Skip to main content

Full text of "Transactions"

See other formats




f K&f 

c cc 











I. On the Connexion of the Finn and Lapp with the other 

European Languages ; by HENSLEIGH WEDGWOOD, Esq. 1 

II. On the Liquids, especially in relation to certain Mutes ; 

by E. F. WEYMOUTH, Esq. * - i; ; >; </! : v .... 18 

III. Miscellaneous English Etymologies; by HENSLEIGH 

IV. On a Zaza Vocabulary ; by Dr. H. SANDWITH. Commu- 
nicated by Dr. GK E. LATHAM 40 

V. On the Derivation and Meaning of f/7nos ; by THEODORE 

AUFRECHT, Esq. . .. .;.., . , 42 

VI. On the Affix of the Welsh Degree of Equality; by THEO- 

VII. On the Nasalization of Initial Mutes in Welsh ; by 

VIII. On the Etymology of the Latin Adverb Actutwn ; by 

IX. On the Languages of Northern, Western, and Central 
America ; by E. Gr. LATHAM, M.D 57 

X. On the Derivation of the Latin Sons ; by THEODORE 

AUFRECHT, Esq. , , . , .,.*. 115 

XI. On the Irregularities of the Versification of Homer ; by 
JAMES YATES, Esq., M.A., F.E.S. . .... . . .119 

XII. On the Derivation of the Latin Otium ; by THEODORE 
AUFRECHT, Esq 143 

XIII. On the Latin Terminations tia, tio- ; by THEODORE 
AUFRECHT, Esq 144 

XIV. On some English Idioms ; by the Eev. J. J. STEWART 
PEROWNE, B.D. Part I 146 

XV. Further Observations on the Connexion of the Finnish 
and Indo-Q-ermanic Classes of Languages ; by HENS- 



XVI. Miscellaneous Etymologies illustrated from the Finnish 
Languages ; by HENSLEIGH WEDGWOOD, Esq. . . . 179 

XVII. On the "Word Distributed, as used in Logic ; by E. 

G-. LATHAM, Esq., M.D 190 

XVIII. Hints on the Thesis " The Old-Friesic above all others 
the l fons et origo ' of the Old-English" ; by M. DE HAAN 
HETTEMA, Juris Doctor, Member of the Friesic Chivalry 196 

XIX. On some Affinities in the Basque Language, with 
Words referred to the Finnish and Indo- Germanic Lan- 
guages ; by JAMES KENNEDY, Esq., LL.B 216 

Scrap. Fanatics. Introduction and Derivation of the Word 218 

XX. On Diminutives. I. English ; by T. HEWITT KEY, 
Esq., M.A 219 

XXI. On the Affinities between the Languages of the 
Northern Tribes of the Old and New Continents ; by 
LEWIS KB. DAA, Esq., of Christiania, Norway . . . 251 

Scrap. Cherte. The word explained 294 

XXII. [On Diminutives. II. Latin ; or] On the Represen- 
tatives of the Keltic Suffix agJi or ach l little,' in the 
Latin Vocabulary ; by T. HEWITT KEY, Esq., M.A. . 295 

INDEX . . . 355 

Notices of Meetings, Treasurer's Cash Account, &c. . . . 359 


Page 112, line 7 from bottom, for vocabulary read vocabulary. 

Page 239, line 25, niding (or nidgef) ' a base fellow ' ; should be transferred 
three lines lower down, so as to fall between lording and riding in the collection 
of English words. 

Page 300, line 17, for lig-neo- read lign-eo-. 

301, 23, cor-ag-an read cur-ach-an. 

~ > 27, tus-sil-ag-on- read tussil-ag-on-. 

302, 7, ag-an read ach-an. 

334, 1, verg-, verg- read ver-ff- } ver-g-. 
341, 5, glomes read fflomes-. 

> 6, in read is. 

- 344, 17, a read . 

- 347, note||, line 9, for aoQurepo 


o o 







[Read January the llth."] 

IN the second volume of our Proceedings (pp. 180-187), 
Professor Key has called attention to remarkable agreements 
between the grammars of the Lapp and Finn, and of the 
Greek and Latin languages. The identity is occasionally so 
complete, that it is truly astonishing how it can have been 
preserved through the series of ages which must have elapsed 
since the Finns and Latins can have separated from a common 
stock, or even have been in such close communication as to 
exert much influence on each other's language. One or two 
examples may be added to those given by Professor Key. 
Thus in Lapp cum or queim, as in Latin, is ' with ' ; mocum, 
tocum, socum mecum, tecum, secum. Lapp ets and Finn Use 
correspond to Lat. ipse; mon ets, ego ipse. The particles 
ek, ke, ak, ka, are used in Lapp to give emphasis to the 
pronoun, in precisely the same way as ce, que, in Lat. Thus 
from iat, hie, ille, is formed tatek, hicce, ace. tabke ; from ka, 
qui, kake, agreeing in form with quisque, but translated ali- 
quis. Lapp yaw is used much 2^ jam in Lat. as a reference to 


certain circumstances affecting the action : mi le jam tat ? 
what then is that? patijam, come then. 

