gloss '^equiparas .i. coequans .i. accom 7 ablo," 
IL. 9' 8, the last three words stand for accusativo et ablativo, 
«nd mean that aequiparas, in £cl. Y. 48 governs the 
/(^ accusative and ablative. 

^ nephglidi (gl. intonsi). This is the nom. pi. msc. of a 
compound of the negative prefix nsph- or neb- (G. C* 861) 
«nd glide or g lit ha the pret. part. pass, of gelim ' I eat, graze,' 
cogn. with Skr. gilati, Lat. gula. 

cetgrinna (gl. nectar). So in Sg. 122^ 2 eSit grinne fino 
(gl. nectar), and in Ir. Gl. 1045 eitgrindi foilei (gl. nectar). 
The dt is the common prefix meaning '* first," and cognate 
with Welsh eyntaf, Ch.-Slav. cina ; but the grinne is obscure. 
propir fedo (gl. thimo), ** proper name of a tree." As to 
propir see (6) supra; fedo (also in Nos. 56, 80, 88) is the 



C^a^ 



C^4) 



L 



316 MR. STOKES : OLD-IRISH GLOSSES ON THE BUCOLICS. 

gen. 8g. oi fid a stem in «= Welsh gtoydd-en "arbor," O.N. 
rf%r, As. wudUf Eng. wood, OHG. witu. It occurs com- 
pounded in fid'hoec infra No. (96). 

(25) fi^i lu9U (gl. damnabis)~/tMt liau, P. This gloss is certainly 
corrupt. The -tUf no doubt, is the pronominal suffix of 
sg. 2, the 'lu may be miscopied for -hi ; but the rest of the 
gloss is obscure. 

(26) huinne (gl. cicuta) i&^huiniM (gl. tibia), Wb. 13.^ 

(27) hron hrachin (gl. pedum )=irom hraehin, P. We should 
probably read hronhaehali cf. Hie est Oingussius cuius cog- 
nomentum Bronbachal, Yita Columbae, p. 21* of the 
Schaffhausen codex, gen. sg. mors Oengusa Branhachlae, 
Annals of Ulster 648. Here bachal is a loan from Lat. 
haculus or bacillus (see Thumeysen Kelto-rom. 38, 39), but 
hron (hrdn?) is obscure. Can it be for brog-no- cognate 
with ppoxo^ ? 

(28) obid (gl. nodis), leg. oidb^ pi. n. of odh M. (gl. obex) MS. 
lat. 11, 411, fo. 125^, pi. ace. udbu, Windisch, Worterb. 
s.v. odb.= Welsh o<?r(/***tuberculum." 

(29) coerca (gl. serta)=tfoA^^a, gl. serta, infra (91), pi. of *coheri, 
a compound of co- and sert (=Lat. sertum\ which occurs 
in the Laws, 1, 12, line 13, deilb rig roda sluagaib sert 
saigid.^ 

(30) dorn (gl. ansa)=(for« M. "fist, hand," Windisch's Worterb. 
Welsh dwrn pugnus, pugillum, Bret, dorn ** main." Lett. 
rfur<f "fist." 

(31) merih (gl. moris), a scribal error for tMribf as it is in P., 
pi. dat. of *mw= Welsh tMr in mer-wydd " mulberry- trees." 
pi. n. mera derga, caera glassa, LL. 117.^ Compare Gr. 
fijopov " the black mulberry." 

(32) grode (gl. frontem) is either the ace. pi. of grwtd " cheek," 
= W. gruddf a neut. stem in 8 (KZ. XXIX. 379), or else 
a derivative therefrom. 

(33) a aru (gl. timpora, U0. tempera) seems a scribal error for 
aracha, the ace. pi. of aire * temple,' gen. arach. 

(34) bled-mil .i. animalia maris (gl. Nerea). The glossographer 
took Nerem to be equivalent to Nerei filiuB " a seamonster." 
See Windisch, Worterb. s.v. bledmall, and add examples of 
the Mid. Ir. ace. sg. and pi. ar in mbledmil muride, IAj, 

1 Sic in MS. The edition has saighit. 



MR. stokes: old-ieish glosses on the bucjolics. 317 

217^ I. norannad bledmila in mara, LL. 108* 25. A com- 
pound of ^/^(^= Welsh hlaidd *' lupus," and mil pi. ace. of 
»ii/N.= Welsh mil **be8tia," cognate with Gr. /i^Xoi/. 
C^S^ benn (gl. comua), ace. pi. of benn horn, peak, which is 
DOW fern., but in Old and Early Middle Irish seems to 
have been a neut. stem in u, dat. benrif Ooid.' 93, gen. benna, 
Fiacc h. 29, pi. dat. hennaib, LL. 251^ 45. The ace. pi. 
for cethora benna, LU. 58* proves that before the twelfth 
century benn had passed over to the fem. d-declension. 
Welsh, Com. ban ' horn, peak ' shews a different grade of 
vowel. 

>^^^) ksib uel ffelbin (gl. stabula). The context is Perducant 
aliquae stabula ad Gortynia uaccae, and the pi. dat. lesib 
may be intended to translate, not ** stabula," but ** stabula 
ad." The nom. sg. is less M. " enclosure," " garth," gen. 
lisSf dat. Ittus, ace. less, pi. ace. lissu, Windisch, Worterb. 
8.V. Welsh llys ** aula, curia, palatium," Bret. lez. As to 
the possible connexion with Fr. liee^ lisikre, see Diez s.v. 
liceia. The etymology of the an-. \e^, gelhin is obscure to me. 

^ ^^^ fema (gl. alnos), ace. pi. of fern, gen. f&ma ** alder " = 
Welsh gwern^ sg. gwemen " alnus," Gaul. Verno-dubrum^ 

. Fr. verne, 

^3 prapir fluminis " the proper name of a river" : see above (6). 
3 luib Serb (gl. apio). Here luib is the dat. sg. of luib, 
gen. lubae, Sg. 65*, 61», Old- Welsh lu in luird (gl. horti) 
«Ir. lub-goirt, Com. lu in lu-orth, Bret. /* in Hotm, and 
^erh " bitter " = Welsh ekwerto, M. Bret, hueru, urkelt. ^svervo-Sf 
which has been connected by Brugmann with NHG. sauer 
±rom suer^ «4r. This gloss is followed by the word herina, 
B. corruption of the Lat. harina, here used to denote the 

,^ demean a^tvp, where poets were crowned with parsley. 

■> omos .i. ligna .i. darchachis .i. calamis. Here two glosses 
(one Irish, the other Latin) have been run together. Head : 
«mos .i. ligna .i. darcha . chis .i. calamis. Here darcha is 
the ace. pi. of the Irish e- stem dair " oak," and '* chis " is 
:fbr the Lat. his (Eel. YI. 72), as miehi^ vecho in Irish latinity 
. «re for mihi, veho. 

J melen (gl. inguina) is a copyist's error for *mleen or mlin 
«=Mid-Ir. bUn ** groin," die weichen, ace. bleoin, blein, 
"Windisch, Worterb. s.v. This supports Strachan's etymology, 
mUn from *mlaknd, cognate with Gr. fiaXaKo^. 



318 MR. stokes: old-irish glosses on the bucolics. 

(42) dresachtach (gl. arguta), better dresachtaeh, as in LU. 106** 
32, derived from dreaaoht '^ ein knarrendes oder quietschendes 
gerausch" (Windisch), LU. 112» 29, and this (according to 
Strachan) from drens- : cf. Goth, drungus 0^07709, ON. 
dri/nr ** roaring," Skr. dhran {dhranati gabde). 

(43) iulhari uel JUsidi (gl. arcades). Here nUhari is the nom. 
pi. of sulbair "gut sprechend, beredt" (Windisch s.v.)= 
Welsh Ay /fl/br, abret. helahar (gl. grecus), urkelt. *8U'lahari'S, 
(where su- is = the Skr. laudatory prefix su-), and JUsidi is 
the nom. pi. of JUsid^JUid, fistith (gl. sophista, gl. catus) 
8g. 16^ 52*, root vid. The pi. n. of JUsid occurs, with 
the / infected,, in centtded issidi * though ye are knowing,' 
Turin palimpsest. A cognate adverb, ind-fissid (gl. scite) 
occurs in Ml. 68® 9. 

(44) todidel (gl. deerraverat), better dodihel as in P. Cognate 
with do-m-ar-aill mihi u^nit, ad-n-ellat quo transeunt, do-in- 
aid-lihea uisitabit eos, fo-n-ind-Ua (gl. euagari), do-n-ella ut 
deuiet, and other verbs collected by Ascoli, Gloss, pal. hib. 
pp. Ivi. Ivii. A Welsh cognate is elaf **ibo," a Latin 
amb'ulare. 

(45) slan (gl. saluus), better sldn (gl. salvus) Wb. S^ 4, from 
an oxyton or paroxyton ^saldno-s^ cognate with Lat. %al-vu-9, 

(46) saithi (gl. examina), Bathi (76), pi. nom. of saithe M. " a 
crowd," "a swarm" (of bees, etc.). 80 in Ml. 90^ 7 in Uathi 
(gl. examina), and see Windisch, Worterb, s.v. saithe. Welsh 
haid ** examen," Bret, het, urkelt. ^satjo-s, 

(47) samre (gl. seria, ace. pi.). This gloss is obscure and pro- 
bably corrupt. 

(48) timch (gl. setosi) is badly corrupted by the copyist. I 
conjecture that it is a misreading of finnich gen. sg. msc. 
oi finnech "hairy," and —finnigh in Corrcenn mac Faitheman 
finnigh, H. 3. 3, p. 59^. 80 the nom. pi. msc. finnich or 
finnig (tri bruitt finnig, LL. 266^ 13), findech, LL. 266^; 
findach, Vision of MacGonglinne. The cognate noun is ^nn, 
a stem in 11, whence the gen. pi. inna finna (gl. pilorum) 
Ml. 72^ 16. Cf. Lat. villus from *vinlus. 

(49) gairri (gl. suras), ace. pi. of gairr ** the calf of the leg,*' 
dat. sg. ina gairr ^ LL. 289* 8, where it is used as synonymous 
with colptha. This is the Irish reflex of Welsh garr ** poples/' 
M. Bret, garr **jambe, tibia," with which Diez connect^ 
the Span, garra and Fr. jarrsL 



I 



MR. stokes: OLD-IRISH GLOSSES OX THE BUCOLICS. 319 



C50) atttun vel rwe (gl. rusco). As to aittun see above (18), 
This ruse (leg. riuc) seems borrowed from, or cognate with, 
Lat. ruscum "butcher's broom." Another rftsc ( = Welsh 
rhing. Corn, rise) means ** bark," **a vessel made of bark," 
and is connected by Diez with Fr. ruehe. See also Thumey- 
sen, Eeltoromanisches 111. 
C^ 1 ) coennich (gl. muscosi), pi. nom. msc. of eoennach " mossy." 
In modern Irish caonnach is '* moss " and caonnachamhuil 
" mossy " : coenna (gl. putamen), Pr. Cr. 33^, seems cognate. 

C ^ ^ ) astaid (gl. turgent), a mistake for, or a sister-form of, attaid^ 
altaitf pres. ind. pi. 3 of attaim " tnrgeo," a denominative 
from aU 'a swelling,' gen. fri m6t in atta, LB. 142,^ a 
masc. or neut. stem in u from a pre-Celtic *azduy cogn. 
with AS. dsi (gl. nodus). 

^ "^) eaindla (gl. tede, i.$, taedae), pi. nom. of eaindel F. borrowed 
from Lat. eandela, gen. sg. co cnocc na caindle, LL. 304^, 
dat. sg. gabait in coire din chaindil * they take the caldron 
away from the torch,' ibid. A compound, rfg-chaindell, occurs, 
LL. 301% where the double / may be due to the length 

. of the preceding vowel. 

^ ^-^ suidi (gl. fuligine), d, ua «* from, by," (96), (122) »uidi 
dat. sg. of suide, a fem. stem in ik cognate with Fr. tuie, 
AS. adtf Eng. soot, or with Gr. avohiri * ashes, scoria.' Now, 
. corruptly, iiigha, suithche, siiithe, 

^^ channachdi (gl. pampineis)'=&Affn(7cAJt P., pi. ace. fem. of ean^ 
nachde, a derivative of eannach, which O'H. explains by 
** sweet-willow, myrtle " : eannac[K\ a canna .i. fid bis il-laim, 
H. 2. 16, col. 95. 
propir fodo (gl. populus) "the proper name of a tree." 

^ See above (6). 

J umnus (gl. fraxinus), miscopied by the scribe for unnius or 
MinniuB F. ind huinnim ardd (gl. alta fraxinus) Sg. 67*. 
Bence the modem uinaeann. Zeuss compares the Welsh 
€nn, onnen. If the nn is from m cf. also Lat. omm from 




C^%) 



*o*iniM. 



octgag (gl. pinus). This is certainly =o(?A<a<?^ F. gl. crann 
giuis, £k. of Lecan 149^ 1 : the kingpost of a house. 
Conn. It may stand for ^l^p'luktakd and be cognate with 

^ OKQ^.fiuhta, fichte, Gr. ttcvkij, Lith. puszis, 

^ ^) oethgaeha 8ardua, Lailu (gl. hs\yLeB)=octhgacha %ardiM lailu, P. 
This is obviously miscopied for ochtach as ardu alailiu '^ a fir 



320 MR. STOKES : OLD-IRISH GLOSSES ON THE BUCOLICS. 

that is higher than (the*) other," the ahiet being taller than 
the pinus. Here alailiu is the instl. 8g. of alaile, 

(60) fit (gl. populus), leg. fid ** a, tree," gen. fedo supra (24), 
(66), and infra (83), (91). 

(61) leie (gl. sine), better leie, imperat. sg. 2 of leicim from 
*leinkffid, a mixture of Ui'i^ and linq (Strachan), cognate 
with Lat. linquo, 

(62) fer airli (gl. age). Here fer seems the imperat. sg. 2 of 
feraim 'I give,* and airli is the ace. sg. airle F. "counsel." 

(63) lecihxi^ (gl. cotibus), leg. lecih^ the dat. pi. of lece F.= Welsh 
lUck ** lapis, scandula, tabula saxea." From *\^p]Ukkuy 
vorkelt. *[/?]f^«4, cognate with Lat. planea, Gr. srXaf, vXa- 
jcci/09, Lett, plakt flach werden. 

(64) ory^< (gl. electra), a mistake for 6rarget " gold-silver," 
(the Paris MS. has orar get). The Irish glossographer took 
electra to mean the mixed metal resembling amber {fKem-pov) 
in colour, which Vergil mentions in Aen. 8, 402. 

(65) eoinnil (gl. ululse) I have not met elsewhere and cannot 
explain. 

(66) elu (gl. cignis) seems a mistake for elaih, pi. dat. of ela 
(gl. cygnus) Ir. Gl. 609, now eala^ sg. ace. mar hela ir-richt 
aiogil gil ' like a swan in the shape of a white angel,' Salt, 
na Bann, 1671, pi. ace. nelu — leg. helu — (gl. olores) infra (82). 
The Welsh alareh, and Com. elerhc (gl. olor 1. cignus) seem 
cognate. Various non-Celtic words have been compared, 
Lat. o/br. Or. eXco, or (assuming the regular loss of p)^ 
Or. iri\eiay Lat. palumba, and Pruss. poalis ^* taube." 

(67) snatha (gl. uitta), hetter endt he (gl. filum) Sg. 54*, dat. sndthiu 
Sg. 64», Welsh y-enoden^ Com. snod (gl. uitta). Windisch 
connects vrj^ and other words in Curtius G.E.*, Ko. 436. 

(68) englem (gl. Maidi)^ englemen (gl. licia) infra (120), is the ace. 
pi. of *tf»y/afm=0'Clery's eanglaim .i. inneach "woof." The 
compound mdr-englaim occurs in Dalian's address to Cerball's 
sword, LL. 47* 51, where englaim seems used metaphorically 
to signify the woof of war. Its etymology is obscure to me. 

(69) immonaise (gl. necte), imperat. sg. 2 of the verb naeeim 
compounded with the preps, imm-fo (GC 883). Naseim 
from ^nadhskd, cognate with the Skr. y/nah (Idg. nedh)^ 
umbr. neeimei ** proximo," Ir. nessam, W. nesaf. Com. nessa^ 
etc. Bmgmann, Umbr. u. Osk. 236, Idg. Forschungen I. 
176. See also Thumeysen, Kelto-romanisches 38. 



MR, STOKES : OLD-IRISH GLOSSES ON THE BUCOLICS. 321 

C^O) honat (gl. bocola), leg. hdnat (gl. bucula), a diminutive of 

h6 ** cow," like siurnat (gl. sororcula), clethnat (gl. tigillum), 

columnat (gl. columella). The same ending occurs in the 

double diminutives fracnatan .i. cailHn, and the proper names 

^anhnatan, Becnatan, Corenatan, B6 (= Welsh huw, 0. Bret. 

h<m) occurs in composition infra in con-ho-chuil, boo-bethin and 

ho-milgB. Cognate with Gr. /3oi)9, Umbr. bu-m ** bovem," 

Skr. gdus, etc. 

C'l) eimaiU (gl. serae) seems miscopied for etermaill dat. sg. 

fem. of *etermall " valde lentus, tardus.*' From the 

intensive prefix eter- (also in eterciany etarmolad^ ettorsande^ 

etar&arad) and mall from *marlo-8 cognate with Ir. maraim, 

X*at. morUy tnoror. 

^* ^) tnda fodh (gl. exuuias), leg. inda fodb "ends of vestures." 

Here inda is the pi. ace. of ind N. ' end,* * head,' gen. 

dat. indf and fodb is the gen. pi. of fodb N. ** vesture," 

** something stript off," ** the spoils of a vanquished foe," 

pi. ace. fodba (gl. manubeus) Ml. 92^ 2. This gloss is (as 

Prof. Strachan has pointed out) a close parallel to the 

Greek original of Eel. VIII. 91, viz. Tovt' avo ra^ 'xXaiua^ 

to Kpaairehov tv\€<T€ A€\0<9, Theocr. 11. 53. As it is 

improbable that the Irish glossator had read Theocritus, 

'We may perhaps conjecture that the ancient Irish, as well 

as the Sicilian Greeks, believed that a lost love might be 

recovered by burning only the ends or hem of the beloved's 

dress. Italian magic, if we may trust Vergil (Eel. viii. 91, 

Aen. IV. 492-497, 507), required the whole of his or her 

raiment. 

adrtth (gl. corripuit)=arfrtf^A, P., for ad-raith perf. sg. 3 of 
a verb compounded of ad- and 'rath^ ''cum variis praefixis, 
prehendere, comprehendere, continere, includere," Ascoli, 
Lex. pal. hib. p. clxxxix. 

eonhochuil (gl. Hylax, " Barker ") ** a watch-dog," eonboeail 
infra {I2i) ^conbuachaiHf Laws I, p. 126, 1. 8, a compound 
of the stem of e^ *' hound " = ici;ft;j/, and bd-chail (= Welsh 
hugaily Com., Bret, bttgel) a compound of bd " cow" (73) and 
eail cognate with the -jco\o9 of fiovK6\o9, the ^0X09 of 
ai7ro\o9. The triple compound eon-bd-chail reminds one of 

:^2^^ ^ce^^ (gl- fors) =^toehet P. tocad, Ml. 35^ 22, VTelsh tynghed, 
^^ athi (gl. examinas), leg. aathi (gl. examina). See above (49). 

nu. Imu. 1891-2-3. 21 



322 MR. STOKES : OLD-IRISH GLOSSES ON THE BUCOLICS. 

(77) .i. fwr (gl. taxos). Here the gloss is meant for " citiso 
(f.tf. cythiso) in the next following line of the eclogue. Ft 
leg. /(^, gen. feuir^ 8g. 68^ is= Welsh gwair. 

(78) gigren (gl. dca%eT)^g%grem P., giugrann (gl. anser) Sg. 64 
a masc. o- stem, pi. nom. giugraind, LL. 297^ 45, gen. el 
giu grand, LL, 265* cognate with Welsh gw grain ** chynelop 
chenalopex, viilpanser, anas scotica." Seems to he a rednp] 
cated suhst., ^gi-gur'annos, 

(79) iter nelu. htin een ehu (gl. inter argutos olore8)=t^ ne 
luineeeUf P. Here iter is^Lat. inter: nelu is a scrihal err* 
for helu ace. pi. of hela * swan ' (66) ; and luincechu is ace. p 
m. of an adj. derived from luinniuc "song/' still living : 
the Highlands as luinneag. 

(80) propir fedo (gl. populus) " the proper name of a tree.*' S< 
above (6). 

(81) clithi (gl. apricis), elit thi, P. the pi. nom. of a participle pre 
pass, from y/kl^ cognate with Lat. ealeoy Lith. szglu, szilti 
Strachan suggests that elithugud 'cherishing/ LL. 160^ 4! 
may be compared. The ' ab ' which follows this gloss stanc 
for ablativus : ** apricis in collibus," Eel. ix. 49. 

(82) folio, inch fedid (gl. omnia fert). Bead folloine 1. fedii 
Here folloine is for folloing 3 sg. pres. ind. act. of folangi 
with an assimilated infixed pron. fedid is the same teni 
and person of fedaim ** fero," no-feidtis (gl. efferebantu: 
Ml. 54^ 12 ; and 1. is the common sign for Lat. vel and Ir. n 

(83) muir (gl. equor), Welsh mor^ Gaul, mori, Lat. mare^ wit 
which Hirt has recently connected Gr. fipv^ ace. fipvxa froi 
*mrogh, 

(84) milberach (gl. simae "flat-nosed") should probably be *mdi 
heraeha "blunt-pointed": i»at7= Welsh moel calvus, glabe: 
and heraeha pi. nom fem. of *herach a derivative of hir- 
Welsh, Com., Bret, her^ Lat. (g)verUf Umbr. beru ** spit, 
and see Bezz. Beitr. XVI. 239. 

(85) y?» tad hirtihd (gl. pinifer) ^ finit adhir thio^ P., shoul 
obviously be pintadbirthidf where pin (pin crann^ O'R.) i 
borrowed from Lat. pinm (so Welsh pinwgdd, Con 
pinhren)f and tadbirthid is cognate with the verb taidbrii 
exhibeo, the enclitic form of ^do-aith-berim, 

(86) mimibi (gl. subulci), leg. mucidi (or muecaidi, LU. 93*), p. 
nom. of mucidy muccaid ** swine-herd," derived from mui 
** pig *' = Welsh, Bret, mock, A compound, rig^muecaid, occur 



fc 



MR. stokes: OLD-IRISH GLOSSES ON THE BUCOLICS. 323 

in LL. 169^ 8. MucCf Welsh moeh, from *mukku^ *muknu, 
cognate with fivxr^py etc. 

(87) Jleaca (gl. ferulas), pi. ace. of JImo *' rod." Flese 
(Gaul. *vliskd^ Fr. JUche) has been deduced from *vlk%d 
cognate with Skr. vrkshd "tree," but seems more probably 
from ^vlt'Skdf cognate with Ir. folt "hair/* Welsh gwallt, 
Gr. aX(709, Ags. weald f KHG. wald. 

(88) propir fedo (gl. ebuli) " the proper name of a tree." See 
above (6) and (24). 

(89) cariaih (gl. bacis), ciraiby Paris MS. leg. ediraih (gl. baccis). 
Here ediraih is the pi. dat. of cdir or cder (gl. bacca) Sg. 22^, 
Welsh cair : compounded in caer-thann mountain-ash, rowan. 

(90) finhondid (gl. uiniator) = finhondiOf P. leg. finhondid (gl. 
uinitor), a compound of fin ** vine " and hondid for ^hongid^ 
noraen agentis to the verb hung or hongim " I break, reap," 
cognate with Skr. hhanj^ hhanajmi. The verbal noun of 
hung, hongim is hiuiinf whence the denominative huanaigim 
mfin-huanaigit (gl. uindimiant) Ml. 102» 12. For the change 
in hongid of n^ to nd^ compare cindis, LL. 86^ 8, 13, for 
eingiSf cmdsiu, LL. 82^ 44=^cingsiu LU. 63* 12, gland and 
glang^ Corm., and see Zimmer, KZ. 30, 63. 

(91) coherta (gl. 8erta)=coerta, P. See above No. (29). 

(92) awh (gl. a), leg. uch^ wehe! vae, Z.' 750. 

(93) sihrase (gl. modulabor) is a redupl. fut. sg. I from V aver. 
The -%$ is the pronominal suffix of sg. I. The Bihra (formed 
like gegna) is from ^svi-sverdm, just as ar-heittet, carm. 8. 
Paul, is from ^are-sveizdionti (Thumeysen, KZ. 32, 670). 
Cognate are Skr. avar, svdrati ** rauschen, besingen," Ch. 
Slav, svirati pfeifen, AS. svarian ** to speak." 

(94) ma acuhrimse (gl. malle). The gloss, if it be wholly Irish, 
means ** if I desire." But I suspect that the ma is a relic 
of the Latin malOf written by the original scribe over malle 
to show the source of this infinitive. If so, a[d^cuhrim'Se 
is a gloss on malo, ajid^ adcohraim (gl. uolo) Sg. 146^, pi. 
3 adcohrat Ml. 89^, 16, pret. sg. 1 nicon ru accohruSf Mi. 
136^ 7, sg. 3 ad-ro'chabair, Trip. Life 202. 

(95) adcichlm (gl. uenabor), the redupl. «-future, sg. 1 of ^adcladim 
whence an adcladat (gl. aucupantesl Ml. 11 2^, 2, «-fut. sg. 
2 adclaissy Trip. Life, p. 88, 1. 28. The verbal noun is 
aelaidj ibid. 1. 25. A cognate verb, ar-claid^ occurs in LU. 
122^, 36 (i tig fir arclaid iasc im-muir ethiar iV) the house 



324 MR. STOKES : OLD-IRISH GLOSSES ON THE BUCX)IJC8. 

of a man who catches fi%h in a sea of ether). The Greek 
KXahaaaofiai ** I rush Tiolently " may perhaps be cognate. 

(96) otididhuoe (gl. comu), ondiohuoe, P., leg. 6nd Jidhucc, Here 
6nd is a compound of the prep. 6y ua (=Skr. dva), see infra 
(122) and the article, and fidhuce (the / being regularly 
inflected after 6nd) is the dat. sg. of fdhoee (gl. arcus) Sg. 
107^ 2. The word literally means "arcus ligneus," being 
compounded of fid supra Ko. (24), and hoee from an Old- 
Celtic *hukk6'f pre-Celtic *hhuff-n6', Skr. hhugnd-y cognate 
with AS. hoga^ OHG. hogoy NHG. Bogen, 

(97) fogau 1. gaau (gl. spicula). Here gaau is the ace. pi. of 
gae Sg. 37^ {gae gona eise gl. fuscina), which seeros cognate 
with Gaul, gaeson, Gr. x^^**^ ^^ x^^^^ ** ^ shepherd's staff," 
and Skr. heshaa ** geschoss." Fogau is the ace. pi. of fogue 
compounded of gae and the diminutival prefix /o-=£r3ro-. 
So in Old-Breton guu-goiou (gl. spiculis .i. telis), Bezz. Beitr. 
XVII. 139. 

(98) tnsnob (gl. liber). Here in is the masc. article and snob 
(corruptly sinoh infra, No. 124) means the inner bark or 
rind of a tree. It glosses suber **cork" in Sg. 64* 10. 
The etymology is obscure. Perhaps it is cognate with Skr. 
8ndy mdyati umwinden, bekleiden. The OHG. muaha *' vitta/' 
has also been compared. 

(99) eethor (gl. palus) = eetor. P., leg. cechor, a sister-form of 
0*Clery*s ceachair .i. lathach ** a slough." See Metrical 
GhssarieBy Bezz. Beitr. XIX. p. 54. Cognate perhaps with 
Gr. Koirpo^ and Skr. qdkan, 

(100) amal (gl. susurro) i^^amar in in-amar, LU. 128» 19, 
amhar .i. ceol ** music'' O'Cl. amor LL. 19* 1 (ba binnithir 
teta mendcrott guth 7 amor each duine inna flaith tweet as 
lute-strings were the voice and song of every one in his reign'). 
Another instance of final I for r is Macculugily Bk. of Armagh 
4^ 2 = macculugir, ibid. 4^ 1. 

(101) suhi vel certe derce ruich (gl. uaccinia). As to suhi v. supra 
at (8). Leree \_f'\ruieh is=inna darcae froich (gl. uaccinia) 
Sg. 49* 10 ; where frdich is the gen. sg. of frdech, urkelt. 
vroiko'^)y cognate with Welsh grug, M. Com. grig, Gr. 
ipetKTj from i'fpeiKTff and derce is the nom. pi. of dere 
** berry," a neuter stem in s, now fem., possibly cognate 

> As to Ir. bru and Bret, brukj brug^ see Thurneyten, Kelto-romanisches 94. 



MR. stokes: OLD-IRISH GLOSSES ON THE BUCOLICS. 325 

with Skr. drdJcshd " vine,** '* grape," an ertveiterung of 
*derkeSf as drakshyami from y/derk to see. The inflection 
of the / of fruich is due to the fact that in Irish, as in 
Greek {fieuea from /lepeaa)^ and Latin {genera from geneea), 
the nom. pi. of neuter stems in # originally ended in a vowel. 

(102) octh alea ih (gl. calathis) = or^Aa /(?flf J, P., is certainly =» 
cathaleaih (gl. calathis) (116). Perhaps the word meant 
is eothalcaih nom. sg. *vothale from ^kutalkd, Etym. 
obscure. It might be cognate with Gr. KOTvXrjy or with 
Lat. scutula, scutella. For the suffix cf. Gaulish Bodalca, 
etc., G.C.* 808, and Welsh madalch fungus. 

(103) aimi draigin (gl. pruna). Here airni is the ace. pi. 
of airne {airnne gl. glandula, Sg. 49^) =■ Welsh einn-eit 
** prunum," M. Bret, irin-enn ** prunellum, prunellus," Cath., 
and draigin is the gen. sg. of draigen ** sloetree," blackthorn 
= Welsh draen spinus, spina, sentis, Bret, dren " epine/* 
which Ebel connected with repxvo^, rpex^oi ** a twig." Cf. 
smera is dime dubdroigin, Silva G. 102. 

(104) maiccini disse uerecunde (gl. transuersa tuentibus hircis) 
=maiccinudis se verecunde et, P. An Irish maiec (or maiccini, 
a diminutive of mace " filius," " puer " ?) and a Latin 
cifiaedis may perhaps be elicited from this corrupt gloss. 

(105) ouislen (gl. stipula), should be euisUn, a diminutive of 
cuisle 'Wein, pipe," anglicised cushla in the phrase octMhla 
or euislenn (Mid. Ind. Mod. Ir. cuisle), whence cuisUnnach 
* piper.' Etym. obscure. 

(106) hrutus (gl. corymbos) is probably a scribal error for Gr. 
/3oT/»v9, just as the gloss glaus (gl. sandix) L. 18^ 39, which 
Hagen and Zimmer^ supposed to be Irish, is a scribal error 
for Gr. r^Xav^. 

(107) eahanrag (gl. male me petit). This should perhaps be 
€0 han-rag ** that I may come with a woman." Here co 
is a conjunction meaning **ut," ** donee," and *han-rag a 
compound of 3fl»=»Aeol. pava, and ^rag, a conjunctive sg. 1 
of the verb whence raga-i, do-rega " veniam." 

(108) suhi (gl. fraga). See above, No. (8). 

(109) /onrf orbeman (gl. in ervo). Ad v. "fundus haeredis": 
fond borrowed from Lat. fundtis, and orheman gen. sg. of 
orhem, pi. n. horpamin, Wb. 2° 14. 

^ Glossarum Hibernicarum Supplementum, p. 5. 



326 MR. STOKES : OLD-IRISH GLOSSES ON THE BUCOLICS. 

(110) bin (gl. bacchare). (117) hoethin (gl. baohare). These 
glosses are respectively Bo ob ethin and boethin in the Fari» 
MS. They seem compounded of b6 **cow," and bethin {beikiu?) 
which may be cognate with Lat. bita, whence KHG. Beete^ 
Eng. best. 

(HI) ciriee (gl. quid ?)'^eurice, P., read cirict, and compare or 
ciric (gl. quid enim ?), ciariee, cericc (gl. quid ergo ?), G. C* 
355. 

(112) ados (gl. si) is =^ad<M, adaa» << although," O. 0} 489, tda9 
da da-gnio (gl. si autem quod nolo illud fiacio) Wb. 3^. 

(113) See (16). 

(114) fobuirge (gl. uiala). See (17). 

(115) See (18). 

(116) See (102). 

(117) See (110). 

(118) bomilge .i. genus oasis (gl. sinnm lactis). The Irish gloss 
seems only on ** lactis ** and to mean " of cow's milk," 
b6-melg, gen. bdmitge, a compound of b6 supra, No. (7Q) and 
tnelg .i. as ** milk," Corm. Tr. p. 107, and s.v. 64^ p. 127, 
just as bdmlachtf Corm. Tr. 20, h=b6-\'mlicht. As melg is 
neuter {cuirm .i. melg n-etha, Egerton 1782, cited in Corm. 
Tr. 107), and its gen. sg. is milge, it must be a stem in «, 
to be added to those in KZ. 27, 292 and 29, 379. Add 
also dere (101). 

(119) See (18). 

(120) See (68). 

(121) horca, milchu uel eonbocail (gl. Kylax). Here we hare 
three words for **dog": horca^orce "lap-dog," Corm. sv. 
Mugeime. Orce co nemib 7 epthib fondiset for beraib eairthind 
"a dog with poisons and charms which they cooked (?) on spits 
of rowan," LL. 120» 12. In milchu (not milchu !) ** grey- 
hound," ** hunting-dog," the mil is = the mil of milrad 
" hunting," dat. milruth, LL. 272* 23. As to eonbocail (leg. 
conbdchuil) v. supra (77). 

(122) ua fordmn (gl. minio). Here ua is=d supra (54), (96), 
and fordinn is a compound of the intensive prefix /or-, and 
dinn dat. sg. of denn ** colour," O'Clery's deann .i. If no dath. 
See the compounds gle-denn, Pelire, Feb. 1 and 13 and dend- 
gorm, LL. 277* 32. 

(123) See (97). 

(124) See (98). 



UR. STOKES : OLD-IRISH GLOSSES ON THE BUCOLICS. 327 



INDEX. 

(The numbers refer to the glosses.) 



Adamn&n, 7. 

adcubrim, 94. 

adas, 112. 

ad-cichlua, 95. 

ad-reth, 73. 

airle, 62. 

aime, 103. 

aittenn, 18, 50, 119. 

alaile, 59. 

amal (=amar), 100. 

ara, 33. 

ard, compar. ardu, 59. 

arget, 64. 

as which iSf 59. 

astaid, 52. 

att, 52. 

ban-rag, 107. 

benn, 35. 

berach, 84. 

bl&, bMthir, 1. 

bled-mil, 34. 

b6-bethin, 110, 117. 

bocc * bow,' 96. 

b6€hail, buachail, 74, 121. 

bo-melg, 118. 

b6nat, 70. 

bundid, 90. 

bron-bachal (?), 27. 

brongide, 2. 

broth, 12. 

brutus for $6rpvs, 106. 

buinne, 26. 

cailech * cicada,* 4. 
caindel, 53. 
c&ir, 89. 
canachde, 55. 
cechor, 99. 
cet-grinne, 23. 
dricc, HI. 
cit, 1. 
class, 15. 
clithe, 81. 
CO, 107. 



coennach, 51. 

coerta, coherta, 29, 91. 

coinnilP 65. 

con-b6-chuil, 74, 121. 

cothalc, 102, 116. 

c(i, 74, 121. 

cuinfec? cuintbechF 16, 113. 

cuislenn, 105. 

dair pi. ace. darcha, 40. 
denn, 122. 
derc * berry,' 101. 
do-di-hel, 44. 
dom, 30. 
draigen, 103. 
dr^sachtach, 42. 
driaide, 3. 

ela, hela, 68, 79. 
englem, 68, 120. 
etermall, 71. 

far-mninethar, 10. 

feadinne, 14. 

fedaim, fedid, 82. 

feraim, fer, 62. 

f^r, 77. 

fern, 37. 

fert, 19. 

M 60, gen. feda, 24, 56, 80, 88. 

fid-bocc, 96. 

fin-bondid, 90. 

fine, 14. 

finnech, 48. 

fisilusu (P), 25. 

fissid, 43. 

flesc, 87. 

fobuirge, 17, 114. 

fodb, 72. 

fo-gae, 97, 123. 

folangim, fo-1-loine, 82. 

fond, 109. 

for -denn, 122. 

froech, 101. 

-^ ^ •= *-* . 






328 MR. STOKES : OLD-IRISH GLOSSES ON THE BUCOLICS. 



gae, 97, 123. 
gairr, 49. 
gelbin (P), 36. 
gel-8ce (?), 115. 
gigrenn, 78. 
glauft f or 7X0^, 106. 
glithe, 22. 
griDDe, 23. 
gniad, 32, 

immo-iiaisc, 69. 
i-n prep., 21. 
in < the/ 98. 
md*end,* 72. 
iter prep.y 79. 

lecc, 63. 

leicim, l^c^ 61. 

less, 36. 

luib, 39. 

luincech, luinniuc, 79. 

lii», gen. losa, 6. 

ma, 94. 

mace, maiocine {?), 104. 

mriiian, 16, 113. 

m^-berach, 84. 

maU, 71. 

melg, 118. 

membrae, 20. 

mer 'midbenry,* 31. 

mil, 34. 

mil-diu, 121. 

mien, 41. 

muccid, 86. 

muir, 83. 

nascim, 69. 

neph -glide, 22. 
neesam, 69. 



6 prep. 54, 96, 122. 
ochtach, 68, 59. 
odb, 28. 

6nd » from the,' 96. 
6r-arget, 64. 
orbem, 109. 
orcae, 121. 

pln-tadbirthid, 85. 

propir, 6, 24, 38, 56, 80, 88. 

-rag, 107. 
ruse, 50. 

saithe, 46, 76. 

Bamre (?), 47. 

see, 115. 

scoth, 5, 17. 

•se, 93, 94. 

Serb, 39. 

sibra, 93. 

Bl&n, 45. 

Bn&the, 67. 

snob, 98, 124. 

srath, 21. 

-su, 25. 

Buib (subeP), 8, 101, 108. 

Buide *Boot,' 64. 

Bulbair, 43. 

tadbirthid, 85. 
tincor, 9. 
toceth, 75. 
tu-crecha, 13. 

ua, prep. 122. 
iicb, 92. 

uilche (?), 16, 113. 
uimiius, 57. 



Xl.— SOME GREEK ETYMOLOGIES. By E. R. 

Wharton, M.A. 

(1) The modem theory that the * prothetic,' or, as the Greek 

grammarians called it (Curtius, Gnindziige,' 720), 'prosthetic,' 

Vowel is in most cases really the first vowel of an originally 

bi-vocalic root can scarcely be rejected (a) where other languages 

Ijesides Greek have a similar vowel, as in ^e/3o9 cpevyofiai o\o<pv9 

besides Armenian erek orcam o\b (Persson, Wurzelerweiterung, 

p. 246, n.), or (J) where two forms can be best explained by 

starting from a bi-vocalic root, e.^, av^w Sk. vaksh- from aveks- 

in a{F)€^w, avpa^ Sk. vd- from ave- in u{F)rjfii, Lat. unguis 

Sk. nakhas from onokhv- in oi/vf : though why one language 

dropt the second vowel and another the first we do not in the 

least know. But in some cases such an explanation is impossible, 

and the word must be regarded as a compound. 

A. Latin in-cdnui in-elutm in-columis and I think in-vilus are 
but emphatic forms of cdmis elutus columis and * vitus * forced ' : 
Un enim saepe augendi causa adicimus,' says Festus. This tn- 
may be identified with the Preposition in meaning ' upon ' (quite 
a different word from in meaning * in,* which goes with cV), Greek 
ava in dva aic^Trrpt^ (=in sceptro). The original meaning was 

* up * and so * upon * (cf . German auf in both these senses), while 
in the Latin Adjectives given above and the Greek Adjectives 
to be given below we have a transition of signification from 

* up ' to 'in a high degree, quite.' The form in the Ursprache 
would be n, represented in Latin by in-, in Greek by au- before 
a vowel,* iJ- before a consonant, as in the following words : 

a'fi\ffXpo9 * weak, gentle,' beside fiXrfxpo^, The termination, 
which appears also in ^BeXv-xpo^ it/eXt^xpo^ jrevi-xp^^t must go 

> In the only place in which it occurs in Homer, Od. 6. 469, aCpri means the 
-mnming breeze, ^w9i irp6 ; and with it in this sense Buttmann rightly connects 
nfiptov 'in the morning, to-morrow.* So in my <£t}ina GrsDca* I have ex- 
plained Homer's ^«if, Lesbian aiws. as from &vo-, an Ablant of ave- in icmu : to 
« people dwelling near the Mediterranean the morning breeze would be the 
iiatoral herald of dawn. The Attic form tns takes its aspiration and accent 
^rom liXios. 

' In some dialects before a consonant also, Horn, hf * up,* Theocr, 6fi'fufiya' 
^KofUy^ (in which the vucalism shows the presence uf a sonant). 



330 SOME GREEK ETYMOLOGIES. 

with xpota x/»«^» 'surface, skin, colour,* so that /8\i;-x/»o« means 
' weak-looking, weakly ' : the root is mle- (not mla-, siuce pXtf- 
X/»o9 occurs in Doric), Sk. mid- 'to wither/ with Ahlaut mlo- 
in Irish hldith 'smooth, soft,' and male- in MeXeov 'useless.' — 
Quite a different word is /3Xof ' slack,' in which the a must 
be due to contraction (Kretschmer E.Z. 31, 295), or we should 
have *pkfj^i as i^eaf or i^ei/f comes from i^eo?, so I would explain 
/SXaf as for */3\a/'if or */3Xo/^iyf, from a simpler form *)9X«/'o«, 
mlVos, Lat. moUis for ^molvis. The root appears in Gothic 
ga-malvjan ' to crush,' and Eng. mellow, 

a'6€<r(/>aro9 * marvellous,' beside $€(r(/>aro9 * divine * : literally 
' struck {i.e. made) by a god,' cf . irp6a'(f>aro9 * made in addition, 
new,' and ^/.0oto« (Hesychius) Si-tpaaio^ * made double.' Ihe 
root of this -(paro9 is ghvn-, ghven-, in Oeivto ' * strike,' /wX^- 
</>aro9 * struck by the millstone,' and, with a transfer of meaning, 
0OJ/O9 * slaughter,' Aprii-(/>aro9 * slain in battle.' — The first element 
of $€<r-(f)aro9 appears with a 'determinative' (Brugmann, 
Qrundriss, 2. 60) in ^09, i.e. *^<ro9 : which, however, can hardly 
go (as Fick thinks it may, Worterbuch* 1. 469) with Lithuanian 
dwas^s ' spirit,' Middle High Qerman ge4wds ' ghost,' for the 
Greek gods were by no means spiritual beings. 

So we have d'fiavp69 ' dim,' beside fjuivp69 (Photius) ; and in 
II. 24. 753 the two readings afiixOakoeaffau and fii\$a\o€trtTav^ a 
word of unknown derivation (the connexion with 6^Tx^**yt Hoffmann, 
Bezz. Beitr. 15. 84, is absurd). So I would explain the d- as 
intensive in 

d-'y€pwxo9j *y€pd-oxo9 ' holding privileges.' 

a-{F)jav\o9 * wicked,' beside Sk. vdttdas ' mad.* On this 
dialectic change of tv to ov see Classical Review 6. 259 : ] 
connect aKo-ffvByrf 'goddess of the sea wave* with Irish tone 
' wave,' d(rv(f>rf\ov (below) with Ti/0X<>f, ffvx^o9 ' long, numerous ' 
with rvx^v ' ordinary,* as a Litotes for * considerable.' Thus alone 
can we fairly explain ^aam and Lat. densus {t.e, ♦dent-tos) beside 
Albanian dent ' to make thick.' 

€L-Kpo9 ' at the top ' (it never means * sharp,* and therefore 
cannot go with aic/9, Lat. acuO), beside (Pa\a'Kp69 ' white-headed ' ' 

^ As Lat. ferio means both * strike' and 'cheat,' and Kpowris hoih 'striking' 
and * cheating* (Ar. Nub. 317), so with 0c{y» I would connect 0fVff{ * cheat'; a 
Doric word, like K6fiaKos ' rogue,' as the a. for 17, shows, with a dialectic ^ for 6 
as in ^t6s for 0(6s (G. Meyer, Griech. Gramm.' 211). 