The formation of the Finn languages is commonly ex- 
plained as if they were composed of two distinct parts, viz. 
the primitive language of the race itself, and an enormous 
importation from the Scandinavian peoples with whom they 
are mixed, with which must be classed numerous words bor- 
rowed from the Teutonic, Slavic, and Lithuanian. It is 
however hardly possible to account on such a principle for 
the whole of the phenomena before us. No doubt a great 
proportion of the analogous forms must be considered as 
directly borrowed from a Scandinavian source ; but after every 
allowance has been made for such an influence, a large amount 
of resemblance will remain, offering the same kind of evidence 
in favour of a remote community of origin, as in the case of 
other related races, as the Celts and Teutons, Celts and 
Slaves, &c. The words common to the Finns and Slaves or 
Lithuanians, are far from being simply or even chiefly the 
names of objects, the use of which may be supposed to have 
been learnt from people in a more advanced state of civili- 
zation, but frequently express actions or abstract notions 
which must be conceived by nations in the rudest condition 
of life. We may cite 

Finn palaan, pallata, to burn ; Bohem. paliti. 

puoli, half, side, middle ; Bohem. pule. 

lentaa or letd, to fly ; Bohem. letiti. 

wedan, wetda, to draw, to lead; Lith. and Bohem. 
wedu, westi. 

Lapp wuoras, old ; Lith. woras. 

jaure, a lake, Juiih. jures (plu.), the sea. 

pak } paka, heat; Bohem. pek, the root of E. bake. 
Nor are we without evidence of a Celtic connexion of similar 


Finn korsi, stipula, calamus ; W. korsen, a reed. 

kannan, kantaa, to bear, carry, hold; W. cannu, to 
hold, as a vessel. 

pullo, thick bark, cork, the floats of a net ; Gael, bolla, 
a net or anchor buoy. 


Lapp buwe, sheep, cattle; W. buw, an ox, kine. 

wele, more ; W. gwell, better, in a greater degree. 

habra, a goat ; W. gafr. 

Finn jalke, footstep, hinder part, behind; W. ol in the 
same sense. 

jdljin, hindmost ; W. olaf. 

jalillen, remaining, the rest; W. olion, things left 
behind, refuse. 

jallen, back again, at last ; W. yn ol, back, back again. 

osata, to hit the mark, to aim right, to be able to do ; 
osattaa, to aim at ; osaella, to try to do. W. osio, to try to 
do; E. to oss. 

sota, war, battle; sotia, to fight. W. cad-, G. cath. 
Lapp kakkel, a distaff; W. cogel. 

Many isolated words are common to the Finn and Scan- 
dinavian languages without corresponding words in the other 
branches of the Gothic stock. The whole of these are broadly 
ascribed by Ihre (than whom there is no more acute or ju- 
dicious philologist) to a Finn origin, and in one important 
instance at least, it seems certain that the course of language 
has run in this direction. 

The Icel. negative is ei, eigi, Dan. ikke, corresponding to 
Finn ei, eikd ; eikd-eikd, neque-nec. Now the Icel. ei is an 
adverb, applying equally to all persons, while Finn ei is 
appropriated to propositions of the third person, being part of a 
regular conjugation, en, et, ei, emme, ette, eiwat, non ego, non 
tu, &c. As conjugations of such a nature were contrary to 
the idiom of the Scandinavians, they seem to have adopted for 
general use the negative of the third person, from the far 
greater frequency with which propositions of that form would 
occur than those of the first and second persons. It is certain 
then, that because a word is common to a Finn and Scan- 
dinavian language, it cannot be assumed that it is necessarily 
borrowed by the former from the latter. 

A considerable list may be made of Finn forms and 
corresponding ones in Greek and Latin, either without inde- 
pendent analogues in the Teutonic languages, or only such as 
are more distantly related than the classical forms : 

B 2 


Finn onki, a fishhook ; Gr. 07*77, 07*05, a hook, a barb. 

onkalo, a nook; ayfcv\o<;, crooked; wyKcCki), the bend- 
ing of the arm. 

kampela, crooked; /ea//,7ruXo5. 

wuori, a mountain ; 0/905. 

myykia, to low; fjLVfcaofjLai. 

uros, male of animals, grown man, strong man, hero ; 
uro-teko, factmn heroicum ; rjpax;. 

kommata, graviter sono ut campana, vas vacuum; 
Kopireiv, to ring, to clang. 

mamma, a breast ; Lat. mamma. 

marketa, to fade ; Lat. marc-escere. 

murheh, grief; Lat. mceror. 

orpo, orphan; Lat. orba. 