> The first element is bhl.|L-, cf. hh|-n- in 4>a\\6s * white ' (Hesychius) and 
Lat. fullo * clothes-cleaner.' ^ 



I 



SOME GREEK ETYMOLOGIES. 331 

(Schulze, Quaestiones Epicae 464), from a by-form of Kapa, On 
these by-forms see Danielsson's Grammatische imd Etymologische 
Stadien pp. 1-56, Johansson K.Z. 30. 347-350, Johannes Schmidt's 
Pluralbildungen der Indogermanischen !N*eutra pp. 363-379. From 
the same root as xapd we may deduce (a) Kotpo^ {i,e, *Kapjoi) 
* thrum,* end or top of the thread ; (Jb) xaprov icpdro9 * headship, 
power,' quite a different word from Kparai-Xeto^ 'rocky,' Gothic 
hardus 'hard' (with which we may put xeprofia 'hard words'); 
{c) K\ypo9 ' lot,' a Dissimilation for *Kpapo9j cf. Arcadian Kpapiunai 
and Khodian *H\o'Kpdp7f9 (Meister, Griechischen Dialekte 2. 104, 
G. Meyer 160), the word thus meaning 'head' as sign of in- 
dividuality, going with uav'K\jpo9 vav'Kpapo9 'householder' (what- 
ever the first element of these words may be) and Hesychius* 
Kpatpa 'head'; {d) Kpatirakri 'headache,' for * Kpaai'irakri ' a fight 
in one's head,' the first element being Locative of *ic^a», while 
from another form ^Kpa-naXtj is borrowed Lat. erapula : 

urfi€ivu)v ' better,' from fievo9 * strength : ' 

a-aK7f0^9 * safe,' quasi ' supported, cared for ' (cf. with Active 
meaning. triceOpov 'careful'); a Dissimilation for ♦a-<rx»y^7«, from 
the root of trxyfiay extv (in ax^Oelu the x ^ retained through the 
analogy of <rx€<i') : 

d-<r7r€px€9 ' hastily,' from tnrepx^ * hasten ' : 

u'<rr€fM/(f>fj9 ' stiff,' from a root meaning • to be hard,' whence also 
trT€fi(pv\a ' pressed grapes,' Sk. Hambh- ' to restrain, hold fast ' : 

d-trv(j)vfKo9 (with Aeolic accent) * insulting,' *rv(/>rf\d9 (see above 
on dj(rv\o9) going with rv<f>\d9 ' bHnd,' 7v(pw&tj9 ' dull,' Tv<po9 
' conceit * : 

d-rapr7fpd9 'baneful,' *TapTJ a Subst. from *Ta^To« Part, of 
Tctpeiu ' to distress ' : 

d'T€U7f9 ' strained,' Lat. in-tentui, from rcipof : 

a-T/»v7€T09 ' swelling,' Lat, turgidus, from a root tver-g- (Friihde 
B.B. 14. 107), an extension of the root tver- (see on aavpwrqp 
sec. 5, and, on the pvj pv^to, p. 11) : 

a-v<rTa\€09 'squalid,' Od. 19. 327, cf. Sk. pwA-'to dry,' Lat. 
su-du8 * dry ' ; from *a-<n;<rro\€09 as uvirpo9 from ^aav7rvo9. 

So, with dv for p- before a vowel, I would explain dv'dehvo9 
' quite dowerless,* dv-de\irro9 * quite unlooked for,' and Hesiod^s 

dp'dirv€varo9 ' quite breathless,' Suidas' dp'd'^(vu}(rT09 di/'d'jrrat(rT09 

(J. Schmidt K.Z. 23. 273). 

B. The same intensive particle appears in several Yerbs : 
Lat. incitd infringe ingemind innovd intremd etc. beside citd frangO 



332 SOME GREEK ETYMOLOGIES. 

etc., Greek aaxapt^M aavalpw 'palpitate ' beside OKapi^w awaip 

aa<f>apa^€u} 'clank' (Theocr. 17. 94) beside <r(/>apay€Ofiai * burs 
and d'Kovu) ' hear ' beside Ko{F)€to ' perceive.' So 

a{f)€i^tt) * sing ' beside oiSa : i.e, aetBu) used absolutely mea 
' make my meaning known,' used with an Ace. ' make know 
celebrate.' In the original signification *know' we have (KofPmai 
B.B. 15. 62) Cyprian tUde 'hear' and the common word cutrOavoft 
(i.e, *a'Fi^ OapopMi) 'perceive': for the transition of meanii 
from ' know ' to ' make known ' cf. Itrropia ' knowledge ' (to el^^vi 
or ' narrative ' (to etSevai ttoicTu), and r^if>/i/w<rKto * know ' besi< 
t^viijpi^ie ' make known ' : 

a-Xi^u) ' heed ' beside Lat. ^Ieg6 in dlligd inUllign neglegd (qui 
a different word, as the Perfects show, from leg6 ' gather *). Fro 
a Neuter Subst. *«I\€709 comes aker^eivo^ 'demanding cautio 
troublesome ' ; which, with Ablaut, and without the intensive c 
appears in Hesychius' Xa'^eiva' h^ipa. From this *ttX.€7o« cor 
further {a) dutr-rfXe^y^ ' cruelly troublesome,' Homeric epith 
of war and death; {b) raurjXe'^j^ 'intensely troublesome,' us< 
of death, with the derived sense of ravvu) 'stretch,' as in tl 
Homeric use with epiBa^ fiaxfjt^t wovov, * to intensify * the strii 
etc. ; (<?) air-ffX€^€W9 * most carefully,' the airo- heightening tl 
meaning : 

d'X€i(/>u) ' smear,* cf. Lat. di-Uhutus ' besmeared ' : 

d-XvicTatw d'Xvtrtru) * am in distress,' going with Xvaaa (t 
^XvK-ja)^ which in Homer means ' martial rage,' the spirit 
a wolf, Xvico?: in II. 16. 156 and 352 warriors are compar 

to wolves, and Theocr. 4. 11 "n-eiaai rot MiXtov xat rw^ Xvko9 axrri 

Xvaaijv shows that the Greeks themselves connected Xvo-o-a wi 
Xt;iro9. With these words go fMpfio-XvrreffOai * to madden as 
hobgoblin, uop^u),^ would,' and fiopfioXvKeTov 'bugbear,* literal 
* hobgoblin wolf ' : 

d'fiaXSwa> ' destroy,' from ♦/4aX6i;?, Sk. mrd^ * soft.* 

afieX^to, cf. Lat. mulged. 

C. In three other Verbs we have the ' copulative ' a- of dOpo 

airav dirXoot, dialectically d' in dS€X<p69 uKoirt^ dKoXovOo^ uXo-^ 

drdXamov ; representing 81JI-, a/j.-a, Lat. aimulf Sk. 8am ' with ' (a£ 
Preposition) or, in compounds, ' together.* This appears as a- 

^ With fwpfi^ go fiSpfwi ' panics * (Hesychios) and I think fitpfitpos * m 
cbievous/ fi4pfiffpa * trouble/ and fwpfAvpwy in II. 18. 402 p6os *(iKtcafo7o k^ 
liopti{tp»Vy rightly explained by Hesychius as rotpd(r<ra»y ' making an uproar * : 
has nothing to do with Lat. murmur^ which would give far too weak a meanii 



SOME GREEK ETYMOLOGIES. 333 

(a) a-fiaprapto 'fail' (the root, on which see Neisser B.B. 19. 120 aq,, 
may perhaps be found in Lat. mora), with which cf. Hesychius' 

afiapetv' a/iapraveiv and Homer's rjfipporov ; as a- in (6) d'fietfito 

* exchange ' beside Lat. mlgrd * remove ' and Old Slavonic miglivH 
'mobile/ and {c) a-fievofiai ' surpass' ('change places with') beside 
Lat. moved. In these Verbs the copulative prefix, like the (of 
course unrelated) Lat. com- in eommaculd eomminud concM oonvelld 
etc., merely * gives intensity to the signification of the simple 
word,' Lewis and Short s.y. cum. With the same force it appears 
in 0-/10X709, which Eustathius says was 'Achaean' for dic/ii; 'prime' 
(as Hesiod Op. 588 uses dfioX^aiif of a ' prime ' cake, /la^a) : I 
would connect the word with mlg- in Lettish milst ' to swell/ 
Lat. mtdttu for *mulctus (Wiedemann B.B. 13. 303 sq.), so that it 
will mean ' swelling, climax,' and wtcrbi afiok'^t^ will be Cicero's 
multd node ' late at night.' 
A copulative, though not an intensive, d- seems to appear in 
d'oaffTfT^p ' helper,' which I would explain as * one who hears a 
divine voice {f6ff<ra, as 6a<ra may everywhere be read in Homer, 
L. Meyer K.Z. 28. 90 : the root appears in Lat. rocd), and in 
obedience to it goes to help his comrades.* Hesychius has also the 
form oaoffTtjpaf referring, according to Moriz Schmidt, to II. 15. 254 
Toiop roi (^d)oa(T7frijpa Kpoviivp cf I^i^9 7rpodijK€f in which case the 
word must necessarily have had a digamma : 

J-<F/3oXo9 ' soot,' quasi ' thrown together, collected,' from fidWto, 
The afi is the same dialectic representative of gj which Tick 
B.B. 17. 323 finds in (jyepeapto^ < life-giving ' and apivpvfii 
'quench' (Goth, qistjan 'to destroy'): and which I find, before 
other vowels than € or <, in (a) dftxfil-apaiva 'a serpent that can 
go either way ' and dfjupi-fffirfria) * dispute,' both from fialvw ; 
and {h) <f)\oiirfio9 ' din,' the clash produced by the impact of one 
heavy body on another, from a root bhleigv-, cf. (pxffiw OXipia 
'rub,' Lat./?^5 'strike,' Welsh W>/ ' catapult.' — Another dialectic 
representative of gv was f, Eur. Phoen. 45 iTre^dpet^iTrcfidpei, 

Hesychius fe\\€«i' ^epeOpa ^elvafiev— pdWeiv pdpaOpa apivwiiev. 
80 I think in {a) ^0X7 ' spray ' from fidXXw ; (b) l^aXo^ * jealousy ' ^ 
beside Lith. geld * pain,' Old High German qudla * torment ' ; 
(c) ^d\l^ 'sea,' and Homer's i7ri-^d(p€Xo9 'stormy,' from fidirrw, 
the sea ' dipping ' the ships, cf. Eur. Orest. 706 pav9 . . . ipayfr^v 



> The & here is Ablaut of i as in KSp6s beside Lat. cera, pAkw beside Old 
High Qerman magOf see Johansson B.B. 16. 306 »q. 



834 SOME GREEK ETTMOLOOIES. 

* the ship sank,' Old Norse kaf * a dive, the deep ' ; (rf) faiw fcZ'w 
quasi ' walk,' and BiXrf^ot ^tjretD quasi * go about/ all from fiatvw : 
(e) poi^o9 'whistling' beside pot^-ho^ (for the termination cf. 

pap-ho^). 

D. Beside the copulative 8i]}-, Greek a- or a-, there appears a 
parallel form so-, Sk. «»-, Greek o-. We have it in Hesychius' 
6'^a<rru)p o^vye? oOpoo9 o^vXov (Schulze 495), Homer's oTrmpo^ and 
II. 2. 765 oTpixa9 o<€T€09 (tho latter, despite Schulze, a miswriting 
for *ov6T€a9, i e, *o-/^€T€a9) ; and I think in 6 8om, Armenian 
a-tamn, each a Singular formed out of a Plural signifying 'the 
united teeth, the rows of teeth,' as perhaps atm^p uarpov Arm. astX 
are Singulars formed out of Plurals signifying *the whole body 
of stars,' the initial vowel in each word being a copulative prefix. 

E. In two important words the initial c- or e- seems reduplica- 
tive, standing for se- : 

e<rO\d9 * brave, good ' =*<r€-<rr\o?, from the root of areWo) * set 
in order,' the meaning thus being * ready, settled, steadfast.' The 
Greeks found a difficulty in pronouncing the combination a-rX, which 
occurs in no old word (<rT\€77i9 * scraper ' appears first in Hippo- 
crates, o<rrX«7f * curl ' in Attic) : they preferred either ( 1 ) to 
aspirate the t, Attic pavtrOXow * convey by sea * beside vavtrroXetv, 
Hom. IfiaffOXfj * whip ' with the same termination as cxeVXiy, Att. 
fiaaOXfj^ * leather ' from the same root as fia<m^ * whip,' and so, I 
think, iffOXo^; or (2) to drop the 9, rXcytyi^ — or the t, Sappho 
fjLaaXrfi * leather,' Doric and Lesbian 6VX09 and Arcadian e<i-Xo9 
(which last must represent *t<rrXo9, not *e<rOX69, or it would have a 
smooth breathing)— or both letters, Att. vauXov * fare ' beside 
Hesychius* pau(r$Xou ; or (3) to change the X to /i, (rTpefyyii ; or 
(4) to insert a vowel, o<rraX«7f (as M. Schmidt reads oardXat^ in 
Hesychius), areXeyr^U. — The same <rT€ XXa; appears, I would suggest, 
in o</>0aX^d9 from ^oir-araXfio^ 'arrangement for seeing,' the first 
element going with o^fia owwira oyfti^ : ^oyjrraXfio^ became 6<pOaXfi69 
as ♦tYrT09, the proper Participle of e^iu, became e(/)069: 

ktrrta * hearth, altar ' (in Homer only in the compounds dvetmo^ 
€0€Vt*o9) ^*tr€'(rTia from a root sti- * stone,' whence <rr?oi/ * pebble ' 
and I think vepl-<ma 'lustration of the Ecclesia' by carrying a 
victim round the altar, and, with Ablaut, Goth. Bta%n% ' stone ' 
and Old Slavonic dlna * stone wall.' On the parallel form Ifrriii 
see next section. 



SOME GREKK ETYMOLOGIES. 335 

(2) The greatest difficulties in Greek vocalism are connected with 
the occasional appearance of i where we should have expected e 
(G. Meyer 67-60, 29; Kretschmer K.Z. 31. 375 «^.), and of v where 
we should have expected o (G. Meyer 61, 62 ; Schulze p. 495 tq.). 

A. (a) Some of the instances quoted of i for e rest on doubtful 
or impossible etymologies : 

alyiXtylr, epithet of wct/m^ in Homer and Aeschylus (Suppl. 794), 
is of unknown meaning (Monro), and can have nothing to do 

with \€Tra9 * rock.' Hesychius' X/^r* irerpa o0' fJ9 vhu)p trrd^ei may 
go with Xetfiw. 

irfr/ia' eU and trrov' ev, quoted by Hesychius as Cretan, are 
too obscure to be deduced from a root sem- : M. Schmidt suspects 
both glosses. 

iX\a9 *rope,' i\X69 'squinting,' tWivoreiWto 'wrap up,' iWofiai 
' move to and fro ' can hardly have anything to do with Lat. voho, 

i7r7ro9 cannot go with Lat. equus, Sk. dgvas : the aspiration is 
quite abnormal, and the Homeric form ought at least to be *iicico9, 
cf. ireXcKKou from *w€\eK'fov, The form ikk99 in the Etymologicum 
Magnum may be Lat. equtu borrowed (with i from tinro9), as 
dxici7ryai09 in Athenaeus is Lat. aquipenais borrowed. 

Kiffffo9 'ivy' cannot go with Lat. hedera from a root ghvedh-, 
or we should have *x^^^^^' 

\iKpi(JH9 'sideways,' with a strange termination, goes with Lat. 
Itcinus * with upturned horns ' and obllquus {i.e. *ob-lic-vos) 
* crooked ' ; not with \expio9f which stands for *X€f-^-io9 and 
goes with Xofo9 (De Saussure, M6m. Soc. Ling. 7. 91, n.). 

X0i^o9 is not directly from x^«'»> *•*• S^jw* ^^^ ^o™ ^Ji 
Ablaut gly*; with sonant sibilant, Thumeysen K.Z. 30. 352; 
the termination is 4J0') cognate with Lat. di€8, 

(y3) In if^in^j i^w iOpi9 itrriri the initial vowel is reduplicative, 
as in iicTi9 {ic7ih€09) lx^v9 (Arm. jukn) tirrafiai {wijofiai) iffOi (Zend 
%dl *be thou') as opposed to ix0€9 eypivxa €<mjKa: in such cases 
it would be absurd to talk of a change from e to i. So in vtatrofiai 
Tiicrto the i is reduplicative, as in nO^mj ^iSwfii rirvaKeaOai beside 

rdravo9 BeStoxa rervKeaOai, Thus 

I'f^vvri * hollow of the knee ' is reduplicated from the root of 

i^u)^=*i'(rB'ju) from the ' reduced ' root of c^ofiai, i6o9, Lat. aeded ; 
and so iSpvw (the « is short) =:*i-<r5-^-va> : 

* l<rT6s, properly Participle of TCa» ' Bet up,' is used as a subst., * mast, loom.' 



336 SOME GREEK ETYMOLOGIES. 

lOpi^* ffTraBttiv and dOpi^' rofjua^ ( 80 M. Scbmidt reads for TaX/wo<?) 
in Hesychius may both go with Opl^w * cut off ' : 

i'trrTrf is the Homeric form of earia, see above. The Homeric 
compoands of e<rrm, and the evidence of Boeotian, Cretan, Hera- 
clean, Locrian, and Sicilian inscriptions (Meister 1. 46), prove that 
neither form ever had an initial F : the Arcadian proper name 
Vistias can have nothing to do with larTff^ and is as obscure as 
the name Yiapaiao^ in the same inscription (Meister 2. 103). 

vttr<rofJLai=i^PL'pa-jofiaif from the reduced root of v€{ff)ofiat, uotrro^ : 

riicrtv=s*ri-TK'TtVj from the reduced root of rexetv, cf. Jtcuriv 
* brother ' from *Tic-n-T«9. 

o 

(7) We have i as Ablaut of je in 8k. vie- vidh- beside ryae- * to 
extend * vyadh- * to pierce,* and so I think in vwep-ncraipotn-o * they 
sped on ' (Od. 23. 3) beside Sk. pra-yaksh" * to press on.' The 
relation of ixrcpo^ ' jaundice ' (for the termination Ha vet, Mem. 
80c. Ling, 4. 230, compares va-repa, rfatr-r^p) to jekv-, Ist, j'ecttr, 
is not quite clear : we should have expected *tVT€^09. 

(^) Homer^S xtpvijfit imviffii vlXvafiai (^trJKiBvafiat beside icipaatra 
wiTatraa iriXaaaa (Txihaaa (^cKedaffaa), Pindar's Kpifiutffu ^ vlrva 
beside Kpifiaaap W€Toi<raty Attic opif^vao/JMi beside ope^ofiaiy owe 
their 1 to the analogy of reduplicated Verbs, they are formed 
after r^t-yvofiai p,i'^vwy^ as Homer's oKipraw (root sksr-, cf. 
aicalptti) is formed after rt-Krw, So Pedersen Idg. Forschungen 
2. 293 says ' the i of trKidvif/u is due to the influence of tanffu 
7 10 fj fill irlfivXfjfii etc. 

Homer's iritrvpe^ beside Lesbian veavpev must owe its i to the 
preceding numeral, Tpta rpi^ rpiT09. 80 (Baunack K.Z. 25. 225 «^., 
Brugmann Grundriss 2. 165 8q,) oicra- in compounds owes its 
-a- to eirra-f Heraclean otcrtv and Elean otttw draw from lirra 
their breathing and labial respectively, 07^009 borrows its consonants 
from €/35o/i09. 

(c) There is no clear proof of any confusion in the Ionic- Attic 
dialect between e and i : the Old- Attic forms Aiuearai — Klviarai^ 
AifXearai — AvXiarat only show different ways of resolving the 
diphthong ei before a vowel, Delian trrXir^'^U is an Assimilation (see 
J. Schmidt K.Z. 32. 321 sq.) for arXeyf^Uf Mipdattou is a very late 
form for the earlier Mei/Battvv on the coins of Mende in Palleoe. 



1 In the MSS. almost always written k^ii^i^iu (Eretschmer K.Z. 31. 375), see 
Aesch. Theb. 229. Eur. Here. Fur. 620. 

' Homer's tripvutii kept its € through the influence of its cognates ircp^ * sell ' 
and Wfn}y, IL 24. 751 ir4py€urx*t ^y^^^* lAff(r«rc, -wiirriv h\6t. 



SOME GREEK ETYMOLOGIES. 337 

But in the non-Ionic dialects the letters interchange so often that 
we can only explain the instances hy supposing that in those 
dialects c was pronounced * close,* half-way to an * open ' «, and so 
might be represented by either letter. Before a consonant we have 
this variation in 

Arcadian lu beside eV, even in consecutive lines of the same 
inscription (Meister 2. 90) : cf. Hesychius i^Kpo^' iyKctpaXoif 

Cyprian «V beside (Hesychius) euavop' tV0€9, t{v)0e beside €{u)0dBe 
(Meister 2. 210): cf. Hesych. ttiXvou' (/)at6v (=»Att. ttcWou) : 

Cretan lu and iv in the same inscription (G. Meyer 58) : 

Locrian 7/1/os (Havet, Mem. Soc. Ling. 2. 168). 

80 the Sicyonians themselves called their city ^exvwu (Meister 
2. 89) : Hesychius has XcKpoi and XiKpol * antlers ' without designa- 
tion of dialect. Before a vowel (Solmsen K.Z. 32. 513 sq.) 
we find this variation in Cyprian, Lesbian, Boeotian, Thessalian 
(excluding Larisa), and Doric : 

Cyprian Oiop and detp, /i\ and /le (Meister 2. 211) : 

Lesbian ^pvaltu and xpvaeo) ; 

Boeotian $169 and deocv, Soxt'ei but KoXeovn : 

Thessalian Alotna and Xiovp : 

Laconian ato^ and Qewva : 

Heraclean Tt^Kpdrto9 but FeT€09 : 

Cretan $169 and deov, dfilwu and d^ewvj itafiev and etofiep. 

We must therefore conclude that, however it was written, e was 
always pronounced * close ' in Aeolic, Doric, Locrian, and Cyprian ; 
and, at lea9t when it stood before a consonant, in Arcadian also. 

B. (a) The derivation of the following words is unknown, 
and we cannot say that the v stands for o : 

irpvX€€9 * champions ' (Hom.), cf. Cyprian wpvXi^ * war-dance ' 
(Hoffmann B.B. 15. 89). 

irpvfiPTf * stern,' irpvfiuo^ 'hindmost': not from w^o, which would 
give just the wrong sense. 

frpvTat/i9 'president': Attic also wporavei'a irporauevu) (Meister- 
hans, Grammatik der Attischen Inschriften, p. 19), Lesbian 
both irpinavi9 and wp6ravi9f the words being popularly connected 

with TTpO, 

vvXtf * gate ' : not from weXtv ' move,' which gives too indefinite 
a meaning. 

7rvfiaT09 Mast': Sk. pii-nar 'back' (Bugge B.B. 14. 68) has 
little resemblance of meaning. 

FhlL Trani. 1891-2-3. 22 



338 SOME GREEK ETYMOLOGIES. 

mrvpaOoi ' stercus ' (Hippocrates), of. Att. tKpvpnBe^. 

ffTTvpti or ff</>vpi9 ' basket ' : Lat. aporta goes rather with tnrapTov 

* rope.' 

T/jyf * new wine ' : Eng. dregs cannot be connected. 

threat * awl ' (Herodotus) : Att. oirrjrtov may take its o- from ovj^ 
' hole.' 

So 7\i50a> * carve ' goes with Lat. glahd * peel,' Ags. eleGfan 
'split,' not with f{Ka(f>a) 'scrape.' In v\o709 *army,' vppaJ£ 

* pell-mell ' (Schulze 495), the first element is a dialectic form 
of trvvj not a ' copulative ' o-. 

{ft) In the following words the v is formative, a stem-ending 
(Brugmann Grr. 2. 104, cf. 91, n.) : 

u'yvpi^ 'gathering' {a'^wv), ofiif^pi^^ irap^f^vpi^f Att. af^vpTff9 

* beggar ' : from a stem ayv^ as uyopa a'yeipto from a stem 070- 
(tt7€-), all cognate with u'^tv * drive.' 

aiavfii/TfT^9 * umpire ' : stem altrv'^ cf. ai(r(^f)af see on a7<rxo9, 
sec. 4. 

ufivhi9 * together ' (the accent, as opposed to that of afioiftrfBt\ 
is from afia)=8tem ufiv- (on the breathing see sec. 3), as apn 
from a stem afio-, 

hiairpvtno9 * passing through ' : stem irpv-, from the root of 

irepau), see vpaatru)^ sec. 4. 

IpinrrtD * I tear ' : stem hpv (Septv), as ipeTno * pluck ' from 
a stem ^pe-, 

iTraaainepoi * one after the other,' of obscure formation: if it 
went with the Homeric aaaorepu) (by-form of uaaov) it would 

be ^eTratraorepoi, 

KpoKv^€i\o9 ' lizard,' Hipponax 119: stem Kpoxv-, as KpoxoBeiXo^ 
from KpoKO'j sec. 4. 

fiapTvpo^ Hom., fiaprvp Find, and Att. : stem /lap-Tv; root mr-y 
as in ftpa-ftcm 'umpire ' (Kretschmer K.Z. 31. 392). 

pu}uvfio9 Hom., vwvvfivo^ Hom. Find., ai/-u>t/vfio9 ctt- 0/4- Hom. 

Find. Att., ev- ttoXv- Hes. Find. Att., 5i- trvv" yjrcvB- varpujvvfuo^ 

Att.: not 'compounds of ovofia* as Kretschmer K.Z. 31. 377 
makes them, or they would end in -p^wv : they are from a stem 
WW (for the ending cf. erv/^o^ ijSvfio^)^ as ouopa is from a stem 
opo' (Bartholomae B.B. 17. 132).* 

* On 6vofia and its cognates — among which Lat. nomen must not be reckoned, 
it cannot be separated from co-gnomen and Old Slavonic zname * sign * — see 
further J. Schmidt K.Z. 23. 267 sq. ; Thumb K.Z. 32. 130 sq. ; Bezzenberger 
B.B. 10. 72; R. Schmidt Idg. Forsch. I. 77; Audouin Mem. See. Ling. 



SOME GREEK ETYMOLOGIES. 339 

o/»-7i;f * quail ' : stems oprv^- and oprvK-f cf ., with other stems 
from the same root, Sk. vartakaa vdrtikd var tiros, all meaning 
• quail.' 

/»o^'(w * swallow * (Ionic) : stem sm-, cf ., with the same * deter- 
^"^iticitive * bh. sro- in Att. potpew (Ablauts are- Lithuanian sriebiu, 
^T"* I-at. 8orhed)f and, with a different determinative, if- in Old 
Sla.vonic srikkati. 

^"^efipvxa * under water * : stem mm-, cf. Lat. mare (Hirt Idg. 
borsch. 1. 475). 

C^y) In pv^tt) 'growl* beside po^u) poxOew, and pvfipo^ * buU- 

*''^**-'"er ' * beside po^fio^, we have different Ablauts : the pv- repre- 

vr- (Frohde B.B. 14. 107), the po- represents VTO-. It 

8 that in one dialect of the Ursprache vr (vl) became m (lu), 

^ another the sonant took the same form as in other combina- 

^ons • ^e have Sk. rue- 'shine' rudk- 'grow* luhh- * be lustful' 

^^side vdrcas * light ' vardh- * grow ' valbh- * enjoy * respectively, 

'^^<T<Tio 'dig' and (nasalised) pvyx^^ 'snout' but Sk. trh- 'to 

**»' \vK09 Lat. lupus but Lith. wilkas, see on ffavpwrrjp sec. 5. 

C ^) In the following cases the v — like the u in Sk. dhur- ush- 

^ beside dhvar- ' injure ' vas- * shine ' hvar^ * be crooked ' 

^pectively — is Ablaut of vo or ve : 

V^'B'iy, gun-,* beside gven-, Goth, qind. Old Irish ben. Old 

^'^onic zena, Sk. janis, and Elean fieueoi [fua^ono rrj r^vvaiKij 

^^^ist^r2. 22). 

,j^ ^'^i-tricvvtov 'skin over the eyes,' skim-, beside skven-td-, Old 
^*^e skinn. 

^ ^ ^* » G. Mtyer Gr. 6r.' 77 and Albanesische Studien 3. 69; Brngmann Grr. 



ii^^^^ fin. and 2. 340; Schulze 201 hq,\ Pereson 227. The forms in Celtic 
j^ ^^actorily explained. The root may appear in iwofuu * blame, disparage,' i.e. 



« '^'i axnmm or ainw, Welsh enw) and Old Slavonic [in*/) have not yet been 



•n.^^***©*' in our parliamentary sense, stigmatize. Arcadian KXe»v((/Aw, Laconian 
^^ f^^6iioVt seem to owe their third vowel o (for u) to a false connexion with 



»^j^^^* e. (Andrew Lang, Custom and Myth, p. 39), a fish -shaped piece of wood 

l_j i-^^^& a hideous noise when whirled round by means of a piece of string. The 

^\^ ^ks themselves seem to have compared the shape of the bullroarer to that of 

^-^.^ ^'i^-neck, Xvy^^ with its long snake-like neck : * the mad bird, the variegated 

X^^^JJ<*^k of the four spokes, bound to an endless wheel/ which Aphrodite in 

fi^^^ar (pyth. 4. 381) brings to Jason to help him in gaining the love of Medea, 

^^u ^^irdly have been a real wryneck, but a bullroarer spun round by means of a 

l^^^^l. From the noise which this would make, not from tbe bird itself, which 

1-^ *^ot a loud cry, came the Homeric i vfw * roar ' ; and, from the use of the buU- 

>^ ^'^r in magical ceremonies, the meaning of Xvy^ as 'charm, spell' (Pind. 

^^* 4. 66) or 'yearning produced by a spell* (Aesch. Pers. 989). 
^^^^ A Velar after or before v becomes Palatal, not Labial (De Saussure, Mem. 
• Ling. 6. 161 9q.), 



840 SOME GRBEK ETYMOLOGIES. 

OvpUf dhnr-, Lit. dUri/s, Arm. durn, Sk. dur-, beside dhyor-, 
Zend dvara. 

kvk\o9 'wheel,* kx(kvlo8, beside kvekvlos Ags. hveoM^^ 
kvekvldt Ags. hveogel^ Sk. cakrds, and kvokvl- Lat. poples (an 
Oscan form, for *quocles, as poplna is the Oscan form of coqulna) 
'ham of the knee' as being rounded (Bugge B.B. 14. 64, 65). 

KvXi^ *cup,' knl-, beside kyel- jreXif (Cratinus, Meineke 
2. 64), and kvol- Sk. kalagas. 

tnvpa^ * spike at the butt-end of a spear,' (i)tiir-, beside 
(B)tver-, see on travpunypy sec. 5. 

vSwp, ud- Sk. Man-, beside yod- Goth, vatd. Old Slavonic voda, 

vTTvo^^ supnos, Old Slavonic 9iin% Arm. khun, beside svepnot 
Ags. svefn, and sydpnos Lith. sapnas (Lat. somnus and Sk. svdpnas 
may represent either syep- or syop-). 

7rt(Tvp€9 7r€<Tvp€9 (soc pp. 8 aud 2), kyetUT-, Lith. keluri, 
beside kyetyer- Lith. ketweri, Old Slavonic c'etverii, Irish cethir, 
and kyetyor- Dor. rerope^. 

So apparently in 

fivXff, mul-, beside myel- Irish melim, Old Slavonic mef/a, 
Sk. mar-f and myol- Goth, tnalany Lith. tnalU^ Arm. malem (Lat. 
mola mold may be either myol- or lliy|-, see below) : 

i/vf, nuk^t-, beside nyokt^- Lat. nox, Irish nocht, Goth, nahts. 
Lit. naktity Old Slavonic noUi, Sk. ndktis : 

0t;X\oi/, bhifl-jom, beside bhycQ-jom Lat. folium. 

Le,y though no extant language has initial my, ny, or bhy, 
the Greek forms here show that such combinations existed in 
the Ursprache. So (Brugmann Grr. 166. 170. 184) y is lost 
after initial bh in vir€p-<j)ia\o^, Lat /?*, Old Slavonic hi * he was ' ; 
and after medial n in Att. <^6va-ra feVo? <f}Oavw, and Old Slavonic 
tintka ' thin.' 

In five of these words we have a further Ablaut, the final liquid 
of the root becomes sonant : 

71/1^7) : gyn** Dor. t^ava, Boeotian pava} 

Ovpd: dkl^-, Lat. /orw, Old Slavonic dvirl.^ 

^ So, I think, ir4x\os * robe ' as being circular when spread out flat on the 
ground ; whence in Latin it was called cydaa, 

' Also, I think, Ionic (not * borrowed into Attic from Boeotian* as J. Adam 
says of ^cb'ovcroY, Classical Review 7. 102) : fidyavaos 'mechanical* — Herodotus 
2. 165 opposes fiayavairi to rh fjLdxtfioy — =*fiayi-avr-jos (for similar shortenings 
in componndssee on f/x'^vs, sec. 4) * quite womanish,* the second element being 
from aMs in the sense of Homer's aUrus * merely, simply.' 

3 But not $(up6s, which was the pivot of a gate, w&Kai (II. 12. 459), not of a 
door. 



eons GKEEK XZTBfll.iKnE&. 

|nail ' : rtems V"T ■■•' i'-T'*-- =- » 
(ame root, Sk, r«r^M« rir-iia ™-r-- 

twallow' (Ionic; : k^tl »»-- i. ^-na 

bh. IT*- in Att- /ioCK« Airjin» ^s- L: 

rM), and, witi t ii:Et?^ar it^i- ^-: — ■.> 

■titti. 

a ' under w»W* ' : s^tTL acK-. -^ --c 

475). 

(Friihde B.B. 11- li". -.;^ — r- 
in one diait^T trf "irt . r=7ri -- ■^ "t_ 
the Bocacl I'^.t ii/^ •un*- ; ^rt ji- - 
hare St rw^- ■ -i-;!^ -»-— r* * - 
M« 'Ii»bt' rw-ei- -r".'*- ^i...'— -f^ 
ig' "'l tfc'^---- •■■ ^■- ■' 

be foll&"Tir *»«• --n' I — -.i^ -^ X 
e ilr«r- ':^;ir.' to- ._::^ *. .- 

c is S^-Ja^. if »» ur -w 

■na-,' fc«~aie C*"- ■■'"■■- "'*■ 

■M, Sk- >»M- Mut i:«aB **--- _- 

22:. 

lar '«ki> vicr ^e ?:;i% W^t- >" . 



I. -r.^ -w Bc 



dialects 



^ m«« conjectatw. 



342 SOME GREEK ETYMOLOGIES. 

(a) in the article, which like our ' the ' was doubtless un- 
stressed, Pamphylian v but Arcadian o : 

{h) at the end of a word, Lesbian awv and awo (the Grammarians 
give heupv as the Aeolic form, Sappho has Sevpo), Arcadian airir 

roTir* aWv* but iXvtrarot Cyprian afrv f^iuoirv wplaerv (never -to), 

Pamphylian i/3tv\da€Tv iwtrfXo^v KmeFep^oSv, So before a final 
consonant, Cyprian Kepafitv^ Nom. Sing, (in other words -ov), 
Pamphylian ftw\TJfA€yv9 and in the same inscription KCKpafUvo^y 
'EotFcBuv^ Nom. and Kovpafftww^ Gen. : 

{c) in the first element of a compound, whether a monosyllabic 
Preposition, Cyprian w-eOrfxc and oviOifKe, both from n-, Att. 
av ; or at the end of a disyllabic Preposition, Lesbian ana;- and 
awo'f Larisaean awv-y Arcadian awv- Karv- ; or at the end of a 
stem, Rhodian 'AyaOv-fifiporov and Tifio-pp68oVf Pamphylian Foikv- 
iroXt^ and N€7o-?ro\€<«. 

Pitch-accent seems to have had nothing to do with this pro- 
nunciation of o, we have v in the oxytone syllable in FoikvvoXi^ 
and may suppose it in *\^aOvnppo7o\'} On the other hand, except 
in the later Lesbian dialect, on which see below, this peculiar 
V seems to appear only in the unstressed syllable.* 

In some sub-dialects o before r, whether stressed or not, seems 
to have been pronounced * close': the instances (G. Meyer 116; 
J. Schmidt K.Z. 32. 394 sq.) are — ^lonic (at Abu Simbel) vfv, 
Cretan vT and StnH, Euboean fji€rviK09. 

(rj) In the later Lesbian dialect — ^that of Theocritus in his three 
' Aeolic ' poems, and of the poetess Balbilla, not that of Sappho 
and Alcaeus or of any early inscription— even radical o seems to 
have been pronounced somewhat * close.* The Greek grammarians, 
who say that the * Aeolians,* i,e, the Lesbians, * turn o into v,* 

give fivtyi9 ^vauou trrvfia rvre v/uu/>a\o9 vwiaOa vpvi9 as the proper 

Aeolic forms, though Pittacus (in Bergk) has ff-rofAa^ a Lesbian 
inscription otc, Alcaeus opvi^ ; so that we cannot be sure that 



^ For ^Kcerhy which is to Korh as (nrh to Inra- (Sappho). 

' Homer s &AAi;5is owes its u to &/uv8if. 

' So in Cyprian Kvp^ina/jLa (HesychiuB) ' wine made from the refuse of pressed 

grapes,' if Meister 2. 220 is right in making tbis&=*Ka>i^({-iri(r/ia, from ire»yo5 

' resin ' -}- xTvta : but M. Schmidt suggests Kvyv-xUcrfui [* juice pressed out '). 

* Cyprian ZoFivat and IvFdpoi seem to contain the same root in two forms^ 
(1) dou- from dou- (Wiedemann, Litiiuisches Praeteritum, 41 «y.), cf. OlJ 
Slavonic darati, and (2^ du-, cf. Lat. duinij with F developt before a vowel, as in 
Chalcidian TapvF6vris, JBoeotian EitFiyopost cf. Epirotic ECfiatripos. Cf. respec- 
tively (1) do(u]- in Cyprian 8c6icoi, and (2) dii- in MBOm (Meister 2. 220). 



SOME GREEK ETYMOLOGIES. 343 

the yarious readings Svpct and vir^wv in Sappho 40 and 4 are 
genuine. In a late inBcription we have vfwitv^f in Balbilla 
(Meister 1. 53) v^if in Theocritus vfioiov vfiapTq^ in Hesychius 
cf vfiaWicv' cf ofiotujify while Theocritus has trrufiay Hesychius 
fivpfivpwv. These facts seem to show (1) that the * close' pro- 
nunciation of radical o was coufined to the later Lesbian dialect, 
and (2) that even in this later dialect the radical o was pronounced 
only slightly 'close/ since in all instances but those just given 
it is spelt o and not v, 

(0) As the Romance languages, descended from the dialect of 
the Roman lower class, make every Latin u into o, so the dialects 
of Western Cyprus (Hoffmann B.B. 15. 49 tq.) seem to have 
made every v into o : the town-name SoXoc (in Cyprus) is in 
Plutarch ^v\oi (Meister 2. 220), we have lOoyUif in an inscription 
from Paphos (Deecke B.B. 6. 71), Hesychius quotes Oopauav 
fioxo'^ <roava (beside OvpB. fivxo^ f<^7^7) ^ Paphian, and his 

p6pfLa^ iirr6ica<T€y Ivtcaifyoteve K6p.ftost KpotrraWov \o(f>Pi^ irewofffiai 
fffior^epov (beside fivpfiif^ iirvKaaev iyxaraipvTeve Kvp,fio9 Kpv<n'aWo9 
Xi/X"** iriirvafiai evitrfivt^epwi) may belong to the same family. 
So also in some Boeotian dialect (G. Meyer 90), 'A/toWoy Ootr/rfH ; 
but Boeotian EtKppotroifaif 'Sofiufyopuff Attic ''0\op,7ro9, Laconian 
Kovoovpeujvy may be due to Assimilation (Kretschmer £.Z. 29. 
412), while Strabo's 'Opfilva for 'Ypfiivri (in Elis, II. 2. 616, 
see Meister 2. 31) seems to show the influence of Sp/io^ * roadstead.' 



(3) Leskien's principle of ' Ausnahmslosigkeit,' that a phonetic 
law has no exceptions, is doubtless true within each dialect; 
but in Greek there were as many dialects as there are in 
English, and every poet and each of his hearers or readers must 
have been familiar with several. So ovis and bos^ words common 
enough, were loanwords in Latin (Havet, Mem. Soc. Ling. 
6. 17 8q.), the proper Homan forms would be *avis and *v6s : 
the Romans said ' sedeo in solio,' though the / in the last 
word was Oscan. In ' the skipper met the shipper in a well- 
equipt skiff' we have four different dialects: in the Windhill 
dialect of Yorkshire the forms m/^n, misel, misdn (myself) are 
used without distinction, and * probably due to importation from 
neighbouring dialects' (Prof. Joseph Wright, Dialect of Wind- 
hill, p. 122). 



344 SOME GREEK ETYMOLOGIES. 

A. Whether an initial vowel should preserve its aspirati 
or not must have depended on dialect in Greek, just as It dC 
in Latin and does in English: the lower orders at Rome, th- 
linguistic progenitors of the Romance languages, must hav< 
dropt all their A'«, and in England it is only the educat 
classes that keep theirs. 

ySv9 and ijBo9 go together: the latter in II. 11. 318 y^icwy i}^o 
' satisfaction from us ' has an ironical sense, which I detect also 
in cognate words with short vowel, viz. {a) a^ov II. 11. 88 
(where it seems to have a digamma), and uBrj or aBtf (in Homer 
only in Ace), * satiety'; (b) aBivo^ or aStvov *to repletion*; 
and (c) the post-Homeric aSpo^ * thick, large,' quasi * sufficient.* 
— In II. 5. 203 ddtft^f also written uSSrji^, may be a contraction 
of *a'fffa^ffif 'without satisfying oneself (Schulze p. 452 sq.); 
and so I would explain dBeu) in Homer's dSTjacte w^j/icoTev, both 
also written aSd-y as for *a-aFa^^o}y * am dissatisfied, displeased ' : 

<t>poifiiov * prelude,' beside irpo-oip.iovy must come from *oifuov^ 
as (ppovpd^ comes from w/)o+*o^o9 * watcher * (^opdu))^ and <ppovBo9 
from *<ppovBoUf i,e. irph ohov^ * ahead on the way,' II. 4. 382. 
Homer's oi^rj * song ' will then be dialectic for *oififj, perhaps 
meaning * connection,' arraiigement of words, from a root soi-, 
Sk. setiM * binding,' «»-, *to bind,' i-^d^ * band.' (Lat. saeeulum 
then must be from some other root.) 

B. In Lesbian we have twep iyjro^ for hvep v^^ov (G. Meyer 91), 
in Larisaean iwep (Meister 1. 224), in Megarian altrtfAvarav for 
aiffvfiifrjTy^ (Kretschmer K.Z. 29. 412 «y.), in Hippocrates both 
ffrpifpi/o^ and <rrpv(l>v69 * hard.' So I would explain Ktvai^o^ 
* wanton ' as for ^Kvyaiho^^ %,$, kvvo^ alhCj e^wv^ * with as much 
modesty as a dog,' the dialectic form being employed to disguise 
the meaning. 

C. The Lesbian representative of y was po (G. Meyer 27), 
e.g, ftpox^t^^ 0po<T€w9 trTpdrayo^y Homer*s rj/upporop beside Att. 
ij^aprou : so /iporo^ (of. Sk. mrlds * dead ') must have been Aeolic, 
the true Ionic word being Ovrjrd^^ Dor. ^j/Stov. So n is represented 
by vo in Alciieus' r^vdc/yaWoi/ beside Att. Kvd<f>a\\ov : * the root, 
as the varying initial proves,' was originally bi-aspirate, ghn-bh-, 

* On va for ^ in Ionic- Attic see Osthoff, Morphologische Untersuchungen y., 
preface: he quotes yydBoSf valoo {i.e. *ifdff-jal), and, for fia from i^, fActrt^ 
beside /ucroAAcUtf. 

' Cf. Hesych. iucaBSv kyaB6vy i.e. the original fonn was •&-xo6rfj, from 
' intensive ' &- (see sec. 1) -i- a root ghadh-, Ahlaut ghadh- in Goth, god* * good.' 




SOME GREEK ETYMOLOGIES. 






^*- ST^^l^^* iQ A.tt. 7i'€^XXoi', and with Metathesis (as in 
Xo9 beside Ags. nafela) ghem-bh-, Old Slavonic %eha 'I tear 
p>i<»ces' (wool being carded for use). 

Qe dialect seems to have used e-, not a-, to represent the 

SOOant : Hesychius has eotrtrrfrrip' aotrtrrfTJp, and efypvJTtfet' 

Mctf the first vowel in each being the * copulative ' prefix, 
®**^^iiially sm-, sec. 1, c. (In a-r^p-virveif literally * is chasing 
"■^^^p," and in H-'^pn * chace,' a-ypeto 'seize,* ^tv-d-ypia * reward 
^^** solving life,' the a- must be a prefix, the root gr-, ger-, as in 
"^^^'^o.. ^w-^pei *take alive/ and Hesych. iyptfvrai' yprft^rai.) So 
I^'** ^^ in Homer* s ipi- and Qepai-rrj^ beside api- and Odpao^, Hesych. 



^^f^^^pa and p,€pyi^€ 'gobble* beside ftdpaOpa ani fidprfo^; and 
• **^^A» in Hesych. ^eXXcii/ fieXepov beside ^dWetv p.a\€p6v. Again, 
^^^li. the consonant coming first (as in fia-revu} r^fvdOo^ xpahlfi 
''^«*-r^9), we have ni»^e in Hesych. /4€<rraf for fidtrra^^ and 
r'**/*^ in Lesbian xperoi for Kpdro^ (G. Meyer 6), Hesych. p€fKpo9 
*" />ck;a0o9 * beak,* Boeotian rpcireBBai beside (in another inscription) 
''R^^^^SBa^ (B.B. 17. 336, n.). 

'*-*- Homer's 6Sp,y &■ Att. 6trp,tjt as "khpr^ro^ = Att. *Airp,tjro9 

(Kr-etschmer K.Z. 29. 420); so Pindar*s KBKahpepo^ 'furnished 

^^5"^ ' -sAtt. Kexaapjiuo^y perhaps meaning * bound with,* and 

goxi:^^ with the post-Homeric K?jho9 ' connexion by marriage * 

^^ite a difi'erent word from the Homeric KijSo9 * care, raoum- 

^•* Homer's xalvvfiai 'surpass,* Perf. KeKaafiai^ cannot be for 

*^^^^M^fjiai or go with ic€ica^/t€j/o9, for Dental + 1/ would remain 

^^^^anged, as in aXoavcvti whvuj ^0vo^ (G. Meyer 280) : I 

^Ici explain it as for *Kdawp,ai (cf. eivvfii for *€tnnfp,i)f from 

^oot kns-, Sk. gans- * to praise,* so that Kaivvp.ai will mean 

*^ praised * for something, Od. 4. 725 irai>7ottj^ aperrjoi 

,^^ **"/*€ I/O 9 iw £iaifaot<ri, 

. '*^* Herodotus (Rhys, P and Q, Groups, p. 16), uses k- for w^ 

, ^^ords derived from the Relative stem, e.g. k6t€ koo ku)^ : 

J ^ his rrpdica 'forthwith* may stand for *7rp6 Tra^ i.e. 7r^o+ the 

/ ^^Umental (Brugmann Grr, 2. 274) of kvo-. Thus irpoKa 

^ be identical with Lat. prope * near,* an Oscan form (see 

' ^ init. on popks) for *proque from *proqua (Brugmann Grr. 1. 

3. — The 'Ku in atnixa must be differently explained, 

^^^ently as kn. Ablaut of «€i/ ' then * (see Persson Idg. Fors- 

6. ^ -I ^vet's (and Benfey's^ connexion of Jiypvirvot with fycfpw, M^ra. Soc. Ling. 
^, ia rightly controverted by Br6al, do. 172. 



346 SOME GREEK ETYMOLOGIES. 

chungen 2. 228) : the first element is *avrt\ Location of avf 
* ipse * (Hesychius), while aiVro? is from the stem of ai;9+a 
'determinative' o (see on $e<T<f}art)^^ p. 1), 

F. Before e or < a Velar ought in Greek to appear as a Dental 
(Bezzenberger B.B. 16. 254 sq,, Bechtel Hauptprobleme p. 356 
9q.) : the rule-right forms of fiio9 and fiia, beside Sk. jtv- and 
/yd- respectively, would be *^/o9 and *hia. The former appears in 
^l6/)09 ' living' (Fick B.B. 16. 287), and I think in f)U^i * hasten, 
am quick ' ; of which the Perfect Active would be *6€6itvtca (as 
that of a<f)tiff%i is a(f)€iM)Kay G. Meyer 659), whence was formed a 
Present hiwKo^ * set in motion ' (G. Meyer 45). The form *Bta I 
detect in 

(a) tiaKovo9 * servant,' quasi /3/a kov&v, * compelled to work ' : 
the second element, as in Hesychius' Kovectf' eir€»ycff0ai, Homer's 
cyKoveuT * hasten,* Att. aKovTrt * without trouble ' (Sohuize 353, n.), 
is from ken-, a parallel form of kyen- in woveiv^ as kel- in iceXXw, 
r€\i/9, Lat. celer, is a parallel form of kvel- in ircXu) * move,' 
Lat. cold : 

(b) Sia-v€ic^i * continuous,' quasi ftia iyexOeny 'brought on by 
force, not to be stopt ' : the second element being an unnasalised 
form (as in Lat. nactus) of the root of ivcf^xeTtf and Lat. nanciacor. 



(4) It is often difficult to decide whether a word is a compound 
or a derivative, or of what elements an admittedly compound word 
is made up. 

a^ffXV^ in Homer always connotes sound : it is used of the cries 
of pain, of a tumult, of the bleating of sheep, and of noisy eating 
and drinking. I would therefore explain it as * dry-sounding,' 
from *afo9. Adj. of afjy 'dryness,' + fjx°^ 'noise,' comparing 
II. 12. 160, Kopv0€9 S* a^(p* avov avreifUf of a * harsh, grating sound ' 
(Monro), and Verg. Georg. 1. 357 aridus . . fragor *a jarring noise.' 
Hesychius' u^ax[_€a] then will be the Doric form ; his d^exj^ is a 
different word, dwo -rov d^rjv exeti^, as the Scholiast on II. 15. 25 
gives the derivation of aj'j/x'/* (which Apollonius Khodius uses 
as= ag'aX609 'dry,' Wackernagel .K Z. 33. 51). 

ai<rxo9 * disgrace '«• *a7<y-<rx-©s ^^om ais- in Goth. aistan*io 
observe,' Old High German Sra ' honour,' at{<ry^w9 * shame, respect' 
(Bezzenberger B.B. 4. 313),^ and ai<r'{F)a 'apportioning,' giving 

1 The 8 is probably terminational, not from 8/8w/a. 



SOME GREEK ETYMOLOGIES. 347 

each his due share of honour. The second element of o7<rxo* 
is the * reduced ' root of e^x**^, so that the whole word means * having 
observation,' getting oneself observed. So 7rdtrx**>=*^^^'*^^f *I 
have sufPering, I suffer,' from iraOo^, iraOe'tu: Hesychius' vaaxq-na 
shows that the Greeks themselves considered vaax' the root. 
There is no proof that *vaO'<TKtM)i with an Inceptive ending, could 
give anything but iratrKU) (which is the Elean form, G. Meyer 269) : 
futryu) is not for *^iy-<rKttff but shows the same stem fiiay' as in 
Homer's fmrr^'dyiceia * meeting of glens,' i,e, ^ty-a-f^-f the root of 
fiiy-vvfii increased by 8, and with the same termination (Brugmann 
Grr. 2. 91) as ai-yrj beside ffi-uyn-du 'look silent' (Kretschmer 
K Z. 31. 471). 

uiftaya 'command' means 'lead up' (Lat. »n-duco, see sec. 1, 
A), cause to do a thing: *wya is an un reduplicated Perfect, 
like oTha, from the strong form (as in dytoy-ost) of the root 
of uyw. So olfiuyyii * wailing,' from otJito«4-*«i»7J/ : for the sense 
cf. tcrvwou uyetu * make a noise,' yeXttrra a76ti' * raise a laugh.' ^ 
So I would explain irpaaau) as *irpa'dyr-ju), ' make progress,' 
whence its Homeric use with a * local ' Genitive, e.p, II. 24. 
264 itfa wpijaawfieu oBoto 'advance on our way,' and so wpayo^t 
* business ' s=*?r/>a-a7-ov : the first element is *7rpa 'progress,' 
a Subst. formed like xp^ ^^^ ofto-KXr/ (Brugmann Grr. 2. 896), 
and standing to wdpo^ vepdia much as hpa- in Spafjui hpato stands 
to Lith. darau 'I do.' 

upya\€09 * difficult ' (the a- must be long, as it is in Att. 
apyov ' idle,' from *a-F^py6si) means * impracticable, not to be 
done,' from a- negative -(-(/')€)j7oi'. For the contraction cf. 

aKtou from *d'F€Kioif, 

appaT09 * unchangeable ' (Plato) = *u'Fpa'T09, a- negative + 
*fpa769 Participle from vy- *to turn,' whence also (a), with a 
termination -mo-, p6^o9 * worm ' (see sec. 3, C), Lat. vermis (from 
•vermis), Goth, vaurms * serpent ' ; with -mn, Lat. vermina * colic,' 
and {h) with a * determinative ' t, Lat. vorto-, Sk. vart- * to turn,' 
Elean Ppardva * stirring-ladle ' (Hesychius), and oprv^ * quail * 
quasi * dancing, turning round.' 

Bt<p0epa * dressed hide ' (Thuc. 2. 75 Beppei^ xal ^KpOepa^ ' skins 

' With the literal sense of * raise * we find Ayw in iurr'fi * raised land, coast, 
river-bank,' cf. Soph. Oed. Rex. 183 fitifiiov hcrdv * the raised altar ' ; so in 
&X^os * weight, what one can lift,' cf. Soph. £1. 119 6,yti9 ohKin awKw X^s 
iiMTtppovoy ix'^o't ^^^ ^^^ Attic use iytiw fxifay *U) weigh a pound, be able to 
lift it.' 



348 SOME GREEK ETYMOLOGIES. 

undressed or dressed*) means 'twice spoilt,' ^i»+0^€</>w, diverted 
from its natural use as a covering for the beast, first torn off 
{^€ppi9 from ^cpUf with a termination -/>(9 as in axpi^ oxpi^y Brug' 
mann Grr. 2. 98) and then tanned. Hesychius has a dialectic 
form hylrdpa, which goes with his yfreipei' (t^Oelpei (G. Meyer 
209 fin.). 

€7X€Xv9 'eel * = *€7xv-x«^v9, * snake with the mouth of a 
tortoise/ x'^^^^' *€yx^^ exactly =Lat. anguU, since eng- becomes 
in Latin ang-^ frangd is for *frengo, cf. Goth, hrikan, — On similar 
shortenings in compounds, where the two middle syllables have 
the same consonant or two similar vowels concur, see G. Meyer 
302 and Schulze 427 : Aeschylus Eum. 52 has fiSeXv/crpowo^ for 
*phe\vicT6-'TpoTro^i and so I would explain (a) waKafivalo^ 'suppliant 

not yet purified * as for *7ra\afio fivalo^ (as na\a/tJ^^iyv=*noXa^o- 

larjhrj^, Fick K.Z. 22. 99), * with a deed of violence, 7ra\a/irf, in 
his memory,' fJivriftrj ; {h) irr6-fiapyo9 * loquacious ' as for *<rTo^'' 
pLaprfo^ ' mad of mouth'; and (e) vwefivr/fivKe, II. 22. 491, of an 
orphan boy, as for *v7r-€-fiuff'i^fivK€ * is bowed down in mind ' 
(quasi *viri^fivK€ /uLCfivtjfievo^) : though for rjfivu) * bow down ' I 
cannot suggest any etymology. 

e55a0o9 'foundation' is marked by Brugmann, Grr. 2. p. 204, 
as the only Neuter word with the termination -bho- : I therefore 
take it as a compound of €^09+001/ arro^aif ' touching the seat' 
or bottom, a Dissimilation of breathing for *€^a(/>o9. 

t7oi^9 ' ready ' may mean ' striving after the way,' oe/u>9 ; 
the first element being (as Prellwitz Etym. Worterbuch der 
Griech. Sprache suggests) jet-, Sk. gat- * seek to reach * (Middle). 
In the Active Sk. gat- means ' to marshal, put in order ' ; and 
with this we may connect cVco?, erv^iov, and (with the root in 
its long form reduplicated) ir^rvfiov • true,' quasi * regular.' 
On the difference of breathing see sec. 3, A. 

evpv9 must bo a compound, or we should have *€«/>i;9,* as we 
have ElXeiOvia beside *E\€v0u} : it seems a contraction of *€V'vpv9 
'full wide' (Sk. urua). So ev0v9 'straight* may be from ev-+0vtt}, 
' rushing well,' going in a straight line. The first element appears 
in three different forms: (I) esu-, €v-, dm 'brave,' with metrical 
lengthening (Schulze 33 sq.) ijm; (2) 8U-, the reduced form, 
Sk. su' ' well,' Greek v- in v/3pt9 beside Ppiapo^^ v^tJ9 beside 

^ The only non-compound word in Greek with v in both syllables is yKvnCst 
apparently a by- form of the *yKvK6s wliich appears in Hesychius* y\vK4i' fiordyti 
ris iStiiifios, 



i 



SOME GREEK ETYMOLOGIES. 349 

Lith. gyjH * I get well (De Saussure Mem Soc, Ling. 7. 89, 
Zubaty K.Z. 3K 52 sq,)-, (3) BU-, Sk. sL 'well/ and I think 
Greek *tJ- in *v0v9, whence by Dissimilation (Osthoff Morph. 
TJntere. 4. 190 8q., though his explanation is very different) XBv^y 
the second element being Ovuj. As Zubaty points out, the 
parallelism of 

8k. sk" ' well ' and dush- ' ill,' 
Zend hu' and dush-f 
Arm. h' and t-, 
Irish. 8U' and du-, 

makes it difficult to separate €v-, as the correlative of Svtr-f 
from Sk. sd-, and put it with either Sk. dy^ 'alive' or dvas 

* favour.' — With ev I would put {a) evre * when ' or * as,' in 
the latter meaning also rjvre, with metrical lengthening: the re 
being superffous, as in avre, 09 re, added on the analogy of 
clauses in which it really meant * and.' Thus II. 23. 62-65 
evre rov vttpo^ ifuipirre . . . ^\0e 5' ewi "^^vxift literally * well 
was sleep seizing him : the ghost appeared/ came to mean ' when 
sleep was seizing him, the ghost appeared;' II. 3. lC-13 evr* 

opeo^ KOfjv^rjtri No'to* Kare')(evev 6fii')(\.rjv , . , W9 Upa rS)v vtto 

voaal Koutffa\o9 wpwr aeWri^ means ' well does the south wind 
bring fog : so rose the dust,' i.e. ' as the south wind brings fog, so 

rose the dust ' ; and II, 4. 277 fAeXavrepou, ijvre irlaaa, (^ialvertai 

* it looks blacker, quite pitchy ' : (A) evxofiai * boast, vow, pray,* 
literally 'use only bona verba* about myself or the gods, the 
same termination appearing in vrj^u} trfiijxu) orci/a^cu rpv\w '^yx'^'f 
arouaxrj, ^tBaxrj (this from the same root as Stju * I will find,' Zend 
rfd- *to know '). 

^X^oBoirew 'quarrel,' II. 1. 518, means 'organise hostilities,' 
from the root of exOo9 exOpds + deky- Old High German ffi-zehdn 
'to arrange,' with which Brugmann Grr. 1. p. 332 puts Betvpov 
'dinner,' quasi ^hevvjov. — Eng. hatred, literally 'arrangement 
(Ags. raed) of hate,' is a somewhat similar compound. 

KpoKoheiXoi * lizard,' an Ionic word (Hdt. 2. 69), = * yellow 
coward,' xpoxo^ * saffron ' + Seiko^, from the colour and shyness of 
the animal. The application of the word to the crocodile must 
have been a Litotes, or joke. 

\ap7u69, properly used of an ox (Ar. Pax. 925), = * with a fine 
hide,' /KI/09. The first element is the intensive prefix Xa-, as in 

XaKaravvr^wUf \aKardparo9, Aa/*o;^o9, representing *Xa6- i,6» *\a<re-f 



850 80MB GREEK ETYMOLOGIES. 

as the similar prefix Xai- in Xaifiafyyo^ Xat^nrodias represents *\a<rt- : ' 
both are from a root las-, Sk. las- ' to be lively/ reduplicated in 
XtXatofiai * I desire,' i.e. ^Xi-Xaa-jofiai.^ 

fievoivaus 'desire eagerly '=*/*€i'o-/^o£i/rtft; (for the shortening 
of a compound see p. 20 on ^7x6X1/9) *am drunk with desire,' 
fievei oiuivfiat I cf. the Attic use of fieOvetp ' to be drunk with 
passion.' In II. 12. 59 fievolueov is wrong both in form (for 
fievoivaov) and meaning (* were anxious ') : Qoebel, Homerische 
Blatter, p. 15 sq. proposes to read fieuoietf au, 

inft^ar€09f the Homeric epithet of \nu3u and Kprihefipovy may 
mean 'such as never was,' 0109 ovvto i^i^vero (as I think the 
post- Homeric awXeTo* 'immense' meant 0I09 ovwio cTrXeTo), mj- 
negative + & Participial form from t^l'^vofiai. So Lat. ingens 
' huge ' means ' quod nondum genitum est.' 

waiytfttf 'sport' (Hdt.) and val^viov 'toy' (Att.) are from 
an Adj. *irai-r^v69 for ^iraih-f^vo^f formed after v^o-r^vo^y the 
second element going with r^ttyvofiai. But the forms Tratyfiotntvti 
irai^oofiai frivatKa owe their guttural to a mistaken explanation 
of irai^us {i.e. ^walh'juj) as for *7raty'jtVf since the -J'a; in most 
Verbs arose from -'xfw: Curtius, Verbum 1. 317, gives thirty 
instances of -^w from -t/'cu, as against nineteen of -^eo from -Sjio. 

TTwfiaXa 'not at all' (Att.) is a negative which was originally 
an interrogative : ttiv fidXa * how, very much how ' ? The two 
words were pronounced and accented as one, to show that the 
fidXa qualified the preceding word and not anything that might 
follow. So TTuf in Aesch. Agam. 1507 is a negation under 
the guise of a question : Sidgwick rightly translates it ' nay.' 
In meaning it differs from wwv ' how ? ' no more than ovru) 
differs from ovrw^ : in each case euphony alone determined which 
form should be employed. So ovvw and ot/ww?, firJTru) and fiyirtuvy 
are used interchangeably : in 

II. 2. 419 ovB* apa TTw oi eTrcKpaiaive Kpoviwv (see Fasi), 
II. 3. 306 oviru) tXi} <yo/t*, 

II. 14. 143 aoi 6* ovirw fidXa Trafy^i; $cot fidKapev KoreovtriVf 
Od. 2. 118 KepScd 0* oT* ov'TTu) T<i/' aKovo/neu ovBe iraXatwVf 

1 For the difference in the final vowel cf. i^>x^'ifo\is iLpxt'tdicrwy (G. Meyer 
81.). 

* \t\irin4vos ' eager ' is not from \i\alofuu hut from ^Kido^cu ' I am much set 




from him *), from the root of it4\w * move.* 



SOME GREEK ETYMOLOGIES. 351 

Soph. Oed. Hex 105 ov f^ap elaelhou 76 irWf 

and again 

II. 4. 234 iJLTjTnM) ri jite^iere OovpiBo^ d\KtJ9y 

Od. 9. 102 ft^WU) T<9 \tO70l0 (/>a*^UJV VOtTTOlO XaOi^Tnt, 

£iOr. Hec. 1278 fn^vto ftMvelri Twhapl^ toaovhe irain^ 

ire might jast as well have had ovww^ {firfTnasi), In many 
pHSsages ovwuj {firfTTo)) may conveniently be translated * not yet ' : 
l)ut in each it is the Verb that gives the connotation of time, 
"the particle denotes only manner (* not at all '). 

tr6\oiK09 * foreign' (^^pap^ap^^y Herodian) must be a comic 
formation from aoKo^ 'ball of iron'-(~^^3 termination of diroiKo^ 
^otK09 iU7oiKo9 avvoiKo^t quosl * lumplsh dweller/ heavf/ citizen. 

<rtf>oSpu>9 'violently' (Od. 12. 124: ff</>odpa and a</}oSp69 are 
j)08t-Homeric)=i* acting for oneself/ from the roots of <r0o9 'their, 
iliis ' and Bpatiff as in o\i7o^^3i/€W ' doing little, feeble.' So I 
"vould deduce fftpe^at^v * eagerly ' (II., in the phrases eirero 
^r<peSap6vf atpeiavov efpeirCf * he followed on his own way ') from 
%he stem of <r0€Tc/>o94-the termination -^ai/d-, fii^Kchavo^y a by-form 

^8 the termination "Tavo-j einjeTayo^, is a by-form of -ruo- in 

vwohpa, in the Homeric phrase viroBpa Ihwv * looking fiercely at 
Inim,' can have nothing to do with SepKofiat, which would be giving 
"^he same idea twice over, and in which case the word ought to be 
"^^InroBpd^f as the Alexandrians rightly had it. I therefore (Etytna 
^SlAtina s.v. odium) take vtrohpa as Instrumental (for the accent cf. 
^»<poSpa from a<t)ohp69) of an Adj. *vir'oB-p69 * with covert hatred,* 
m the root of ohvop^t (Schulze 341) 'am angry,' Lat. odium, 
rm. ateatn * I hate,' Old Korse otul * fierce ' (^.y. otul augu 
fierce eyes'). With odium goes alrdx 'fierce' (Lat. Consonant 
aws 22, see Thurneysen K.Z. 32. 562) ; so that, if I may coin 
he Latin word, v7ro^^a=*subatrociter. 

a'x/'ov (apparently Keuter) * paleness ' is in Homer the colour 
f fear, II. 3. 35 wxpo^ re fiip eTXc wapeid^, Od. 11. 529 
XprjaavTa xP^°f ^^ ^ coward : it may mean ' egg-colour,' as 
^^^ellow as the yolk of an egg, to wxpov tov wov (Aristotle). 
"iThe first element is the root of tv{f)6p (a post-Homeric word), 
XLat. ovum, Old Slavonic aje (which last proves the root to be 
^-, not 0V-) ; the second is a by-forra of xp^"^ * colour.' From 
^^XP^^ ^^^ later (first in Hippocrates) formed an Adj. tvxpo^ 
* pale, yellow.' 



\ 



352 SOME GREEK ETYMOLOGIES. 

(5) Some other words may best be given in alphabetical order- 

ai€r69 * eagle' (■■*ai/'eTj9, as Pergaean alfiero^ shows), Aratus' 
aiijr69,^ may mean ' mighty one,' going with aitfro^f epithet of 
Hephaistos in II. 18. 410, aia Mand' (the 'mighty' earth), 
and aiaifr/9 'everlasting.' The two last words Johansson, B.6. 
18. 4, puts with ai€i\ alwy, and Sk. dyiis 'living': the common 
idea then will be * full of life, strong.' 

aiv69 * terrible ' may originally have meant * bitter, cruel,' ^hfi-jo^ 
(cf. paivw from */iafijtOf Guth. qiman)y m- Ablaut to om-* Lat. 
am&rua 'bitter' (on the first vowel see Latin Yocalism 5), u}fio9 
* raw, cruel.' 

3xuiv 'javelin' may go with HkvKo^ 'acorn ' and mean ' made of 
oak.' So Schrader £.Z. 30. 461 connects aiyauetf 'spear' with 
Eng. oak, 

avlpo7?fra ' manhood,' occurring three times in the Iliad, is of 
course unmetrical: it may be corrected in several ways. (1) The 
most impossible of all is Clemm's, who reads *Bporrjra from *vhpo' 
Tfj'Ta {a-vhpo^)'. this in two places gives an un-Homeric caesura 
Kara reraprou tpo-^aiov (Monro, Homeric Grammar,^ 367. 2), 
\iwovaa *hpor?i7a Kal ij/SijVf and in the third, II. 24. 6 TroOcwt^ 
*BpoTiJ7(l T€ Kal p,€uo9y wlll not cvcu scan. (2) The idea that 
*dBpoTfjra could be a * reduction ' of *dvBpoT^ra lacks support : 
dfiportf is epithet of vv^in II. 14. 78, and ufifiporo^ in Od. 11. 330, 
but the sense is different, d/Sp6rrj (like u^ifipoTtjf G. Meyer, 179) 
is ' neu componiert ' from fiporo^, to mean * void of men, unpeopled,' 
^KaO* fjv ppoTol fiij (f>oirwiTty Schol., eu rf fiporo9 ov irpoeiaiu 
Eustathius), and so in Aesch. Prom. 2 a^porov eU ipi^fuav (as 
Dindorf rightly reads for aparoy : Hesychius has afiporoy' dwdyOpiO' 
vov).'^ (3) I would therefore read *dpor^ra, as a parallel form 
to dperrjv^ the dp- in each case representing nr- (beside d-vrip). In 
II. 2. 651 ^EtvvaXii^ dvSp€i(/>6inij we may read *i/>€0oi/Tiy, with the 
same stem as dpeTtj, 

dinaKa7o9 ' sturgeon ' was a Scythian word, Hdt. 4. 53 : if this 
means Slavonic I would connect the word with ant- in dml 
'against,' Lith. ant 'up to,' and akv- in Lat. aqita (the Ur- 
Slavisch form would be *aka), and explain it as ' going up 



* For the varying quantity of the second vowel cf. iLpy4ri ipT^ri, iurKtO^is 
dcrmfO^f, wptrrowayiis fitwiTT^s (Schulze 473 n.). 

' From ikfipirri Goebel, Homerische Blatter, p. I »q.y deriveB h^pord(m in 
Tl. 10. 65 ijJ\ -Kws iifipord^ofjLty &AA^Aoiiy, * walk oy night to each other's hurt' : 
Hesychius has ifiporrjtrcu *■ to meet by night* 



SOME GREEK ETYMOLOGIES. 353 

stream/ since accordlDg to Pliny (Hist. Nat. 9. 60) the stargeon 
swims with its scales turned forward. Lat. aquipSnsis may meaa 
t;he same thing, though I cannot explain the second element. — 
Another Scythian word, ilax^ ' hazelsap,' Hdt. 4. 23, may go 
"^rith Polish oskola * birchsap,' from a stem askhv-. 

uyrpov 'cave' may mean 'cul de sac,' and go with Sk. dntas 
* end,' Goth, andeis. 

avXa{ * furrow,' avXrj *court' (within a high fence, Homer's 
J3a0erf9 avX^^), avXov * flute,' avXtvy * hollow way ' or * windpipe,' 
xiaay all go together, the common idea being ' hollow.' The stem 
'^lien is au-l-n- from aye-l-n-, p. 1, whence also ye-l-n- in Lat. 
«?«i//i« (Persson 230 : i.e. *veln6s, with * pretonic ' a), Sk. vdnl 
* reed ' and vdnds * arrow made of reed.' ^ 

fipe<po9 ' babe ' stands to fipaxv9 * small ' much as iXafppov, see 
ote p. lly stands to i\axv9 : the root of fip€<po9 must be mreghy-, 
that of Ppaxv9 is mrghy-, Goth, ga-maurgjan * to shorten ' 
C<Johansson K.Z. 30. 442 sq.). I detect a third form, mfghy-, in 
A^^p<pi^o9f II. 24. 316, 'the little one,' according to Pliny (Hist, 
^^at. 10. 7) the smallest but one (the fieXaudero^) of the six kinds 
^>:£ eagle. 

BeXeap 'bait,' Laconian ftX^p (Alcman 130), must mean 'dropt' 
^i^to the water, from fidXXto (not, as Meister 2. 204 says, from a 
^oot gvel- meaning * to split, tear'). In Od. 12. 252 the gram- 
^HArian Callistratus read IxOvtn to?? oXif^otai SoXoy Kara BetXara 
C* bait') pdXXtvp, for etBara (Schulze 102). 

S€<rjrdTrf9 and Sk. jaspatis * master of the house ' owe their ^ to a 
X>opular connexion with the words for * lord,' 7rofft9 (* husband ') 
pdiis : the proper form, as Old Slavonic gospodt * lord ' shows, 
gyispod-, the -S- perhaps appearing in Setnrol^u). The further 
^^rivation is obscure : the word indeed may be un-Aryan. 

Sexofiai in Homer and Att., beside Si ko fiat in Sappho Pindar 

It. and 'BoK' in compounds in all dialects, owes its x to ix*^9 

^ word of cognate meaning : t.^., to use Tick's convenient 

Expression, lexo/mt * rhymes ' with ^x^fuu (Middle). — With 

^oc- may go Boxfidi (t.tf., *BoK-<r-fAd9f as 7rXoxP'd9 is for *7rXoic-<r- 

/4t>9, De Saussure, Mem. Soc. Ling. 7. 91^) 'aslant,' a metaphor 

^ftxim a beast turning to 'receive' the hunter, II. 12. 147 (of 

^ These must be quite different words from Sk. vdnl * music, tone * and vdndi 
* music, hundred-stringed harp,' with which Johansson, Idg. Forsch. 2. 56 n., 
pats alt\6$ 

* Cf. ^xf^s * cleft,' XL 23. 420, for ♦^«7-a'-/ii<Jf, from f^tyyOfu. 

FUl. Tram. 1891-2-8. 23 



354 SOME GREEK ETYMOLOGIES. 

boars at bay) av^p&v rjSe kvpujp he^aTai KoXoavpTov loma hoxfiits '9 
aiffffovre vepl <Tif>iiTtv ayvvrotf vkift^, 

^pt9 * contest ' may mean * spear- work,' from *^pFt9f derv — 
Ablaut to dorv-, ddm, Sk. daru * piece of wood * : cf ., witl^ 
short vowel, derv- Lith. derwd * pinewood,' dorv- hovpara, don^- 
iopv * spear/ This *SrjpFt9 then became *hrjppi^^ tijpt^, thought 
by ordinary Greek laws it should have become ^Sepfi^f *^€vpi9; 
much as *fiifv<r69 became *fi7fvtf69 (Lesbian /t^i'i/o?), fiifvo^f though, 
by ordinary Greek laws it should have become *^i'<ro9, ^fieuro^. 
In other words, the law that rv became pp was earlier in 
operation than the law that erv- became epF, as the law that 
ns became w was earlier in operation (Brugmann Grr. 1. 611) 
than the law that ens became ei/v. 

€tKfj * at random ' seems a Litotes for eUorwt ' as we should 
have expected, simply, naturally,' Soph. Oed. Rex 979 cltcij 
KpaTitrrov ^rjyy oiray^ Svvatro ti9. It is then Instrumental of an 
Adj. ^eiKo^f seen in eiKo-^oXetif 'to aim at random,' cognate with 
€otK€ ' it seems.' 

ciTTctp * to say ' may mean * to clear up,' veikv-, cf. Sk. vt'e- 
' to sift, separate, examine.' 

€7ri'<rrafiat * know ' seems formed from the Adj . iTrKrrrjfttvv 
(Od. 16. 374) 'knowing,' literally * setting oneself fo a thing.' 
The Subst. iTrtim^firf • knowledge ' appears first in Hippocrates. 

kpfirjv€V9 * interpreter ' must be formed from *Epfiyp (Ace. of 
*EpitirJ9)f taken as a stem: Hermes was the god of speech, Xo^^ion 
(Lucian). So Ziyi/, Ace. of Zevs (II. 8. 206), being taken as a 
stem produced in the Tragedians the forms Z^i/a Z71/09 Zrfvi. 

^rpov * abdomen ' (* wind ' in our pugilistic sense) stands to 
Old Slavonic vitrii * wind ' as Lat. venter (see Etyma Latina) 
stands to ventus: yrpov is from (a)Ye-, urj^j with a termination 
-trom, as venter , stem ventri-, is from vent- (a Participial form 
from the same root, Brugmann Grr. 1. 612), with a termination -ri-. 

KoWaia * a cock's wattles' may mean * beauties,' *Ka\Xaio9 Adj. 
from *Ka\\y by-form of ica\\o9 (as evxy of ev^o^), 

Kkoto^ k\w69 * wooden collar,' i.e. *ic\a>/"-i-os, shows the same 
stem k|v- as Lat. cldva * wooden club.' 

kwXDu) * hinder ' is a Dissimilation for *kv\viv (as xivicvtv * bowl ' 
for *icvicvu), Lith. kuktu) : with a short vowel the root appears in 
KvWotv * cripple,' and Sk. kunis * crippled in the arm ' (Fortunatov 
B.B. 6. 216). 

\ao9 ought in Ionic to be Xy/o'?, as it is in Hipponax, and perhaps 



SOME GREEK ETTM0L061ES. 355 

ciice was ill Homer (Monro, Horn. Gramm.' p. 390). From \7fd9 I 
'would deduce (a) Xrflov * crop,* the produce of ' common ' land, and 
ijf) Xfjt^ Xfftrf Xeia * booty,* public property before it was divided 
smong the combatants, cf. Xela^ uBa<rra Soph. Aj. 54. 

fiardu) 'linger' (II.) and ^dnj 'folly' (Tragg.) are from mn-tos 
Tart of fieyw, with the same transition of meaning as appears in 
3ng. dwell and dull. 

fieiwv ' less,* for *firi'ju)v (G. Meyer 391), goes with Sk. mo- 'to 
'measure,' and so means ' more measured,' fierptwrepo^, not so 
::imraen8e. 

fiTf-yvWf Dor. fia vvw 'make known,' is from mn- Ablaut of mn- 
:^ Lat. mens, Sk. maiis ' thought.* For the transfer of meaning 
£t>m ' think ' to ' declare ' see on aetSiVf p. 4. 

fjLOf^o<rT0K09f epithet of Eileithyia in Homer, of Artemis in 

'3*heoeritus, cannot mean, as Brugmann Grr. 1. 204 makes it, 

* caosing pangs,' from ♦/40701/9 Ace. Plur. of fioyo^ : tiktw is not 

x:ised metaphorically in Homer, and such a use would be peculiarly 

inappropriate in connexion with the occasion. Liddell and Scott 

^^nghtly translate it 'helping women in hard childbirth,' protectress 

"^^ojv fior^i^ TEKovawv \ tho fii'st clcmeut is an Adverb *fio*^'o^ (with 

^lie same termination as wdp-o^) from a stem /io7-, whence with 

^ll.ocative ending, and the same 9 as in lifupi-^f we get fi6yi9 ' with 

^.ifficulty' (accented like the Subst.), and, with a determinative 

^ «- (see on Oetrtparo^y p. 2), fi6yo9 ' labour.' 

fivwyfr 'gadfly, goad' means ' flylike/ stinging as a fly does: 

^:roin fiva, the Attic form of fwia (Theophrastus, Hist. Plant. 5. 77, 

"^^here Liddell and Scott wrongly make it the name of a plant), 

'*^- u^a, Prellwitz derives it from fwta in the sense of ' buzzing ' ; 

^xit this will not suit the meaning ' goad.' 

f^eoXt/? ' fresh ' is not a compound of dXiaKOfiai^ but a by-form of 

^€aXo9 (as hayftiXi'i^ is of hayfriXo^y aXovprfy^ of dXovpf^o^)^ from a 

^^"ubst. *i'€a ' youth,' whence also veavi^, veatfla^. For the termi- 

^^«tion cf. diranjXd^, ffiyijXd^, 

tfovff09=*v6ir-Fo9 (Eretschmer K.Z. 31. 471), which I would 
Connect with v€{ir)ofiai 'come': thus Od. 9. 411 vovffo9 A<o9 
^*^ean8 'the visitation of Zeus,' and Soph. Ant. 421 Oeia voao^, of a 
"^^hirlwind, ' the visitation of heaven.' 

irtupXa^etv ' to foam ' must be from an Adj. ♦7ra0\o9 (or *7ra0\a9, 
^^^ lyvfivd^ beside 71/^1/09), from the same root as ire^t^ 7rofK/>6Xv^ 
\ l>ubble.' It cannot be, as Prellwitz makes it, a Reduplication 
"«m €<pXaBov * they burst,' which would give *7rai<pXd^€iy (Brug- 



356 SOME GREEK ETYMOLOGIES. 

mann Grr. 2. p. 1084) : Tra^txpaivo} 'shine' must be formed on an 
Adj. *7rafi'<f)avii9 ' all shining/ whence also ira^Kpavowi/, while 
irairralvu) * look round ' must be from a root kvnkyth- (or whatever 
the last letter may be), kvenkvth-, whence, without the nasal, Sk. 
eaksh' * to see ' (Fick B.B. 18. 134). — So TerpefLalpu) must be from 
an Adj. *-T€'Tpefiav69y not straight from rpifiiVf or we should have 
*Tnp€fiaiyu)f like riratviv ; and rerpaivu) * pierce ' from an Adj. 
*T6-T^ai/os, or we should have ri-rpaivu) (the form used by Theo- 
phrastus). 

7r€7rvvfi€tf09 ' wise,' i/6ov TreTn/vffOai ' to have understanding ' 
Od. 10. 495 (see Goebel, Homerische Blatter, p. 24), go with 
wonriwia ' am busy,' the common idea being that of strength : 
the root is kvneu-, Old High German pi-hniutan 'to glorify,' 
9ri/vf * meeting-place of the Ecclesia,' quasi * enclosure, strong- 
hold.' Thus wviw * blow ' must originally have been used of 
the wind blowing strong, and then transferred to the breathing 
of human beings. 

jreTmjtv^' Bi* affOeyeiav koi SetXiaif TrcjrrwKw^f says Hesychius : 

in Homer the word always means 'fallen,' ^.y. Od. 14. 474 
ifjro T€VX€<ri Tcirnfun€9 xeifieOa * WO lay on the ground, with 

our shields over us,' and II. 2. 312 (of young sparrows) ir€7a\oi9 
vTTo^rejTTj/arres * at the bottom of the nest, under the leaves.' 
It has no more to do with in/fffffa) * crouch ' than has KaTairrt'fniv 
in .11. 8. 136, of horses falling under the car : horses do not 
crouch down when they are frightened, but struggle to get away. 

9r»77t; * fountain ' must go with 7ri;709 * big,' the Homeric 
epithet of horses and waves, and Sk. pajrds * strong ' : it means 
a place where the water is strong enough to force its way out. 

7rl0rjKO9 vlOwv * ape ' must go with mOo^ * jar,' and means 
'rotund, pot-bellied.' 

Tpoxvv means * wholly ' in Homer, as it is allowed to mean 
in Apollonius Ehodius: it has nothing to do with 701/t; (which 
would not account for the x)>^ ^^ H- 9. 570 irpox^n) KaOe^ofiitn/ 
means 'sitting right down,' not 'sitting on her knees,' which 
would be an impossible feat. I would deduce the word from 
*7rp6£, formed from 7rp6 as irepi^ is formed from jrepi^ and, I 
think aTra^ from *a7ra Instrumental of ♦asro?, t.^. spikYOS, from 
8q|l- 'together' (see 1, C) : for the sense of. Lat, prOrsus 'utterly,' 



> Sk. prajnus^ quoted by Fick, Worterbuch* 1. 432, is unauthenticated, and 
at best only means * bandy-legged.' 



SOMB GREEK ETTMOLOOIES. 357 

from prd. The second element is the -nn which appears in 
Cyprian omt 'this,' Arcadian raw 'these things,' Goth, thannu 
* 80 then,' and I think waw * altogether ' {Le, tt'^-w^ the first 
element going with Lat. pen-ttus, Sahler K.Z. 31. 371), see 
Persson Idg. Forsch. 2. 251 : *7rpd^-w became wpox^ by De 
Saussnre's law, Mem. Soc. Ling. 7. 90, as ^\v^'V09 (cf. Zend 
raokhshna ' shining ') became Xvxt^o^. 

prjr^o9 * rug * (Hom.), peyo^ (Anacreon), pe^iv * dye' (Epicharmus), 
and a\o-vp^y9 * dyed with sea purple ' (Att., see Schulze 498, «.), 
must go witii pyiTffw 'beat the ground' (H. 18. 571), cloth being 
stamped on to make the dye penetrate : so German walken ' to 
clean cloth ' is the same word as Eng. walk. The root must be 
"▼reg-, Ablaut vrj- in pa<T<Tw ' push ' : Sk. raj- ' to be red ^ cannot 
\)% connected, if only because the meaning would be too narrow. 

aarimj * chariot * may be a Persian word, standing for *^ariv7j 
^as ffarpa7r7f9 stands for *(aTpa7r7f9, Old Persian khshatrapdvafi' 
• viceroy '), and going with Sk. kshat-tdr- * charioteer.' 

ffavpanyp 'spike at the butt-end of a spear* must mean 'twirler,' 
^^Tom a Verb ^travpou), itself from a Subst. travpo^, originally, I 
"Vrould suggest, meaning ' a stirring-stick,' and hence coming to 
^:Kiean ' a lizard,' which when motionless looks like a piece of wood. 
^I'his ffavpo9'=*rFap-Fo9f from a root tvr-, whence also Lat. trua 

* stirring-ladle' (on the ru see p. 11 med.), orpvyw ' urge,' orpaXiw^ 

* quickly ' ; Ablauts (1) tur- in ropivri ' ladle,* ^ a Dissimilation for 
^^rvpvinf, as KOKKv^ is for ^icvKKv^f Lat. cuculus ; (2) tvor- in Old 
^^^orse thvara ' stirring- stick ' ; (3) tver- in Ags. thviril ' chum- 
Ife^andle,' Eng. twtrl^ Old High German dtoeran 'to mix up,* Sk. 
^^ar- ' to hasten.* From a by-form stver-, stnr- (see Schrijnen, 
^^henom^ne de 1' S Mobile) comes trrvpa^j p. 12. 

ffuffiMf which in Homer always means ' dead body, carcase,' 
^^ay go with <ru)9 ' safe,* and mean ' remnant, what has escaped being 
^Qten by dogs or birds ' : H. 3. 23 ivtrre Xewy ix^-P^ fie^akijg ivl 

r6affat9 ' being,* iirn6ffffai9 ' finding ' (both in Pindar), seem 
"^^^ point to an Aeolio Verb *76aiTafii ' I do so much/ from roauo^, 

ipvXa^ ' guardian * may originally have meant ' the man in 
^lie bouse/ oiKerrf^, bhn-l- being Ablaut of bh5a-l- in Old Norse 
^l 'lair' (Wiedemann Lit. Praet. p. 137, despite Kluge 



* Fick, Worterbuch*, 1. 499, adds rVp6t 'cheese'; but this is not made 
^y stining, though butter is. 



358 SOME GREEK ETYMOLOGIES. 

K.Z. 26. 97), bh5-l- in (fiivXevw *lurk,* 0afXa9 *in his den' (of 
a bear). So 0t;Xt; 'tribe' may originally have meant 'house' 
in the sense of Lat. getu, 

*X^PV^f used by Homer in the forms x^pv^t X^PV** X^PV^h must 
go with x^^P ^^^ mean 'belonging to a handicraftsman,' x^P*^V^i 
as opposed to a warrior. Thus it is used coDtemptuously, II. 1. 80 
fiaffiXev^i ore x*^^^"-*' ^^^p* X^PV*^ (' * ^^^ fellow'), Od. 15. 324 
oTa re toc9 ayaOotfft irapahpuyuxTi x^PV^^ > ^^^ ^^ ^ ^^^ I^euteri 
II. 14. 382 iffOXa fiev etrOXo^ ebvvey x^P^I^ (* those fit for an 
artisan ') Se ^ct/joi't BotrKev. In Od. 14. 176 oori X^PV"^ ^rarpov, 

' not like a handicraftsman beside his father,' we have a Genitive- 
Ablative of comparison, such as Brugmann Griech. Gramm.' 

183 fiuds in Thucydides' 7r6\ep,ov a^toXoffanarot^ ru>v vpoyeyetnf' 

fievtvv 'most notable in comparison with those before.' The 
Comparative of *x^PV^ ^® x^P^^^^ ^^ X^V"^" * more fit for an 
artisan,' or, as in II. 14. 382 above, 'more like an artisan.' But 
in another view battle was called ' handwork,* x^Pf^V (with the 
' reduced * root xj-) ; and with this I would put x^pp^^^<^^ * a stone 
used in battle as a missile.' 



359 



XIL— RARE WORDS IX MIDDLE ENGLISH. By 

the Rev. Prof. Skeat, Litt.D. 

[Read at a Meeting of the Philological Society, Friday, June 3, 1892.] 

Abos. For this word, see Boose. 
Aker. See Ker. 

Boose, a cow-stall. The earliest quotation in N. E. D. is dated 

1440. But it occurs a century earlier, in Allit. Poems, ed. Morris, 

B. 1075 ; where it is misunderstood, and ignored in the glossary, 

owing to the preceding indef. article being written closely against 

it, as is common in M.E. The mysterious word ahos means a hos, 

i.e. a boose. It makes excellent sense. Speaking of Christ's 

3iativity, the author well says: **Wa8 never so blisfull a bower 

«s was a boose then, Nor no shroud-house so sheen as a shepen 

there." Shepen is a stable. 

Borken, barked. It is curious that Stratmann's Diet, does not 
happen to give the pp. of the verb herken, to bark as a dog. Yet 
darken, which is the right form, occurs in the first line of bk. i. 
;pr. 5 of Chaucer's Boethius. It so happens that it is missed in 
^r. Furnivall's index, and Dr. Morris's text has the corrupt 
:>^ading broken. Matzner gives an example of borken from the 
IKing of Tars, 1. 400 ; but he misses the example in Chaucer. 

Bredes. In Allit. Poems, B. 1405, we have: ''Bumes berande 
^he bredes upon brode skeles," i.e. men bearing the roast meats 
^pon broad dishes. I note this because bredes is not in the 
glossary, and the side-note says it means '* bread." See brede in 
Stratroann. 