orwitseta, to deprive of parents, to disinherit ; Lat. 

panen, panna, to place ; Lat. pono. 

porsas, a pig ; Lat. porous. 

oras, a boar ; Lat. verres. 

ihminen, a man; Lat. homo (homin). 

waimoinen, womanly ; Lat. femininus. 

werma, firm, trusty ; Lat. firmus. 

wermasti, firmly ; Lat. firmiter. 

waras, a thief; Lat. fur ; Russ. vor. 

warkahin, secretly; Lat. furtim. 

wiho, greenness, green fruit; wihanta, wiheria, wi- 
heriainen, green; Lat. viridis. 

wihota, wihertaa, Hung, virit, virtil, to be green, to 
flourish; Lat. vireo, vires co. 

wilu, frost; Lapp jdla, cold; Lat. gelu. 

ajaa, to drive; Lat. agere, 

ryokia, to belch ; Lat. eructo. 

krapista-, Lat. crepare, crepitare. 

krapistus', Lat. crepitus, crepitaculum. 

papu, beans, pulse; Lat./6; Hung, bab-, Pol. bob. 

potty, dust, snow driven about by the wind; Lat. 
pulviSy pollen. 

ohra, barley ; Lat. hordeum. 


Finn kaula, the neck ; Lat. collum. 

ulwoa, to howl ', Lat. ululo. 

sarpa, a reed ; Lat. scirpus. 

kara or sara, sedge ; Lat. carex. 

sarawisto or sarawikko f Lat. carectum. 

salawa, a willow, sallow ; Lat. salix. 

salawisto, salawikko ; Lat. salictum. 

salata, to hide, conceal ; Lat. celo. 

Hence sola, anything hidden, the locative case of which, 
salaan, is used adverbially in the sense of secretly, in a hidden 
place, clam. 

Finn salainen, clandestine ; Lat. clandestinus. 

Jj&pppalen, in the presence of; Lat. palam. 

Ymnpyytid, to seek, to invite ; Lat. peto. 

puhdas, pure ; Lat. purus, putus. 

puhtaus, purity, cleanliness. 

suola, salt ; Lat. sal. 

kallo, the scalp, forehead, skin of the forehead ; Lat. 
calva, callus. 

jaa-kallo, crust of ice covering the ground; Lat. 
callum, applied to the hard surface of the ground. 

kallokas, shoe of raw hide ; \Lat. calceus. 
Lapp kallok, shoe of skin of reindeer ; J 

Finn kakistaa, kikottaa, to chatter as a pie, laugh loud; 
Lat. cachinno. 

ddri, margin, edge; Lat. ora. 
Lapp hapos, a horse ; Gr. tTTTro?. 

pir,pira, about, around ; Finn piiri, a circle ; Gr. Tre/ot. 

walla, but; Gr. aXXa. 

ivuoke, form, likeness; Gr. euco?. 

wuokak, like, equal ; wuokas, fit, convenient ; Gr. eoifca, 
to be like, to be fit. 

ara, early, soon; Gr. vjpi. 

aina, ainak, only, single, simple; Lat. unicus', Lith. 
wenas, wenokas. 

all, high ; Lat. altus. 

aletet, haletet, to fly ; Lat. ales (alit), bird. 

air, aira, copper ; Lat. as (<er) ; Lith. waras. 


Lapp pullistaa, to puff up, to swell ; pulli, a flask ; Lat. 

pullikoitsita, to speak in an inflated manner ; Lat. am- 

buola, puola, a bit ; Lat. bolus. 

ruopses, red; Lat. ruber. 

taibet; Lat. debere, oportere. 

kona, kuna } ashes; Lat. cinis', Gr. KOVW. 

wade, a ford ; Lat. vadum. 

juomits, a twin ; Lat. gemellus ; Fr. jumeau. 

jarbes, round ; Lat. orbis. 

kawak (flexuosus, curvus) ; Lat. cavus, hollow. 

muorje, a berry; Lat. morum, a mulberry; Wallach. 
mour, a blackberry. 

mostos, out of humour, sorrowful ; Lat. mcestus. 

harret, to growl; Lat. hirrire. 

kattjett, to fall ; Finn kadota, to perish ; Lat. cadere. 

sanahet, to endeavour; Lat. conari. 

sarwa, sarwes, an entire reindeer ; Lat. cervus. 

The Finn sarwi, Hung, szaru, szarv, a horn (whence szarvas, 
horned, also a stag), show the radical meaning of Lat. cervus, 
and at the same time bring Finn sarwi into correspondence 
with Gr. /cepas and Lat. cornu. Other words which primd facie 
we should suppose to be borrowed from a Scandinavian source, 
are found also in Hungarian, a language which has not un- 
dergone the same mixture with the Gothic tongues, and may 
thus put forward a fair claim to be considered as part of the 
original stock of the language. No one would doubt that 
Lapp garde, a hedge, inclosed place ; gar dot, to hedge ; garden, 
a farm ; Finn kartano, a yard, court, were borrowed from Sw. 
gard, a yard ; garde, a field ; gar da, to fence, and the like, 
were it not that the Hungarian has kert, a garden ; kerit, kertel, 
to inclose; keritek, kertelez, an inclosure, curtilage, hedge. 
So we have Lapp waret, to keep, to guard ; Finn warrota, to 
watch, observe, wait for; wartia, a watchman, guard, appa- 
rently borrowed from Sw. wara, to observe, and its deri- 
vatives, but the same foot is preserved in Hung, vdr, to wait 
for; vdr, a fortress; varta, a watch or guard. The Finn 


mesi, meden, honey, agrees on the one hand with Lith. medus, 
Slav, med, Lat. mel ; and with W. medd, E. mead, a drink 
prepared from honey, while on the other it is shown to be a 
genuine Ugrian word by the Hung, mez, honey. So Finn 
west, water, lake ; wesinen, wetinen, watery, wet ; wettya, to 
become full of water ; Hung, viz, water, river. Lapp nikke, 
nekke, the neck; Hung. nyak. Lapp lapa, the sole of the 
foot ; Finn lapa, a blade, as shoulder-blade, blade of an oar, 
might be supposed to be borrowed from a Gothic root corre- 
sponding to E. lap, flap : but the Hung, has lap, flat side, 
plate, leaf; lapoczka, shoulder-blade, spatula, shovel. 

The Finn languages are extremely rich in words ex- 
pressing different kinds of sounds, and there is hardly a page 
in the dictionary without some word translated by parum 
crepo, strepo, strideo, susurro, murmur o. As a specimen may 
be cited kohista, kolista, komista, kopista, korista, tihista, 
tikista, tirista, titista, wikista, wilista, winista, hohista, ha- 
wista, jumista, morista, nirista, porista, sohista, &c., with 
almost every possible combination of the two consonants, and 
every variation of the vowel by which they are connected in 
the radical syllable. Of such words as the foregoing, many 
are represented by similar forms in Swedish, German, or 
English, but very many have no corresponding terms in those 
languages. Now as long as direct imitation is a living 
principle in the use of a word, the primary cause of the 
articulation is apparent on the face of it, and there is no 
occasion to seek the origin in another language in which the 
same image may be represented by a similar sound, unless 
overwhelming evidence of borrowing be forced upon us from 
other quarters. 

The syllable slam is used in Swedish and English as well as 
in Lapp to represent a loud noise ; Sw. slamra, to jingle, jabber, 
to talk idly (Widegren.) . In Lapp slam, a noise, nialme slam, 
strepitus verborum (nialme, the mouth), uksa slamketi, ' the 
door was slammed/ janua cum strepitu claudebatur ; slamem, 
ruin, fall. Here the imitative force of the word is as manifest 
in Lapp as in English. And there seems as little reason for 
supposing that the word must have been borrowed by the 


Laps from the Swedes, as vice versd. The same root seems to 
be truly represented by the Latin clamo, clamor t as we have 
seen many instances in which a Finnish s corresponds to a 
Latin c. In the same way it is probable that there may have 
been no direct borrowing in any of the following examples : 

Finn natista, leviter crepo ut mus rodens ; G. knattern, to 
patter; Dan. gnaddre, to grumble, growl. 

Finn naputtaa, leviter ico, crepito ; napista, napsaa, leviter 
crepo, murmuro, strepo ut dentes in manducando; G. knap- 
pern ; Sw. knapra, to gnaw. 

Finn narrata, strideo, crepo ut cardines januse ; G. knarren ; 
Sw. knorra, to murmur, grumble, growl. 

Firm, porata, porista, vociferor, ebullior; Du. borrelen, to 
purl, to bubble up, or in Flemish to vociferate; Port, bor- 
borinha, vociferation; Finn poret, a bubble; O. E. a burble. 

Finn hossottaa, leviter ferio, e. g. vestes vergis ; Fr. housser, 
to switch. 

Finn hikka, singultus, hiccough ; Fr. hoquet ; Sw. hicka. 

Lapp suokket, sjuoketet, to sob ; Sw. sucka. 

Finn huiska, scopa minor lavationi apta, a whisk ; huiskata, 
hue illuc cursitare, huiskua, hue illuc j actor ut arbor vento, 
huiskuttaa, hue et illuc moveo, quasso, ut canis caudam; 
huiskutan wettaa, I splash water about ; huiskutus, quassatio. 

Finn humata, humista, to hum, to sigh as the wind among 
trees; Icel. umra, kumra } to murmur. 

Finn huutaa, clamo, vocifero, to hoot-, huuto, clamor, 
vociferatio, rumor, fama vagans. 