Cheyisaanoe, resource. This is given in the New Eng. Dic- 
tionary as the right reading in the Bom. of the Eose, 3337. I 
tlraw attention to it because it afiPords us a conclusive test as to 
the genuineness of the Eowley Poems. It so happens that all the 
existing editions have cherisaunee, by mistake, though the original 
I'rench text has ehevisaunee. The editors thought it meant 
« comfort/' and so explain it. Hence, by a second misprint, arose 



360 RARE WORDS IN MIDDLE ENGLISH — PROF. SKEAT. 

the form oherUaunie, and even cherteaunei, duly explained as 
** comfort " in Kersey. Ghatterton fell into the trap, and began 
his poem of ^lla with " Some eherisaunei 'tys/' i.e. it is some 
comfort. 

Beooped. Rightly explained in Stratmann, with a reference 
to Rom. of the Rose. It means ** cut, or slashed/' said of shoes. 
The explanation in Halliwell is wrong. He refers us to eoppidy 
which he explains by 'peaked.' Under eoppid, he refers to 
eouped in P. Plowman ; but this means ' slashed.* Of. also coped^ 
Libeaus Desconus, ed. Kaluza, 1. 143. 

Beg^e. A copy of the Romance of Sir Degare occurs in the 
Percy Folio MS. iii. 26, where the name is explained as meaning 
" almost lost." Dr. Furnivall notes that an O.F. degarer answers 
to Low Lat. deuagari, as deg aster does to detiastare. This does not 
work out correctly ; for, by the same rule, degarer would answer 
to Low Lat. deuarari, as the g is for u. But the fact is, that, 
as in other cases, the real prefix is des- (for Lat. </««-), which 
became de- in later French (mod. F. di-). The full form would 
be des-garL This only differs from O.F. es-gari in the use of 
the prefix deS' instead of es- (Lat. ex) ; and we know that these 
two prefixes were of equivalent value. The O.F. esgare. meant 
lost, abandoned, strayed, out of the way ; cf . mod. F. egar^ 
lost. The English explanation ''almost lost," just represents 
a not uncommon sense of esgari, viz. strayed, applied to one 
who has lost his way for a time, but may hope to recover it. 
The etymology is now apparent, viz. from the prefix des-, de- 
(Lat. dis')t and the verb garer, to take heed, from the O.H.G. 
tvardn, to observe, heed, be aware of. The Eng. tvarg is from 
the same Germanic root. Cf. Ital. sgarraiOf mistaken, from the 
same source. 

Bray, a squirrel's nest. This word occurs in Drayton's Quest 
of Cynthia, st. 51 ; W. Browne's Pastorals, bk. i. song 6 ; and 
in Cowper, in a piece called A Fable. The A.S. spelling would 
be drag ; cf. dag for A.S. d4Bg. It seems to me that the sense 
of * nest ' would very well explain a passage in B6owulf, 1. 757, 
where it is said that Grendel was scared by Beowulf, and wanted 
to get away to his own haunt ; ** wolde on heolstor fleon, s^an 
deoflan gedrag^^* he wanted to flee to his hiding-place, to seek 
the devils' dray. The explanation of gedrag in Grein is not 
at all clear. There seem to have been two forms, gedrag and 
gedreag, which may have been firom different roots. 



RARE MTORDS IN MIDDLE ENGLISH — PROF. SKEAT. 361 

Eynes. This form appears in the glossary to the AUit. Morte 
Arthure, with nine references. Wholly failing to understand 
it, I applied to Mr. Bradley, who correctly informed me that 
it is a ghost- word. The n should be u ; and all the forms quoted 
are variants of the mod. E. eaves, a border, margin. See evese 
in Stratmann, and eaves in the New English Diet. 

Fausere, Fasonre, Yasnre. In Libeaus Desconus, ed. Kaluza, 
1. 1919, we have a curious and unexplained word. The passage 
relates the effects of an earthquake : — 

" The halle-rof unlek, 
And the faunsere ek, 
As hit wolde a-sonder," 

i,e, the hall-roof unlocked or came apart, and so did the faunsere, 

as if it would part asunder. The MS. can also be read asfauusere, 

which is, in fact, a better form; and the various readings give 

Qs the forms fasoure, vasure. I wish to suggest that the 

word here meant is an old form of the mod. F. voitssure, for 

which Littr6 gives the old forms vousure, vossure. In mod. E. 

architecture, the corresponding term is voussotr, applied to a 

somewhat wedge-shaped stone used for forming an arch. The 

proper sense of the word is the curvature of an arch, or the 

vaulting of a roof ; Gotgrave explains vousure as "a vaulting 

or arching.*' The point is, that it gives precisely the very 

cense required; ''the hall-roof fell apart, even where it was 

made of vaulted stonework." The sb. is formed as if from a verb 

wousser, answering to a Low Lat. *volutiare, from uolutus, pp. of 

-moluere; that is, it is from the same root as vault, and has much 

The same sense. 

Ferannt. The gloss, to Morte Arthure explains feraunt as 

•seemly.' But Bradley rightly explains it as 'iron-grey,' O.F. 

^errant, as an epithet of a horse. Hence it is not from A.S. faran 

* to go,* but from Lat. ferrum * iron.' Ferrand occurs, in romances, 

^8 a name for a horse ; from its colour. 

Pnatted. In Weber's King Alisaunder, 6447, there is a des- 
csription of a monstrous race of men, with very long faces, and 
^ars an ell long; " &nd fnatted nose, that is wrong." The word 
la misprinted fuatted, both in the text and glossary ; but such 
^ form is imposj^ible; there is no such diphthong as ua in M.E. 
Wrong means simply distorted, or crooked. As to fnatted, it is 
the Danish fnattet, given by Molbech, meaning afflicted with 



... * 



•y "■ ' 



862 RARE WORDS IN MIDDLE ENGLISH — PROF. SKEAT. 

a disease called fnat^ which was a skin disease, a kind of itch. 
So in Swedish dialects, fnatta is to scratch a place that itches 
(Rietz). 

Fodding^. In Kyng Alisaunder, 1. 48, we find: " Wyse men fond 
also there T welf foddin^ to thes yere, The yere to lede hy right 
ars (arts)." Weber* s explanations are seldom right, but in this 
case he has seen his way. He makes f adding to be the A.S. 
fadung^ disposal, arrangement; hence, a division of the year, 
a month; and he explains to lede by "to guide." This explana- 
tion is borne out by a passage in the A.S. Leechdoms, quoted 
in Bosworth and Toller, s.v. gefadung. Cf. fade, ready (?), in 
Sir Tristrem, 1. 153. The word appears neither in Matzner nor 
in Stratmann. 

Fuatted. See Fnatted. 

Gele. This word occurs in the Pearl, 931. Morris explains 
it by ** spy, see " ; this suits the context, but there is no authority 
for it of any kind. Mr. GoUancz says it either means ** hesitate, 
delay," from A.S. galan; or perhaps it is from Icel. gala, to 
comfort, to soothe. The latter will not suit the context. It 
is clearly from A.S. galan, in the intransitive sense of ** tarry." 
The passage is — ** And by thyse bonkes ther I con gele, I se no 
^ygyig nawhere aboute," i.e. And wherever I have lingered 
beside these banks, I see no building anywhere. (The MS. 
needlessly repeats And before / se,) 

Oessenen. In Morte Arthure, 2521, we find the line — **He 
bare, gessenande, in golde, thre grayhoundes of sable." The word 
gessenande is not explained. I take it to be present part, in -ande 
from a verb of which the stem is gessen-, and the infin. mood 
is gessenen. It must be French, because it is a term in heraldry. 
I take it to be a verb formed from the M.E. sb. gesin or gesine, 
Cursor Mundi, 3906, Gov. Mysteries, p. 150. This is adapted 
from the F. gesine, which Cotgrave explains as * a lying-in,' 
though the related verb gesir merely means to lie down. If 
we give to this verb gessenen the same simple sense, it makes 
it equivalent to the F. coucher; and the pres. part, becomes 
equivalent to the pres, part, couehant, lying down, a well-known 
heraldic term. Then the sense becomes : *' he bare, on his shield, 
or, three grayhoundes couehant, sable^ This is perfectly in- 
telligible and consistent. 

Harrawnte. This word occurs in the Morte Arthure, 2449. 
It is necessary to quote the passage. 



RARE WORDS IN MIDDLE ENGLISH — PROF. SKBAT. 363 

* Thane come the herbariours, harageous knyghtez, 
The hale batelles on hje harraumte therafter.' 

Sarageous is eaid to mean 'violent/ though its etymology is 
not clear. JSarrawnte is supposed, in the glossary, to be a verb. 
I take it to be really a present participle, representing the O.F. 
haranty pres. pt. of harer, which Cotgrave explains by * to set 
on a dog.' The original sense of harer was simply *to cry out 
aloud, to shout,' as it is merely the O.H.G. haren, *to cry aloud, 
to shout,' in a French form. I explain the passage thus : ' Then 
come the harbingers (or vanguard), fierce knights ; and the com- 
plete battalions (or squadrons) come after them, all shouting 
aloud.' On hye^on high, aloud. 

Himland. This word, spelt hymhnde^ occurs in the same line 
as hope, q.v. It is clearly a present participle, but has never 
been explained. I would explain it as ' abounding in hummocks,' 
from the same root as hummock. At this rate, himland hilU would 
mean rolling hills, hummocky hills, which suits the whole context 
admirably, as the thing described is a rough ride over uneven 
ground and various obstacles. Probably the vowel t is due to 
a mutation of u ; compare E. pit from Lat. puUtis, I think 
we may compare it with E. hump and Low G. hiimpel, a little 
heap ; Lat. cumulus^ Gr. icvfia ; Ital. cima^ a mountain-top. 

Hope. In the Morte Arthure, 1. 2503, we find: "Thorowe 
hopes and hymlande hillys and other." Mr. Bradley explains it 
by * valley,' which is practically right, and refers us to A.S. hop, 
with short o. I wish to point out that there seems to be two 
distinct hopes, one with original short o, and one, perhaps, with 
long 0. Mr. Bradley has got hold of the right one, etymology 
and all; but Grein and Bosworth are at fault. Sievers has: 
** hop, recess," which I take to be the right A.S. form, but with 
a wrong sense. I would set it all right thus. (1) A.S. hop, strong 
neuter sb., pi. hopu, explained by Jamieson as in use in Scotland 
in the forms hop^ hope, and as meaning a sloping hollow between 
two hills, sometimes a rather deep glen. The o is short, because 
the plural terminates in u ; ^ee Sievers, A.S. Gram. § 239. Only 
two A.S. examples are known, both plural, and both in Beowulf, 
viz. fen-hopu, or sloping hollows with a fenny bottom, and moT' 
hopu, sloping hollows on a moorside. This explains the form 
hopes in the present passage, and doubtless occurs in some 
English place-names, such as Hope, near Castleton-in-the-Peak, 



{ 



364 RARE WORDS IN MIDDLE ENGLISH — PROF. SKBAT. 

Hopton, etc. The other word I take to be of Norse origin; 
it also occurs as hope in Lowland Scotch, but it has quite a 
different sense; it means a recess or haven, and occurs in place- 
names on a sea-coast, such as Kirkhope in Orkney (Yigfusson). 
This may well be the Icel. hdp (with long o\ a recess, a haven ; 
and may be connected with E. hoop^ i.e. a circular bend. I think 
we ought to keep the words separate, and to correct the quantities 
given in Orein and Toller. 

Ker. In the glos. to Sir Gawain and the Grene Knight, we 
find ^^ Kerre^ rock, 1431." Stratmann explains it better by 
** marsh." The line is : " In a knot, bi a clyffe, at the kerre-syde,^* 
i.e. at the side of the marsh ; the same marsh, or pool, is called 
a floiche (flush) in the preceding line. I notice this because the 
very same expression occurs again, only a few lines above, at 
1. 1421 ; but it is there printed aker syde, with the explanation 
of aker as "field, plain" in the glossary. But the alliteration 
proves that the word begins with k. The line is — ** Sone thay 
calle of a quest in a ker-syde." See Carr (2) in the New Eng. 
Dictionary. 

Keve. This word occurs twice in the Pearl, 320, 980. Strat- 
mann suggests " turn," but with a query. Mr. GoUancz points 
out to me that it is the Icel. kej)'af to dip, to dive, to sink. The 
sense seems to suit sufficiently. In the former passage, the 
dreamer is told that, before he can go to heaven, his '^ corse in 
clot mot calder keve," i.e, his corpse, made colder, must sink 
into the clod. The latter passage is less clear. The dreamer 
lifts up his eyes, and sees, far above, the heavenly Jerusalem, 
brightly shining "byjonde the brok, fro me warde keued," t,*. 
beyond the brook (of death), that dipped down away from me. 

Lanen, pools ; K. Alis. 3856. Given in Stratmann, s.v. La^e, 

Laye. This word occurs in Morte Arthure, 3721 ; and is not 
explained. But it is the same as la^e, a pool (A.S. laya, a lake), 
explained in Stratmann flmd in Matzner, with references to other 
passages. 

Lorayn. In the Morte Arthure, 2462, there is a mention of 
** launces with loraynes " ; the word loraynee is not in the glossary. 
Halliwell and Stratmann give lorein, with the sense of **a rein." 
So here, launees with loraynes may very well mean lances provided 
with thongs for throwing them. Cf. Lat. lorum, a thong. The 
Eoman iaeulum was furnished with an amentum ; Ovid, Met. xii. 
321-3. 



RARE WORDS IN MIDDLE ENGLISH— PROF. SKEAT. 365 

Kaohes. In Morte Arthure, 2950, we are told how Sir Gawain 
attacked his enemies ; one of his feats was that he '* metes the 
machea of Mees, and melles hym thorowe," i,e. he meets (some- 
body), and smites him through. The ' somebody ' is here called 
''the maches of Mees," which has not been explained. When 
we remember that a was then pronounced as the a in path^ 
which only differed from the sound of ar when the r was 
properly trilled, we see that maches is an error for marches. 
This is the O.F. tnarehiBy given in Littre as the usual spelling 
of the word which exists in English as marquis. Again, 
Cotgrave explains that marquis meant, *' in old time, the 
governor of a frontire town." Hence "the maches of Mees" 
means the governor of MeeSf where Mees is the name of the 
town. Further, as the said marquis was a follower of the 
Buke of Lorraine, it is an easy guess that Mees represents a 
frontier-town of Lorraine, which happens to be Metz, as rightly 
explained in the Index. Hence we arrive at the romantic fact 
that one of the governors of Metz had the distinguished honour 
of being slain by Sir Gawain. Note that the ch sound of marches 
is still heard in the E. marchioness. 
Mansell, Masnel. See MasueL 

Kasnel. We must add to the M.E. Diet, the word masuel^ 
a little mace, a kind of weapon; O.F. maguele (Gode&oy), also 
spelt massuelCf masuele. Cotgrave gives: Massue, * bl club,' which 
is the same word, without the dimin. su£Sx. It occurs twice, 
in Rich. Coer de lion, ed. Weber, 351, 5660. In the latter 
place Weber has spelt it masnelf with n for u ; and in the former 
place he actually has mansell, probably a misprint for masnell, 
"which means masuellf as before; In both places the line scans 
1)etter with the right form. In 1. 35 1 read : ' Forth he took 
« masuelf^ in three syllables. In 1. 5660, scan the line : ' By 
that o I ther syde | his mas \ UeV The Low Lat. form would be 
muucucella^ as it is a fem. form ; see Maxuca in Ducange. 

Mes, a good position for taking aim. This word occurs twice 
in the Bom. of the Rose, though it does not appear in the editions. 
Xi so happens that in both cases (11. 1453, 3462) it occurs in the 
same phrase, viz. at good mes; and in both places it has been 
turned by the editors (except Thynne) into at goodness, though 
this yields no sense. Mes is a real word, and an old Anglo- 
French hunting term. It represents a Lat. missum, and signifies 
a good place for aim, a good place for a shot. Thus, in a lay by 



366 RARE WORDS IN MIDDLE ENGLISH — PROF. 8KEAT. 

Marie de France, entitled Gufpemar, 1. 87, a man tries to shoot 
at a deer. "Trair voleit, si tnes eust," i.e. he wished to pull 
the bow, if he could get a good shot. And, in Gaimar's Chronicle, 
there is a most interesting example, with reference to Sir "Walter 
Tyrrel and King William Rufus. It is said of Tyrrel that when 
**le grant cerf a mes li vint, Entesa Tare," i.e, when the great 
stag came within his aim he drew the how. The arrow missed 
the stag, as we all know, and killed the king. The E. phrase 
at good mes represents an A-F. a hon mes^ i.e. within one's aim, 
in a favourahle position for a shot. When this is seen, both 
passages become easy. In the former, the writer says that the 
God of Love came after him with a bow — 

" Right as an hunter can abyde 
The beste, til he seeth his tyde 
To shete, at good mes, to the dere." 

In the latter passage we have the lines — 

'• Suffre, I rede, and no boost make 
Til thou at good mes mayst him take." 

The original French has en hel leu, in a good position, in the 
former passage, and en hon point in the latter passage, with the 
same sense. 

Odam, son-in-law; King Alis. ed. Weber, 2081. Weber ex- 
•plains it wrongly, though rightly comparing G. Eidam. I note 
it because it is difficult to find in Stratmann, where it is entered 
under dthum. 

Faleis, Palis. There is a word in Chaucer's Boethius which 
does not seem to be explained with sufficient clearness. We know 
that M.E. pakis usually means 'palace'; but there is another 
paleiSf also spelt palis, in Chaucer, which Dr. Morris and Dr. 
Fumivall explain by ' pale ' ; and the Lat. original has fuUlum. 
* Pale ' is not quite right, nor does it explain the form ; the 
right sense is * a set of pales,* i,e, a paling, or a palisade, a kind 
of stockade. The word is not plural, but singular; it is the 
O.F. palis, paleiSf mod. F. palis; whence the verb paltBBer and 
the sb. palissade. The Low Lat. forms are palitium (whence 
F. palis) and palaeium (whence O.F. paleis), I find that paUis 
occurs three times in Libeaus Desconus, ed. Kaluza, 1556, 1791, 
1862 ; and in the first instance, at any rate, the sense of ' stockade ' 
suits better than 'palace ' ; the various readings in 1. 1556 present 



RARB WORDS IN MIDDLE ENGLISH — PROF. SKEAT. 367 

the forms palyMd, i.e. provided with a palisade, and palyd^ i.e. 
provided with pales ; which renders my suggestion highly probable. 
The word is not in Stratmann. The same word occurs, spelt 
palaySj in Gawain and the Grene Knight, 769. See Pyked below. 
Fartlet, a woman's ruff. I have not seen any satisfactory 
etymology of this word. In the Century Dictionary it is con- 
nected with partlstf a hen. I believe that the two words were 
originally quite distinct, but were gradually confused in spelling. 
Partlet, a hen, is spelt P&rteloU in Chaucer; and it is asseited 
in the Century Dictionary that this PertelUe was a feminine name. 
But the other partlet is remarkable for having, originally, no 
r in it. The M.E. form is patelet, which occurs in Bob. Henry- 
soun's Garmond of Gude Ladeis, st. 7, as the name of an article 
of female attire. This is precisely the O.F. paieleUe, given by 
Godefroy and Cotgrave. Cotgrave has it in the sense of a part 
of a bridle ; but Godefroy gives it as a dimin. of patUy properly 
a bird's foot, but occasionally used in the sense of a band of stuff, 
or a flap of a garment. I think the right etymology must be 
looked for in this direction, though I do not quite see through 
it. Perhaps we receive further light by consulting paitlatttB and 
paitclayth in Jamieson. Skelton has both patlet^ in his Magnificence, 
1. 2100, and partletteSy in his Maner of the World, 1. 163. In 
Fairholt's Costume, ii. 15, s.v. arming Doublet^ we have a re- 
markable mention (in 1513) of '^arming patlettB of white satten 
quilted and lined with lynen cloth, for my lord to wear uuder 
Iiis harness." 

Feohelyne. This word occurs in Morte Arthure, 1341 ; the 
sense being unknown. 

llie passage is one where a certain emperor threatens to deprive 
J^rthur of all his possessions. He says he will not leave in Paris 
(which then belonged to Arthur) so much as ''his parte of a 
pechelyne." 

The O.F. pescher, to fish, occasionally appears without s, like 
the Mod.F. pScher; see Godefroy. I take pechs to be from O.F. 
pecker f to fish; and I ex^loXn peche-lyne to mean ''a fishing-line," 
i,e. a line to catch fish with. This suits the context. Arthur 
is to lose all his possessions, and his men will have nothing to 
eat ; they will not even have a fragment of a fishing-line to catch 
a stray fish with. The threat is sufficiently expressive. 

Fisane. This word is also spelt pesane, as in the Morte Arthure, 
3458, and pusane. See pusane in Stratmann. Add to the references 



368 RARE WORDS IN MIDDLE ENGLISH — PROP. SKEAT. 

there given pysane^ LibeauB Desconus, ed. Bitson, 1618, and 
pitaine^ in the same, ed. Kaluza, 1708; and, probably, pesant, 
conjecturally explained by * head-piece ' in Fairholt's Glosuary of 
Costume in England, and dated 1579. It seems to have meant 
a gorget, or neck-piece, fastened below the helmet. Bradley 
proposes to connect it with the O.F. gorgerette pisainns^ mentioned 
in Godofroy, s.t. pisaitif i.e. an adj. formed from Fisa, in Italy. 
Gk>defroy also gives the adj. pisanes with the same sense, and with 
the example elms pizane, which I take to mean * helmet of Pisa.' 
I conclude that the word is really formed from the place-name 
Fisa, Milan was likewise celebrated for cutlery and armour; 
cf. E. milliner. Eitson's Glossary to his Met. Eomances gives 
a quotation from Grose : "3 coleretes pizaines de jazeran d'acier." 

Fotenere, a purse. In the Percy Folio MS. iii. 47, we find 
the spelling poteuere; and in the same, ii. 305, it is potewer. 
But the right spelling is potenere, whence the other forms result, 
by mistaking n for u, and then altering i^ to tr; as, no doubt, 
the scribes did. See Fatvtenere in the Prompt. Parv., and "Way's 
note; and the note on pawtenar in Skelton, ed. Dyce, ii. 105. 
Godefroy gives O.F. pautonniere, s.f. a purse ; which is clearly the 
same word. Ducange discusses it under FantonartUf Fanlaneria, 
and Fautoneria. He thinks it refers to a beggar's scrip; from 
O.F. pautonieTf a servant, beggar, rascal. 

Pyked. In Gawain and the Grene Knight, 1. 769, we have 
a description of a park ** with a pyked palays, pyned ful thik " ; 
and in the next line we are told that this ** palays " extended 
for more than two miles. This line has never been explained. 
Fyked and pyned are both explained wrongly in the glossary, 
and palays is not explained at all. Possibly the editor took palays 
to mean ''palace," flmd this threw him out. But palaces are 
not usually two miles long ; hence we must take palays in its 
other sense of "palisade" or ** fence"; see Falays above. Then 
pyked means furnished with pikes or spikes; see this meaning 
in Stratmann. Lastly, the y in pyned is short; it stands for 
pynned, i.e. pinned in, enclosed, fastened, penned up ; cf. pindar^ 
and see Stratmann. We know that this is right because, in the 
same MS., in the poem on Patience, 79, we find — *' Pynez 
me in a prisoun, put me in stokkes," where Dr. Morris rightly, 
as I think, explains pynez by fasten, or shut in, i,e, pen or pin 
up; though the sense "torture" is possible. Hence the line 
means that the park had a spiked paling all round it, the spikes 



RARE WORDS IN MIDDLE ENGLISH — PROF. SKEAT. 369 

being pinned or fastened very thickly or closely together. This 
gives an exact sense, and suits the context. The misspelling 
pyned is due to omitting a stroke over the n. 
Pyned. See Pyked above. 
Quemes. See Ternes. 

Badevore. This difficult word occurs in the Legend of Good 

Women, 2352. The only suggestion of any value is that given 

in Urry's Glossary, viz. that it stands for raa de Vore^ i.e. serge 

made at a place called Tore. Tyrwhitt remarks that "there 

is a town in Languedoc called La Faur; but I know not that 

it was ever famous for tapestry." Further investigation shows 

that this explanation is certainly correct; the difficulties that 

arise all vanish on examination. In the first place, as Urry's 

Glossary tells us, such phra.ses as JRas de Chalons^ and ras de 

Gennes, were really in use. My own difficulty was a phonetic 

one. I could not see how the b in the Old French ras could be 

ignored in the pronunciation ; particularly when we notice that 

this very word ra9 produced the E. word rashy as explained by 

I^ares, of which more anon. But Mr. Mayhew pointed out to 

me that, in O.French, s before a consonant vanished in the 

second half of the eleventh century. Hence, when r<M de Fore 

was used as a complete phrase, it regularly became radevore in 

^glish. The next point refers to the place Fore. This is 

dearly, as Tyrwhitt says. La Faur, in the province of Languedoc, 

«nd in the modem French department of Tarn, at no very great 

^listance from Toulouse. It appears that silk and serge are still 

Uade at this very place; see Engl. Cyclopsedia, s.v. Tarn, We 

"thus obtain, as the final result, that radevore means, precisely, 

** rash made at La Vaur." We have now only to enquire into 

'the sense of rash. The F. ras means, especially, serge or satin. 

The English raeh^ says Nares, is **a species of inferior silk, or 

^ilk and stuff manufacture." One of his quotations speaks of 

** velvets, satins, sylkes, rashe, and other stuffs " ; and the Century 

3)ictionary quotes from Middleton : " 't is good stuff, indeed ; 

It is a silk rash." This explanation precisely fits the passage 

in Chaucer. Philomela had learnt, in her youth, how to weave, 

in an embroidery frame, such rash as was made at La Vaur. The 

i^earest mod. E. equivalent seems to be ** serge." 

Sigh-clout. This word occurs in the Percy Fol. MS. ii. 323 ; 
See note 5. It is explained by Dyce in a note at p. Ixix. as a 
clont for straining milk ; from ««>, to strain. Dr. Fumivall remarks 

FhU. Trans. 1891.2-8. 24 



370 RARE WORDS IN MIDDLE ENGLISH — PROF. SKEAT. 

that be only knows sile in this sense. But ite, to strain milk^ 
is given in Hal li well, who, curiously enough, quotes the very 
passage here in dispute, and refers us to Palsgrave, who has : 
** I sye my Ike, or dense." It is a particular sense of A.S, st^an, 
M.E. sijetif to sink. 

Stele. In the Allit. Poems, C. 513, the author laments that 
some people don*t know the difference between their right hand 
and their left, nor yet between ** the stele and the stay re." 
Among these we have to include, very possibly, William of 
Shoreham; and neither Morris nor Stratmann have seen any 
difference. They explain both words as meaning the step or 
rung of a ladder. I believe this will suit some passages, and 
that the senses of the words were confused (as our author hint«i) 
even in the fourteenth century. The stay re is really the stair 
or step of a ladder. But the steles are the two uprights, the 
handles by which the climber holds ; called by Chaucer the stalkes. 
Steel in pro v. E. still means a long upright handle, as of a 
besom or of a pitch-fork. Stratmann (s.v. stale^ which, though 
differing in form, has the same senses) obligingly refers us to 
the very passage in the Ancren Riwle which settles the question ; 
there was no ambiguity at that date. At p. 354 we are told 
that ignominy and pain are the two stales of the ladder (the 
two leddre-stalen) that are upright to the heaven ; and between 
those stales {stalen) are fixed the tindes^ steps, or stairs. 

Stivour. (Not in Stratmann.) In Weber's King Alisaunder, 
2571, we have: ** Mury is the blast of the styvour^ Weber 
explains it as **an ancient wind-instrument," but it certainly 
means the player on such an in8tr\iment, just as, in the next 
line, luirpour means a player on a harp. Otherwise, his note is 
correct; the instrument was called estwe^ and is mentioned in 
the Eoman de la Eose, 21308 ; see my note to House of Fame, 
1218. See £stive in Godefroy. Cf. Lat. stipula^ in Virgil, 
Eel. iii. 27. 

Talle ne in tueh. In Allit. Poems, ed. Morris, B. 48, we 
read that a ragged man would be turned out of a nobleman's hall, 
and forbidden to enter again, on pain of being set in the stocks, 
**thagh neuer in talle ne in tuch he trespas more." I take 
this to mean : ** though he should never again do wrong either 
in tale or in touch," i,e. by word or deed. I see no difficulty, 
especially when we notice the curious uses of touch in Sir Gawain 
and the Grene Knight, by the same author, and further note 



RARE WORDS TN MIDDLE ENGLISH — PROF. SKEAT, 371 

thaty in that poem, the words tale and touch are alliterated, 
1. 1301. Dr. Morris explains ttich by 'cloth/ from the G. Tueh; 
but I decline to equate the Eng. ch with the Ger. eh in this way ; 
the M.E. for * cloth ' was touk^ i.e. if it be related to M.E. touker, 
a fuller (Stratmann, s.v. tuken). The editor further explains talle 
by tuly^ which means scarlet. This I cannot accept either. ' The 
spelling talU for tale is like the spelling waUe for wale, to choose, 
B. 921. 

Tayt. In AUit. Poems, ed. Morris, B. 889, the word tayt 
means "joy." The glossary says "fear," but see Stratmann. 

Teneling. This word, in the glos. to the Grene Knight, is an 
error for teueling ; see Stratmann, s.v. teveling, i.e. sport. This 
word is entered in Stratmann under tavelin^ and is easily missed. 

Temes. In Rich. Coer de Lion, 2009, is the line : * Ternei and 
qtiemes he gave him there.' Weber's Glossary indulges in a bad 
shot as to the sense : * Ternes and quernes, thrusts in fencing, or 
blows with the broadsword.' The context shows that both senses 
are impossible ; for King Richard, who dealt these blows, had 
nothing in his hand but a truncheon, which is neither a broad- 
sword nor suitable for thrusting. The fact is, these are terms 
in dice-play. Temes means double three, and quernes means 
double four; neither are in Cotgrave, but he gives * Qutnes, two 
cinks, or fives, on the Dice.' Littre, s.v. terne, quotes from 
Villon ; the passage shows that amhesas (E. ames ace) meant double 
ace, and ternes double three. The passage is jocular. King 
Richard, with his truncheon, gave his enemy a double three, 
and after that a double four. The fourteen blows near finished 
him ; "he thought he should be dead." He could not have 
survived fourteen thrusts of a broadsword, delivered by the 
Richard Coer de Lion of romance. 

Thacces. In the glossary to the Allit. Poems, we find " Thaecey 
a blow, C. 325." This is due to a most curious misconception. 
The alliteration shows that the word should begin with a vowel ; 
and, in fact, thacces is merely the two words the acces run together. 
Hence thacces must disappear from under M, and take its place 
under a. The line is — " For when thacces of anguych wacz 
hid in my sawle " ; i.e. when the attack of anguish penetrated 
to my very soul. 

Thnlged. In Gawain and the Grene Knight, 1859, we read: 
"Thonne he thulged with hir threpe, and tholed hir to speak." 
Thulged is not in Stratmann; and the glossary says: **thulged= 



372 RARE WORDS IN MIDDLE ENGLISH — PROF. SKEAT. 

tholged^tholed^ endured." This cannot be right, because thoUd 
occurs in the same line properly spelt; and we cannot thus 
account for the spelling thulged. The fact is, that ge represents 
a /-sound, resulting from a palatalised d ; cf . the frequent pro- 
nunciation of dsw as Jew. Thulgen represents A.8. thyldgian^ 
to bear patiently, from the same root as tholian. See gethyldigean 
in Bosworth's Diet. 

Tipen, to overturn ; cf. mod. E. tip up. Stratmann only gives 
ti'ppen, with short «. But the word should rather be tipen^ with 
a long «. In the only example quoted there is but one p, and 
the vowel is written y. ** Type doun yonder toun," i.e. overthrow 
that town ; Allit. Poems, iii. 506. So again in the Percy Folio 
MS., Death and Life^ 194 (vol. iii. p. 64): — ** Trees tremble 
for feare, and tipen to the ground." Cf. Lincolnshire type or 
tipe^ to tip up, in Peacock's Glossary, and tipe^ in Halliwell. 
These point to a lost Germanic strong verb, tlpan. 

Totez. In Allit. Poems, ed. Morris, B. 41, a ragged man is 
described as having *'his tabard to-tome, and his tote% oute." 
Dr. Morris says that totez is merely a form of "toes," which 
I cannot accept. Stratmann gives tote^ sb. ?toe; totez^ pple-y 
A.P. ii. 41. Of course the plural of "toe" cannot be 
past participle ; * pple * is probably a misprint for pi. (plural). 
The word is surely the Low G. tote^ a peak. Hexham has : 
''^ een Tote, a teat; de Tote van een schoen, the beake or lap of 
a shoe; een Tote-pot, a pot with eares, etc. Cf. M.E. toten, to 
peep out; Am ton toteden out, his toes peeped out, Piers. PL 
Crede, 425. I translate tote by extremity or end; the sense 
is, "the ends (probably of his toes) peeped out." It is a mere 
coincidence that toe begins in the same way. A.S. tOtian, to 
peep, is quite distinct from td, toe. Again-tote^ a peeping behind 
one, occurs in this same poem, B. 931. 

Trashes. In Allit. Poems, ed. Morris, B. 40, is the line — 
"With rent cokrez at the kne and his clutte trasches." "Rent 
cokrez " are rent cockers or leggings. Clutte means clouted, 
patched, as explained in the New E. Diet., s.v. Clouted. Trasches 
is explained, with a query, by Morris and Stratmann, to mean 
trousers. Surely the words cannot possibly be identified. Trashes 
is the plural of trash, still in use ; and one sense of trash is rags. 
Cf. Swedish trasa, a rag, a tatter ; slita i trasor, to tear to tatters ; 
hans kldder aro utsletne i trasor, his deaths are worn out to 
rags or tatters (Widegren). 



t 



RARE WORDS IN MIDDLE ENGLISH — PROF. SKEAT. 373 

Treieted. Not in Stratmann in the sense required. It occurs 
in Gawain and the Grene Knight, 960. The i stands for 7; the 
sense is " variegated " ; see Burguy, s.v. tresgeter. In the 
glossary it is misprinted treieted; but the sense given, "adorned," 
is correct. 

Troched. Both in Allit. Poems, B. 1383, and in Gawain and 
the Grene Knight, 795, some well-built towers are described as 
troched. The glossary tells us that this is an architectural term 
of uncertain meaDing; Stratmann suggests "ornamented." The 
word occurs frequently in the Venery de Twety, in Reliq. Antiquee, 
i. 151, where it is applied to a hart who has thrown out tines 
from the tips of his antlers. The divisions of the antler are 
given in this sentence: **whan an hert hath fourched, and then 
auntlere ryall and surryall^ and forehed one the one syde, and 
troched on that other syde, than is he an hcrt of .X. and the 
more." The engraving of an antler in the Century Dictionary 
helps us here. The fourches, or forks, are the long projections 
on the one side of the horn, and the troches are the clustered and 
shorter projections on the other, near the tip. If we now turned 
to Cotgrave, we find : " Teste de cerf trochee, troched, or whose top 
is divided into three or four small branches." And again : 
** Trocheure, the troching on the top of a deer's head, or the top 
troched." Hence troched, as applied to a stag's horn, means tufted 
at the tip with small tines. It is a term of the chase, and of 
French origin. I do not suppose it was a term in architecture, 
but was applied to a tower poetically. The sense is, clearly, 
'* adorned with small pointed pinnacles." 

TJnkek. This word occurs in the Seven Sages, ed. Weber, 955. 
By comparison with 1. 2251, it is seen to be an error for unlek, 
i.e. unlocked, opened. Weber explains it by * unopened,* whereas 
it means precisely the contrary, as the context shows. The 
spelling with k is due to anticipation. Similarly we find eikerklik 
for sikerlik, in the same, 1373. 

Windren, to paint up or trim the eyebrows. In the Rom. of 
the Rose, 1018, we read: **No wyntred hvornvA hadde she." This 
is one of the words which Tyrwhitt does not explain; nor is 
it in Morris or Stratmann. Only two lines below the infin. mood 
occurs as wyndre, " It neded nought To tcyndre hir or to peynte 
hir ought." And this is the better form. This verb to mndre 
represents guignier in the P. text, of which an older form must 
have been wignier; and it is obvious that the E. word is merely 



374 RARE WORDS IN MIDDLE ENGLISH — PROF. 8KBAT. 

the O.F. toignier done into English, and treated with an excrescent 
d, after n, just as the F. »on has become E. iound. We thus 
get the sense at the same time ; for guignier meant to disguise 
or paint or trim up, and is usually joined with farder^ to paint, 
with a like sense. Hence the passage means that the damsel 
did not paint or trim her eyebrows, because it was needless 
for her to do so ; they were not capable of improvement* 

Writhe. I wish to draw attention to a passage in Chaucer's 
Boethius, bk. i. met. 4, where both the best MSS. are wrong. 
The reference is to Mount Vesuvius, that (according to MS. C, 
piinted by Fumivall) *^writith owtthorw his brokene chymynees 
smokynge fyres." FumivalPs index explains writith by **work- 
eth," which is impossible, and was obviously suggested by the 
reading mreheth in the MS. printed by Morris. In this case, 
as in other places, the black-letter editions by Caxton and Thynne 
have an older reading, viz. tpritheth, which is perfectly correct. 
The reading writith is due to the mere omission of an A; the 
reading wircheth is due to the transposition of r, thus giving 
wirthethf with the usual confusion of t and ^, thus giving wircheth, 
The sense is that Mount Vesuvius writhes or twists its smoking 
fires out of its broken chimneys, which is very expressive. How 
Ch. came to use writheth is obvious when we refer to the Latin 
original : Torquet fumificos Vesevus ignes." It is, perhaps, worth 
while noticing that Chaucer again uses writhen away to translate 
Lat. detorqueri ; bk. v. pr. 3 ; 1. 4452 in Morris's edition. I 
have lately found that MS. li. 1. 38, in the Camb. Univ. Library, 
likewise has the correct reading wrythith. 



375 



XIII.— THE OLD ENGLISH ALLITERATIVE LINE. 
By Prof. H. Frank Heath, Ph.D. 

IMead at a Meeting of the Fhiioiogieal Society^ June 2, 1893. j 

The references to Beowulf throaghout this paper are taken from M. 
Heyne^s edition, 1888; the references to other poems from Grein^s Bibliothek 
der A-S. Poesie, 1857. 

When Dr. Pumivall first asked me to read a paper before this 
Society I had recently formed a new theory about the construction 
of the Old English "Expanded Lines," ** Longer rhythms" as 
Guest caUs them, Streckversey or Schwellverse^ as they are named 
by German scholars. It struck me that I might make this theory 
the subject of my paper, and Dr. Fumivall approved of my 
suggestion. But as time went on, and I began to think my 
paper out, I remembered that I held heretical views about the 
construction of the Normal alliterative line, views which, so far 
as I am aware, are only shared in England by one friend and 
fellow- student, and that it would therefore be necessary for me 
to at least give an introductory section upon the Normal line, 
as I conceive it, or rather as my most honoured master, the late 
Prof. Ten Brink, conceived it to be constructed. 

Then came the publication of Prof, ten Brink's fragmentary 
posthumous contribution to Paul's "Grundriss" upon O.E. literature, 
containing a very condensed account of his theory of O.E. metric, 
and without the explanatory notes which he had intended to add. 
I found upon enquiry that men of the very highest rank amongst 
English philologists had found some difficulty in understanding 
this cramped and incomplete presentation of the case, and I felt 
that, this being so, I could not better honour the memory of my 
master, nor better advance the cause of truth than by an ex- 
position of his views, as I understand them to have been, after 
hearing a course of lectures from him on the subject, and taking 
an active part in the Seminar which he held in connection with 
it. The first part of this paper, therefore, lays little claim to 



376 THE OLD ENGLISH ALLITERATIVE LINE. 

originality, other than that small measure of it which lies in 
the method of presenting the subject. 

First, then, I hope you will allow me to remind you of one 
or two general truths, about which there is no doubt, but of 
which we must grasp the full importance, if we are not to 
wander from the right track into the many pitfalls which beset 
the study of this very difficult subject. To begin with — What 
is the essential difference between Prose and Verse? The 
difference, of course, is that verse is speech with a definite 
Rhythm^ whilst prose has properly no definite Rhythm; though 
corrupt prose styles such as that found in some of Aelfric's 
Homilies show a very near approach to a Rhythm both continuous 
and definite. 

Rhythm, of course, means regular movement causing a sensa- 
tion in one of the organs of sense ; movement, the law of which 
can be appreciated by the senses. This is universal. In dancing, 
in music, in a swinging pendulum, when travelling in a railway- 
train, we have this phenomenon called Rhythm. 

** Wer theilt die fliessend immer gleiche Reihe 
Bolebend ab, dass sie sich rhythmisch regt ? 



Des Monschen Kraft im Dichter offenbart." 

— Faust. Prol. 

Rhythm, then, more exactly, is a regular sequence of Mavements, 
each growp of Movements constituting the sequence being divided into 
equal periods of Time {morks) of which one period receives a pre^ 
ference according to some principle, 

Now Metre is Rhythm realised in speech. Metric is, there- 
fore, only a part of Rhythmic. Metric is the particular , Rhythmic 
the general. All poetry is necessarily rhythmical, and the ordinary 
distinction between rhythmical and unrhythmical measures is due 
to the fact that the classical languages made use of a particular 
form of Rhythm in their verse. There are, I need scarcely 
remind you, three varieties of Metre, (I) The Syllabic, where 
every rhythmical time-period coincides with a syllable. This is 
nowhere found in strict use, but the Vedas are a near approach 
to it. (2) The Quantitive measure — in which the time-period 
which receives the preference (Thesis) has a definite duration in 
Speech relative to the other time-periods. Thus the length of 
time taken to pronounce the long syllable of a dactyl is exactly 



THE OLD ENGLISH ALLITERATIVE LINE. 377 

twice that taken in speaking each of the short ones which follow. 
That is why a spondee can be substituted. This is the measure 
of Greek and Latin. (3) The Accentuated measure. Here the 
relation of stressed to unstressed syllables in the line cor- 
responds, on the whole, to the relation between time-periods, 
or mords, which receive a preference and those which do not, 
i,e, the stressed syllables coincides with the Tact or beat of the 
rhythm. 

We see then that Rhythm is the essential, and it may be 
the same though the metrical realisation differs. Thus, in the 
classic Elegiac verse, the rhythm of the Hexameter and that of 
the Pentameter is the same, but the Metre is different ; or, again, 
the Ehythm of the lines in a stanza of the Nibelungen-Lied is 
the same throughout though the Metre differs. 

This being so, I should like to assert the two following dicta 
with regard to any theory of O.E. Metric. (1) No explanation 
of O.E. verse can be satisfactory which does not prove a unity 
of Bhythm. (2) No explanation is satisfactory that does not 
give us a Prosody which a poet could easily carry in his head. 

Now the most elaborate system of recent times, and the one 
universally adopted in England, is that laid down by Prof. Sievers 
in the tenth volume of Paul and Braune*s ** Beitrage," and more 
recently in Paul's Grundriss, and in his '' Altgermanische Metrik," 
Halle a/S, 1893. Sievers has undoubtedly added very consider- 
ably to our knowledge of the O.E. verse- system, and his work 
marks a great advance on the results of Wackemagel and Max 
Bieger, but does it stand the tests I have just formulated? I 
think not. Sievers believes, as you all know, that there are Jive 
Types with Variations : — 



Typus A. ^ X I -1 X 
Typus B. X - I X -1 

Typus C. (a). X ^ I -1 X II ^^. 

w.x ^ I 6 X ir'?j;s: 

Typus D. (a). ^ I - H X 
Typus E. ^ H X I - 



378 



THE OLD ENGLISH ALLITERATIVE LINE. 



Examples from B^wulf. 

A. (•) jomban jyldan, 1. 1 1 1 1 (^) lebfne >^en, 1. 34. 

B. (») aie I don >a, 1. 34 1 1 C^) ond Hal | ga tU, 1. 61. 

C. (o). (•) oft Scyld Scefing, 1. 4 1 1 (}) ond jrtmhelmas, 1. 334. 
()3). (») of feorwejum, 1. 37 |l (^) in ^earda^um, 1. 1. 

D. (a). (») beam | Healfdenes, 1. 469 1 1 (^) f^nd | mon cynnes,!. 164. 
()3). (») fyrst I for* jew&t, 1. 210 || (^) holm | iip setbaer, 1. 619. 

E. («) healeeraa | ms^et, 1. 78 || (^) weor^myndum | }ih, 1. 8. 

Here I can find no unity either of Rhythm or Metric. Here 
are half-lines of Trochaic and Iambic form, and that even in the 
two halves of the same line ; cf . aledon ]?a | [ let^fne ]?ebden | . 
Not only that, but in Typus C we find an Iambic foot immedi- 
ately followed by a Trochaic one, «.«. absolutely no Ehythm at 
all. It is quite impossible to speak such a line without either 
lengthening the first stressed syllable or putting a pause after it ; 

thus the line must either bexliixl— XorX^Al.1 X. 
If we analyse Typus D in the same way we should get 

u-^ I uH- I ^ X, i.e. 2%ree Chief Accents; and similarly 
-A 
with E. 