Finn hurrata, hurista, susurro, ut aqua fluens vel apes 
volantes, to whirr ; Sw. hurra, surra. 

Finn kummata, kummista, to sound as a large bell ; kimista, 
acute tinnio, to chime kumina, resonance. 

Another argument in favour of a connexion of very old 
standing between the Finn and other European languages, 
may be drawn from the numerous cases in which it enables us 
to explain words without apparent derivation in their own 
language. One of the cases of Finn sama, the same, is 
samalla, in the same ; samalla muodolla, in the same manner ; 
but samalla alone is used ellipticaily in the sense of ' at the 


same moment/ agreeing with Lat. simul. A somewhat different 
modification of the same root in a widely different language, 
gives Malay samo-samo, together, from samo, the same. 

Lapp kastas, wet; kastatet, to wet, to baptize, seems to 
indicate the idea of washing, as the origin of the Bohem. 
cisty, clean, pure, chaste (whence cistiti, to cleanse, and 
cisterna, a cleansing or washing place, a cistern), and of the 
Lat. castus, chaste. 

The name of the domestic cock, Finn kukko, Hung, kakas, 
is derived, like that of so many animals, from the sound by 
which we imitate his cry; Finn kukkua, cuculo, cucurio; 
Lith. kukti, to crow, to hoot; Bohem. kokrhati, to crow. 
From the upright strut of a cock, the term is then applied 
to whatever cocks or stands up, as a cock of hay, &c. In 
Finn kukku is the pile in heaped measure ; kukkelo, kukkura, 
the top of a mountain, affording a plausible explanation of 
Lat. cacumen. 

Lapp kukke, long ; kukketet, to prolong ; kukkehet, to think 
or find it long ; kukkelastet, to remain long, to delay ; Finn 
kokottaa, to expect, wait for, delay, exhibit a root which might 
easily pass into Lat. cunctari. The origin of these words seems 
to lie in Finn koko, a heap or pile, applied in a secondary sense 
to the structure or stature of the body, whence ko'okas, tall, 
great. The local cases of koko are used in the sense of the 
Lat. con, together, as pane kokoon or ko'olle, bring into a 
heap, place together; tulewat kokoon or ko'olle, they come 
together ; and as the second k is actually lost in one of these 
forms, it is not difficult to suppose that kokoon may be the 
exact equivalent of Lat. con. 

Again, koko, in composition, is used in the sense of totus, 
omnino; koko-kyla, the whole village; koko-mies, a complete 
man ; kokona, kokonansa, wholly, entirely ; kokonainen, whole, 
unbroken. Thus the Lat. cunctus might be derived from the 
same root with cunctari and with the preposition con. 

The expression of relations of place by reference to parts of 
the body is worthy of remark. From Finn korwa, an ear, is 
derived the expression for nearness, by the side of, ' locus 
juxta quid, ut aures juxta caput ; ' on tien korwalla, it is by 


the side of the road, literally in the ear of the road ; korwainen, 
by the side of, about. In like manner, hantd, a tail, is used 
in the locative cases in the sense of behind, and probably 
explains the origin of that word. Kaypi Mnnassani, he conies 
at my tail, comes after me ; juokse sen hdntddn, run after him ; 
hantyre, a follower; hdnnittdd, to follow any one close, in- 
sector quern quasi ad caudam. 

Finn rataan, radata, to squeak, creak, crepito ut mus, 
currus, affords a plausible derivation, as well of the rat, the 
squeaker, as of Finn ratas, Lat. rota, a wheel, the creaking 
of which, before the use of grease, would be a most obtrusive 
characteristic. The plural rattaat, as Lith. ratal, is used in 
the sense of a chariot, whence perhaps Lat. rheda. The 
origin of Lat. carrus and E. car, carry, may in like manner 
be found in Finn karista, strideo, crepo; G. garrezen 
(Schmeller) ; Icel. karra, to jar, to creak. 

Finn kalkkata, to clank, sonum edo crepantem ut ferrum 
in cudendo, suggests a natural origin of Gr. ^aXico^, brass; 
and Lapp maret, to roar, rush, murmur, of mare the sea, the 
TToXu^Xotcr^o? Oa\a<Tcra. 

Lapp suokket, sjuoketet, Sw. sucka, to sigh, correspond to 
Lat. sm^-ultus; and in the same way Finn tomu, sonus 
gravis, tumultus, pulvis, to Lat. ftm-ultus, where the same con- 
nexion may be noticed as in our own language between kicking 
up a dust and making a disturbance. Finn tomista, to 
make a deep sound, to make a dust ; tohu, strepitus, tumultus, 
pulvis. So G. getummel, confused noise, hurly-burley, bustle. 
The syllable torn is used in other languages as representative 
of a heavy sound, as in the Indian tom-tom ; a drum, and in a 
list of onomapoietic words given by Dr. Latham in 'The 
Varieties of Man/ as spoken by the half-breeds in Oregon, is 
turn, a heavy noise ; tum-wata, a waterfall. From the same 
source is doubtless W. twmpio, Fr. tomber, to fall, tumble. 

The feelings of discontent, grief, anger, are naturally desig- 
nated by words derived from the murmuring sounds uttered 
under those emotions. Thus from G. jammern, to wimper or 
wail, is jammer, grief; from murren, to grumble, murrisch, 
peevish, morose. So in Finn morista, murista, to growl, to 


be discontented; murrus, mentis indolis murmurans, indig- 
natio, tristitia; murahtaa, subito murmuro ut canis, obmur- 
muro ut homo iracundus ; murheh or murhet, sorrow, grief, 
distress, corresponding to Lat. mcereo, mceror. In like manner 
Finn surrata, to whizz or buzz, Sw. surra, to buzz, to 
murmur, lead to Finn suru, grief, sorrow ; surra, to grieve ; 
surrua, surkua, to be sorrowful. Analogy then would lead 
us to suppose that ira might be connected with hirrire, to 
snarl, which loses the initial h in irritare (properly, to cause 
to snarl), to provoke, and in support of such a supposition 
may be cited Finn hyrista, to hum or buzz ; harista, arista, 
to snarl, to snort with anger, to be angry and surly; 
haristaa, to cause to snarl, irritare; ari, iracunde hirriens, 
iracundus, morosus ; drina, hirritus, murmuratio, iracundia. 

From Finn muu, other, is formed muutoin, otherwise; 
muuttaa, to transfer to another place, to change to another 
form, to change clothes, horses, countenance ; G. umandern, 
ver ander n. Hence may be derived Lat. muto in analogy 
with Gr. aXXacro-ct), to change, from aXXo?, and G. andern, 
from ander. 

The sound of catching the breath, as in sobbing or choking, 
is imitated by the syllable nick or nack. Thus we are 
informed by Lieut. Burton (Pilgrimage to Medina, i. 222), 
that to ' nakh/ in vulgar as in classical Arabic, is to gurgle 
ikh ! ikh ! in the bottom of one's throat till the camel kneels 
down. With an initial s, snickup or sneckup was formerly 
used in E. for hiccough. In Hung, we have nyog, to sob, to 
groan, nyekken, to make a bleating sound ; Lapp niakket, to 
sob, to hickup, and in Finn nikka, a sob ; nikottaa, to cause 
to sob ; nikistaa, to choke, to suffocate, halitu privo, strangulo ; 
nikahtua, to be suffocated. Then, as the cessation of breath 
is the first sign of death, to stop the breath and to choke, are 
frequently applied to any kind of violent death. Thus G. 
wurgen, the equivalent of E. worry, of which the primary sig- 
nification is to choke or strangle, is also used in the sense of 
killing, massacreing, cutting the throat ; einen schaf wurgen, 
to kill a sheep. So Dan. qucele, to strangle, choke, smother, 
is the equivalent of E. kill; A.-S. qualstow, a place of death, 


cwylan, to die. In the same way it is probable that Lat. 
necare, to kill (generally as we are told applied to putting to 
death without a weapon), is analogous to Finn nikistdd, to 
suffocate. And that the primitive sense of the word was 
never entirely lost sight of, is witnessed by the use of necare, 
negare, in Mid. Lat. (Diez, from the Burguiidian laws) in the 
sense of drowning, whence It. annegare, Fr. noyer. From the 
same root is probably the name of the water-demon, Lapp 
Nik, Finn Nakki, Icel. Nikr, Sw. Ndcken, ' genius fluviorum, 
homines cupide aquis submergens/ often supposed to be the 
origin of ' Old Nick/ the familiar designation of the devil. 
But that expression, as has elsewhere been pointed out, is 
really derived from a different development of the root in 
Pl.-D. Nikker, the executioner, ( the old executioner/ from 
Flem. necken, A.-S. hncecan, to slay, in which the meaning of 
the word has undergone the same transition as in Lat. necare. 
The same fundamental image would supply a satisfactory 
designation of the word neck, which we must then suppose to 
have been first applied to the throat, from the guttural sounds 
imitated by the syllables nik or nak ; so in G. gurgel, the 
throat, from the gurgling sounds which it produces. The 
diversion of meaning in G. nacke, Fr. nuque, to the back of 
the neck, need cause little difficulty. 