When we turn to the Variations the matter is worse. I only 
give the Variations of A as a specimen : 



A. ^ X u X ^^r 

^ XX I -1 X 

^ XXX I ^ X 



(Bw«) ji xxxx I - X 

{SSSr) ^ xxxxx I ^ X 



All X I ^ X I ^ X 
occur X I ^ XX I ^ X 

with X I .n XXX I ^ X 
Ana- X I ^ XXXX I ^ X 
knisis. X I :L XXXXX I .1 X 



Further, there are five Variations of A caused by the addition 
of a third and even a fourth (subordinate) Accent to the line in 
different positions. 



THE OLD ENGLISH ALLITERATIVE LINE, 379 

I have already shown that this scheme does not satisfy my first 
requirement. I do not think any one will contend that it satisfies 
the second. No poet could possibly carry so complicated a system 
in his head. To compose with such a Prosody would imply at 
once the greatest crudity and the greatest subtlety in the ancient 
* scop.* 

Old English verse is accentualf and can only he explained from ths 
Rhythmical standpoint. Even Sievers, in his article in Paul's 
Grundriss (s. 870, Bd. II.), acknowledges that the Germanic 
verse is to be traced back to a four-accent type, with a constant 
Iambic Ehythm, such as we find in a number of the Yedic songs. 
Eut he believes that at the time when the earliest Germanic 
poems were composed, musical recitation had completely dis- 
appeared, and, further, that the poets had lost all sense of 
Ehythm. His arguments to prove this are, to me, completely 
inconclusive. And the constant recurrence of such phrases as 
'swutol sang scopes,' and reference to the *gl^o-bekm' as one 
of the chief joys of the banqueting hall would be meaningless 
if we are to suppose that simple recitation was the method of 
delivery. Besides, though Moller^ has failed to show that 
'Beowulf' was originally written in a four-lined strophe, yet 
he has, I think, conclusively proved that a tendency is traceable 
in O.E. poetry to fall into a four-lined stanza-form as a re- 
miniscence of an older phase of composition. If this be so, it 
is clear that the Germans must have used the musical delivery, 
and that at the time of the popular epic (as we know it), even 
if song had given way to intoned recitation, which is quite 
possible, yet the old sense of rhythm still survived. 

To turn to the constructive side of my argument, the following 
is Ten Brink's scheme of the O.E. Ehythm, viz. One Fundamental 
Type, with Four Variations, as follows : — 

F.T. X)XXX XXXX II 

Var. 1. x)ixx XXXX II 

Var. 2. \)yO(J>^ YXyJk 1 1 

Var. 3. x)i^X)C XXXX 

Var. 4. x)i^x5c x>:xx 



> H. Moller ''das A-£. Volks-epos in der urspr. stroph. Form,'* Kiel, 1883. 



380 THB OLD ENGLISH ALLITERATIVE LINE. 

Katalectic F.T. (a). x)5cxA x5c(x(* II (^)- x)xXX xicXX II 

A A 

„ Var. 1. (a). X)xxie x5cfx(* II ()3). X}^)x^ X>[XX IP 

A A 



„ Var. 4. x)xX)i xiixik 

A A 

The O.E. verse, therefore, ia the short or half-line shows 
four beats of two Mor^s, each of the form XX , though there is 
evidence that in earlier times the half-line consisted of two 
beats of fottr Mor^s each, e.g. I XXXX I . The following facts 
point in this direction, (a) That even in historical times the 
Rhythm of the verse always distinguishes two Chief from two 
Subordinate Accents, (J) That when the Accents are not arranged 
as Dipodies the Prosody seems to require that the type of line 
in question should be more clearly expressed in speech. 

It will be seen that in all the above cases the Rhythm is the 
same. In the second half-line the first of the two Chief 
Accents is the stronger, and bears the alliterative letter. Indeed, 
it is the strongest Accent in the whole line. Anakrusis is 
oftener lacking than not, which caused Sievers to look upon 
the Anakrusis as not belonging to the verse.' Examples of 
the various forms are the following : — 

F.T. jehedde under heofenum, Beow. 605». 

Var. 1. A 8^ >e wdSterejesan, Beow. 1261*. 

Var. 2. A ofer je5fen6s bijanj, Beow. 362*. 

Var. 3. A s'i'jemund^ jesprdnj, Beow. 885^. 

Var. 4. A ^^1 y^^ jeswCnj, Beow. 849». 



It will be noticed that the unaccented syllables between the 
second and third Accents in the last two examples are lacking. 



^ This type occars only in the second hemistich after an akatalectic first. 
' Two-syllabled Anakrusis is sometimes found, and probably also three- 
syllabled occasionally, though this is denied by some. 



I 



THE OLD ENGLISH ALLITERATIVE LINE. 381 

In fact, it is seldom that lines occur in which all the unaccented 
syllables ^re expressed, and, to put the matter shortly, Old English 
verse, as we shall see, in common with all Germanic verse, only 
requires that every ^* prefer enced Time-period^^ (Thesis or Outes 
Zeittheil) shall correspond with the beginning of a syllable. Thus, 
in the Eundamental Type, all the unaccented syllables may be 



wanting, thus: — /^ j^"^^^ A ZY^^^^^ Beow. 11% or /^ breost 



y yj I \j A • 

hord blod hr^ow,* Beow. 1720». 

The syllables able to stand in the Thesis or verse-tact are 
determined partly by Lachmann's Law, which deals with the 
word-stress, and partly by the law of Sentence-stress, which, 
in O.E., was not logical, as now, but purely conventional. 

Lachmann*s Law, I may remind you, is as follows: — That 
there are two grades of Stress in words besides the unstressed 
syllables, though the Chief Stress is not different in kind from 
the Sub-stress. Further, that after a long syllable, bearing a 
Chief Stress, follows a sub-stressed syllable ; after a long syllable, 
bearing a Sub-stress, follows a sub-stressed syllable ; also after an 
unstressed syllable (short or long) follows a Sub-stress. On the 
other hand, after a short syllable, bearing either a Chief Stress or 
a Sub-stress, follows an unstressed syllable. 

This may be formulated thus : — 

^ ± k 

u H Sd 
6 H 

Scholars have denied that Lachmann's Law applied to the 
spoken language. But this is * a priori ' improbable, for it 
neglects the very close connection between verse and the spoken 
tongue in Early Germanic times — besides which, if it only applied 
to the verse, this highly conventional system of accentuation 

^ The dot • expresses a short pause. 



382 THE OLD ENGLISH ALLITERATIVE LINE. 

(as it then would be) would certainly strike an audience as nn- 
natural and fail to catch on. But it can be proved that Lachmann's 
Law applied to the O.E. spoken language. 

In words of the followiug form : — C id ; ^ ^ 5 ^ \j li ; 
i id ^ ; it can be shown that the syllables in the second place 
are either weakened or disappear. In words of this form: — 
w 2f ^ » this must occur even in compounds in which a chief 
accent had fallen there in the simple word, e,g. jedtwe (Armour) 
< ja+tawu (cf. jd-mol for ja-mal, which shows that 5a- was 
accented), or, again, fra^twe (Ornament) < fra+tawu. 

Other examples, which prove the validity of this Law, 
are : — the disappearance of the Eeduplication in the Praet. of 
Red. vbs., e.g. het < hehat, and the early disappearance of the 
connecting vowel in the Praets. of weak verbs, e.g, tealde for 
talde < talida ; sealde for salde < salida. In these last cases, as 
also in -^eatwe and frcetwe^ the disappearance of the middle-syllable 
insured the survival of the suffix-syllable, which was of the 
utmost importance, by transferring the Sub-stress to it. 

The conventional Law of Sentence-stress, as is well known, 
was shortly this: That certain Parts of Speech naturally took 
a stronger stress than others. (1) The Substantive, certain 
notional Adverbs, and the Pronouns self, begen, and perhaps 
o^er. (2) The Verb. (3) Pronouns and Particles. In addition 
to this, when two words of the same rank stood together, the 
first took the Chief Stress. Thus we should say * Shakespeare's 
mother ' but * the mother of Shakespeare,' i.e. Logical stress. 
In O.E., on the other hand, they said * ficjlafes beam * but 
* beam Ec^lafes.' In the second half-line, however, a certain 
amount of freedom was allowed in this respect. Thus in Beow. 
1129^ **wuno(ie mid Fin," the chief accent falls upon 'wimode,' 
and not on * Fin,' because the first syllable of the verb is the 
alliterative one. 

I can now proceed to give the rules for Quantity and Accent 
in the O.E. verse. 

Rules for Qaantity, 

1. Absence of Anakrusis is replaced by a Pause. 

2. When the unaccented syllable (senkung) is wanting after a 

Chief Accent, its value is replaced by lengthening the 
preceding Chief- accented syllable, e.g. X)^(x)x* 



THE OLD ENGLISH ALLITERATIVE LINE. 383 

3. When the nnaccented syllable (senkung) is wanting after a 

Sub-accent, its value is replaced partly by lengthening 
the preceding syllable, partly by a Pause. 

4. Syllables to be lengthened must be long. Here is to be noticed 

that Germanic verse occasionally allows the freedom that 
syllables which in the spoken language are short, are, for 
the purposes of versification, treated as long, eg, words 
of this form u — cr , when they come at the end of 

a verse and are preceded by 6 o, are accented 6 "^ 
instead of CS — cr. The word 'eyning^ is specially to 
be noticed in such lines as fe6rh cynln^es, he^fon cynln^^s 
(where the * n,' instead of beginning the second syllable, 
is counted as closing the first, and consequently lengthened 
in accordance with the universal custom in O.E.).^ Also the 
Pres. Parts, berende and wesende in lines such as swe^rd 
berende and raed berende and cniht wesend6, Beow. 372^. 
(5) In conclusion, a rhythmical More can in general be filled by 
either a long or a short syllable — though in practice this 
freedom is very restricted— especially by the action of 
Lachmann's Law. 



Bules/or Accent^ i.e. Relation of Verse Accent to Word Stress. 

I. The two chief Accents fall upon the two most strongly stressed 

syllables, and in the second hemistich the alliterative 
letter should coincide with the strongest stressed syllable 
in the line — except in cases like "wunode mid Fin" 
referred to already. 

II. (a) Every syllable capable of bearing a stress, if it comes 

before a less stroDgly stressed syllable which fills a Mor^ 

{i,e. is not capable of being slurred), or if it comes at the 

end of a line, must bear a Verse Accent. Thus the second 

syllable of ' Burgendum ' and the third syllable of 

*Gifica' in the line ' BUrgindutn G'ijicd: Again, 'f61c^ 

ft 
jeondf erd6 1 1 ' 

^ Cf . wis-dom ; see Ten Brink Chaucer's Sprache and Verskunst, § 87. 



884 THE OLD ENGLISH ALLITERATIVE LINE. 

{h) Every syllable capable of bearing a stress is accented 
when it follows a less strongly stressed syllable which fills 
a Mor^ (t.tf. is not capable of being slurred), e,^. Edrmanric 
G6tum. 
III. If a syllable capable of bearing a stress stand between two 
which are more strongly stressed, it may or may not 
receive a Yerse Accent. If it is a sub-stressed syllable^ 
and a Suffix, it can only bear a Verse Accent when, by 80 

doing, it forms a Dipodie, ^.y. ' E^rmanrilc G^tum.' The 
'an' must be unaccented, because, if accented, it would 
not form a Dipodie. On the other hand, in leu&6 J^eod^n, 
the syllable -ne does bear a sub-accent. 

Of the normal line scanned, according to these rules, there 
are plenty of examples — the vast majority are such. I have 
already given an example of each variation, and I have no time 
for more. 

Katalectic Lines. — One of the Sub-accented syllables may be 
replaced by a pause — but never a Chief-accented. As a rule it 
is the fourth verse Accent which is wanting. Only the Funda- 
mental Type and Variations 1 and 4 can be Katalectic. A 
Katalectic verse can end either with (a) an unaccented or (/3) an 
accented syllable. In the former case the Pause is a short one, 
in the latter long (cf. scheme supra, p. 9). 

Examples — 

F.T. a. fr^imcyn wi'tan /^ Beow. 252». 
Hr'flntfng nama /^ Beow. 1458^. 
p, m^m6nde mod /^ Beow. 60*. 
w o'rd hord onl&c /f Beow. 259^. 

It is to be noticed that in case p an unaccented syllable occurs 
between the second and third Accents almost invariably. And 
when this is not the case then there is invariably Anakrusis, 
e.g. Gecjfste \i a Beow. 1871*. 

Var. 1. a. we J^urh h^ldne hi'^e /^ Beow. 267*. 
}?8Bt W8BS g'6'd cyning /^ Beow. 11^. 
p. }7ffet fram ham jefr^gn /f Beow. 194*. 
6nd Httlga til a Beow. 61^. 



THE OLD ENGLISH ALLITERATIVE LINE. 385 

The omission of an unaccented syllable between the first and 
second Chief Accent is much commoner in the second hemistich 
than in the first. 

Var, 4. a. swdtol sang scopes /^ Beow. 90*. 
Ill II 
wop up ^hafen /^ Beow. 128*>. 

)8. f^rst f6r]7 jewat /f Beow. 210». 

h'61m st^rme we^l a ^eow. 1132^. 

Here two things are to be noticed, (I) This type of verse 
is not so beautiful in form as those above, for a Pause, following 
upon a Sub-accent, is not natural. The poet is, therefore, careful 
to ensure that the Sub-accent coincides with a Chief stress. (2) 
In case ^ there is always an unaccented syllable between the 
second and third Verse Accents. 

In var. 1 only^ the first Sub -accent is sometimes absent, thus : 
' I " " ' , and this form generally occurs in the second half- 
line, e,g, A jew'aden hsbfde, Beow. 220*>. 

How can we distinguish i \ n h i from n '» i | /, i.e. 
var. 4, Katatalectic ? 

Easily, for in the case of var. 4 " " ' | ', the last Verse 
Accent, as we have seen, always coincides with a Chief StreiSf 
whilst in this particular variety of Katalectic, var. 1, the last 
Terse Accent always coincides with a Sub-stress (Nebenton) ; 
moreover, in this case, the line always begins with an unaccented 
syllable, which looks like Anakrusis. 

Finally, there is an occasional Katalexis of the Fundamental 
Type where the first chief -accented syllable in the line is 
lengthened so as to cover the following unaccented and accented 
Mor^s, thus: 

x)xxx x>(xx 



Ex. or astelidffi (Coedmon's Hymn). 

ham jesoht^, Widsi^ 7^. 

// 
j^rymm jefrunon, Beow. 2^. 

jM-jewsedu, Beow. 2618^ 



Phil. Trans. 1891-2-3. 25 



386 TUB OLD ENGLISH ALLITERATIVE LINE. 

The prolonged syllable at the opening, it mast be noticed, is 
not only always long and bears a Chief Accent, but it is generally 
followed by a Continuant Consonant. In addition to which, 
this particular form of verse nearly always occurs in the second 
hemistich, which makes the matter easier; for the first accent 
in this hemistich is naturally the strongest in the whole line. 
This form of line, I may say, presupposes and points towards 
a musical form of delivery — to song, for it would be difficult 
to speak such a line without destroying the Ehythm. 

It is possible, of course, to find shorter lines than even these, 
but all scholars, including Sievers and Eieger, look upon them 
as corrupt. 

Hyperkatalectie lines proper , i.e. lines ending in an extra 
unaccented syllable, there are not many of. Most of them 
ai-e to be reduced to the right measure by slurring, which was 
apparently allowed freer play in the Caesura than elsewhere, e.g. 






W6rd w»ron wynsume, Beow. 613*. 
E^rd on eMsdipe, Beow. 1728*. 
Selltce saedracan. Beow. 1427». 

Expanded Lines, called by German scholars SchwelUverse or 
Streck'Versej are hypercatalectic, in that they possess a larger 
number of Verse Stresses than the normal line. Either, or both, 
hemistiches may be expanded. The half -line is either (1) expanded 
by a half; the method in the great majority of cases, or (2) 
it is doubled in length, which is a peculiarly characteristic form 
of the expanded line in the younger Genesis. 

Now is it possible to discover any principle in the way in which 
these verses are expanded ? Because if not, then it will be im- 
possible to say which part of tho verse is expansion, and which 
not, and it is equally impossible to see how such a line was 
understood to be expanded by an Old English audience. 

Ten Brink does not seem to have come to any final conclusion 
about the matter. Certainly the view taken by my old friend. Dr. 
Gregory Foster, in his Studies on Judith (Q. and F. 1892); a 
view with which Ten Brink was then inclined to agree, I believe, 
does not coincide with that which he seemed to hold in the 
winter of 1889-90. 

I will quote Foster's theory verbatim, as it gives what was 
practically Ten Brink's last opinion on this matter. 



THE OLD ENGLISH ALLITERATIVE LINE. 387 

** Such lines," he says, " are * expanded ' by prefixing to each 
hemistich a chief-stressed syllable, thus raising the number of 
chief-stresses in each hemistich to three, and one or more either 
secondary or un-stressed syllables. In the first hemistich, this 
prefixed syllable should alliterate, but not in the second hemistich, 
in which the syllable having the chief-letter becomes the second 
chief-stressed, but remains the first, and only, rime-letter of the 
hemistich." 

Now, though as a matter of fact, this account of the construction 
fits the majority of the lines in Judith, yet even in a poem, the 
author of which has shown such marvelous skill in the use of 
these expanded lines (as Foster has proved), and who, in other 
respects, was evidently a most polished and skilful versifier — 
even in this poem there are at least two lines which will not 
fit in with Foster's 'theory, and he remarks that ** the irregular 
form of these lines in the Cynewulf poems is remarkable." 

Besides which, one does not see why in such a line as bealoful 
his beddes nefisan, Judith 63*, the word bealoful should be ex- 
pansion rather than ne6s^. There is nothing in "the reason of 
things, according to this theory, why the first element should 
be expansion. Foster makes no attempt to explain the con- 
struction of hemistiches which are douhU the normal measure. 

Now, in my opinion, no theory about these Expanded Lines — 
80 far, at any rate, as they are found in the older verse, the 
poetry composed in recitation and not on parchment — can be 
the right one, unless it explains to us not only their actual 
form, but how an audience appreciated that they were listening 
to an expanded line. If the expansion was not heard by the 
audience, as in some way an integral portion of the line, the 
additional syllables would strike the ear as an impertinence, and 
as absolutely destroying, not only the Prosody, but the rhythm 
of the line.* Of course the rhythm of an expanded line must 
be the same as that of a normal line in character^ though it should 
be felt to be an extension of the normal phrase. The law, 
then, if one is to be found, must be one which appeals to the 
ear, not to the eye. 

Now, as we all know, the unity of the two hemistiches depends 
upon the alliteration. The first half-line, usually, though by 



^ They would in fact have an effect similar to the irregular Terse of Southey 
for our ears. 



388 THE OLD ENGLISH ALLITERATIVE LINE. 

no means always, has two alliterative letters; the second half 
only one. The existence of only one rime-letter in the second 
hemistich ensures the two halves heing felt to form one whole, 
and makes it clear in the majority of cases whether we are 
listening to the opening or close of a line. This heing the 
case, what would he the most natural place for a poet to make 
an addition to his hemistich ? Surely if the expansion is to take 
place in the first hemistich, it would come between the normal 
hemistich and the verse pause. If the second hemistich is to be 
expanded it would receive an addition between the verse pause 
and the opening of the hemistich. In this way the ear could 
never be in doubt as to what was expansion and what not, nor 
as to which hemistich the expansioa belonged to. The alliterative 
letters would fall in exactly the same places as in the normal 
line ; the expansion would receive no rime-4etters. On testing 
the expanded lines in the O.E. corpus from this point of view 
I have found that the vast majority of expanded lines conform 
to this type, which may, therefore, be taken as the normal, or 
Type I. Every line (about which there is no doubt that it is 
expanded), in Bdbwulf is constructed according to this type; 
but no expanded lines occur in the oldest part of Bebwulf. 
Here is a line expanded in both hemistiches, though it is not 
necessary that one should always be expanded because the other 
is, indeed, as a rule, in Bebwulf only one hemistich is expanded. 



i I beagd 1 1 
ne >6rfte hfmj "Si l^&n o^wltan 1 1 



Beow. 2996. landes ond Wcenra 

t It* It 



n 



Other examples are : — 

Beow. 1167*. 8Bt fotum sat frean I Scyldingd || 
Beow. 1708*. freode, 8w& wit fdr^um I sprfecon 
Beow. 2174». wraetllcne w^ndiir- 1 ma^^um || 

Two things are to be noticed here — (1) that when the first 
hemistich is expanded it always has two alliterative letters, at 
any rate in the older poetry. In Judith, which is modelled 
upon the old heroic poems, there is only one exception to this. 



THB OLD ENGLISH ALLITERATIVE LINE. 389 

tiz. 90**. (2) There is no objection to the second element of 
a compound forming the expansion of the first hemistich, especially 
when there is no unstressed syllable at the opening of the second 
element. The necessary pause, which must then take place 
between the two elements, helps to mark the commencement 
of the expansion. Several examples of lines, constructed like 
Beow. 2174», are to be found elsewhere, e.g, the Rood Poem, 
47», 63», 68* ; Judith, 7», 1 1», 346». 

All the expanded lines in the Seafarer (except one, which I 
shall come to presently), all the expanded lines in the Wanderer 
(except one, which I shall return to), all the expanded lines in 
the undoubtedly older portion of Wtdsi^,* and the two examples 
of expansion in the Battle of Finnsburg (11. 13 and 24), are 
constructed upon this normal type. Line 24 in the Battle of 
Finnsburg is interesting, because it is impossible to scan it at 
all upon Foster's theory, whilst it is quite easy on mine :• — 

1. 24. Si'gef^rh^ is mtn nama /^ I cw£% h^. 

Two unaccented syllables together {doppelU Senkung) is by no 
means unknown in the first hemistich.^ 

To sum up — Every expanded line in the national epic and the 
lyrics (I have examined) are Type I. with two exceptions, which 
are Type II. 

To turn to the religious epic — the work of the Gaedmonian 
and Cynewulfian schools: 

In Genesis A. all the expanded lines are of Type I. There 
are only three slight irregularities, viz. 2866*^, which is expanded, 
and only has one instead of two alliterative letters, and 2142^ 
and 2182^, which have the alliterative letter in the expansion 
instead of on the first Chief Accent of the normal portion of 
the hemistich. 

Gen. A. 2142^ nis tr^ruld feoh, J ^e fc me ajan wille. 
Gen. A. 2182^. /est^J m^nte^ 'Inje^lincum. 



^ It is doubtfal whether 9^ is expanded ; if it is, then this is another 
exception. 

' Miillenhof rejected 11. 75-87 and 131-134 as later interpolations, and 
I agree with him. I think we shall see that there is ground for regarding 
U. 59-64 with considerable suspicion. 

3 Cf. ^esawon ]>a ^fter w»ter6 || Beow. 1426*. 



390 THE OLD ENGLISH ALLITERATIVE LINE. 

It seems impossible, in either case, to consider the expansion 
as at the end. If the expansion is at the end, then they belong 
to Type III. 

In Gen. B., which was written about two centuries later (latter 
half of ninth century), and has been proved to be based on the 
* Heliand,' there is a very large number of expanded lines. 
Many of them belong to what I shall call Type II., and I pass 
these over for the present. Of the remainder, all, except two, 
are expanded in accordance with Type I. In two cases (338* 
and 488*) there is only one instead of two alliterative letters. 
The two exceptions to Type I. are 287^ and 301^, but it is 
doubtful whether 287^ is an expanded hemistich at all. L, 301^ 
seems to be a case of what I shall call Type III., unless we 
may assume Elision of the ' e ' in ' hsefde ' ; in which case we 
might possibly construe it as an unexpanded hemistich.^ 

Of the four expanded lines in Exodui all are of Type I., 
viz. 11. 570-73. 

There are many more expanded lines in Daniel, If we omit 
the clearly corrupt passages, every expansion is constructed accord- 
ing to Type I. There might be some doubt about 453", but this 
is, in all probability, a case of three-syllable Anakrusis. 

In Christ and Satan, lines 260^ and 261* belong to Type II., 
but all the rest belong to Type I. If 1. 89* is not corrupt, there 
is only one alliterative letter instead of two. Line 205* seems 
irregular, but the ' mid ealra,' at the opening of the line, has 
evidently crept in from 1. 203, and it is, therefore, a normal 
line. To sum up then: In the Ctedmonian poems there is only 
one probable case of Type III. — ^at the very utmost only three 
in all. Of Type II., if we except * Genesis B.,' there are only 
two hemistiches, both of which occur in * Christ and Satan.' 

"We now come to Cynewulf and his school. All the expanded 
lines in the 'Christ* are clear examples of Type I., with the 
exception of five. 1163^ and 1515^ may possibly be examples 
of Type III., but possibly they are hemistiches with two-syllable 
Senhung — though it is true that this phenomenon is rare in the 

1 I. hyld h8efd6 his (^ ferltfren^ or 

II. h'yld h^fde hU ferl6reii6. 

But case II. we should have two syllables together unaccented, a doubtful 
possibility in the second hemistich ; besides, it occurs in the middle of a passage 
of expanded lines, and this speaks against the second method of scansion, though 
it is true that 302^ may also be a normal hendstich. 



THE OLD ENGLISH ALLITERATIVE LINE. 391 

second hemistich. 1385^ I can make nothing of in any way unless 
we are to assume three-syllable Senkung between the first and 
second Stresses. 1496* and 1425^ seem clear cases of Type III.* 
In the Elene, as in the Christ, there are but few alliterative lines, 
but they are all Type I. 582» is slightly irregular, in having only 
one alliterative letter. Lines 163 and 610, which Foster takes 
to be expanded, seem to me normal lines. 

In the Dream of the Rood the expanded lines are of two types — 
of the second Type there are about ten lines (quite certain), 
and about as many again where we cannot be sure whether they 
are Type I. or Type II. The remainder are clear examples of 
Type I. Three lines, 10^ 47*», and 83% offer some difficulties, 
but there are no clear cases of Type III. This is significant, 
as Ten Brink places this poem as the first of Gynewulf's religious 
epics. 

The expansions in the Andreas are all Type I., but there are 
only a few. 

In Guthlac, however, there are no less than eleven clear cases 
of lines constructed on Type III. (the least perfect form of the 
three), e.g. 163% 212», 347% 348», 436», 437»? 440% 673», 674*. 
713*, 350». This is very extraordinary, and it is worth notice 
that Ten Brink has placed this poem among the last of Cynewulfs 
productions. 

Finally, let us turn to the ^Judith,'* which Foster has clearly 
shown to belong to the Cynewulf school, ana to be later than his 
work. The poet, as Foster has proved, makes a most skilful use 
of the expanded line, and there are, relatively, a large number 
of them. The large majority are clear cases of Type I., but 9* 
and 65* seem evidently cases of Type III., and, on the analogy 
of those, I am inclined to interpret seven other cases, 11% 17% 
19% 20% 33% 270% and 242». But I shall return to these. 
Suffice, for the present, that I believe there to be nine cases of 
Type III. in Judith. 

Type II. This type is the double of the normal line, instead 
of half as long again. Here it would be impossible for the whole 
expansion to occur between the close of the normal line and 
the verse pause in the case of the first hemistich, because the 



> p. 1306*, 1378% 1410% 922% 1060», 1360% and 1666*, which Foster 
conriden to be irregular expanded hemistiches, I look upon as normal un- 
eipanded ones. 



392 THB OLD ENGLISH ALLITERATIVE LINE. 

line would then sound like a normal line without alliteration in 
the second hemistich. 

The expansion, therefore, takes place in the best examples, 
half between the two rime letters and half at the close. This 
in the first hemistich. 

The result, therefore, is equivalent to two. hemistiches with 
one alliterative letter in each. As a rule the second alliterative 
letter occurs in the accented syllable fourth from the end, but 
sometimes on that third from the end, and occasionally on the 
second from the close. In the second hemistich the whole ex- 
pansion occurs between the verse pause and the normal line. 
In the later poetry there are examples of the whole expansion 
occurring at the beginning of the hemistich, even in the first 
hemistich, but this is very exceptional. 

There are no Type II. lines in Beowulf or Finnsburg, ue. in 
the national epic 

In ' Seefarer ' is one clear example of Type II. in the first 
hemistich of 1. 106. 

d^\ bf% se %S hfm | his iryht^n ne ondrsbd6%. 

Here if, as Sievers assumes, ' ne ' can bear a stress, the 
Sub-accent should fall upon ' ne.' 

The second hemistich of 1. 106 is an expansion of Type I. A 
curiously similar line to this occurs in the Wanderer^ 1. 112 — 
the only one : 

Ti\ bfS 8^ ^e hfm | ^rt^we jeheald^^. 

The second hemistich is also of Type I. 

Now we come to the * Widsl^.' Here are two groups of very 
long lines which can only be explained as Type II., but they 
are very irregular in construction, viz. 59-64 and 76-87. 

An example from each group is : 

60. Mid Gefdum 'fo wses | and mfd ^inedum || 6nd mid 

Gefflegum. 
84. Mid M6rdum ic wees | 'and mfd Persum 1 1 ^nd mid Mj^rgfingum. 

Even if this scansion is right there are several irregularities. 
Now Miillenhof rejects 11. 75-87 and 131-134 on other satis- 
factory grounds, as later interpolations. 



THE OLD ENGLISH ALLITERATIVE LINE. 393 

It is certainly very tempting to reject group 5d-64, which is 
exactly parallel in style and in syntax, and almost, though not, 
perhaps, quite so vile in yersification. 

If this is justified, we only have two examples of Type II. 
in the older lyric poetry. Of these, that in the Wanderer occurs 
in the second half of the poem, which Ten Brink, upon other 
grounds than those of metric, considered to he hy a later hand, 
and I agree with him. We have thus reduced our ten little 
niggers to one. I suspect there will soon he proof that will 
dispatch even this one. Type II., indeed, is not made any con- 
siderahle use of till we come to Genesis B., where there are a 
large numher, and if we rememher the source of the English poem 
we shall not feel surprised I think. Most of the examples are 
regular, hut I have noted that the alliteration occurs in an 
irregular place in II. 322* and 708*. 

322*. L&gon ^a 6«r6 | Fynd on ^am rfr6 || 
708* is a similar case. 

In some cases it is difficult to say whether we have a line of 
Type I. or II., though the constant undouhted cases of the latter 
make an a priori presumption in its favour. 

Example 260*. Wi% >6ne hi'hstan | he^fnes weiad^nd 

or 
Wi« >one hehstan he^fii^s (^wealdend. 

In CTirist and Satan there are two hemistiches of Type II., 
viz. 260^ and 261*. 

260^. ffod se^Ifa hfm | rice heald6«. 
261* is a regular example. 

In the Cynewulfian poems we only find this type in " The 
Dream of the Rood," hut here there are a good many, though, 
again, it is sometimes hard to say whether we have a Type I. 
^ or Type II. 

An example of a second hemistich, as I have not yet given 
a regular one, is : 

66^. ciirfon hfe %&t I on 3e6rhtan stane. 



The conclusion to he drawn is that this is not so old as Type I., 
and prohahly arose after poems were composed on parchment and 
not by ear. 



.1" V 
t 

\ ■ 



394 THE OLD ENGLISH ALLITERATIVE LINE. 

Type III. is the least perfect and the rarest. These could 
only arise when poets tprote, and did not compose in recita- 
tion. In the first hemistich expansion is hy a half, hut at the 
beginning of the hemistich instead of at the end. In the second 
hemistich expansion is at the end and not at the beginning. But 
this is rarer than Type III. in the first hemistich. The expansion 
can only be recogniaed as such by the Syntax. 

Examples from the ' Guthlac ' : 

163». cwfedon ]7fi§tJ he 6n }?dm be5rj6 || 
440^ ealles %uj ^ses wlte &wiinn6 || 

Example from ^ Christ ' : 

1425^. LtSj on heardum /^ I stSne. 

If this is not right, then there is no alliterative letter. 

In Judith there are several examples of Type III., but they 
are peculiar in having alliterative letters in the expansion as a 
rule. This one has only one alliterative in the hemistich. 



9". Gi'rwdnJ up sw&s6nd6 || 
19». FillleJ flett sl'ttendum || 

I do not see how else to scan this with 9» in view. 

In conclusion, we find none of these lines in the national epic, 
nor in the old lyric, and in the CsBdmonian poems there are also 
none (for, of course, I do not include Genesis B.). This type, 
therefore, does not occur till the Cynewulfian time, i,e, a time 
well within that when composition took place on parchment, and 
the invariable appeal to the ear both by poet and public was gone. 
Cynewulf seems to have grown more lax in the use of this type as 
time went on, for in Guthlac are many more than in any poem 
of his I have examined. In Judith, too, as we should expect, 
since it was written after Cynewulfs time, there is a com- 
paratively large number of these lines, but I attribute the poet's 
use of the first alliterative letter in the expansion to his study of 
the older schools of poetry. He saw a line like Beow. 2173* — 

"Wr&tlfcne wfindor I ma^^um || 
and did not recognize that 'ma%%um' must necessarily be the 



THE OLD ENGLISH ALLITERATIVE LINE. 395 

expansion. The appeal to the ear was gone — his imitation was 
a purely mechanical and therefore superficial one, and if we are 
to scan many of his lines we must assume the word with the 
first alliterative letter to he in the expansion at the beginning of 
the line. 

In conclusion, I need only say that if this conception of the 
construction of the expanded line be correct, the occurrence, or 
non-occurrence, of the various types in O.E. poems may serve 
as a relative time test in deciding their dates. 



( 



396 



XIV.— ON GAELIC PHONETICS. By J. H. Staples. 

{£ead at the Meeting of the Philological Society ^ Friday, March Zrd, 1893.) 

The speech sounds represented by the symbols and characters in the 
accompanying tables, and for which the test words in Tables II and 
III are examples, are those of a dialect of Argyll, having much 
in common with Scotch Gaelic dialects elsewhere, and therefore 
offering a somewhat representative utterance, but the nasal form 
of the peculiar Gaelic vowel in " aon,'* though prevalent in other 
districts, is not used in Argyll. There the vowel in " aon " would 
be the same as in "caomh," see Table II, 2nd column, Nos. 7 
and 11. Also the pronunciation of n in the combinations 'en,' 
* gn,' see foot of Table III, though permissible, would in Argyll 
be rather old fashioned. The use of the 'n' sound in these 
combinations is very prevalent in many other Highland districts, 
but in Argyll nasal 'r' as in ''mna," see lowermost space. 
Table III, No. 3, is substituted for ' n.' 

In the following observations a Gaelic word is in ordinary 
spelling in inverted commas, single letters or letter-groups being 
alluded to thus 'bh,' phonetic spelling and single phonetic 
characters in parenthesis thus («), using Prof. Sweet*s modifica- 
tions of alphabetic types, with exceptions described at head of 
Table I, and the following modifying marks are also used in 
connection with such types : viz. nasality (•"), thus (5) ; forward 
position (•), thus (a); backward position (.), thus (r) ; half length 
(• ), full length (:), thus (fi:fenta-xk) **fireantachd"; extra or 
mere rounding without amounting to full consonantal (w) by 
(«?), thus (uM'l) '*ubhal." I write phonetically (p), (t), (k) for 
Gaelic *b,* *d,' *g/ when not voiced by the nasal prefix, but 
those letters represent sounds not quite identical in their voice- 
lessness to the English voiceless stops, for a slight breath recoil, 
usually used with English (p) (t) (k), is quite or nearly absent 
from Gaelic *b' *d' *g,' and on the other hand Gaelic *p' *t' 



GAELIC PHONETICS. 

TABLE I. 



GAEI 



The symbols beginning at the left are Melville Bell 
horizontally above, thus I instead of Is, so as to include if 
in ordinary or modifications of ordinary type placed immedi 
Primer of Phonetics with the following exceptions : q instea 
Yoicelessness is indicated thus — n' instead of nh. Where an 
nasal sign is repeated in the same square immediately below 
proceeding inwards and downwards and with consonants begin 

Vowels. 



I i 
I 


i 
1 


1 A 
1 


r i 


T i 
T 




[ e 


1 e 
I 


] a 


C e 

C 


X 


> 3 « « 
3^ 3 












J » 




i u 

i 


f p 

I 

1 

f 


f ii 
ft 


1 u 


{I oi 


\ 


1 


« 

^ 


\ 


} 






7 


i 







UND8. 



modified by Professor Sweet, but I use the sign for nasality 
two or more symbols or characters if convenient. The letters 
the right of the Y.S. symbols are also those used in Sweet's 
bhe voiced back open, n inst-ead of n for the front nasal, and 
Don-nasal sound may also occur nasalised the symbol with the 
Sweet's order is reversed, beginning with the narrow front and 
p and nasal sounds. 

Consonants, 





F 


m 








T 


n 


L 


N 


d 


9 












1 


n' 










D 


b 






TD 


d 


ffl 


J 


a 


g 


1 




D 


P 






D 


t 


Q 


C 


a 


k 




) w 




e 


3 


s 


z 


(1) 
(1) 


r 




• 

3 


e 
e 


q 








z 


/ 


s 


8 


O 


r' 


O 


9 


c 

««« 
c 


X 


e h 










CO 
CO 


1 


CO 
CD 


\ 






• 








CO 


1' 









GAELIC PHONETICS 

TABLE II. 



VOWELS WITH DIP 



Test words for short, half long and long, and for na 



1. I ftr, «; ftbr, /. 

2. X ts, 8 ; pwb, /. 

3. [ le, 8 ; U, I. 

4. [ fear (man), « ; (eur (grass), /. 

5. T ciod, 8, 

6. \ Beadh 8. 

7. 1, aithr^flch, «. 

8. 1 laogh, I. 

9. ] agt/s, 8. 

10 (3*" ^*t^reaeh, « ; «/reach, /. 

( 3 ^^^ (palm of hand), s ; bds (death), I. 

11. J cal, /. 

12. f fltbdh, 8. 

13. {I ttgh, «. 

14. "E saor, J /; moghaX^ L 

15. I tt//g, i / ; an t saoer, /. 

16. \ {eabhas, ^ I, 

17. \ loch, «. 

18. 1 gwl, 8 ; wr, /. 

19. 1 gwr, 8. 

20. } robh, «; gh^obh, J /; I6n (marsh), /. 

21. J bho, 8. 

22. } BO, 8; dg, I. 



1. 1 



mti 



2. 


C If«i 


3. 


I mal 


4. 


T iionj 


5. 


i^ 


6. 


X tak 


7. 


1 aon 


8. 


1 a«tl 
\ fnat.] 



9. f bnoD 



10. 


{I o« 


11. 


{ ca< 


12. 


fr 


13. 


Id] 


14. 


1 ID 



15. } U\ 



JS AND TRIPHTHONGS. 



ral. 



nh, /. 



1. l'\^ iasg; 2. I] bnVrgh ; 3. If iiodh ; 

4. I| m/onn ; 5. li &u. 
6. I^ g^fllach ; 7. r^ ieoil. 
8. [I leig/its ; 9. [^ kachd. 
10. OX ^flfch. 11. Cj» b^o. 



1. U^T b^ann. 



12. "[T a chionn. 
, . 13. I^T ann ; 14. 1^1 sonn. 



Ige)«; 
ind)/. 



15. Jlkib; 16. 3C g^^l ; 17. jTcdl. 



18. {I 

19. ^f strooidheil 

20. If hruidhmu. 



,/. 



21. II bwtdhe; 22. l]^ uan ; 23. i] Iw^th 



2. ly: fttatm. 



24. pi ofgh. 



GAELIC PHONETICS. 

TABLE ni. 



CONSONANTS 



Lip. 



1. F m\. 



Point. 



2. T siw. 



Bloi 



5. 1 shndLinh, 



6. D am b\ mi. 



10. Di b\ mi. 
ID pioi. 
ID! o6air. 



M joiob. 

SD 80/?. 

SDfi paj^achd. 



7. (D an (/an I^t© 3 n. 



11. oi (/an. 
ID rurf. 
IDI ear/ar. 



DC ^arbh Dd3(DX> 
fiO sla/. 
SDfi ba^a. 
0-«D lea^. 



14.>^^a. 15. o siuMal. i 16. cdi- oir ^(i)i-; o)-! ro ; iir. i 17. e an «o. 1 



21. > /ear. 



22. o mo shron. 



28. CO /e; cu/. 



30. CO a shhi. 



23. earn fear «G 



1. > /w/iath. 2. o aw/mil. 



3. (Dm;/a. 4.ooamA/uadh. 



Special combinations of consonants in Gaelic not foun 
necessary connections in the above table — cxo ^/ath ; oico rf/uth 



[ TEST WORDS. 



Front. 



3. £i Qvin. 



Back. 
4. J iongsi. 



Throat. 



8. ffl an (/ealanach 



9. Q an ^ael 1^a]|]co. 



12. Q(D (/ealanach 

IQCD idlT IlQO\(D. 



13. ai ^ael. 
la og. 
\Q\ agad. 



QO ^earr Qojo). 
cuifl? acioo. 



Qi-s cinn ; a-ifi cul. 

OQi- trie ; ca-i boc. 

OQi-c iceadh ; CQ<2 eucoir. 



U8. 19. (I) boW/ieach. 



20.€»^a ^/i nan ; eghahhadh 



las. I 25. o chioaaich. 



26cymochridhe;Ccho[r each 27. 2 ri h-am 



i 



29. 00 /it ; /a; mall. 



o.coDomh mil/. 6(1) ou/Ache 



7. e dhoaih. 

8. c c/iomnuich. 



)8t of the languages of Western Europe, and not described a 
dluth; Di tn\xth; ©t an tnuth ; ai cno; on gnoh; Qi an ^«ob 



-J-- >'' 



J. H. STAPLES — ON GAELIC PHONETICS. 397 

' c * are connected with much more distinct breathings than the 
English stops. These breathings I represent by (h), and in some 
cases by (9) and (x). As in *bl' *br' *gl' *gr,' the stops are 
usually more Tocal than in other cases, they are represented thus 
(pbl), (pbr), (kgl), (kgr). 

Although the Gaelic rule of spelling goyeming the last and 
first Towels of succeeding syllables in the same word seems 
generally known, and though readers are familiar with the use 
of the digraphs *bh,' *dh,' *gh/ *th' and * mh/ as medials and 
finals, yet the appearance of the written words seems to the 
stranger of very confused phonetic significance, and the English 
student puzzling over the redundancy of letters forgets the, 
I think, greater irregularities of his own orthography. Gaelic 
writing was changed in Scotland some time in the last century 
in orthography ft'om what had before been common both to the 
Iri^h and Scotch branches. The inaugurators of the change 
aimed at a better representation of the separate Scotch develop- 
ment, but were restrained by regard to customs they were used 
to, and by a desire to preserve the history of words or inflections. 
I shall try to show that, considering the types allowed it and 
the rules its writers felt bound by, some study of its system, 
which is at least ingenious, combined with a careful record of 
its actual speech sounds, and if possible with a comparison of 
them in different dialects, may 'be of some value to the philologist. 

§ 1. One of the first things to be remembered is, that the vowel 
letters added to meet the rule above referred to, have no phonetic 
value, but are simply used to meet pedantic exigeucies, see § 12. 

§ 2. The Gaelic language remarkably illustrates the fact that 
speech sounds have a modifying influence on one another, vowels 
on vowels and on consonants, and especially in Gaelic certain 
consonants appear to exert a modifying influence on vowels, and 
in this language the influencing sound apparently once existing, 
now dead in speech, has remained in writing and left its mark 
in the character of adjacent sounds. The Latin alphabet, in- 
sufficient as regards this, as well as some other tongues, to 
represent all the speech sounds, vowel and consonant digraphs 
have been adopted by its writers to represent sounds for which 
the types furnished no single representative. 

§ 3. The consonant sounds spelt by * d/ * t,' * c,' * ch,' * dh,* 
' gh,' * 8 ' and sometimes * th ' and * sh ' differ in character according 
as they are, or presumably have been, influenced by the contact 



398 J. H. STAPLES — ON GAELIC PHONETICS. 

of the vowels * a,' * o,' or * u ' on the one hand called " broad," 
or by * e ' or ' i * on the other called ** narrow," see § 12 and § 15. 

§ 4. The speech is also nice enough to distinguish between 
advanced and retracted *r,* *1,' *ir (\), 'n' and *nn' (n), and 
as in the other cases, the influencing vowel, ** narrow " for the 
advanced, "broad" for the retracted, may be now extinct in 
speech, but have left behind the consonant position appropriate 
to its contact: see § 12 and § 21. 

§ 5. All the actual combinations of vowel letters are * ai,* * ao,' 
«ea,' *ei,' *eo,' * eu,' *ia,* *io,' *iu,' 'oi,' *ua,' *ui,' 'aoi,' *eoi.' 

* eai,* * iui,' * uai.* The spelling * ae ' only exists as a modem 
phonetic abreviation of " Gaidheil " as ** Gael." Of these vowel 
compounds **ao" always represents one of two simple vowels, 
see Table II, 1st oolnmn, Nos. 8 and 14, 2nd column, Nos. 7 
and 11, and *ei' represents only (e). There seems a repugnance 
to use the letter ' e ' by itself between consonants except before 
final *th.' The compounds * ai,' *ea,' *eu,' *io,' *oi* and *ui' 
only occasionally spell diphthongs, more usually simple vowels 
of which the first letter indicates the sound, often modified by 
the second letter whieh, always decisive of the character of a 
succeeding consonant, often only serves that use. The compounds 
*eo,' *iu,* spell sounds, of which the second letter indicates 
the character, and the first only marks the pronunciution of a 
preceding consonant, or if diphthongs be spelt the first vowel is 
short or weak, the second long or strong. The spellings 'ia' 

* ua ' are customarily diphthongs. The combinations * aoi ' only 
occasionally spell a diphthong ending in (i), otherwise only a 
simple vowel sound, * eai ' is practically obsolete, * a * being 
dropped, * eoi ' spells a diphthong in which * i ' only " narrows " 
a final consonant, *iui' is commonly a simple (u) between "narrow" 
consonants, ' uai ' spells a triphthong unless ' i ' serves only to 
" slender " a consonant, see Table II and § 14. 

§ 6. The vowel sounds spelt by many of these combinations 
are often much modified from what they otherwise would be 
by the presence in the same syllable as finals or medials of ' bh/ 
*dh,' *gh,' *th,* and 'a,' 'ea,' *o,' and *io,' specially by 'U,' 
and 'nn' as finals in monosyllables. Almost all vowels are 
nasalized by the presence in the syllable of * m,* * mh,' * n,* or 

* nn,' and other vocalic sounds and one unvoiced one are nasalized 
by the samie means, see Table II 2nd column and Table III 
lowermost space, and § 13, § 14 and § 18. 



J. H. STAPLES— ON GAELIC PHONETICS. 399 

§ 7. Particularly initial * m ' and final * 1 ' modify very strongly 
the sound of ' a ' in monosyllables, the first by greatly advancing, 
the second by deepening, and so furnish vowel sounds which exist 
in no other circumstances, see § 17. 

§ 8. Another important factor, which seems to have helped 
to decide the state of Scotch Gaelic speech, is the habit of carrying 
to a comparatively illimitable length one vowel or syllable, almost 
always an initial, and of dispatching with the utmost possible 
speed the adjoining syllable, so that short vowels and short 
syllables tend, by being unemphasized and excessively shortened, 
to be broadened or mixed or entirely silenced by the preceding 
long sound, and this, combined with an evident tendency to 
blur over or silence the consonants represented by the ' h ' digraphs, 
when medial or final, helps to explain the fact that many 
apparently dissyllabic words are monosyllabic or even monoph- 
thongic in speech, see § 20, § 21 and § 22. 

' Th ' generally ' h ' as initial, is always silent as a final, except 
sometimes after ** narrow " vowels when it is (9), and also often 
as a medial, chiefly when heard then as dividing a front from 
a back or mixed vowel when it also becomes (9). * Fh ' is always 
silent, with few unimportant exceptions being (h). ' Sh ' existing 
only as an initial with the value of (h) or if absorbing and 
silencing *i * as (9), see § 12 and § 14. 

§ 9. Although Gaelic spelling has generally preserved the 
separation of syllables which once existed separately, by keeping 
the consonantal digraph marking a boundary now sometimes 
extinguished in speech, yet on the other hand there is a dislike 
to the contact of many consonants as medials or finals, and a very 
short but very distinct vowel is inserted as a glide which writing 
omits, and which differs in character according to adjacent sounds. 
This occurs, for instance, between * b ' and * 1,' * 1 ' and * bh/ * 1 ' 
and 'g,' *r' and * b,' 'r' and 'g,' 'r' and 'bh,' *r' and * ch,' 
*r' and *m,' 'n' and * m,' 'm' and *s,' see §23. 

§ 10. The breath glides, which are such a noticeable feature 
before the voiceless stops spelt by * p,' * t,* and * c,* taking the 
place of these vowel glides become in most dialects between ' r ' 
and * t * (s) and sometimes (f) : see § 24. 

§ 11. A process of phonetic decay has evidently destroyed many 
of the consonants in particular words, and in some dialects has 
extinguished some short second or final syllables or final stops 
and nasal consonants usually distinctly touched. But it may be 



400 J. H. STAPLES — ON GAELIC PHONETICS. 

noted that the letters * 1,' ' r,' and * ch ' ** broad " as (x) " narrow " 
as (9) have always full value given them, see § 20 and § 21. 

Examples of the usual normal sounds of the vowels when least 
influenced by adjacent sounds are found on Table II Nos. 1, 
2, 3, 4, 8, 10, 14, 18, 20 and 22. Instances of values of vowel 
digraphs on Table II Nos. 1, 2, 4, 5, 12, 15, both in Ist column. 

It may be noticed from an examination of the vowel digraphs 
and their values that * i ' is always as " narrowing " vowel the 
last in a syllable, and unless it be intended that the vowel in 
the succeeding syllable be sounded ** narrow," ' e ' begins the 
syllable harmonizing with ' i ' in the first: see § 12. *a' and 
*u' are very often used indifferently as ** broadening" vowels, 
' a ' and ' ' may sometimes be used indifferently in some mono- 
syllables often indicating a difference of dialectal pronunciation, 
see § 23. 

Examples are given below in Gaelic spelling and phonetic 
writing illustrating the foregoing matter as marked in paragraphs. 
Many vowel digraphs differ in different dialects thus ** geal," 
**beul," **fior" in some (kjaj), (pial)» (^er), in others {ke-]), (p*l)f 
(fi : r). 

§ 12. The words '^basaich" (pa: siq), "bais" /^pfl://, "laidir" 
(Xd : cj«f) ; Gaelic initial * 1 * is usually (X), but (X) medial or final 
is written ' 11 ' ; ** eadar " (tftcr), ** ait " {a : cq), '* staid " (sta : 09) ; 
final * d ' and ' t ' '* narrow " are both (09) ; " foid " (fdC9), ** ciod " 
(khift), *»cuid" (khuc9), "rud" (rut), "nut" (fuht), *'deagh" 
(cje:q), **an deigh'* (^njje:j), **deas*' (cjw), *'deis" /cjeryy, 

eadhon" (<?qen), "boidheach" (pdjcx), *' toiseach " (thd/cx), 

thoisich" (hd/19), "sios" //i:s;, **fios" (fis), "fior" (fi:r), 
"fir" (fir), "chiosaich" (9ysi9), "fichead" l^qet), "dachaidh" 
(taxi), **nithean" (ni9tfn), "shiubhal" (9uw^l), "mo shuilean" 
(me hu.'len) furnish some examples of § 1, § 2, § 3, § 4, § 8, and 

§11- 

§13. In "feabhas" (fios), "gheibh" (jow), "ghabhaa" in 
Deeside (qo:l), " leam " (lom), common Argyll pronunciation, 
are instances of a rounding effect of medial or final * bh ' or * m,' 
see §6. 

§14. In "an" (?n), "ann" (?5n), "gleann" (kgl^^SN), 
"beann" (p^?5n), "call" (kh/itiX), "mall" (matiiX), "sonn" 
(scQn), "ionnsindh" (jflNsi) or (lesi), observe diphthongal effect 
in many of these words with round raised ending connected 
with terminals ' nn ' and '11,' § 6. The character the nasal gives 






J. H. STAPLES — ON GAELIC PHONETICS. 401 

to the vowel in the article seems to remain with other use of 
the vowel by itself as feminine article and possessive pronouns 
(e) and this is usually the sound of the pronoun *' e " which 
apparently, in some dialects, seems considered '* narrow " in the 
beginning, and broad in the end to judge from its use in connection 
with the substantive verb "is" and emphatic suffix, thus "is 
mise" (is miye), "is tusa" (is tus^), "(iysesan" f/e%eiiji 
" (iysise " ffi/ej. 

The words I'tuais" ftyxt/J, "fhuair" (hucf), "fuaire'* (fucfe), 
"buail" (puel), are instances where the final *i,' otherwise making 
a triphthong, only narrows the adjacent consonant, see § 4 and § 5 ; 
"dh'ith mi" (jiq mi), "tfiiiif" (tfuf), are instances of the sound 
of final * th * after a narrow vowel, and of the mere " narrowing " 
effect of * i' respectively, see § 4. 

§15. In "tigh" (taij), "rinn" (reiw), the round commencing 
diphthongs seem related to the final consonants. In the word 
"rinn" the diphthong may be due to its derivation. 

§16. In "liadh" (Xieq), "seadh" //eq; "lagh'* (\6?q), 
"aghaidh" (eqi), "Hath" (\i^), the 'dh,' * gh,' and *th' seem 
to exercise an effect unnoticed in "fear" (f^r) monophthongic 
or " each" {eey) diphthongic, see § 6. 

§17. In "math" (m»), "mac" (m»xk), "mam" (m»:m), 
and in "al" (f;;l), "cal" (kh»:l), "fal" (f»:l), we notice the 
respective effects caused seemingly by the initial and final con- 
sonants as these sounds obtain under no other circumstances, see § 7. 

§ 18. In "dubh," "subhach" "cobhair," "treabhadh," "ubhal," 
"cuibhrionn," the value of *bh ' has generally died away or amounts, 
and in some dialects only, to a slight rounding, thus " ubhal " 
(utt^l). In "sabhal" *bh' only rounds the vowel thus (s^ul). 
In " aoibhneach," " aobhar," " cabhag," " cubhaidh," " seirbhis," 
" scriobhar," " toibheum," " uaibhreach," * bh ' still remains (v), 
or in some dialects and in some cases (/9), see § 6. 

§19. In "deimhin," "diomhair," "ionmhuinn," 'mh' is (v) ; 
in "amhuil" (w); in "amhairc," "amhluadh," "kunhach," 
"amhghar," " samhradh," "geamhradh," *mh' in most dialects 
only nasalizes adjacent sounds and perhaps slightly rounds vowels 
thus (at^^ua), (s^ft^re'q), see § 6 and § 8 and § 9. 

§20. In " dioghaltach," " glaodhaidh," " chladhaich," "fiad- 
haich," "bodha," the medial 'gh' or 'dh' now usually only indicate 
division between what are or were two syllables, and in some 
dialects the long first is extinguishing the short second, " saoghal ' 

Phil. Tram. 1891-»-3. 1^ 



402 J. H. STAPLES — ON GAELIC PHONETICS. 

is now, only in some places^ dissyllabic but usually monopbthongic 
(sp:1), ''laghach" still dissyllabic or only dipbtbongic in Argyll, 
is in Deeside monopbtbongic like '^saogbal" (Xp^x), ''leigbis" 
fXei/J sbows that * gb ' can now only dipbtbongize tbe vowels 
in some dialects, wbile in otbers 'gb' unpronounced as a con- 
sonant still breaks tbe word into two syllables. 

§21. In "ratbad" (ra:et), "brutbacb" (pbruiex), dissyllabic 
compare witb "bruacb" (pbrue'x), monosyllabic, "atbair" 
(abef), dissyllabic in Argyll, but purely monopbtbongic in Deeside 
as ''bratbair" and "matbair," wbicb, botb tbere and in parts 
of Argyll, are (pbra:f), (m2:f), we can note decay in value of 
* tb,' wbile * i ' in later words only " narrows " * r,* see § 8 
and § 4. 

Tbe value of all dig^pbs is no doubt very irregular, wbile 
" eadbon " is («q^*n), " meadbon ** is (m^^'n) or merely mtf:n. 

§ 22. Tbe words " urrainn " (uiriw), " faodainn " (f^ttN), 
**dealanacb" (cjalenax), but in Deeside (cjalnax), "maduinn" 
(mflltiir), sometimes (mat^), second syllable witbout a vowel, 
"Dombnull" (t5:X), only in far West dissyllabic (tSix), illustrate 
tbe decay of tbe second sbort syllable and its final absorption 
in tbe first long one, see § 8. 

§23. In tbe words "tarbb" (tbar^v) or (tboru), "falbb" 
(falcv) or (falu), "balg" (palck) or '*bolg" (pdlek), "lorg" (X^r^k), 
"ainm'' (?:nim), "earb" (<?r^p), "orm" («rdm), ^'trioblaid" 
(trtptfric9), **aimsir" («ni/*tr), "dorcbadas" (tdr^x^'^'s), syllables 
are added or inserted not apparent in tbe spelling, see § 9. 

§24. Wbile in '*olc" (tflxk), "tore" (tbtfrxk). and "tearc" 
(c^^rxk) tbe voiceless (x), taking tbe place of tbe vowel glide, 
adds no syllable. Tbese voiceless glides between 'r' and H' 
appear thus "ort" (Sfst), "cuairt" (kbuef*+/l) and '*mun cuairt" 
(mu]t)guef'-t/t), tbe latter illustrating tbe vocalizing effect of 
preceding nasal, and 'r' unvoiced by tbe voiceless f/J glide is 
witb it pronounced simultaneously, see § 10. 

Exceptions to some of tbe rules mentioned are tbe words '' is " 
(ts) not fi/Jf and "so" f/^J not (so), and tbe preposition "de" 
is in many dialects (te) not (cje), and " iad " is usually (rt) not 
(iot) or (i^). " De " and " do " prepositions and possessive 
pronouns "do" and "mo" are all pronounced witb tbe same 
vowel (e). Tbe rudimentary difference between "de" and "do" 
prepositions appears in tbe forms of tbeir respective prepositional 
pronouns in combination witb "e"; witb "de" being "detb" 



J. H. STAPLES — ON GAELIC PHONETICS. 403 

or *'dheth" (cje) or (je), with "do" being "da" or "dha" 
{to) or (qa). Sometimes the i of "iad" appears as if at the end 
as {ec<;) the proper pronunciation as if written ** eaid." 

The forms ** deanadh " (cj?:n?q), •* deanamh " (cj?:new) or 
(cj?:na)y acknowledged forms of the present participle of " dean," 
customary in different dialects respectively, also " ugh " sometimes 
** ubh," in one dialect (u:q), in another (u:), in another (u:w), 
in another (u:y), seem very suggestive ; also the participial suffix 
"adh" in "cathadh" (kha-^q) in the West, is in Deeside (u) 
thus (khahu). 

The remarkable word "oidhche" (SiJ^e) in Deeside, an almost 
purely phonetic pronunciation, is in the "West (01390), while a 
Kerry Irishman gave me his pronunciation as (ij^e), by its nasality 
apparently points to a quondam * n * which stiU exists in the 
adverbial locution "an nochd" (en nSxk) and seems to show a 
process of degeneration both of consonant and vowel sounds. 

Evidently Gaelic has undergone much phonetic decay, but it 
still retains with remarkable purity and length a very large 
number of vowel sounds, both front and back. It has an un- 
doubted preference for whispered consonants, and as it uses the 
back open (q) for the modified (d) or (t) instead of (%), and 
(h) for modified (th) instead of (d), we may consider it has also 
a preference for back consonants, which is borne out by every 
"chd" being (xk) and by its using (q) and (j) as "euphonic" 
iosertion between the vowel ending prepositions " de '* or " do " 
and following nouns beginning with a vowel, and also as prefix 
before vowel-beginning verbs thus "foghnaidh na dh'fhoghnas 
ged b'ann de dh'aran 's de dh'im" (ffiini nif qd:ncs ket p'?Sn te 
qareH ste ji:m). 

The preference for whispered sounds is somewhat balanced by 
the evident favor shown to nasal sounds almost always strongly 
voiced. 

A few of the initial lines of one of the most generally admired 
poems by a well-known Gaelic poet, a.d. 1724-1812: "Donna- 
chadh Ban Mac an t saoir" (Mac Intyre) with ordinary spelling 
and phonetic rendering. 

A Mhairi bhan hg 's tu 'n oigh th'air m'aire 
Ki'm hheh bhi far am bithinn fhein ; 
O'n fhuair mi ort coir cho mhr 's bu mhath learn 
Le posadh ceangailt o'n chleir ; 



404 J. H. STAPLES — ON GAELIC PHONETICS. 

Le cilimhnanta teann's le banntaibh daingeann 
8' le snaim a dh'fhanas nach treig : 
'Se t' fhaotainn air Idimh le g^dh gach caraid 
Einn slainte mhaireann a'm clir6. 

a fafi van o:k stun o:ij b^ mare 

ri m veo: yi for ^m bi:ir be:n ; 

n bu^r mi orst kbd:f xo m5:r s pu y& lom 

le pbo:8i' k9^i)ilc9 on x^e:r; 

le Ith^iicnenU C9^ilN slep^dKtiy taii)eir 

sle snaim S qanes nax tre:k : 

/e MtN er Xaiv le kgra kax kbaftc^ 

raiN 8la:nc9e yairSs em yre : . 



Appendix II 



■♦«- 



NOTES 



ON 



THE ORTHOGRAPHY OF THE ORMULtJM 



BY 



ARTHUR S. NAPIER, M.A., Ph.D. 



NOTES 

OH 

THE ORTHOGRAPHY OF THE ORMULUM* 



-M- 



FoB the guttural and palatal spirants Orm retains the Old 
English 2 (in the former case adding an h), while for the guttural 
stopped consonant, as in god, he employs a sign which is represented 
in the printed editions by g. But in the printed editions the sign 
g is not restricted to the guttural stopped consonant : it is used 
also in words like egge (= Modem English 'edge'), where the gg 
had the dzh sound. In other words, the printed editions of the 
Ormulum make no difference between egge (='edge') and eggen/n* 
(= to ' egg on'), though the pronunciation of the consonants in the 
two words was, in Orm's time, the same as now, i.e. dzh in the 
former case and a stopped g in the latter. But on examining the 
MS. I found that, though the editors make no difference, Orm did. 
The letter with which he always denoted the guttural stop (as 
in 9od, gladdy eggenn, &c.) is perfectly distinct from the sign which 
he used to express the dzh sound (as in egge * edge,' seggen ' to 
say,' &c.). The latter, which in ike following remarks, as well 
as in the transcript of the fieu^simile, is denoted by g, has the 
form of the continental g — : cf. bi^^enn in the facsimile, lines 2, 
16, 18, 20, 41, 47, and Mgg«nn, line 6. The former, which I shall 
denote by g, may be described as a sort of compromise between the 
Old English 2 tuid the continental g : it has, in common with this 

^ These notes are reprinted, with Dictionary dat^ from about laoo. 

slight alterations, firom the Academy, A still earlier instance may be fonnd 

March 15, 1890. in the Lindisfiune Gospels, Mark zy. 

' The earliest quotation for this 1 1 , where eancitaveruni is glossed by 

word given in the New English ge-eggedon. 



2* THE OBTHOG&APHY OF THE OBMULUM. 

latter, the closed apper part, thus differing from the Old English 2 ; 
but it has, in common with the Old English 2, the straight horizontal 
top stroke, which projects to the left as well as to the right of the 
letter — : cf. ^odd^ lines 4, 9, 10, la, 38, ftc, 6tgmn«rft, line 
13, ftc This straight horizontal top, especially that part of it 
which projects to the left, is its most characteristic feature, and 
serves to distingnish it from the g, frt)m the round top of which 
a short sloping stroke extends to the right, there being no stroke 
whatever to the left. The absence of any stroke to the left of the 
top of the g at once distinguishes it frx)m the g. Except for 
the one or two isolated instances mentioned on page 4, Orm 
never confuses the two signs, but always uses them correctly, g 
denoting the guttural stopped consonant, and g the dzh sound« 
I give a few instances — ^the pronunciation, g or dzh, is added in 
brackets, the number which follows denotes the number of times 
I have met with the word in question in the Ormulum MS. : 
«gg0 * edge ' {dzh~4) is in each case written with gg ; e^enn * to 
egg on ' {gs), «ggt»»ng (^i) are in every instance spelt with gg. 
The verb 5tggenn 'to buy' {cUA'iS) is always written with gg, 
being thus invariably distinguished from 5tggem» 'to dweir 
(y-20). The verbs leg^emi *to lay' {dzhr-i)^ and seggenn *to 
say' (0^^-33) are in every instance written with gg, while the 
Scandinavian <ngg * flEtithful ' (5^3), ^^^^^Brrle^e * love ' (gr-a) are 
spelt with gg. 

If any proof is needed that Orm's Mggenn, &c. really had the 
dzh sound, it is afforded by the use of the sign g in the Romance 
word ^yn (OrmtUum, ed. Holt, I. 245, /wrrA enoterr gyn, * through 
wise art '). This gyn or gin appears in other early Middle English 
writings, meaning, as here, ' skill, art,' or ' a mechanical contrivance, 
a machine.' It also got to be used in a bad sense, ' cunning,' and 
' a snare,^ surviving in the latter meaning in the Modem English 
' gin.' It comes from the Latin ingenium^ through the medium 
of the Old French engin^ Some writers have, it is true, regarded 
it as Scandinavian, and brought it into connexion with the Old 
Norse ginna, * to deceive.' But the pronunciation of the Modem 
English word entirely precludes the possibility of a Scandinavian 



THE OBTHOG&APHT OF THB ORHULUH. 8* 

origin. Moreover, the varioua Middle English meaniogs are more 
eaaily and naturally explained from the Bomance engin than from 
the Norse gitma. 

The Bomance origin of ^g/n then being admitted, its initial 
consonant must, in Orm's time, have had the dzh sound ; thence, 
as we hav« every reason for supposing that Orm did not use this 
sign for more than one sound, we may assume that, wherever it 
occurs, it had the value of dzh. The fact that in words, whether 
proper names or not, borrowed from Latin Orm always uses 5 
before back vowels and g before front vowels, serves as a further 
confirmation of this. He writes qua^^^an, go/t^e, and OMguM^iMt, 
but e^ppU and magy (= magt). 

The later language shows that, when ng was originally followed 
by % or y, the g underwent the same assibilation as the gg ; instances 
are ' hinge,' ' singe,' &c., so that one would expect to find Orm in 
such cases writing n^ and not ng. Now, wherever ng is preceded 
in native English words by e, an t otj must have originally followed 
the ng (to this there are very few exceptions — ^the preterite hmg 
'hung' is one), so that wherever the combination eng occurs, we 
should expect to find it written 0r»g, unless the g was immediately 
followed by some consonant which protected it from assibilation, as 
in ennvjUsAy I^nngw, or in the case of Scandinavian words. Accord* 
ingly we find hUen^e ' belonging to' (1. 3230) written with g. The 
other words in question are {heKjefimgett, ' angel ' ; henngedd, &c., from 
henngmm, 'to hang, crucify'; ^rermgdenny 'thronged': wmgess, 

* wings ' ; strengenn, * to strengthen '; genge, * a company' ; gmigenn, 

* to avail, assist * — ^but they are always written with 3, never with g. 
In the case of {heh)enn^eUy the explanation is simple enough. In 
all the cases except the nominative and accusative singular the 
g was protected from assibilation by I {ennvjfeM), and the influence 
of these forms protected the g in the nominative and accusative 
singular ; but, in the case of the other six words, no similar expla- 
nation is possible, so that, unless we are prepared to adopt the 
unlikely assumption that in Orm's dialect assibilation only took 
place in the case of ggt but not in the case of ng^ we are driven to 
the conclusion that all the six words are of Scandinavian origin. 



4^^ THE OBTHOG&APHT OF THE ORMVLUM. 

In his article on the Scandinavian loan-words in the Ormulum 
(Paul and Braone's BeUrdge zur GesMehU der deutschen Spraehs 
und LitercUn/Ty x. i) Brate comes to the conclusion, on quite other 
grounds, that henn^dddj/frenn^dennf and toen^eaa, are Scandinavian ; 
but the remaining three he regards as native English, because of 
the lengthening of the root vowel before ng. But, if we suppose 
8tren^enny genge, ^jenynn to belong to an older stratum of loan- 
words borrowed previously to the lengthening of e before ng, 
that objection would lose its force. 

The following is a list of the words in which I have found the 
sign g used ^ : bi^^enn * to buy ', oMggenn ' to atone for ', hilengej 
egg« 'edge', E^ppte, gyti, feggenn, ifagy, «egg«nn, m>/«egg«nn«. 

That in a few isolated cases Orm should have let the wrong sign 
slip through his fingers, writing g for g and conversely, is but 
natural. Amongst the hundreds of instances of the word "^odd 
which I examined, I only found a single example of the misspelling 
^odd (line 2161). Line 3995 the MS. has glutermesse for v^ltaerr- 
nesae. Conversely in 1. 8772 e^ippte is written instead of the usua 
e^ppte, of which I have noted twelve instances. 

A glance at line 10 of the facsimile will shftr that Sweet's 8tat« 
ment {History of Engl. SotmdSf p. 160) that the Ormulum only h 
/, requires modification. A list of some of the words in which 9 
found, is given by Holt, I. p. Ixxz. 

It will be also noticed that the of the combination eo 
generally been erased, and that in most cases the has been af 
added by another hand. 

' The words with ^ in the Latin this Latin part of theAfiS., emp 

portion of the MS. (i. e. the portion the continental ^ in all cases, 
between the Dedication and the Pre> ' Onn regularly writes set 

face, which contains Latin texts) are, with an initial s. In three in 

of course, left out of consideration, as (U. 8439, 918a, 9188) the s h; 

Orm does not use the sign g at all in altered to g. 






N 



- "- * 



// 






— — _ . •'• 



T^ WtMV?' g"-. 



u^t^^m% ' '» 



Ormulum MS. 



Part ( 



\ 



_« 



/ 




%» 









i V . 




PHILOLOGICAL SOCIETY. 

(MEETING AT UNIVERSITY COLLEGE, LONDON, W.O.) 

1893. 



COUNCIL, 1892-93. 

Praident. 
PROF. A. S. NAPIER, M.A., Ph.D. 



Vici'Presidmtt, 

WHITLEY STOKES, D.C.L., LL.D. 

THE REV. RICHARD MORRIS, LL.D., M.A. 

HENRY SWEET, M.A., Ph.D., LL.D. 

JAMES A. H. MURRAY, LL.D., M.A. 

THE REV. PROF. W. W. SKEAT, Lirr.D., M.A., LL.D. 

THE REV. PROF. A. H. SAYCE, D.C.L., LL.D. 

HENRY BRADLEY, M.A. 

Ordinary Members of Ccmneil. 



J. BEUZEMAKER, M.A. 
E. L. BRANDRETH, ESQ. 
TALFOURD ELY, M.A. 
C. A. M. FENNELL, Litt.D. 
H. HUCKS GIBBS, M.A. 
I. GOLLANCZ, M.A. 
PROF. P. HEATH, Ph.D. 
T. HENDERSON, M.A. 
PROF. W. P. KER, M.A. 
PROF. T. DE LACOUPERIE. 



R. MARTINEAU, M.A. 
W. R. MORFILL, M.A. 
J. PEILE, M.A., Litt.D. 
THEO. G. PINCHES, ESQ. 
PROF. J. P. POSTGATE, M.A. 
PROF. C. RIEU, Ph.D. 
PROF. RIDGEWAY. 
W. H. STEVENSON, ESQ. 
PROF. J. STRACHAN. 
MAJOR TEMPLE. 



Treasurer, 
BENJAMIN DAWSON, B.A., The Mount, Hampstead, London, N.W. 

Eon, Seerstarjf, 
F. J. FURNIVALL, M.A., Ph.D., 3, St. George's Square, Primroee Hill, N.W. 

Bankers. 
MB88B8. BARCLAY, RANSOM, & Co., 1, PaU MaU East, S.W. 



Em T&ANCB Feb £1 Is, ; Subscription, £1 \s. a Year (due every Ist of January), 

OR £10 10«. FOR LIPS. 



Fublishers of the Transactions, 
KEGAN PAUL, TRENCH, TRUBNER & Co., Limited, London. 



IV Members of the Philological Society, Nov. 1893. 

1875. Dr. C. A. M. Fennell. Barton Cottage, Cam- 
bridge. 
1877. ♦Donald W. Ferguson, Esq. (Messrs. Kegan Paul, 

Trench, Triibner & Co., Paternoster House, Charing 

Cross Road, London, W.C.) 
1888. T. o Flannaoile, Esq. Pendennis House, Dunbar 

Road, Upton, E. 
1872. Gaston Philip FoX, Esq. 34, De Vere Gardens, 

Kensington, W. 
1842. Danbv P. Fry, Esq. 138, Haverstock Hill, 

N.W. 
1847. ♦Dr. F. J. Furnivall. 3, St. George's Square, 

Primrose Hill, N.W. {Eon. Sec.) 



1859. ♦H. Hucks Gibbs, Esq. St. Dunbtan's, Regent's 
Park, N.W. 

1879. Dr. J. Hall Gladstone. 17, Pembridge Square, 
Bayswater, W. 

1892. I. GoLLANcz, Esq., 154, Houndsditch, E.C. 

1862. Dr. Clair J, Grece. Red Hill, Surrey. 

1869. The Rev. Walter Gregor. Pitsligo Manse, Fraser- 
burgh, Aberdeenshire. 



1868. Prof. John W. Hales. 1, Oppidans Road, Primrose 

Hm, N.W. 
1862. ♦Sir Reginald Hanson. 4, Bryanston Square, W. 

1879. ♦Prof. J. M. Hart. Cornell University, Ithaca, New 

York, U.S.A. 
1889. Alex. S. Harvey, Esq. 3, Dr. Johnson's Buildings 

Temple, E.C. 
1893. Prof. Frank Heath, 29, Douglas Road, Canonbury, 

N. 

1880. ♦Prof. H. R. Helwich. 29, Neugasse, OberdobUng, 

Vienna. 

1881. T. Henderson, Esq. 1, St. Michaels Road, Bedford. 
1849. The Right Rev. Lord A. C. Hervey, Lord Bishop 

of Bath and Wells. The Palace, Wells, Somerset- 
shire. 

1868. J. N. Hetherington, Esq. 4, Lansdowne Road, 
Netting Hill, W. 

1854. ♦John Power Hicks, Esq. Clifton Lodge, Blomfield 
Road, Maida Hill, W. 



Members of the Philological Society^ Nov. 1893. v 

1864. ♦Shadworth H. Hodgson, Esq. 45, Conduit Street, 

Regent Street, W. 
1876. C. R. Hodgson, Esq. 42, Queen Square, W.C. 



185-. Martin H. Irving, Esq. (care of Messrs. Robert- 
son & Co., Warwick Square, E.C.) 



1878. C. S. Jerram, Esq. 134, Walton Street, Oxford. 



1892. ♦Prof. William Paton Ker. 95, Gower Street, W.C. 
1882. R. N. Kerr, Esq. 



1886. Prof. Terrien de Lacottperie. 54, Bishop's Terrace, 

Fulham, S.W. 
1869. *The Hon. and Rev. Stephen Willoughby Lawley. 

Spurfield, Exminster, Exeter. 
1893. Dr. Karl Lentzner, 113, Iffley Road, Oxford. 
1890. Prof. Jas. Alex. Liebmann. Rondebosch 7, Cape 

Town, South Africa. 
1862. ♦©. Logan, Esq. 



1883. *The Rev. A. MacDiarmid. The Manse, Grantown- 

on-Skey, Scotland. 
1890. W. Stuart Macgowan, Esq. 1, Montague Lawn, 

Cheltenham. 
1892 George MacLean, Esq. 27, Montague Street, Russell 

Square, W.C. 
1886. W. C. G. Macpherson, Esq. Howrah, E.I.R., 

Bengal, India. 
1867. Prof Russell Martineau. 6, Eldon Road, Hamp- 

stead, N.W. 
1842. C. P. Mason, Esq. Parkside, Caterham. 

1873. The Rev. J. B. Mayor. Queensgate House, King- 

ston Hill, S.W. 

1884. ♦F. D. Mocatta, Esq. 9, Connaught Place, W. 
1854. ♦Lord Robert Montagu. 41, Queen's Gate, S.W. 

1874. ♦W. R. Morfill, Esq. 4, Clarendon Villas, Park 

Town, Oxford. 
1862. The Rev. Dr. R. Morris. Harold Wood, Essex. 



VI Members of the Phihhgical Society^ Nov. 1893. 

1871. Sir Charles Murray. The Grange, Old Windsor. 
1868. Dr. James A. H. Murray. Sunnyside, Banbury 

Road, Oxford. {Joint-Editor of the 8ociety*s 

Dictionary.) 



1886. Prof. A. S. Napier (Presidenf). Headington Hill, 

Oxford. 
1892. H. A. Nesbitt, Esq. 7, Newburgh Road, Acton, W. 



1881. T. L. Kington Oliphant, Esq. Charlsfield, Gask, 

Auchterarder. 
1874. Owens College, Manchester, (care of Cornish, 

33, Piccadilly, Manchester.) 



1873. Prof. Arthur J. Patterson. IX. Lonyay-utca 11, 

Budapest, Hungary. 
1892. Arthur Paul, Esq. 4, Berkeley Road, Crouch End, 

1866. Dr. J. Peile. Master, Christ's College, Cambridge. 
1889. Miss C. Pemberton. Jagerhaus, Meran, Tirol, 

Austria. 
1886. Theo. G. Pinches, Esq. 62, Newman Street, W. 
1880. *Prof. J. P. Postdate. Trinity College, Cambridge. 



1888. Prof. R. Quattrocchi. Arezzo, Italy. 



1884. The Rev. J. Richardson. (London Missionary 
Society, 14, Blomfield Street, E.C.) 

1882. *William Ridgeway, Esq. Gonville and Caius 
College, Cambridge. 

1869. Prof. Charles Rieu British Museum, W.C. 

1889. M. L. Rouse, Esq. 3, The Manor Way, Black- 
heath Park, S.E. 



1879. ♦The Rev. A. H. Sayce. Queen's College, Oxford. 
1892. John Sephton, Esq. 90, Huskisson Street, Liverpool. 
1884. J. G. E. SiBBALD, Esq. The Admiralty, Whitehall, 
S.W. 



Members of the Philological Society, Nov, 1893. vii 

1863. ♦The Rev. Prof. Skeat. 2, Salisbury Villas, Cam- 
bridge. 

1880. *Eu8tace S. Smith, Esq. Bonner Road, Victoria 
Park, E. 

1871. ♦!. B. Sprague, Esq. 26, St. Andrew's Square, 
Edinburgh. 

1889. The Rev. W. G. Spurrell. St David's, Pembroke- 
shire. 

1893. J. Herbert Stamp, Esq. Trinity College, Cambridge. 

1886. J. H. Staples, Esq. Bruachdryne, Braemar, N.B. 
1879. The Rev. Dr. Thomas Stenhouse. Stocksfield-on- 

Tyne, Northumberland. 
1893. W. H. Stevenson, Esq. 7, Tooting Bee Road, 

London, S.W. 
1858. Dr. Whitley Stokes. 16, Grenville Place, Cromwell 

Road, South Kensington, S.W. 

1887. Prof. Strachan, M.A. Woodbank, Marple, Cheshire. 
1882. *Mrs. A. Stuart, jun. 19, Regent Terrace, Edin- 
burgh. 

1869. ♦Dr. Henry Sweet. Cambray, South Park, Reigate. 



1883. Major R. C. Temple. (H. S. Eong & Co., Com- 
hilL) 

1881. Henry Walter Thomson, Esq. Hazelbank, Syden- 
ham Hill, S.E. 

1866. Samuel Timmins, Esq. Spring Hill, Arley, 
Coventry. 

1891. Toronto Public Library. (C. D. Cazenove, 26, 
Henrietta Street, Covent Garden, W.C.) 

1887. Edw. Treoear, Esq. Government Buildings, 
Wellington, New Zealand. 

1886. Trinity College Library, Cambridge, (care of 
Deighton, Bell, & Co., Cambridge.) 

1871. Dr. E. B. Tylor. The Museum House, Oxford. 



1892. Universtty op Minnesota. Minneapolis, Minne- 
sota, IJ.S.A. 



1864. ♦E. ViLES, Esq. 16, Wetherby Gardens, South Ken- 
sington, S.W. 



viTi Members of the Philological Society, Nov. 1893. 

1873. M. J. Walhouse, Esq. 28, Hamilton Terrace, N.W. 
1880. Richard Ware, Esq. 88, Heath Street, Hampstead, 

N.W. 
1851. ♦Dr. R. F. Weymouth. CoUaton House, Brentwood, 

Essex. 
1863. Henry B. Wheatley, Esq. 2, Oppidans Road, 

Primrose Hill, N.W. 
1882. *Thomas Wilson, Esq. Rivers Lodge, Harpenden, 

St. Albans, Herts. 
1870. Nicholas Wilcox Wyer, Esq. 3, Matford Terrace, St. 

Leonard's, Exeter. 



Bankers : Messrs. Barclay, Ransom, and Co., 1, Pall Mall 
East, S.W. 

Publishers of the Transactions : Messrs. Keg an Paul, Trench, 
TrObner & Co., Limited, Paternoster House, Charing 
Cross Road, London. 



STEPHEN AUSTIN AKD SOH8, PRIMTERS, HERTFORD. 



Tbeasubsb's Cash Accouirr, 1892. 



QO 



c^ 



^ 



o 


too 


<o 


00 


00 


^ 


o 


>♦ 


CO 


»*« 


<o 






-^ 


00 




00 


od 




w^ 




CO 

GO 


00 





«« 



eg 



■Si 






s 








B 
08 

00 

I 

uo 

o 

• ^ o S O "^ "■ 

o §«° 2^ — 3'-' 



08 



B I 
^« 

a> 

Ps o 

P CO 

o 

3« 






0(0 
O O 



CO 
O 



(O 



>♦ 


o 


•* 


o 


(O 


« 




o 

CM 





o 

H 

<1 







03 »rt 

•2a3 
S 

9 
e 

M 





CO 

O 

•< 

Q 



<« 



O 
P4 

w 

00 



o o o o 

^ (O o o 
•^(0 04 



<0 CO 

•-< eo 



o» 



CM 

o 

00 



u 





00 

o 
H 



eo 



CM 

00 C 



OS 



CA 



oM 

00 • ^— ^^ 

^ O 08 



OS « "^ 



i 



o •> •» » 
r «. • fc 



o 
o 



ii 
s 

o 
S " 


c8 



c^ 



r^^ 



H O 

Hfa 



CQ 






04 

8 

e 

S a 
*« o 

O no 



s 

8 

o 

,0 

ja 



r 



o 

8 

2 



'S 


CO 

o 


g 


00 


'e 


f-i 


§ 


t-4 


M 


f-l 


o 








3 


-< 


Q> 


s 


^ 





p^ 



I 



405 



XV.— THE ACCENTUAL ELEMENT IN EARLY 
LATIN VERSE, WITH A NEW THEORY OF 
THE SATURNIAN METRE. By W. M. Lindsay, 
M.A., Jesus College, Oxford. 

[Bead at the Meeting of the Fhiloloffieal Society held on Friday, March 2, 1894.] 

§ 1. The Nature of the Latin Accent. 

Wb have two means of ascertaining whether the accent of a dead 
language was one of pitch or one of stress. We have the 
phenomena of the language itself on one hand, and we have 
the statements of native Grammarians, so far as they are trust- 
worthy, on the other. Both these means of evidence point to 
the pitch-character of the ancient Greek Accent. The words 
of the language do not show that Syncope and Reduction of 
ITnaccented vowels, which are the characteristic effects of a 
stress accent; the Greek Grammarians' accounts of the Accent 
of their own language point in the same direction. Modem 
Greek has, however, a stress-accentuation, so that the Greek 
Accent must have changed its nature in course of time, though 
at what precise period the change took place, it is difficult to 
say. No douht the nature of the accent differed more or less 
in different parts of Greece ; and the accent in one dialect may 
have allowed stress to predominate over tone at an earlier period 
than another.' Accent is taken into account in Greek Metre in 
the verse of Bahrius, a contemporary prohahly of Augustus, and 
author of a verse-translation of ^sop's fahles. 

When we turn our attention to Latin, we are confronted with 
the difficulty that, while the Latin Grammarians often speak of 
their accent in terms properly applicable only to a pitch-accent, 

' In the N. Qreek dialects, for example, as in the N. Greek dialects of modern 
times, the stress-accent is stronger. (See Hatzidalds in Kuhn^e Zeiteehrift, 
:. 388.) 

FW. Trans. 1S91-2-3. 27 



406 THE AOOBNTUAL BLBMEKT IN 

all the features of their language point to its having been a stress- 
accent. The reduction of the accented vowel (e.g. abi^o, etc., 
but Greek inraf^w, etc.), the Syncope of syllables following the 
Accent (e.g. ohjurgo from oljurigo^ ealdut from ealidus, etc., etc.), 
all indicate unmistakeably the presence of a stress-accent. And 
the difference of its accentuation from Greek, though not a single 
Grammarian definitely informs us of this difference, comes out 
clearly in the treatment of Greek loan-words, especially in the 
language of the less educated Romans. Greek ^o(/>ia became 
Sofia, a stress-accent replacing the pitch-accent, with the result 
of lengthening the accented vowel ; Greek eiBwkov became iddlum. 
Instances like these show that the Romans had much the same 
difficulty as we have, in pronouncing Greek words with a short 
accented paenultima, or with an accented ante-paenultima and 
long penult The difficulty would not be so great for a Roman 
as for us, if his stress-accent, like that of his modem descendant, 
the Italian, was not so strong as ours ; nor would it be so much 
felt at an earlier period, when the distinctions of quantity were 
more vividly marked than in the later Empire. The Hungarian 
language, where the sense of quantity is equally vivid, accentuates 
the first syllable of every word, without detracting from the 
quantity of vowels in the following syllables. No doubt, too, 
the nature of the stress-accent would differ in various parts of 
Italy in ancient times, as it does to-day (see Meyer-Liibke, Ital. 
Gram. § 122, p. 71). In Praeneste, if we are to believe such 
indications as the spelling MGOLNIA for the name Magolnia on 
inscriptions, and perhaps the form eonea for eiconia, a stork. 
Syncope was carried to greater lengths than in Latin, and the 
stress of the accent must have been stronger. But that the Latin 
language of all periods, at which we have definite knowledge 
of it, was a language of stress- accentuation, is proved by all 
the evidence at our disposal, and disproved by nothing except 
the silence of the Grammarians. The same tendency to Syncope, 
which before the literary period produced undeeim out of *unO' 
decern, is seen working in the Early Literary time in words like 
ohjurigo (Plant.), ohjurgo (Plaut. and Ter.), and in the Augustan 
Age in ealidtu and ealdut (the form preferred by the Emperor 
Augustus, Quint. 1, 6, 19), while virdis for viridie asserted itself 
still later, and the same tendency still shows itself in modem 
Italian {e,g. Ital. gridare from Lat. quiritare). And hand in hand 
with Syncope goes the reduction and change of unaccented vowels. 



EARLY iJkTIN VER8B. 407 

How then are we to explain the absence of comment on the 
part of the Grammarians ? We must, I think, take three things 
into consideration. First, that the study of Accentuation, and 
all the terminology used, came to the Romans from Greece. 
The word aceentut itself was nothing but the Greek word vpoai^hta 
in a Latin dress; and not only the terms employed, but the 
descriptions of the phenomena of accentuation are taken directly 
from Greek authorities. In the second place, the contrast between 
their accent and the Greek would not be felt so markedly by 
Koman Grammarians as it would by us, whose accent has so 
much stronger a stress than the Latin or modem Italian, a 
consideration which makes it less surprising that they did not 
remark on the essential difference between the two systems of 
accentuation. And thirdly, the Greek Accent itself had probably 
at the time of these Grammarians already entered on that process 
of change which ended in the stress accentuation of modem 
Greek. The Ghreek writers on accentuation would, no doubt, 
go on using the terminology of the earlier phoneticians without 
perceiving that their terms and descriptions were no longer so 
applicable to the actual phenomena as they had once been; and 
if the Greek contemporary theorists on Accent misused the 
terminology in this way, a Roman imitator might be excused 
for carrying the misuse a little further, in applying the same 
terminology to Latin Accentuation. Indeed the writers on Latin 
Grammar were seldom Romans by birth; they were usually 
Greeks, and would have the same difficulties in describing the 
Latin Accent as a Frenchman in describing the strong stress- 
accent of English. These considerations may explain how it is 
that only a few statements of the writers on Latin Grammar 
are rid of the terms 'high' and 'low' (instead of 'strong' and 
' weak ') accent, such as the remark of a fifth century Grammarian, 
that the accented syllable in a Latin word is the syllable which 
would be heard at a distance, when the others were inaudible 
(Pompeius, p. 127 K. : "finge tibi quasi vocem clamantis ad longe 
aliquem positum . ut puta finge tibi aliquem illo loco contra stare, 
et clama ad ipsum . cum coeperis clamare, naturalis ratio exigit 
ut unam syllabam plus dicas a reliquis illius verbi ; et quam 
yideris plus sonare a ceteris, ipsa habet accentum . ' optimus,' 
quae plus sonat? ilia quae prior est . numquid hie sonat 'ti' 
et ' mus ' quemadmodum ' op ' ? Ergo necesse est ut ilia syllaba 
habeat accentum, quae plus sonat a reliquis, quando damorem 



408 THE ACCENTUAL ELEMENT IN 

fingimus." The same language is used by Servius (fonrth century) 
in Don, p. 426, 10-20 K.). We may then believe the Latin 
Accent to have been in the main an accent of stress, like that 
of modem Italian, though like it (and the accent of the Romance 
languages generally), it may have been accompanied by a higher 
tone than the tone of the unstressed syllables. 