Finn painaa, to weigh down, to be heavy, to press ; paino, 
weight, pressure ; paini, depression, curving downwards, seem 
radically connected, not only with Lat pondus, a weight, but 
pando, pandare, to bend, weigh down ; pandus, curved, and 
also with Trovo?, labour, the lifting a weight being the most 
obvious type of labour in general. The term is in Finn 
also applied to exertion of force, as in ponnistaa, to do any- 
thing with great exertion, to string a bow, G. spannen, in 
which probably the same root is contained. As over-exertion 
becomes highly painful, TTOVO? is used in the sense of pain, 
suffering, distress, grief. The word pain itself is probably 
from the original sense of the Finn root painaa to press, 
whence A.-S. pinan, to torture; Du. pyn, ache, pain. To 
pine or languish is to suffer pain. Pain, in the sense of 
punishment, from Lat. pcena, Gr. iroivy and punio, to punish, 


are radically distinct, being derived from the custom of 
making reconciliation by paying the price of blood, from Gr. 
<ovo9, offering an example of a phenomenon which has been 
frequently pointed out, where the convergence of meanings 
originally widely different in words of similar sound has ended 
in the coalescence of the words themselves. 

The mention of iroivr] as the price of blood, suggests a 
much more natural derivation than the one usually given 
of the A.-S. wera, the weregild or penalty to be paid to 
the relations of the slain man, in Finn weri, Hung, ver, 
blood, making weregild the precise equivalent of the G. 

The Lat. puniceus, purple, Gr. <omf, red (<om<7<ra <Xof ), 
are commonly supposed to have reference simply to the 
peculiar dye in which the Tyrians or Phoenicians excelled. 
But this may perhaps be an early instance of false etymology, 
as Gr. <oti/o9, <omo?, blood-red, seems to point to a con- 
nexion with <ovo?, blood, bloodshed, similar to that of Hung. 
voros, red, with ver, blood. A like connexion may be seen in 
Finn puna, red colour ; punistaa, to stain with red ; puna-tauti 
(tauti, illness), dysentery or discharge of blood. The primary 
origin may perhaps be Goth, fon, funins, fire, whence funisks, 
fiery, may be compared with Gr. <omf ; and a similar relation 
may be observed between purpureus and Trvp, fire. 

From Finn madan, mataa, to creep or crawl ; Hung, masz, 
to creep, is formed mato, matikko, a worm, maggot, explaining 
Icel. madkr, a mawk or maggot, and G. made, a maggot, 
mite, as well as E. moth, a designation which would first be 
applied to the larva by which the mischief is done, and se- 
condarily to the winged insect into which it changes. 

Lapp sjuddet, to hum or buzz, explains Sw. sjuda, G. sieden, 
E. seethe, to boil. 

Finn pulata, to splash, as a duck in diving, or fish in 
jumping; pulahtaa, to spring as a fish, to dive, to fall into 
the water, analogous to G. spulen, to wash, to rinse, and 
probably to Sw. spilla, to spill or splash over, seems the 
origin of pula, an opening in the ice, and W. pwl, E. pool, a 
piece of water. 

From Finn tiukkua, to pipe or make a shrill sound, is 


probably derived Lapp tjuk, the young of birds or quadrupeds, 
as of dogs or cats, Hung, tyuk, a chicken, fowl. Hence might 
arise provincial E. tyke, a dog, originally a young dog, then 
an affectionate expression for the animal independent of age. 
The colliers in the north always speak of their bull-clog as 
( the pup/ A.-S. bridda, is a young bird; It. piccione, pip- 
pione (whence the E. pigeon], is properly a young pigeon, 
from the peeping sound of a young bird. 

Lapp wikke, a fault, wikkalati, guilty ; Finn wika, a bodily 
defect, injury, moral fault ; wikainen, guilty, seem to preserve 
the origin of E. wicked. 

Finn karsta, soot, and hence dirt, filth, explains G. garstig, 
nasty, filthy. 

The syllable mut or muk is widely taken as representative 
of a low inarticulate sound, the least audible sound, whence 
G. mukken, mucksen, Lat. mutire, muttire, Gr. ^v^co, JJLV^W, 
to utter such a sound. The analogues in Finn are mutista, 
mytista, mussito, susurro, whence mutina, a murmuring (ex- 
plaining mutiny, a murmuring among soldiers), and mytiainen, 
culex minor, from the humming of the gnat, leading to the 
derivation of G. mucke, a midge, from the other form of the 
root, muk. The name of the gnat is probably derived, on the 
same principle, from the syllable nat, which is used as repre- 
sentative of a low indistinct sound, in Finn natista, leviter 
crepo ut mus rodens, and in Dan. gnaddre, to grumble, growl. 
From muk is formed Finn myhkia, mussito, clam loquor, su- 
surro, Dan. mukke, to mutter, Finn mykaista, to hush, to forbid 
one even to mutter; mykystya, to be silent; mykka, dumb 
(as Lat. mutus from mut) ; mykkyri, homo taciturnus vel 
occultus; G. mucken, to keep a surly silence. Hence a 
numerous class of words applied to doing a thing secretly, as 
G. meuchel-mord, clandestine murder, assassination; Sw. 
i mjugg, secretly, underhand ; le i mjugg, to laugh in one's 
sleeve ; E. hugger-mugger, clandestinely, privately, and con- 
sequently shabbily, in a disorderly manner, agreeing very much 
with Finn myhky-mdhkin, temere, sine ordine. The addition 
of an initial s gives Sw. y smyg, smygwis, clandestinely; 
smyga, to slip in, to do a thing secretly ; smyga sig pa nagon, 
to spy one, explaining Fr. mouchard, a spy; sra^-handel, 