§ 2. Eules of Latin Accentuation, 

The law of Accentuation which prevailed in the Classical and 
subsequent periods is that known as the ' Paenultima Law,' a 
very simple one, namely, that the accent falls on the ante- 
paenultimate syllable, if the paenultima be short; on the 
paenultima itself, if long, e.g. dicdresy dceores, The accent never 
falls on the final syllable, except in the case of apocopated words 
like illie for iUlc{e\ horunc for horunc{e). But there are traces 
in Latin, as Corssen proved, of an older Accentuation, by which 
the accent fell on the first syllable of every word. Cases of 
Syncope like undecim for uno-decem, and of Yowel-reduction like 
infringo from in and frangOy conCldo from cum and caedo, triennium 
from tri' {tres) and annus, point unmistakeably to an earlier 
*unO'decein, *c6ncaido, *triannium, etc. ; so that at some time or 
other the Indo-European Acccent law was replaced by a new law, 
namely, the accentuation of the first syllable. A chauge of the 
same kind seems to have taken place in the Teutonic languages 
(see PauPs Grundriss Germ, Philol. I. p. 839), and probably also 
in Celtic (Thumeysen in Hevue Celtique, vol. vi.) ; and in some 
languages of the present day, such as Lettish, this uniform ac- 
centuation of the first syllable prevails. At what precise period 
the change, no doubt a gradual one, from this earlier system to 
the Paenultima Law began and completed itself, it is difficult to 
ascertain. But there is some evidence that it was still incomplete 
in one particular in the period of the Early Drama; for the 
metrical treatment of words like faeilius, mtdierem (^ ^ ^H) in 
the plays of Plautus and Terence indicates, as we shall see, that 
the pronunciation of such words in their time laid the accent 
on the first and not on the second syllable, and the syncope of 
the second syllable in halneae (in Plautus still halineae) points 
to the same thing. A line in which the metrical ictus falls on 
the second syllable occurs very rarely in their plays (see my 



EARLY LATIN YBRSE. 409 

article in Philologui^ LI. p. 864). And a word like dmidius 
(from m^dius) must have been still accented on the first syllable 
about 250 b.c, for the change of unaccented ^ to If is not found 
on the oldest inscriptions. But though we cannot fix the time 
when Latin words passed from the old to the classical accentua- 
tion, when, for example, sdpientia became sapiintia, Umpestatihm 
became tempestdtibuB^ we can guess, partly from the analogy of 
other languages, partly from the inherent probabilities of the case, 
what the nature of that change was. A long word like sapient i a 
UmpeatatiJnu must have had at all periods a secondary as well 
as a main accent; it could hardly be pronounced otherwise, as 
we can see from our ' own pronunciation of such words as 
' ch^racterfstical' (with secondary accent on first, main accent 
on fourth syllable). So that sdpimtia would be more accurately 
written sdpih^tia. The change from the old accentuation to the 
new would be, in reality, nothing but a usurpation by the 
secondary accent of the prominence of the main occent ; sdpientia 
would become s^pihttia, timpestdtihus would become thnpestdtibtu. 
A secondary accent (perhaps the media prosodia of Yarro) is in- 
dicated for the first syllable of words like armatura by the 
Eomance forms, which treat the vowel of the first syllable in 
the same way as they treat accented a. Italian Fiorentino beside 
Firenze may point to the secondary accent having been stronger 
in the first syllable of Lat. Florentinue than of Lat. Florentia 
(cf. Ital. seppelire, scellerato, etc., with doubling of the consonant 
which follows the vowel with secondary accent). (See Meyer- 
Liibke, Chramm. d, Romanieehen Spracheny I. p. 501.) 

The Latin accentuation of the sentence, as distinguished from 
the accentuation of words by themselves, may be determined 
with a fair amount of accuracy, partly by the help of the remarks 
of the Latin Grammarians, partly from observing the phonetic 
changes of Latin words in the Roman tongues, where an accented 
word or syllable is not subject to the same laws of development 
as an unaccented, partly from the analogy of other languages. 
Examples of Sentence-enclitics are: (1) Enclitic Particles like 
jtt^, r^, n^ (Interrogative). These were always written as ap- 
pendages of the preceding word, e.g. Caesarque, Ciceroque, atqve 
(weakened in pronunciation to *atcy ac)) (2) the various parts of 
the Substantive Yerb. The unaccented nature of erat, erit, etc., 
is shown by Eomance forms like Ital. era and Span, era (Lat. erat), 
O.Fr. ert (Lat. erit), for an accented e would have taken another 



410 THE AOCEMTUAL ELEMENT IN 

form, such as Ital. *iera, Span. *yera (of. Ital. niega, Lat. ne^ai). 
£8f est are written in the MSS. of Plantus, Yirgil, etc., as ap- 
pendages of a Perf. Part. Pass., amatut {amatu^s), amatust, 
amatumst for amattues, amata est^ amatum est. It need hardly he 
said that the extent to which these words, and, indeed, all 
Sentence-enclitics, were suppressed, would depend on the caprice 
of the speaker, on the nuance of thought, on the style of 
composition, etc. No hard and fast rule can he laid down ahout 
them, just as no rule could he made for the use of "s ' for ' is,' 
"re 'for 'are 'in English. A sentence, for example, of Cicero, 
ending with the words Imtum est, is quoted hy a Grammarian 
as an instance of a sentence ending with a monosyllahle (Mar. 
Sacerd. p. 493 K.). (On the rules of Latin Accentuation, see 
my articles in the Classieal Bevtew, Y. pp. 878 and 402.) Of 
Word-groups the Grammarians mention some, e.g. res-publica, 
jus'jurandum, etc., and the Eomance languages point to others, 
e.g. fortS'faeere (O.Ital. forfare, Fr. forfaire), ad-illam-haram 
(Ital. allora, Fr. alors). 

The theory of Bentley and Hermann that the accent was shifted 
a syllahle nearer the heginning of a word when the final syllable 
was elided, so that in a line like Ter. Andr, init. : poeta 
cumprim(um) anim(um) ad scribend(um) appulit, the accent 
would fall on the first syllable of scribend(um), is now generally 
abandoned. There is no evidence to prove it, and what evidence 
there is goes against it. But the versification of Plautus points 
to the retention of the accent of the simple word when an ap- 
pended -qw, -ne is elided, e.g. pr6spereq{ue)t surHiptasq(ue), and 
though Servius (ad .JSn. X. 668, etc.) declares that tatUon, 
Pyrrhin, etc., are properly perispomenon, because they stand 
for tantd-nSf PyrrhUnSy he tells us in a note on another passage 
(ad .^£H. YI. 779) that vid^ [i.e. vides-n^e)] was the actual pro- 
nunciation of his time. 



§ 8. The Accentual Element in Latin Poetry, 

The Latin Accent, we have seen, was an accent of stress, and 
difi^erent from the Greek accent of tone or pitch, though its stress 
was not strong enough to overmaster the quantity of a vowel. 
An educated Boman pronounced drdtor with the stress of the 



BABLY LATIN VERSE. 



411 



Toice on the second syllable,^ but without impairing the long 
quantity of the initial o ; the final o of pond retained its length,' 
in spite of the fact that the stress of the voice fell on the first 
syllable ; the first syllable of piper had the stress-accent along 
with the short quantity,' and so on. It was thus possible for 
the Komans to imitate the quantitative metres of the Greeks, in 
which the harmonious or metrical element of the line consisted 
of the regular arrangement of long and of short syllables, — ^the 
Dactylic metre, for instance, composed of Dactyls (— u u), or 

on occasion Spondees ( ), as in the Dactylic Hexameter (with 

six metrical units), 

— KJ KJ I — KJ KJ I — KJ KJ I — U U I — y_\J I — y 

or as in the Dactylic Pentameter (with fire metrical units), 



— U KJ l—KJ \J 



— I — u u — u u 



KJ 



the Iambic metre, composed of Iambi {y .), varying with 
Spondees, or even with Tribrachs (v/ u u). Anapaests (o w .) 
and Dactyls, as in the Iambic Trimeter (with three metrical units). 



!£ - I u - I Vd - I u - 



ii-|ub£ 



the Trochaic, composed of Trochees (— u), varying with Spondees, 
or even with Tribrachs, Anapaests and Dactyls, as in the Trochaic 
Tetrameter Catalectic (with four metrical units), 



«.u 


— u 


-» u 


— u 


_ u 


— u 



-U -w 



u 



^y. vo\\dKt9 5' avarpeirovfft koi fiAX* ev fie^Korat* 

All these quantitatire metres the Eomans could imitate without 
being driven to that substitution of accented syllables for long 



^ drator was a mispronmiciation of the time of GouMiitias (fifth century a.d. 
See Cons. p. 391 K.). 
' ponSf however, came to be the aniyersal pronimciation in conrse of time. 
' P^P^ ▼*> another 'mispronimciation of Conaentiiu* time (Cons. p. 391 E.}. 



412 THE ACCBNTUAL BLEMB19T IN 

syllables and of UDaccented for short, which we see in English 
imitations of Greek dactyls : 

Thfs is the forest primeval ; the murmuring pines and the hemlocks. 

But a stress-accent like the Latin could hardly be kept from 
asserting itself in Roman poetry; and, as a matter of fact, we 
find that the changes which the Greek metres underwent in the 
hands of these Boman imitators, are, in many cases, due to an 
attempt to reconcile the natural stress-accent of the words with 
their ictus or metrical beat. 

The Dactylic Hexameters of Yirgil and Ovid are hardly allowed 
any other ending than (1)^1^^, iii^, e.g. moenia Romae^ or 
(j^) JL yj* \j -^ \jf ®-8' fna^nus Apollo; the Dactylic Pentameters 
of Ovid are practically confined to the ending ^ u u(») ^ \jf 
yj ^f 6*S* praeeipitata formtj moenia maffna forent; and if we 
seek for the reason why these endings were preferred to Greek 
endings like those just quoted ('Ax'X^oy (four syll.), and ovre 
\tffv fA€Ovto)f we shall hardly find a better one than the harmony 
of ictus and accent in the favoured endings (£. v/ v/i ^ H moinia 
E6mae\ il u* sj -^ ^ mdgnw Apdllo; ^ ^ v/(>) iL w> u ^ 
prakcipitdta fdrentt moinia mdgna fdreni\ as contrasted with 
the conflict of ictus and accent in the rejected endings. 

Similarly with the Roman imitations of the Iambic and Trochaic 
Metres of the Greek Dramatists. The Dipody Law of the Greek 
Iambic Metre prescribes that in the latter part, the (metrically) 
important part, of each Dipody of an Iambic line, a Spondee shall 
not be substituted for an Iambus, so that 

Qoa hr^ hihr\^ | fiat tt^v ifiav | tov Kaphiav 

could not be changed to oaa hrj \vnov fuu jc.t.X. ; for the substitu- 
tion of a Spondee for an Iambus at the end of a Dipody would 
spoil the iambic character of the line. Plautus found himself 
prevented by the large number of long syllables in the Latin 
language of his time from observing this rule, but while he admits 
spondees into these feet, he excludes those spondees which would 
bring the ictus and the accent into conflict. 

vin e^mHU \ mus ? tilam ego dU \ earn et til meUm ? 

is legitimate, but 

vin cUnmUlem ? | vin tUam ego, etc., 



EARLY LATIN VERSE. 413 

is avoided, as uniambic, because the conflict of the natural accent 
ammiitem with the metrical ictus 'fnutim was felt to bring into 
unpleasing prominence the irregular formation of this important 
part of the line. Change the second word to a word in which 
there is not this conflict, say tnutHer, and the line becomes 
rhythmic again : 

vtn tnUtuer ? | vin tttam ego, etc., 

as it is with e'6nmut'imu9f etc. 

In a Trochaic line the correspondingly important parts of the 
line, the parts which give the line its trochaic character, are 
the flrst parts of each Dipody. These, in the lines of the Greek 
Tragedians, may not show a Spondee for a Trochee. In the lines 
of Plautus they may indeed show a Spondee, but the clash of 
ictus and accent is avoided. A line like 

v'irginem hdheo \ grdndSm, dUte \ eiasam atque 'inlo \ cUi \ Urn 

satisfied the Eoman ear, but not 

virginem haheo \ ddtdtam, etc., 

where the natural accent of the word, ddtatam, would not harmonize 
with the ictus dOtd-. 

In a Greek Anapaestic line the second foot of a Dipody may not 
substitute a Dactyl for an Anapaest unless the first foot be a 
Dactyl, e.g. Kamreae KorOave | , but in the Anapaestic Cantica 
(or Choruses) of Plautus a Dactyl is allowed even when the first 
foot is something else, provided that this Dactyl irregularly 
substituted for an Anapaest has an accent not in conflict with 
the ictus. Thus we find Dipodies like _ jl, _ ^ v-r | i?«w»»i 
eaerdU | m, or y_, ^ il , _ ^ ^ | ubiet^mque est Upi \ dum, though 
passim eaerulus \ , with conflict of natural accent eairultu and 
metrical ictus caeriilus is avoided. Similarly in the Bacchiac 
lines of Plautus the second foot of a Dipody, if it substitutes 
a Molossus (_ il il) for a Bacchius (^ Z Z)t may not show the 
irregular long syllable with the stress-accent upon it, and in his 
Cretic lines, while a second hemistich like Amph. 200 : 

ff " . /' ^ " 
dlspirtU{%) Ordinis 

is allowed to begin with a Molossus (^ __ ^), composed of the 
flrst three syllables of dispertitif instead of a Cretic (Ji \j jl), 



414 THE AOCBKTUAL BLEMENT IK 



.// // // _ »i 



a change to dlspSriis drdlnia would not be allowed, where the 
accent falls on the irregular long syllable, the middle syllable 
of ditpMu, 

All these changes of the Greek schemes of metres are then 
so many instances of the Eoman stress-accent asserting a claim 
to recognition. Another clear case of the kind is the treatment 
by the Latin Dramatists of Tribrach words and word-endings. 
While a word composed of three short syllables may be used 
in the Greek Drama without regard to correspondence of accent 
and ictus (0^. vbarof, dyaOoy may, equally with jraTipa, take the 
place of an iambus and the ictus v^aT09, arf'aOov), a Eoman 
Dramatist eschews gwiira and the like, evidently because the 
conflict of ictus {jjeviera^ with accent {gSnera), was in such 
words intolerable to the Roman ear.' 

These examples of the regard paid to the accent of words by 
Roman poets might be considerably increased in number, and 
must not be supposed to comprise the whole evidence of the 
existence of an accentual element in Latin poetry. But taking 
them to be sufficient for our present purpose, we may proceed 
to enquire into the extent to which this accentual element 
prevailed. 

The Latin Dactylic Hexameter, as we have seen, favoured 
endings like moenia Romae, magnus Apollo, which bring accent 
and ictus into exact correspondence. How far was this corres- 
pondence aimed at in the rest of the line? That an exact 
correspondence in each of the aix feet would usually give the 
line an unpleasantly monotonous sound, we can see from that 
line of Ennius, which incurred the ridicule of Lucilius : 

tparm hastu longit campus »pUndet et horrct, 

though it can occasionally be used with efPect, as in another line 
of Ennius, expressive of the measured pulsation of oars : 

poste reeumhiie, vestraque peetora pellite tonm 

* lie back and let the oars strike into your chests.' The reason 

1 Prof. L. liiiller m\nB de Be Metriea\ p. 154, quotes from Seneca, Med, 
450 (an Iambic Trimeter or Senariiu) : 

fugi'muB latOH fitgimue . hoe non eet novum. 

The combination of three short syllables in place of an iambus is not objected 
to in itself. Plautus begins iambic lines with JEt ita {Cure, 639} and similar 
tribraohs, where there is no clash of accent and ictus. 



BARLY LATIN VBBSB. 415 

of the monotony is not far to seek. Since the Latin accent 
attaches itself to the long paenultimate or antepaenultimate 
syllable of a word, its correspondence with the metrical beats 
of a dactylic line tends to make all or many of the feet of the 
line end with the ending of the word, and leaves the line without 
these Caesuras, or divisions of words between the metrical feet, 
which weld the line into a graceful whole. Yirgil, accordingly, 
while insisting on the correspondence of ictus and accent in the 
last two feet, aims rather at Caesura in the earlier part of the 
line, e.g. 

diver I $i eir \ eunupiei \ unt • hoe \ acrior \ idem,^ 

and avoids not merely lines in which the first two feet are marked 
oft from the rest, as Ennius {Ann, 31 M.) : 

vifres | ▼S'taque | corpu^ meum nune demerit anne, 

but even those whose fourth foot is wholly contained in one word 
and ends with that word, $,^, £nn. {A. 42 M.) : 

quanquam multa manus ad \ caSli | eaertUa templa, 
Enn. {A. 215 M.) : 

Brundmum pvilcro prae \ cinctum \ praepeU portu^ 

or uses them designedly with archaic efPect, e.g. Am, I. 33 : 

taniae molts enU Eo \ manam | eond&re gmtem. 

But it is the Comedies of Plautus which supply the best material 
for deciding the question how the Greek metres were changed 
in their tranference to Eoman soil. For while we have only 
fragments of the poetry of Ennius and the other early imitators 
of the Greek Dactylic metres, we have practically the whole 
of the works of Plautus, the first adapter (if we exclude Livius 
Andronicus, and Naevius, his older contemporaries) of the plays 
of the New Comedy, plays which, in their original Greek form, 

^ Afanost identically the same beginning is used as the first half of an Iambic 
Senarins by Pacurins (Tragg, 224 fi.) : 

divorsi eir \ eunupicimut hor \ rorptreipUf 

but the older Dramatist is careful to keep ictns and accent in agreement {divorsi 
fiVreumip'tcimut, div6rt% drcuvupieimut). 



416 THE ACCENTUAL ELEMENT IN 

were, no doubt, familiar to many of the educated Eomans of 
the day. And another piece of good fortune has preserved to 
us the whole of the plays of Terence, a Comedian of some fifty 
years later, who, like Plautus, used in his plays the actual spoken 
language of his time, and is free from the suspicion which attaches 
to some extent to the Augustan poets, but especially to the poets 
of the Silver Age, of the use of artificial diction, of obsolete 
forms, quantities, and possibly accentuations of words. The 
Iambic and Trochaic metres, which Plautus and Terence employ 
in their dialogue scenes {div&rhia or devmrh%a\ are the metres 
which come nearest to the language of conversation and of 
prose literature,' so that by comparison of the lines of Terence 
with those of Plautus we can trace not only the development 
of the spoken language in the second century b.c, but also the 
course taken in adapting these Greek metres more and more to 
Roman requirements. And for our present purpose, the investi- 
gation of the accentual element in early Latin poetry, these 
writers are all important for this reason, that, since their diction 
is patently the diction of the ordinary life of the time, and their 
dialogue-scenes reproduce all the varied phases of everyday con- 
versation, with its tones of banter and innuendo, of entreaty and 
refusal, of threatening, command, and deprecation, they afford 
us an opportunity of testing how far what is called the ac- 
centuation of the Sentence, as opposed to the accentuation of 
the single word, was regarded in Latin poetry. Lines like Plant. 

Stieh, 185 : v'ini iUo ad e'dnam : i'icfade, 

promHtte v'ero ; nX ^rav'dre: est eSmmodUm. 
volOf inquam, fieri: nSn amtttam qu'in etis (Iamb. 
Sen.), 
Stieh, 93: nUn iedeo isti (MSS. ietie); v8i sed'Ste: ego eedero Vn 

BuhsHllio (Troch. Sept.), 
8iich.eS5:J^gone? Tilne. Mihine? n bins, etc. {Troch.). 
Merc. 947 : iki valutsti? quid, par'dntes met valtint? tarn gr'itioit: 
Bene voc'ds, hen'igne dteis . eras apUd te, nUne domi, 

^ Since the ictus falls on long syllables in aU except trisyllabic feet in Iambic 
and Trochaic lines, and since the Latin Accent attaches itself to long (at least 
long paenultimate) syllables, there are clearly more chances in Latm than in 
Greek poetry of the ictus and accent lighting on the same syllable. But if 
we make the experiment of applying the Latin accentual laws to the Trimeters 
of the Greek Comedians, we find that the coincidence of accent and ictus is 
not at all so frequent as in the Senarii of Plautus and Terence. Some other 
influence is clearly at work in the latter. 



EARLY LATIN VEBSE. 417 

have clearly tbe iptutima verba of actual Eoman conversation, 
and would be uttered by tbe actor witb tbe same gesture and 
tone tbat would accompany tbem in everyday life. Did, tben, 
tbis intonation impress itself on tbe metre too? Do tbe 
empbasized syllables receive tbe ictus or metrical beats of tbe 
line, and are tbe subordinate words and unaccented syllables 
relegated to tbe tbeses. An examination of tbe plays will, I 
believe, make it certain tbat tbis, so far as is consistent witb 
tbe quantitative requirements of tbe line, is invariably done; in 
otber words, tbat tbe lines of Plautus, wbile tbey are in essence 
quantitative, being Latin reproductions of tbe quantitative lines 
of Dipbilus, Pbilemon, and otber Greek Comedians, take all 
possible regard of tbe accent of tbe several words, and aim at 
reproducing in tbeir metrical arrangement tbe intonation of 
ordinary discourse. Tbe famous dictum of Eitscbl {Prolegg, 
cb. zv.) : cum quantitatis severitate summa accentus observationem 
quoad ejus fieri posset, conciliatam esse, bas never been success- 
fully impugned.^ Tbe more we learn about tbe metrical and 
prosodical usages of Plautus, and about tbe sentence-accentua- 
tion of tbe Bomans, tbe more we are inclined to regard Ritscbl's 
statement as too weak ratber tban too strong, and to believe tbat 
if we bad full knowledge of tbe actual accentuation given to 
tbe sentence by a Roman of tbe time we sbould find it reproduced 
witb great fidelity in tbese early comedies. Our knowledge of 
tbis subject is necessarily defective. Tbe Grammarians of tbe 
Empire tell us, for example, tbat unde^ tbe Relative, was an 
unaccented or subordinate or enclitic word in tbe Latin sentence, 
wbile unde^ tbe Interrogative, bad tbe accent ; and tbeir statement 
can bardly be doubted, if we consider tbe accentuation of the 
corresponding words in otber languages, of our own ' wben ' for 
example in its use as Relative and as Lidirect Interrogative in 
tbe sentence ' I sball see bim wben be comes back, but I don't 
know wben be is coming ' ; tbey tell us also of a distinction 
between quiif tbe Indefinite Pronoun, as in tiquis^ nequis, and 
quis tbe Interrogative, tbe former being an enclitic or subordinate 
word, tbe latter an accented. But tbey do not add, wbat tbe 
analogy of otber languages would lead us to exx>ect, tbat tbese 
enclitics received an accent wben tbey preceded otber enclitic 

^ The objections of Prof. W. Meyer {AhhandUmgen d. Bayeritehen Akademie, 
xvii. p. 1, Munich, 1884} have been answered by Prof. Langen in the FhiMopui, 
vol. xivi. See also my article in the Joum, Fkil, XX. p. 136. 



418 THE AOCBNTUAL BLBMINT IN 

wordSy that, for example, unde Bel. and guii Indef. were attered 
with a certain stress of their own in phrases like ilnd(e)'luhet ^ 
{Epid. 144), nsquis sit arbiter (the ending of an iambic Senarius, 
Fom, 178), where they precede the subordinate verbs lubet (cf. 
guUubet, qudiubet), and fit. And while they tell us that Inter- 
rogatives were accented, they do not say whether the accent of 
the Interrogative or of the Noun was stronger in a question like 
^M hie homo^it?, which in Plautus always bears the ictus on 
the quit and not on the first syllable of homo. It is seldom that 
they condescend to such minutiae as the difference of meaning 
conveyed by the accentuations siquando and (the usual) aiqudndo, 
a difference pointed out by Bonatus in his note on Terence, 
JSun. 437 : 

icin s'fquando ilia mentionem Phaedriae 
faeit: 

'siquando' et prima syllaba acui potest, et media, tamen variat 
sententiam, and, no doubt, correspoDding to our ' ff ever,' as 
compared with * if 6ver.' They are, as we have seen, silent about 
the secondary accent in long words like tempestatibtia, Florenlinu8f 
and they are equally silent about the secondary accent which 
must have fallen on a subordinate word like unde Rel. in collo- 
cations like und{eyadvenimu», und{eyadveni. In the absence of 
complete information about the accentuation of the spoken Latin 
sentence, it is impossible to compile exact statistics of the number 
of lines in which this or that accented word or syllable has not 
the ictus, and so it is difficult to demonstrate by figures the 
wonderful extent to which agreement of ictus and sentence-accent 
prevails in Plautus and Terence ; though, I believe, that anyone 
who takes the trouble of reading a play or two with his attention 
directed to this point will not be able long to retain any doubt 
on the matter. 

The term * subordinate * is preferable to ' enclitic * in speaking 
of Latin sentence-accentuation. Quintilian censures the rule of 
Hellenizing Grammfirians of the Empire that disyllabic Latin 
prepositions are ' accented on the last syllable ' before the noun. 
He points out that what really happens is that they are joined 
with the noun into a compound word, or word-group, wiiich 

^ On this suppression of a final i in unde^ nempe, atque^ etc., before a word 
beginning with a consonanti see SkutMh, Fonekungtn I. 



BARLT LATIN VBR8B. 419 

takes the accentoation of any ordinary word: eireum'hUora for 
example having one (main) accent on the antepaenultimate 
syllable, like eireumlitiOf etroumdiieiio, eircumtpieio. Priscian 
(Y. 67, p. 188 H.) objects similarly to the statement that qui$ 
the Indefinite Pronoun in siquisy numquitf etc., is an Enclitic like 
TI9 in etrii, and prefers to call siquii a compound with the natural 
accent of a compound word. This fusion of qualifying words 
with the words which they qualify, of dependent words with 
the words on which they depend, is the cardinal point of Latin 
sentence-accentuation. 

Among the subordinate words of the Latin sentence were Con- 
junctions, like et, sed, ut. Eitschl, in his Prolegomena (p. ccliii. 
tqq.), called attention to the fact that they are carefully kept 
in the theses of the line, as in li^m, 2 : 

sequar . Bed Jlnem fore quern dieamneieio; 

and in the first 500 lines of — ^let us say — ^the Amphitruo an instance 
of ui ('that, as') with the ictus can hardly be found, except 
in phrases like Utlubet (v. 396), where it would have the main 
accent of the word, iit-vidliiur (v. 334), where it would have a 
secondary accent. 

The strong stress of Interrogative, and the weak stress of 
Belative and Indefinite Pronouns is reflected in the versification 
of Plautus and Terence, in their treatment of such a word as 
unde. If we examine the instances of unde Relative and unde 
Interrogative in their plays,^ we see that in the great majority 
the Relative stands in thesi and the Interrogative %n arei. 
Similarly quie Interrog., a word which naturally stands at the 
beginning of a sentence or line, will be found in this position 
ffir oftener in Trochaic lines, where the ictus falls on the first 
syllable, then in Iambic, where the ictus falls on the second 
syllable of the line. Prepositions were fused with a following 
Noun into a compound word, and would be entirely without stress 
in a group like ad-caiuamj as in a compound Verb like ao-ekeo; 
but before an enclitic or unemphatic Pronoun they would doubtless 
take an accent dd-me^ dd-eum like Greek ny>09 /te, Engl. ' to him,' 
' for him,' Early Irish for-m on me, fort on thee. This treatment 
of the Preposition before an unemphatic Pronoun is reflected in 
every page of Plautus and Terence, who hardly ever relegate 

1 The list will be found in Skutsch, For$ehuHgeH I. § 5. 



420 THB AGCBNTUAL ELEMENT IN 

it to the thesis unless the Fronoon is elided, e.g. ad-m^e) 'idvmiSf 
and it is probable enoagh that the elision of the Pronoun restored 
the Preposition in actual pronunciation to its unaccented state. 
A good example of the treatment of the group when the Pronoun 
has emphasis is seen in Astn. 772 : 

abs ted aeeipiai, tihi prapinet^ tu hiboi. 

The stressed and unstressed uses of the Latin Personal Pronoun 
have produced two series in the Eomance languages, e,g. Italian 
me, te emphatic; mi, ti unemphatic. These unemphatic forms 
are joined to the verb, e.g. prestatemi il libro ' lend me the book.' 
If something analogous took place in Latin (and I see no reason 
for doubting it), an ictus like that of ierv'd-me 'save me' in 
Cure. 628 (trochaic) : 

PhaHldromef HhseerS, eerv'd me, etc., 

cannot be quoted as an example of the conflict of ictus and accent. 
The accent of the Imperative standing alone would be «^d, but 
the word-group servd^me would take the same accentuation as 
a word like eervamus, eervdtxBy servdte. Though it is rare for 
two iambic words to be allowed to stand together in a line of 
Plautus, presumably because this involved a double clash of accent 
and ictus, we find {Amph, 991) a line beginning with the words 
patir voMme^ where we may suppose the ictus to have fallen 
on a different syllable from the accent in the first word only 
(pdter, but vocdt-me). An example of the emphatic pronoun 
is Men. 1076: tU erus et; tu eervom quaere: til saketo: tU vale. 
The corresponding double series of Possessive Pronouns in 
Eomance, e.g. Ital. mio and mo (e.g. ma-donna), point to an 
emphatic and unemphatic variety of the Latin Possessive, which 
I find reflected in the versification of the Comedians. If we 
contrast, for example, Capt. 261 (trochaic) : 

ut vos hie itidem illi (MSS. illie) apud vob mens eervatur filiue 

and its emphatic meus on the one hand, with the subordinated 
Possessive in a line like Baoch. 251 : 

heu edr meum et cerebrum, meohule, finditur, 
we shall not be inclined to agree with an opponent of Ritschl's 



EARLY LATIN VERSE. 



421 



dictum, who finds in the pater of Adelph. 983 (trochaic) an 
instance of conflict hetween accent and ictus : 

fdeiet, 3 vir 8ptume, 3 pat'ir mi feativiuume. 

The emphatic Demonstrative is seen in phrases like it ego ium, 
ego it sum, which in Flautus always have the ictus on is, as 
contrasted with the ordinary unemphatic usage, e.g. propt'er-eos^ 
prae^-eos, or in a line like Poen. 394 (trochaic) : 

oeului hHjuB, lippitudo mea^ mel hiijus,fel meum, 

although the different metrical treatment of iUl and iUio, both of 
which were in the time of Plautus used either as the Adverb 
' there/ or as Dative of the Pronoun ' to him,' has been obscured 
by the scribes of the MSS. of Plautus, who have usually changed 
illi Adv. to the classical illie, and illic Dat. to the 'doublet' 
reserved for this sense by the classical writers, illi, e.g. C(^t, 278 
(trochaic) : 

quod genui illi ett unum pollens atqus hanoratissumum 

where the MSS. offer illie, the classical form, but where the metre 
requires illi. That Plautus used hoHlne, harUne but hdrum, Kdrum 
before a word beginning with a consonant has been established 
by Studemund {Jahrh, Phil, cxiii. p. 57), though this usage too 
is often obscured by the MSS. In Pseud, 69, for example : harilnc 
ffolUptatum, the Ambrosian Palimpsest alone has preserved the 
true reading harunc, while the Palatine MSS. have altered it to 
the more familiar harum. 

The Latin Grammarians have not given us a list of the sub- 
ordinate or auxiliary Verbs in Latin. We have, however, many 
proofs, if proofs be needed, that the Substantive Verb belonged 
to this category, and its fusion with a preceding word is indicated, 
not only by the spellings of the best MSS. of Plautus, amatust 
{amatus est), amatumst {amatum est), ematast (amata est), pulcrast 
{pulcra est), pulcrumst {pulcrum est), etc., but by its metrical 
ictus in the line, e.g. placitai-sunt {Sec, prol. 21), salvae-sunt 
{Pseud. 1036), pauper-sum {Aul, 88). The usual place of the 
disyllabic forms of the Substantive Verb after a Participle or 
Adjective is at the end of a line, e.g. Asin. 320, salvi-sumus. 

That volo in such a phrase as volo scire was a subordinate 
Verb, like 'will' in English, is seen not only from its invariable 

PliU Trans. 1891-2.8. 28 



422 THE AOCBNTUAL ELEMENT IN 

scanBion 9old scire, but also from its almost invariable ictus 
volo^etrey which surely reflects an accentuation volo-seire. The 
same Verb shows its subordinate character in qttatUumvtt ^ as much 
as you wish/ quamvis (like quantumlibet 'as much as you please,' 
quaml%het\ and we can hardly be wrong in refiiBing to regard 
these common endings of Plautine lines, faetUm volo, faeias volo, 
as instances of the clash of accent and ictus. Dare, hahsre, and 
facers were in certain uses as much Auxiliary Verbs as our * have/ 
'do/ and line-endings like coetHm-daho (^eoquam), miMam'/aee 
(^dimiite) must be judged in the same way sa faetUm-volOf faei^' 
volo. They had in all probability the same subordinate character, 
when combined with a noun to express an action which might 
be expressed by a verb alone, e.g. Jidcm dare 'to promise,' done 
dare ' to present,' operam dare * to attend,' just as we throw 
the stress on the Noun ' noise ' and not on the Verb ' make ' in 
the phrase 'to make a noise,' and this subordination of the Verb 
would imply in Latin its fusion with the Noun into a word- 
complex, fidht'dOf dond'data {donS data at the end of a line, 
Ter. JSun, 564), operdm daham {operam dabam at the end of a line, 
MeauL 110). If this was so, there will be a coincidence, and 
not a conflict, of accent and ictus in such iambic lines as Ter. 
adelph. 473 : 

fldem dans,ytfraiM ee illam ducturum domun 

Phorm. 492 : nondum mihi eredis . Hariolare . Sin fidem do . 

Fabtdae ! 
Andr. 243 : ttane ohetinaU operam dat ut me a Glyceric mieerum 

abetrahat ?, 

or in such trochaic lines as Plant. Mil, 455 : 

ddfidem, $i omittie isto me intro ituram, quojubes. 

Among Subordinate Nouns, like our 'thing,' 'kind,' 'part,' 
in such phrases as ' something of that kind,' ' some parts of 
England,' we may reckon in Lat. ree (cf. qudre, quamdbrem^). 



^ The common phrase ret divina (in early Latin also res dina, like ditem for 
divitentf obliscor for obliviscor) *a sacrifice/ as in an old inscription {C.I L. XI. 
4766) once written as one word REIDINAI (Gen.), like retpubliea, once as two 
words RES DEINA. 



EARLY LATIN VERSE. 423 

modus (cf. qudmodOf guemddmodum), locus in the Adverbial word- 
groaps uhi'loei?, intered-loci (Donatas ad Ter. Eun, 255 (Iambic): 

dum haec loquimur, 'interea loci ad maeellum uhi adventamus)^ 

dies, in quotidie^ postridxe (for posteri dis, like erastini die), and 
so on. 

The normal ictus of the phrase ei-rei operam-daham in Plautns 
is ei ret opsr'dm daham^ and of the various cases of mala-res (the 
equivalent of malum ' punishment/ ' evil in store ') is mala res, 
malae ret, maVim rem, mals re. Mala crux seems also to have 
been treated as a compound Noun in the phrase % in malam erucem 
*go and be hanged,' for the phrase may be qualified by an 
Adjective, i in maxumam malam erucem. The accentuation maldm- 
erucem is reflected by the ictus maVdm erucem. Other cases of 
word-groups, composed of an Adjective preceding and a Noun 
following, are probably hona-fides, mala-fides (hence hon'dn fid!i, 
Most, 670, though forming two iambi, is allowed to end a line, 
because the accent fell not on the first but on the second syllable 
of bona), magnum malum (cf. magnopere), bonus vir, the last two 
phrases showing normally in Plautus the ictus magnUm malum 
(a common ending of a line), bifne vir, biinii* vir (final s usually 
did not produce 'length by position' in the pronunciation of the 
time of Plautus). Also some Numerals with following Nouns 
like trisviri, viginttviri, possibly septhitridnes, to which class we 
may perhaps refer expressions of time like trigintd-dies, vigintl- 
dies (cf. our 'fortnight,' 'twelvemonth'), or of value like trigintd- 
minae, vigintl-minae (cf. our ' sixpence,' * twopence *). The normal 
ictus in the Dramatists is trigintd dies, vigint'i minae, etc., although 
iriginia, viginti, when not used in such collocations, show the 
ictus which we should expect, trig'inta, viginti. 

These examples may suffice to show how the sentence-accent 
of Latin conversation asserts itself in the dialogue metre of the 
Early Drama, though it would not be difficult to double their 
number. Perhaps as strong a piece of evidence as any other 
is the fact that certain colloquial phrases, which evidently come 
direct from the streets of Borne into the lines of Plautus and 
Terence, show uniformly the same metrical accentuation, and that 
too, though there is often no necessity for this from the prosodical 
nature of the words. The phrase of everyday life nil moror 
'I do not care' appears in all the plays of Plautus with the 



424 THE AOCENTTJAL ELBKEKT IN 

ictua almost invariably on the first syllable, not on the second; 
and there is no reason to doubt that a Roman^ of that time 
at least, would pronounce the word in this way with a single 
accent, and that on the important word of the phrase, the word 
nil. The common phrase ' I wish to know ' volo scire might, 
from its prosodical nature, its arrangement of short and long 
syllables, take a metrical accentuation volS 90,^ and we should 
expect to find iambic lines beginning volS setre i^itur, voU scire 
aiitem, etc. But the almost invariable metrical accentuation in 
Plautus is void Bcirey with an ictus which in all probability 
conforms to the usual pronunciation void scire, the subordinate 
Verb volo being fused with its Infinite into something like 
a compound word. The normal ictus quid-istic? or guid isi'ic? 
> well I well!' qui'd istue? or quid istiic? 'what's that?' agrees 
with what the Grammarians tell us of the accentuation of the 
Interrogative {quis, but Indef. quis unaccented), and of the 
oxytone nature of istle, isiiic (for ^istod-ce). The phrase of 
endearment voluptas mea is always scanned by Plautus volUptas- 
meay with a shortening of the second syllable, that is only found 
when the next syllable has the natural accent, e.g. volUptdtiSi 
volUptdtem; and the accentuation voluptds-mea agrees with the 
subordinate character of the Possessive Pronoun (cf. pat^r-mi, 
frat'er-mif maUr-mea). The subordination of the Personal Pronoun 
is seen in vac miserotnihij a common expression of disappointment 
or despair, which bears in the lines of the Dramatists the ictus 
on the last syllable of mtsero and on the Interjection, vae miserS 
mihi, and which would probably be accented vai-miserd-fnihi. 
Another exclamation, eu^ae eugae ! ' bravo bravo,' bears invariably 
the ictus eugae etlgae, and would no doubt have in ordinary 
utterance the accentation eugiaeyeiigae (Greek ei^c). Similarly 
quid ita? quid agis? quid ais? etc., seem to reflect quid^ita^ 
quid-agis, quid-ais, and so on ; quit hie homost ? shows the 
same subordination of homo {quis-hio-homost) as scelus-viril 'the 
wretch ! ' of vir {sceW-virt). 

As regards the other laws of Latin Accentuation, the laws 
of the accentuation of the single word, as opposed to the various 
modifications to which its accent is subject in a sentence, we find 
here too a wonderful amount of agreement between ictus and 
accent in the lines of Plautus. His greatest difficulty must, of 
course, have been with iambus- words like firG, firUnt, and it 
must have been increased by the large number of long final 



EARLY LATIN VERSE. 425 

syllables in the language at the beginning of the second century 
B.C., e.g. -d^, 'H, 'It in 3 8g. Act., -Or, -dr, -ir in 1 Sg. Pass, 
of Verbs, -5r in Nom. Sg. of Nouns and Adjectives ; for Plautus 
and his contemporaries still pronounced eoquat, coquet, coqudr, 
eoquir, etc., though, after his time, the difficulty of keeping the 
long vowel sound before a final -t,^ -r changed the pronunciation 
to eoqudt, coquet, eoqtULr, eoquh'. The Latin Accent, as we have 
seen (p. ), is excluded from the final syllable, so that the 
word-accent was ferd, ferUnt, eiqudt, etc. But the nature of 
the Iambic and Trochaic metres allows no opportunity for the 
ictus to fall on a short syllable immediately followed by a long 
syllable. The x>ermi8sible ictus-forms in an Iambic line are 
(Iambus u II > Spondee ^ !L > Tribrach kj u u » Anapaest 
\j \j !L} Dactyl ^ \j yjf Proceleusmatic \j vj \j kj)* in & Trochaic 
(Trochee !Lyjt Spondee !L — > Tribrach (j \j Uf Anapaest 
\j \j ^f Dactyl l.yj \jf Proceleusmatic vi \j kj u)* There is 
no room for an ictus like ^ . to reproduce the accurate pro- 
nunciation of a disyllabic word like ffro, with short accented 
first syllable and long final syllable. 

This discrepancy furnishes the opponents of Eitschl's dictum 
with their strongest argument. How is it possible, they ask, 
to ascribe an accentual element to the lines of Plautus, when 
by the very nature of the metre used, the accent of those 
numerous iambic disyllables of common use would resist adapta- 
tion to the ictus or beat of the verse? Another argument, as 
frequently used as this one, is that an Iambic line must end 
with an Iambus (i^ jl), which implies an ictus on the final 
syllable of the line, whereas an accent on the final syllable or 
a word is contrary to the Latin practice ; and these two arguments 
are regarded by many persons as conclusive. And yet to my 
mind the very fact that Plautus evidently takes the greatest 
possible pains to overcome these two difficulties in the way of 
reconciling ictus and accent, is one of the strongest proofs that 
he aimed at such reconciliation. Let us examine his methods 
of dealing with these Iambus-words ? 

(1) In the first place, he avails himself of the tendency of 
Latin pronunciation to shorten their final syllable, a tendency 
which asserted itself most strongly in those ending with a long 

^ Similarly our ' note ' has a shorter Towel -soond than ' node.' 



426 THE ACCENTUAL ELEMENT IN 

vowel followed by -r, -t (e.g. am6r, amdt), and in many words 
which, ending with a long vowel, were closely joined with other 
words in ordinary rapid utterances, e.g. csLY^-faxu, domi-restOf 
domo-r^ni, dabo-pldgam, dedH-pldgam^ cito-^rrtf, modo-r^i, prob^- 
f actus f datO'plagam (O.Lat. datdd)} 

(2) Or, secondly, he so places them in the sentence that their 
accent is diverted from the first syllable, whether to the final, 
e.g. honaSquef honaine, honaive, honai-aunt, malai-reSf mald-Jide, 

fidem-dat^ or to a preceding word, e.g. hdc-modo (like qudmodo), 
aliqud-modo, coctum'dabo, viginti-minae. 

(3) Thirdly, he elides their final vowel. 

The usual place to which he assigns an Iambus-word with 
conflict of ictus and accent is before a pause in the sentence, 
e.g. Trin. 1-2: 

Sequere hoc me, gmta^ ut milnus fUngarU tuam. 
Seqdor. sedftnemf^re quern d'ioam tiHscio, 

where the disagreement would probably not be so marked. 
(Similarly before the last Dipody of an Iambic Senarius.) At 
the end of a line iambic words are preferred which would in 
ordinary utterance have no perceptible accent, e.g. tuam in the 
line just quoted, which would be as much an unaccented disyllabic 
as the -ecio of nhcio, and the common endings already mentioned 
{sahai) eumtts, (factum) volo, {operdm) daham, etc. Of trisyllables, 
tribrach- words, as has been mentioned, are never used with an 
ictus on their second syllable, e.g. genera; dactyl-words are 
occasionally allowed with an ictus of the kind by Plautus in 
the first foot of an iambic line, but this is never permitted by 

^ That this shortening of a final long syllable under the influence of a preceding 
short syllable (what is called the Law of the Brevis Brevians, i.e. brevis (syllabf^ 
brevians (sequentem syllabam)) was not a mere metrical licence on the part of 
Plautus, but a more or less faithful reproduction of ordinary pronunciation is clear 
from the statement of Quintilian (1. 6. 21). He tells us that in uncouTentional 
talk the second syllable of ave (have) * hail ! ^ood-day ! ' was shortened, though 
a few punctilious persons took pains to give it its proper Ion? sound. Cave was 
reduced not merely to eavS but to eau'^ to judge from Cicero s story of Crassus, 
on his setting out for Parthia, mistaking the cry of a fig-seller Cauneaa ! Cauneas ! 
(sc. ^cm vendo) for cave ne eaa ! cave ne eae ! as well as from the spelling cautis 
for cave sis (JuTenal, 9. 120). A similar reduction of have to hau may be inferred 
from Phaedrus' fable (App. 21) about the man who mistook for this salutation 
the caw of a crow. Some of these shortened forms forced their way into Augustan 
literature {putd Hor., cavS Ovid, viden Virg., modd, citS, and always hene^ maiSy 
etc. We can understand why these last two Adverbs should have succumbed 
earlier and more thoroughly to the shortening tendency, if we consider their 
frequent use in phrases like ntale/teio, bMefieio, etc. 



EARLY LATIN VERSE. 427 

Terence, who evidently felt an ictus like pect3raj perd'ita to 
involve so violent a conflict between ictus and accent {pictora^ 
p&dita) as to be shunned at all costs; molossus words are not 
infrequent with ictus on the first syllable, when their final syllable 
is elided, e.g. Ter. Andr, init. : 

poeta quUmprim{um) 'dnm(um) a<^8cribend(um) 'ippulit, 

this discrepancy of ictus and accent being regarded as less 
unpleasant than a double discrepancy like seribendUm, Of quadri- 
syllables, words of the form of the First Paeon (u u u — ) ^^ 
the Proceleusmatic {kj \j \j \j)i e-g* bdltniae, bdlln^d from 
Gk. fiakaveiov, which, as we have seen (p. 2), retained their 
early accentation of the first syllable (whence later balneaSy balnea) 
have the ictus also on this syllable with very few exceptions 
in Plautus and even in Terence * ; words of the type _ ^ _ y , 
e.g. Florentine advenite, are by the nature of the Iambic and 
Trochaic metre restricted to the ictus FWrenttnm advenite, etc., 
which corresponds with their accentuation Fldrentinm ddvenite, 
and much the same is true of words of the type u u — — > 
e.g. Bepelirey acSleraioSf or sepelire^ seeler'itoe {skpelire^ ecelerdtos) ; 
choriambic words, e.g. tnterea, consilium, dimidius, have usually 
the ictus Jnterea, consilium, dimtditu, which corresponds to their 
accentuation under the Paenultima law, though the ictus d'imidius, 
etc. (like the older accentuation of the first syllable, p. 2), is 
not at all infrequent. 

These are the chief points in the case for the accentual element 
in the Iambic and Trochaic metres of Plautus and Terence. 
They constitute in my opinion too formidable an array of 
testimony to be disregarded. And anyone who, in spite of all 
this evidence, finds himself unable to believe that the question 
of accent was ever allowed to enter into the composition of the 
lines of the Roman Comedians is confronted with this further 
difficulty that the Eomans themselves evidently read the lines 
of these authors as if ictus and accent were very much the 
same thing. We see this not only from the comments of 
Donatus on the pronunciation of words in some lines of Terence 
(e.g. siqtMndo, see p. 4 ; intered-loci, see p. 6), but from the 

i But other combinatioiu of four syllables of the kind are freely used with ictus 
on the second of the group, e.g. £i etiam {£pid. 524), An adiit? (Mere. 981)» 
where ictus and accent would not be in conflict. 



428 THE ACCENTUAL ELEMENT IN 

express statement of Aolus Oellias (second century a.d.), who 
mentions an appeal to a line of Plautus, ending alwrum affatim estf 
for the accentuation affatim, and to a line of Terence, ending 
exadversum loco, for the accentuation exddversum (the usual 
accentuation of his own time being affatim, exadvdraum), and 
quotes the testimony of a pupil of Valerius Probus, that that 
celebrated Grammarian (of Nero's time) read the line of Plautus 
with that pronunciation of the word. 

At the same time I do not believe that anything is so likely 
to dispel one's doubts about the accentual element in the 
Comedians' dialogue metres as the perusal of one or two of 
their plays after a study of the rules of Latin accentuation, 
particularly sentence-accentuation.^ "We should be prepared to 
find a difPerent state of things in the choric metres of Plautus, 
the bacchiac, cretic, anapaestic, etc., for the passages composed 
in these metres were not spoken but sung, and do not like 
the dialogue passages aim at a close reproduction of the tone 
and diction of ordinary conversation. The Cretic and Bacchiac 
metres, with their abundance of long syllables, especially favour 
forms Hke cavi, mihi, modO {Mc volo scire t'i Cure, 134), which 
were not the forms employed in common speech ; but their ictus 
are not more difficult to reconcile with the natural accents of 
Latin words than are the ictus of the Trochaic and Iambic 
Metres; and we find the accentuation of the Paenultima Law 
fairly conserved in these Cantica. Still it is a significant fact 
that Terence, after making trial of Bacchiacs and Cretics, gave 
up the attempt to adapt them to Latin. The Cantica, however, 
composed in the Anapaestic Metre, seem often to ride rough- shod 
over the accentuation, although it is possible that further research 
into the nature and laws of the passages so scanned, and in 
particular the discrimination of Dactylic from Anapaestic lines, 
may remove some of the harshness of what are usually regarded 
as Anapaestic lines.' A cardinal point of the metre is its 
readiness to substitute not only Dactyls for Anapaests, but dactylic 
words for anapaestic words, a substitution which is as natural 
in Greek as the substitution of a Spondee for a Dactyl, but 

> It has been remarked that some of the lines of the Early Tragedians read 
almost like lines from a Christian (accentual) hymn, e.g. Ennius 163 K. : 

magna t'impla eaelitumj [| commixta stellit sptendid'ia. 

' It is difficult, for example, to believe that uaut sum in alto, diciHt Fromctheut 
were actually so uttered by a lioman. 



SARLY LATIN VERSE. 429 

which inyolves in Latin the use of ictns-forms like pBctHra, 
perd'iCa, not to Bpeak of proper'da, quid^a^Ss ? It is no wonder 
that Terence refused eyen to make experiment of Anapaestic 
Terse in any of his Comedies. 



§4. Hie Saiumum M^tre. 

The earliest Latin imitations of the Greek metres differ in 
two respects from their Greek originals : (1) in the use of 
Alliteration ; (2) in the regard for the natural accent of the 
words. This suggests that the native Latin metre, the Satumian, 
which was used before, and for some time after, the adoption 
of the Greek quantitative poetry, was (1) Alliterative, (2) 
Accentual. The Alliteration was one of its main features will 
be evident to anyone who takes even a cursory glance at the 
Satumian lines preserved to us (about 150 in all), but its accentual 
character (though almost implied by its alliterative side) has been 
recognized only recently, since the investigation into the metres 
of the different Indo-European nations has shown the isolated 
position of the Greeks in their use of an entirely quantitative 
metre. The Roman Metricians of the Empire, when they came 
to write adaptations of the Greek treatises on Metre, and found 
themselves compelled, after a description of the Hexameters of 
Virgil, the Pentameters of Ovid, the Sapphics and Alcaics of 
Horace, etc., to give some account of the native metre used 
at the time of the Funic Wars by Livius Andronicus and 
KaeviuB, were quite at a loss to fit it into the Greek metrical 
scheme. They tried to account for it as a composite of Iambic 
and Trochaic metre, an Iambic Dimeter Catalectic with a Trochaic 
Dimeter Brachycatalectic, though they were obliged to confess 
that hardly more than a single line could be suited to this 
Procrustean scheme. Their model line is the epigram of the 
Metelli on the poet Naevius: — 

ddhunt mdlam Jf^tiUi \\ Naevl6 pditae, 

which has become the stock example of a Satumian. It is 
strange that, although the Roman metricians frankly confessed 
their inability to suit the actual extant lines to quantitative 
rules, the Quantitative theory of the Satumian metre should 
have found acceptance in modem times ; for once we leave 



430 TUB ACCENTUAL ELEMENT IN 

this single example, dahunt malum, etc., we find line after line 
which resists a quantitative scansion, 

e.g. Rundis atqttS Porpuremfilii Temu, 
laqu^ atuUm ad caelum sustulit auas tm. 

For a time it was supposed that these final syllables might have 
been long in Early Latin, just as eurdt^ audit, eanddr had their 
final syllables long in the time of Plantns. But Comparative 
Philology has now taught ns that they were short at all periods 
of the* language, their I.-Eur. forms being -6s, -q^, -dm. The 
-que indeed of Latin atque, iaque was so short a syllable that 
in ordinary rapid utterance the vowel was lost altogether, even 
before a word beginning with a consonant, alque becoming ae 
(for *ate), neque becoming nee, etc., while -urn of eusum, etc., 
was in classical poetry invariably, and in anteclassical poetry 
usually elided, though Flautus and Ennius sometimes reckon it 
as a short syllable like the -urn of eireum in the classical compound 
eircHtnit or circuit. Final -a of the Nom. Sg. of A-Stems, and 
the Nom. Ace. PL Neut. of 0-Stems may have been long at 
some remote period, but in the period of the earliest literature 
it is invariably short, and is scanned as a short syllable by Livius 
Andronicus, and Naevius themselves in their other poems in the 
Greek quantitative metres, so that, e.g. vita, can hardly be a spondee 
in a line of a Scipio epitaph, written c. 130 b.c. : 

quoiei vita defeeit, non honos honored 

It is, however, the recent investigations into the versification of 
Flautus that have given the coup de grdoe to the old theory of 
Satumian Metre. Flautus never uses a tribrach word with 
metrical ictus on the second syllable. How then can we make 
the beginning of yv. 96, 101 iambic? 

subigit omne Loucanam opsidesque abdoueit, 
facile facteie euperasee gloriam maiorum. 

Flautus avoids the use of a spondee with metrical ictus on its 
final syllable in the even feet of an iambic line. How then 
can we scan : consul censdr aidilis. 



^ Dr. Reichardt, the latest champion of the QuantitatlTe theory, allows that in 
127 lines 63 cases occur of a short final where a long syllable is required, as 
against 66 cases of a long final ! {Jahrb, Klau. Phil, (Suppl.) xix.). 



EARLY LATIN VERSE. 431 

And in general, as we have seen, Flautus does all he can to 
bring metrical ictus into correspondence with the natural accent 
of the words. If this care is shown in imitations of Greek 
metre, a metre purely quantitative, can- we believe that the 
native Latin poetry brought ictus and accent into conflict in 
the first half of, I would say, every line : 

eonsSl cena&r aidilit 

annSa gnatiis vigintt 

quoiei vita defeeit 

GnaiuHd poire prognattu, etc., etc. 

Even the more plausible form of the Quantatitive theory, according 
to which Iambic might be at will replaced by Trochaic rhythm, 
and Trochaic by Iambic (e.g. RUncm atque being trochees, not 
iambi), fails to meet these objections, and its author^ has now 
relinquished it for a (more or less thorough) accentual theory. 

A host of arguments might be brought against the Quantitative 
theory, if it were necessary. The single fact, however, that every 
extant Satumian line begins with an accented syllable, is almost 
enough to show that the metre is governed by Accent, and is not 
a composite of a quantitative Iambic and a quantitative Trochaic 
line. For how otherwise can we explain the fact that, while the 
iambic lines of the Dramatists begin again and again with trisylla- 
bles accented on the second syllable {e,g, Flautus in the first scene 
of the Miles has v. 1 Cur'dUy v. 4 Praestringat^ v. 29 Conisus^ v. 39 
Facete^ v. 40 Novhse^ v. 41 Curdmque, v. 44 IHg'inta, v. 57 Uirtilte^ 
y. 72 Uidetur, v. 74 Latrines), we have not a single instance of 
such a beginning of a Satumian line ') ? The natural order of the 
words ' aedile, consul, censor,' the order of the prose Scipio epitaph 
(I. 31 aidiles eoeol ceaor) is not retained in the Satumian epitaph : 
eonsol censor aidilis (v. 94), an inversion which may reasonably be 
referred to the necessity of beginning the line with an accented 
syllable. Aud an examination of the lines in detail will show that 

' Prof. Zander, of Luod, who gives me this information by letter. 

^ T. 67 Pleriq%te omnet is an exception that proves the rule, for Plaatos and 
Terence always throw the metrical ictus in this phrase on the first syllable of 
plerique and the first syllable of omne*^ pointing to a pronunciation of tne word- 
group, like pleriqu{e)-onm«s. v. 105 (on a Scipio epitaph) AetaU quom parua 
shows the common spelline aeUUe insteiEid of the older form, here required by the 
metre, aevitate, by a similar graver* s mistake as the substitution of aetemtu for 
aeviUrmu on an iambic epitaph of Diocletian*s time (Orell. 6017) : divini vis est 
aetemi iemporis. 



432 THE ACCENTUAL ELEMENT IN 

each line not only begins with an accented syllable, bat has three 
accents in the first half -line, two accents in the second, account 
being taken of secondary accents in a degree which, we can believe, 
corresponded to their prominence in actual pronunciation.^ 

This Accentual theory of the Satumian Metre has the merit 
of suiting each and every extant Satumian line.' But stated 
in this form, it seems to me hardly sufficient as a metrical 
scheme. No doubt the Saturnian poetry was stigmatized by 
Horace as horridw tile nuiMrtts, and grave virus ; but could it 
have been so devoid of harmony as lines would be which 
answered only these two requirements, (1) of beginning with 
an accented syllable, (2) of showing three accents in the first 
half, two in the second half ? A good many lines of prose 
could satisfy such a test, the opening sentence of Cicero's First 
Philippic, for example: Antequam d^-republica, || pdtres con- 
scrfpti, Dicam-ea quke-dic^nda || hoc>tempore arbitror ! I , and 
even with * Alliteration's artful aid ' we can hardly suppose such 
a metre to have been favoured by a poet like Naevius, who 
was familiar with the Greek style. 

The counting of syllables, which is a feature of the poetry 
of various Indo-European peoples is an essential element in 
Komance metre. The commonest Romance line, the Decasyllabe 
as it is called in French, the Endecasillabo in Italian, gets its 
name from the number of syllables used, viz. ten with masculine 



^ A five-syllabled word always counts for two accents, e.g. 

Y. 104 fndgna idjnintia || multdaque uirtlUetf 

90 ded^t Ihnpestdtebus || aide meretodj 
66 dnerdriae onitstae || ttdbant infliutris, 

and a four-syllabled woxd (at any rate of the forms — 2£ — ^ and 
bi — w ^) does the same at the beginning of the line, e.g, 

T. 44 immoldbat auream J uietimam pulcram, 
60 »up4rbiUr eontemptim || e6nterit Ug%6n€9f 

91 C6rfUliu9 Lucius |) Sctpio Barbdtut. 

A word at the beginning of a line or sentence would naturally have more pro- 
minence eiven to it than in the middle. For the double accentuation of quin- 
sy llables m other parts of the line than the beginning, the only strong instances 
are y. 129, eomplurimae^, y. 130,^HfiuirtMm||, whose first syllables alliterate witii 
neighbouring words, and would receive extra stress on tnat account, and the 
proper name Seipione, y. 87. 

2 The last line of the epitaph on Naeyius, quoted by Oellius from Yarro, should 
probably be read : Uquier lingua Latina || oblitutunt M6mae, 



EARLY LATIN VERSB. 433 

jending, eleven with feminine ending, e,g, Ariosto Orl, Ftxr, 
IV. 62: 

di vera pudicizia k un paragone, 

where the rhythmical factors are (1) the two accented syllables 
(the sixth, -ei-f and the tenth -yo-), (2) the number of syllables 
(eleven, with elision of -ta h before the initial vowel u- of un), 
(On the Eomance Metres see Stengel in Grober's Gh^ndriss, 
vol. ii.). I believe this counting of syllables to have been 
an element of the native Latin metre, the Satumian. Seven 
syllables are the quota of the first half-line, six of the second, e.g. 

dabunt malum Metelli || Na§uio poetae ; 

and this number will be found to occur in the various lines, 
the laws of elision being nearer those of Flautus and Ennius 
than those observed by the classical poets, in that a final long 
vowel or diphthong or a final syllable ending in -m was not 
wholly elided, but was left in 'prosodical hiatus' (reckoned as 
a short syllable ^) ; and that this treatment coresponded with 
the actual pronunciation we may argue from such forms as 
pri'hendo (from prae hendo\ eireii{m)-it (from eireum it), $.g. 

V. 25 topper eitl ad aed%» || 
40 Uqtie swUm ad caelum ||. 

The monotony is varied by occasionally allowing an extra short 
syllable in positions where in current pronunciation it would 
be completely, or partially, suppressed, to count with a preceding 
short syllable as a single syllable, e.g. 

V. 66 II cdpUihus optrtie 

99 11 glori{a) atq{ue) tng^nium, 

where eapitihus, ingenium are allowed in place of trisyllables, 
just as EduIus {Ann, 267 E.) allows eapilihue to take the place 
of a dactyl at the beginning of a hexameter line : 

eapitthu* nutantis pinoa rectoaque eupressos, 

^ Monosyllabic enclitics were an exception. Their treatment varies in Plautus, 
4.g, in Cos. 591 : 

eum hoc J c&m istaCy euwque amiea etiam meoy 
eum stands in prosodical hiatns, though usually elided. 



434 THE ACCENTUAL ELEMENT IN 

or as Horace has eomilium {C. III. 4, 11), prineipium {C. III. 
6, 6) in place of trisyllables.* 

Further I find that the accents of the line are systematically 
arranged. The extant verses show us clearly two types, one 
(the A-type we may caU it) )ix(») "kx^ X*X II XXX. XXX 

ddhunt mdlum MeUlli || Naeuto poHae^ 

the other (B-type), which is much less common, xx(»} XX> 
XXX II XXXX» XX 

priin{a) incidit Oreris || Prosirpina piur. 

What the historical genesis of either type may have been is 
another matter, but it seems to me that what was present to 
the mind of Livias and Naevius in their actual use of them 
was an endeavour to secure (after the first two feet of the line, 
where the accent was fixed) a regular alternation of accentual 
rhythm, a ' rising ' accent (e.g. MeUUi) being followed by a 
* falling' (e.g. N(Uu%o\ and vice verad (e.g. Cereris by Proserpina). 
The scarcity of our material does not allow us to discover 
precisely to what extent variations from these normal types were 
permitted. A variety of the A-type has in the second half-line 
II XX^X* XX) ^*S- (Mtlocktus siimmi; a variety of the B-type, which 
lacks one of the requisite number of syllables, but satisfies the 
requirements of alternation of accentual rhythi^, has in the same 
part of the line || XXX» x'x» ®-8f« fi*^^ uirum. But that these two 
types were deliberately followed by the writers of Satumian lines 
(more successfully, as is natural, by the poets, Livius and Naevius, 
than by the chance writers of metrical inscriptions) is, I think, 
put beyond a doubt by the actual extant lines. The theory might 
be tested, if the words were by any piece of good fortune dis- 
covered, which are wanting in certain fragmentary lines. For 
example, a fragmentary line of Naevius, quoted by Festus 
(472 Th.) : quod hruti nee-%atia sardare queunt, will, if my theory 
be right, lack four syllables at the beginning so accented xxCO XX 
(perhaps 

dietm Br^tM, quod IrUti || neesatis sardare 
quhtnC). 

* That this ' resolution * was recognized as a licence, a permissible departure 
from the re^lar number of syllables, is 8hown by the limits within which it Is 
confined. Two such * resolutions * are avoided in the same half-line, probably 
even in the same line. 



XARLY LATIN VBKSB. 435 

Eules of Caesura are often given for the Satumian Terse, but 
improperly, for they are dependent on the three main rules of 
accentuation, number of syllables, and alternation of accentual 
rhythm, and are implied by them. The only caesura that really 
formed of itself an element of Satumian metre was the break 
between the two Imlf-lines ; the others, viz. the occasional break 
between the first and second 'feet' of the first half-line (e.g. 
dabunt malum), and the usual breaks (I) between the second and 
third 'feet' of the first half -line (malum Meielli), (2) between 
the first and second ' feet ' of the second half-line {Naeuio poetae\ 
are merely the conditions under which a certain number and 
arrangement of accents, combined with a certain number of 
syllables, can be secured for the line. But Rhyme (in the final 
syllable or syllables of each half-line)^ was an ornament much 
sought after, so that the regular structure of the verse seems 
occasionally to be sacrificed for the sake of securing it, ^.y. 

V. 47 {Lritf pdpHldtur vd$lat, || r0{m)-h6sUuin caneinnat^ 

instead of the usual rhythm: iirit, vdstat, piSpHldtur, like v. 131 
Fundii, fitgat^ prdstemtt. And there are possibly traces of an 
arrangement of lines, usually or occasionally, in distichs, any 
peculiarity of rhythm in the first line being * echoed ' by the 
rhythm of the second e.g. 

TV. 18, 19 ndmque nitUum piius || mdcerat humdnum 

qudnde mdre saiuom I uU-eUeui iunt-mdgnae, 

though the number of quotations by the Grammarians of anything 
but isolated lines is so small that we have not data enough to 
warrant a certain inference. 



Naevius, who died at the end of the Second Funic War, was 
succeeded in the domain of Epic Poetry by Ennius, whose in- 
fluence effected the replacement of the native metre by the Qreek 



* The final syllable was, of course, unaccented. Similarly rhyme in Commodian 
and S. Auznstine is of unaccented finals, while in Romance poetry it is accented 
syllables which bear the rhyme. 



436 THE ACCENTUAL ELEMENT IN 

quantitative Hexameter.^ The Sattumian disappears altogether 
from onr ken in the course of the second century B.C.; and in 
the plebeian epitaphs of the later Republic and Empire the 
common metres are the Iambic and Trochaic. (A firequent ' poor 
man's epitaph/ for example, as frequent as our ' Affliction sore 
long time he bore ' is : 

noU doUre^ mater^ wentwn nuum, 
proper avit aetas ; hoe vduitfatui mmts. 

* Dear mother, do not weep for me. 
Short was my course ; 'twas fated so to be.') 

But if the new theory of Prof. Stengel (Gr6ber*s OrundriM der 
Romaniiehm Philoiogie^ II. p. 19) be right, the Satumian metre 
must have remained in popular usage down to the Romance times, 
though no trace of it has been left in literature or in epitaphs. 
{Inser. Neap, 8829 : 

rogo te, mi viator, noli mi nocere 

may be a Satumian, but is too doubtful to quote.) Prof. 
Stengel, who refers the common metre of Spain, the fourteen- 
syllabled line, with accented seventh and fourteenth syllables, 
to the Latin Trochaic Septenarius (Tetrameter Catalectic). sees 
in the commonest Romance metre, the D6ca8yllabe of France, 
the Endecasillabo of Italy, a direct development of the Saturnian. 
This line he shows to have had at an earlier stage more than 
ten (or eleven) syllables, and to have had an accent on the sixth 
syllable, as well as at the end of the line ; and this internal 
accent he ascribes to the accent at the end of the first half-line 
of the Satumian, e,g. 

dahunt malum Metelli 
V. 1 1 partim errant nequfnunt 
V. 4 argenteo polubro. 

He supposes that the Paenultima law of accentuation had often 
the effect of weakening the accent at the beginning of the line, 

^ There seems to have been a later edition of the Odyssea and the Bellum 
Poenieum in hexameters, perhaps for the use of Roman school-boys, for some of 
the lines quoted by the grammarians have unmistakeably this form, s,ff. PriBcian*s 
quotation (I. p. 419 K.) from the Odyuea : 

cum socios nostros mandiaset impius Cyclops. 



EARLY LATIN VERSE. 437 

ArginUo becoming argenieo with no perceptible secondary accent 
00 the first syllable, so that the accentuation at the beginning 
of the line came to be regarded as unfixed : 

pdrtim hrant 
argintsOf 

while the accent at the close of the half-line (and of the whole 
line) continued to assert itself : 

partim errant nequknurU || 
arg&nUo poliihro ||. 

However that may be, it is certain that the popular poetry of 
the Republic, as well as of the Empire, was markedly accentual. 
The few examples that*have been preserved are mostly Trochaic 
Septenarii, being lampoons of soldiers on their general (on the 
day of a triumph such abuse was thought to avert the Nemesis 
that attends undue prosperity), e,g, the lampoon on Pompey 
(quoted by Sacerdos, p. 462, 1 K.), of accentual, somewhat 
halting, rhythm : 

qu6m non pudet 6t rubet non est homo sed ropio, 
or the more rhythmical one on Caesar (Suet. Com. 80) : 

ecce Ga6sar nunc triumphat qui sub^git Qdllias.^ 

The recognition by the Christian writers of this accentual metre, 
and the process by wliich the 'poor man's poetry' became the 
poetry of the Church and the Nation, are matters which lie 
outside the province of this article. I will only mention that 
the recent investigations of the poetry of Commodian suggest that 
a better knowledge of the exact pronunciation and accentuation of 
the language of his time will explain the rhythmical structure of 
the hexameters of the ' Instructiones ' and ' Carmen Apologeticum.' 



* The call of the ' retiarius ' gladiator may be a (looee) accentual Trochaic 
Tetrameter : 

n6ii te p^to, plscem p^, quid me f (igis, O&lle ? 



FhU. Trani. 1891-2>3. 29 



438 THB AOCSMTUAL ELEXSHT IK 



APPENDIX. 
Satummulm Ysbsb — A. Lm oj all thx (ookflxib} Exzakt Lnrxs. 

A-Type )Jx(,) XX, XXX D XXX , XXX (<» D XXXX> Xx) 

B-Type xx(,) XX, XXX D xxxx, XX (<>' XXX , xx) 

(Lillet of doubtful reading are enclosed in brackete. Where two short sjllablee 
replace a single sellable, the quantitiee are indicated. Departures m>m the 
1188. are pat in italios). 

{Liv. Andr,) 1. Yfmm mfhi Cain6na || (nsece venutam. 

2. M6a pu^ra, quid-v^bi || ex-tuo-ore supra. 

3. Ndq(ue) tarn t6 oblftuB-sam, || La^rtie noster. 

4. Arg^nteo polubro, || aureo ecliitro. 

5. Tuque mlhi narrdto || omnia di86rtiin. 

6. Mdtrem < prooi > procftum || plurimi ven^rant. 

7. Qu&ndo dfes advi^Dtet, H quem-profdta Mortast. 

8. AdtiEL Pylum deventee || aut ibi-omm6ntan8. 
[9. Tiiiicque r^mos jussit || religdre striippis.] 

10. Ibid^mque vir-summus || T&trieolsi adprimuB. 

11. Pdrtim 6rra]it, nequinunt || Gra6ciam redfre. 

12. Apud-nympham, Atldntis || fOiam C&lypBonem. 
[13, fgltur d6mum Ur(x{-[cor *] || prae-pavore frixit.] 

14. Utrum g6nu(a) amploctens || vfrginem ordret. 

15. rbi mdnens Bed6to, || donicam yidSbis. 

16. M6 carp^nto yeh^nte || mMm-domum venfsse. 

17. Sfmul dacrHmas d^-6re || no6geo det^rsit. 

18. Ndmque DuUum p6ju8 || mdoerat hum^um. 

19. Qudmde mdre sa^vum, || v{«-et-cui sunt-mdgnae ; 

20. Topper < eas > confrfngent || importunae undae. 

21. M^rcurius cumqu(e)-6o || fflius Latonas. 
[22. NSxerant mult(a) int^r-se || n6xu nodorum.] 

^3. Ndm divina Moii6ta8 11 fflia m^-docuit. 



^ ear probably belongs to the context, and should be remoyed from this line. 



EARLY LATIN VBRSE. 439 

24. T6ppeT fdoit hominee || ut-prius f6enmt. 

25. T6pper ciH ad-a6di8 || T^nimus Circd't'. 

[26. Cdmis atitem vindrnqne || qu6d-3iba]it ancldbant.] 
27. S^mota puer, Satiimi || fflia, regfna. 



{yaevtui) 28. Eomm sictam 8eqii6ntar || multi mortdles. 

29. ubi foras ciim-aiiro || flioo exfbanU 

30. Multi im e-Tr6ia || str^nui vlri. 

3 1 . Jdmqu(6) emi- m^Dtem fortuna || f(6cerat qui6tem . 

32. riierant Bfgii(a) expr^asa Q quomodo Titani, 

33. Blc6rpores Qigdntes || mdgiiiqa(e) Atldntes 

34. Euncus &tqu6 Purp^us || fQii T6rras. 

35. Sllofcolae homdnes Q b411iqu(e) indrtes. 

36. Bldnde docte percontat || A6neam quo-pdcto. 

37. PHiD(a) ino6dit C^ris || Proserpina puer, 

38. Defnde pollens sagfttis || (nolutus Arqnit^nens, 

39. S^nctuB D^lpfais progn&tus || Pythias Apollo. 

40. fsque susom ad-ca61um || sustulit suds-res. 
[41. Amiilias < ac-multis > || gratuldtur dfvis.] 

42. Postquam dvem asp^zit || t^mptilo Anchisa, 

43. Sdcr(a) iii-m6Dsa Pendttim || ordine ponuntur. 

44. Immoldbat auream || yfctimam pulcram 
[45. Sfmul toocia || porricerent 6xta.] 

46. Trdnsit Melltam Bomdnus, || {nsulam integram 

47. Urit, popilldtur, vdstat || rem-hostium coDcinnat. 

48. Yfrum prdetor adven^'t, || auspicat auspidum. 

49. Consent 6o yenturum || obyiam Po6num. 

50. S\ip6rbiter oont^mtim || content l^glones. 
[51. S^ptimum-dectmum dnnum || (lico sed6nt««.] 

i[52. Id^-quoque pacfscunt, || mo6nia siquo^.] 
[53. Liitdtium ^oncfUant, || captfvos plurimos]^ 

54. Sicllt6nsis paoiscit || obsides ut-r6ddant. 

55. E'i y^nit in-m^ntem || hominum fortunas. 

56. OnerdiY&e onustae 11 stabant in-flustris. 



1 



^ Perhaps i?/unmoi Mrp^iVot. 



1 



i 



440 THE ACCENTUAL ELEMENT IN 

57. Pldriqu(e)-6mne8 siibtguntur || sub-^umju- 

dfotum. 

58. E^8-diy(na8 edfoit, || pra6dioit ^ cdstus. 

59. S^nex, h^tas pt^tdte, || adlociitus summi 

60. K^gis frdtrem Neptunom, || regnatorem mdram. 

61. Summe d6um regnitor, || qolanam g^nilfsti? 
^ 62. S^8equ(e) (iperlre || mdyolunt ibfdem, 
( 63. Qa4m oum-stiipro redlre || dd-suospopiiliris. 

64. Sin (lloB d68erant || fortfssimos vfros, 

65. Mdgnum stiipram populo || fieri per-g6nti8. 

66. N6ctu Troiad exfbant || dtpttibus op^itis. 

67. Fl^ntis dmbae, ^b^dntes || Idcrimis cum-multi8. 
[68. Atque pHus piriet || lucusta lucam] 

69. Ferant pulcras cret^rras, || aureas lepfstas. 

70. Magnae m^tus tumiiltas || p^otora poasidit. 

71. N6vein 16 via conc6rde8 || fOiae sorores. 

72. Pdtrem sunm supr^mum || 6ptamuin app611at. 

73. Scopas dtqae verbenas || sdgmina Bumps^nmt. 

74. Sfmul dlM-Ulunde || riimitant int^r-se. 
[75. Apiid-empoilum in-cdmpo || bostium pro-mo^ne] 

76. Sfmul daon(a) eoroiii || portant ad-ndvis ; 

77. Mfli(a) dli(a) in-fsdem || inserinuntar. 



( 



78. Ddbunt m^um Met^lli || Na6vio po6tae. 

79. rmmortdles mortdles || 8i-for6t-fas fl^re, 

80. F16rent dfvae Cam6nae || Na^vinm po^tam. 

81 . rtaq(ue) postquam est O'rcbo || tr&ditas thesauro. 
[82. LcSquter Ifngua Latfna || oblitf-sunt Eomae.] 

83. Summas opes qui r^gum || regias refregit. 

84. Occursdtrix artfftcum, || p6rdita spintumix. 



{Inscriptions) 85. Hone oino plofrume || cos^ntiont E <6mai> 

86. Dillonoro optumo || fufse viro, 

87. Luciom Scipione. || fflios Barbdti 

88. CoQSol, censor, aidflis || hfc-fuet apud-vos 

1 Of. Ter. Andr, 777, prov6l9am . . pirvolvam. 



EARLY LATIN VERSE. 441 

89. H^ c6pit Corsica || Aleriaqn(e)-urbe 

90. DMet Tempestdtebus || aide meretod. 

91. C6rn61ius LuciuB || Scfpio Barbatus, 

92. Gnafvod pdtre progn&tus, || fortis-vir s^ptensque, 

93. Quoius-forma Tirtutei || parfsuma fuit, 

94. Gonsol, censor, aidflis || qu6i-fuit apud-vos, 

95. Taurdsia Cisauna || Sdmnio c^pit, 

96. Sdbtgit omne Loucdnam || 6psides-qu(e) abdoucit 

97. Qu(ei)-aptce insfgne Dialis || fldminis gesfstei, 

98. Mors perfecit tu(a)-ut- Assent || omnia br6via, 

99. Hones, fdma, virtusque, || gl6ri(a) atqu(e) in- 

gentum 

100. Qutbils 8(ei)-in-16nga rfciifset I tib(e) utier vita 

101. Eactle fdcteis siip^rdses || gl6riam roaiorum. 

102. Qu&-re lubens t(e}-in-gremYu, || Scfpio, recipit 

103. T^rra, Publi, prognitnm || PiibUo, Com61i. 

104. Mdgna sdpientia || muMsque virtutes 

105. Adtate qnom-p^rva H posidet hoc-saxsum, 

106. Quolei vfta defecit, || non-honos,^ hoDore, 

107. Ts hie situs qnei nunquam || yfctus-est virtutei, 

108. Annos gndtus yigfnti || is-l(oc)ei8 mandatus. 

109. N^-quairdtis honore || quef-minns-sit mandatus. 

110. Qu&d r^-8ua difeidens H dspere afleicta 

111. Pdrens tfmens heic vovit, || voto-hoc soluto, 

112. Decdma facta poloucta; || lefbereis lubentes 

113. Donu ddount H6rcole; || mdxsume mereto 

114. Semol t(e)-6rant se voti || crebro condemnes. 

115. Ductu aiispfcio || imp6rioqu(e) ^jus 

116. Achdia capta, || Cormto delete, 

117. Eomam r^dleit triumpbans. || ob-ha8c(e)-re6- 

ben(e)-g^stas, 

118. Qu&d in-b611o voverat, || hanc-a^em et signu 

119. H^rculls-Yictoris || imperater dedicat.' 

120. H5c est-fdctum moniim6ntum |j Madrco Caicillo 



^ Doubtful accentuation. But cf. Plant. Amph, 379, effo f«m, non iu, Soaia; 
Bud, 136, Veneriparavi . . non mihi, 
> Or is this line an Iambic Senarius ! 



442 THE ACCENTUAL ELEMENT IN 

121. Hospes, gr^tum-(e)8t qa(om) apud-meas || resfi- 

Kstei seedes. 

122. B^n(e)-i6in-gera8et villas; || donnias stn^-qura. 

123. Gbnlegium [quod est] acfptum || aetdtei ag6dai, 

1 24. piparum ad-veftam qaolundam || festosqae dies. 

125. Qu^i Boueis-astutieis || opfdque Yolgdni 

126. Gondecorant Baiplsume || com v( via loidosqae, 

127. Ququei hue dederunt || imperat6ribu8 summeis, 

128. LTtei s^sed lubentes || b6n(e)iovent optdntis. 



129. Uno cbmplurimae || cooB^ntiunt gSntes 
ISO. Fopuli piimdrium fufsse vfrum, 

131. Fuudity fugat, prost^mit H maxumas llgYones. 

132. Mdgnum num^rum triumphat, || hostibus devf ctia. 

133. Duello m&gno dYrlmdndo H regibus sublg^ndis. 



{Prophme8,etc,)lS4. Terra pastern ten^to || sdlus hic-man^to 

[135. Cdve in-m^ man^e || suS-flumine sinasi] 

i[136. Turn tii audax insfste || hostium muris.] 
[137. Memor qudm per-tot-dnnos || obsides uibem.J 
[138. Apollini vov6ndo8 || c^nseo ludos.] 
[139. Qui quot^nnis oomiter || Apollini ffant.] 
[140. Hoc si r6cte faxitis, || gaud^bitis semper.] 
141. Qudrnvis nov6ntium || duonum n^gvimdte. 
[142. K6quid fraudis stupHque || fer6cia p^^t.] 

The text and scansion of these lines I have discussed in two 
articles in the American Journal of Philology (xiv. p. 139 and 
p. 305 )y where the whole question of the Satumian Metre is 
treated with fuller detail. The Satumians of inscriptions must 
not be supposed to be any more conformable to the rules of the 
metre than the Dactylic and Iambic quantitative lines on in- 
scriptions of the same kind. We may gauge the technical skill 
shown in the Satumians in the Scipio Epitaphs (w. 85-109) by 
the Elegiacs on another epitaph of the family (C/.Z. I. 38); 
the Satumians of Mummius' Dedication (w. 115-119) will be as 



EARLY LATIN VERSE. 443 

far removed from those of Naevius as the Dactylic Hexameters 
on another of his Dedicatory inscriptions ( CLL. I. 542, e.g. 

eogendei dissolumdei tu ut faeilia faxseis) 

are from the lines of Ennius; the inscription of the Faliscan 
cooks (v7. 123-128) seems to be as unmetrical as, e.g. CLL, 
I. 1027, in 'iambics,' where the metre is destroyed by certain 
additions, which I bracket, in each line : 

hospes, re$isis, et hoe ad grumum \ad laevam] aspiee^ 

tihei eontinentur ossa homints honi, [tntserieardiSf amantU^ pauperis]. 

rogo te, viator, monumento huie nil [male feeerif\. 

The halting Satumians of the prophecies, most of which are 
mere conjectural restorations of Livy's prose paraphrases, may be 
capped with equally bad hexameters from the Sortee in CLL. 
I. 1438 sqq., e.g. 

I. 1438 eonrigi vix tandem quod eurvom est factum erede. 
I. 1448 laetus lubens petito quod dahitur . gaudehie semper. 



444 



XVI.— CONTRIBUTIONS TO THE HISTORY OF 
THE DEPONENT VERB IN IRISH. By J. 
Strachan. 

[Bead at the Meeting of the Fhilologieal Society held on Friday ^ June let, 1894.] 

The object of this paper is not to investigate the origin of the 
r deponent, which Old Irish shares with Latin, and its relation 
to the Indo-Gennanic verbal system, or to discuss, except inci* 
dentally so far as they have any bearing on the subject proper, 
the theories that have been put forward concerning the origin 
of these forms. ^ Whether it will ever be possible to get 
beyond conflicting theories, and to arrive at any certain or even 
probable account of the genesis of the type, may be reasonably 
doubted. But, taking the deponent as it exists in the oldest 
records of the Irish tongue, it should not be an impossible task 
to trace, with more or less exactness, its history within the Irish 
language itself, to follow the old forms in their life and decay, 
and to search out the starting-point and follow the development 
of any new types. 

The degree of precision with which such an investigation can 
be carried out must depend on the nature of the documents on 
which it is based. Where there is a continuous series of dated 
documents, each of which represents faithfully the language of 
its time, the course of the enquiry will run smoothly enough. 
In Irish, however, the student does not find himself in this 
fortunate position. For Old Irish we have trustworthy documents 
in the Glosses and in fragments of Irish preserved in the oldest 
manuscripts. In the later language we have metrical compositions 
like the Saltair na Ranny where the original forms were to a great 
extent protected by the metre. But with the mass of Middle 
Irish texts, some of them very ancient in their origin, preserved 

* Cf. "Windisch, TTeher die Verhalformen mit dem eharakter r., etc., Abbandl. 
der Sachs. Gee. d. Wiss., phil.-hist. CI. x. 447 sq. ; Zimmer, Ueber doe italo" 
keltische Fatsivum und Beponena^ Kuhn's Zeitschrilt, xxx. 224 sq. ; Thunieys«>n« 
EZ xxxi. 62 sq., Indo-German. Forsch. i. 460 sq. ; Brugmann, Orundrite, 
ii. 1388 sq. 



THE DEPONENT VERB IN IRISH — J. 8TRACHAN. 445 



in manuscripts of the eleventh centary and later, it is very 
different. In the transmission of such texts there was a tendency 
to replace old forms by modem ones. The probable extent of 
such corruption is not to be estimated d priori by the date of the 
manuscript. We shall find instances in which later manuscripts 
preserve the old forms better than earlier ones. And in the same 
manuscript different texts will be found to have suffered in a 
different degree. Thus, in the Book of Leinster, a manuscript 
of the twelfth century, the old deponential forms are well pre- 
served in many of the old texts, but, for instance, in the famous 
Tain B6 Citailhge^ the Cattle-raid of Cooley, they have almost 
vanished. In such a case as this we must suppose that in the 
one instance the old forms were more or less faithfully copied, 
in the other the language has been approximated to the language 
of a later period that the old tale might delight new generations. 

Such corruptions of necessity make the problem more com- 
plicated. Already in the Glosses the deponent verb has begun 
to pass into the active; and in a particular active form from a 
deponent verb found in one of these old texts, preserved in 
a manuscript of a much later date, it may be impossible to say 
whether the active form was original or whether it was introduced 
later. The remedy here lies in widening the field of observation ; 
for by the examination of a number of texts of the same character 
individual peculiarities may to a great extent be eliminated. 

Another difficulty is the difficulty of chronology. In some 
cases the date of the composition of a text may be accurately 
or approximately fixed from the internal evidence supplied by the 
subject-matter. In other instances it might seem to be fixed 
by the name of the author. Unfortunately we must be very 
careful in accepting such statements; there is, throughout Irish 
literature, too much of a tendency to fasten to the productions 
of later times the great names of the past. A third means of 
fixing approximately the date of composition is the language. 
But here again the development of the Irish language in its 
successive stages is a subject that has yet to be worked out. 
Until that is done we must rely in this matter, to a great extent, 
on general impressions, which accurate observation may prove 
to be false. The following pages may serve as a starting-point 
for the history of one particular form in Irish, the deponent 
verb. The results of such an adventure on a fresh field must 
of necessity contain mi;«^ f£ai ji^iU ;^6iKe to be modified or 



/ 

f 






446 THE DEPONENT TBRB IN IRISH — J. STRACHAN. 



corrected by farther obserration and discussion. Something will 
have been accomplished if the way has been made smoother for 
those who come after. 

The main subject falls, naturally, into two parts. The first 
part contains a collection of materials for the history of the 
deponent verb; the second treats of the history of the deponent 
based on these materials; a third part will deal with some new 
Irish developments, whereby certain forms of the active verb 
have taken to themselves endings borrowed from the deponent 
inflexion. Fart I., though it will consist chiefly of dry lists of 
forms, is a necessary preliminary to Part II. For the Old Irish 
of the Glosses the Grammatioa Celtica furnishes extensive col- 
lections of material, though the further publication of Irish 
Glosses, particularly of the Milan Glosses, enables considerable 
additions to be made to the material gathered together by Zeuss 
and Ebel. For the Old Irish of the profane literature and for 
Middle Irish there are no collections of any extent; here the 
work had practically to be done from the beginning. To ransack 
even as much of this literature as is already generally accessible 
would be an endless and a profitless task. What has been done 
has been to take a number of representative texts and to examine 
them carefully. This may lead to the omission of some deponent 
forms that might have found their place here; but, on the other 
hand, the history of the deponent is not to be learned from a 
multitude of scattered forms huddled together from a variety of 
heterogeneous sources. Apart from the glosses the material 
has been got to a great extent from the two oldest Middle Irish 
manuscripts, the Leahhar na kUtdhre (eleventh century) and the 
Book of Leinster (twelfth century). The former has been ex- 
amined throughout; of the latter I have read through all that 
seemed to be of importance for this investigation. From these 
two sources have been selected such texts as appeared most 
illustrative. To these have been added texts from other sources, 
such as the Sanas Cormaio and the Saltair na Itann, two works 
which we shall find to be of the highest consequence, since 
between them lies the breaking up of the deponent inflexion. 

In the arrangement of the material collected from these texts 
that order has been adopted which seemed to put the facts in the 
clearest light; it does not of necessity correspond to the order 
of the date of the original composition of the several texts. In 
the present state of our knowledge it is impossible to arrange 



THB DBFONBNT VERB IV IRISH — J. STRACHAN. 447 

the oldest stratum of texts chronologically, and something hesides 
has been sacrificed to convenience. The language of the Old 
Irish Glosses has been treated as a whole. Each of the other 
texts has, as a rule, been treated separately. This involves repe- 
tition, but that is a much less evil than would have resulted 
from the mixing up even of the material got from closely similar 
texts. We have seen the corruptions to which these documents 
have been exposed in the course of their transmission. But they 
did not all suffer in an equal degree. For example, to put together 
the forms of the deponent verbs found in the LU. and LL. versions 
of the Tain would be to convey an entirely false impression. 
Besides, it will be found that the presentation of such forms 
from a number of texts of various kinds and of various ages 
will be the most effectual way of illustrating their history. 



I. MATERIALS FOR THE HISTORY OF THE DEPONENT 

VERB. 

1. The