- secret dealing, smuggling ; smuga, a hole, corner, hiding- 
place ; Icel. smiugr, smuga, a hole, a crack, narrow opening ; 
smeigia, Dan. smb'ge, to slip on ; Icel. smocka ser inn, to slip 
into something just big enough. Hence smockr, a sheath, a 
tube, and E. smock, a dress that you slip into. 

From Finn holata, holista, to give a hollow sound (einen 
dumpfen Laut erregen), as that of the flowing of water, or 
murmuring of a crowd, holo, anything hollow, Ao/o-puu, a 
hollow tree. So from kopista, to thump, to sound hollow 
(dumpf tb'nen), are formed kopina, sonus ex pulsu, and kopano, 
caudex arboris cavus pulsu resonans, which seems essentially 
the same word with Lat. campana. The corresponding form 
in Gr. Koiravov, a pestle, is applied to the instrument which 
gives the blow, instead of the body which receives it. The 
nasalized form KOfiirea), to clang or ring, 

KO/JL7T61, %a\KO<i 67Tt (TTt]Q (T (T L 0afcV05, 

leads to Mid. Lat. campana, as a modification agreeing very 
closely in sense with Finn kopano, to which it answers in 
form, in the same way as Sp. timbal, a kettledrum, to Arab. 
tabl (Burton, Pilgrimage to Medina), atabal. The name ori- 
ginally given to a drum, like those of the South Sea islanders, 
composed of a hollow block of tree, and, in a more advanced 
state of the art, to the instrument made by stretching a skin 
over the mouth of a brazen vessel, would naturally be pre- 
served when the sound was produced by striking against the 
metal itself, when the kettledrum would become a bell. The 
usual derivation of campana, from bells being first used at 
Nola in Campania, is a most improbable one, even if the fact 
were true. They plainly would not have been known by that 
name in Campania itself, and if the instrument had spread in 
such a manner from a single centre, the Campanian name 
would probably have travelled with it. But the whole story 
is in all probability a myth, founded solely on the fact that 
bells were known by the two names of Nola and Campana. 
Now as bell is from the imitative root which gives Icel. belia, 
boare; G. bellen, to bark, and E. bellow (templorum campana 
boant, Due.), and G. glocke, E. clock (originally a bell), from 
the root which survives in Fr. claquer, E. clack, Bohem. hluk, 
din, noise, so doubtless nola is from G. knall, a loud noise, as 


the report of a gun, crack of a whip, &c., the E. representative 
of which (knell) is appropriated to the clang of bells*. 

Other modifications of the root hop, as representing a 
sounding blow, are Gr. /cvjj,/3o<;, /cv/jufii}, any hollow, especially 
a hollow vessel, cup, basin, boat ; /cv/jL/SaXov, a cymbal ; and 
in Finn kopio, vacuus, resonans ut vas vacuum ; koppa, cavum 
quid, a cup ; Lapp kappe, hollow ; kopera, excavatus, concavus, 
curvus ; and as another form of kopera is kowera, the p passing 
easily into a w in Finn, we are brought through the Lapp 
kuowat, to hollow out, kawat, to crook, to bend, kawak, 
flexuosus, curvus, to the Lat. cavus, as an offshoot from the 
same root. 

The E. worth, W. gwerth, price, has a plausible derivation in 
the Finn wero, the equivalent of Lapp wuoro, vicis, a turn or 
time (whence wuorom, by turns, sometimes), although the Finn 
word is not given as having that signification in the nomi- 
native. But in what is called the elative case, werosta, it is 
used as Lat. vice, for ' in the place of, instead of/ and hence 
comes to signify, what is of the same value with. Thus, 
I eat cheese instead of bread; I take corn instead of 
money ; I stand in the place of a man, i. e. I reckon as a 
man ; ancient custom stands in the place of law, has the force- 
or validity of law. The adjective weroinen is in like manner 
applied to what stands in the place of, is of the same value or 
estimation with, and hence werta, what may supply the place 
of or be compared with anything, what is equal in respect of 
quantity or value, worth-, sen werta, so much; kouraan 
werta rahaa, a handful of money (koura, the grasping hand) ; 
werteinen, par, aequalis; wertaan, werrata, to compare; wer- 
taus, comparison, parable. 

The Lapp waro, merx, wares, Finn wara, copia, opes, goods, 
might appear simply borrowed from Sw. wara, merchandise, 
but the origin of the word is shown so clearly in the Finnish, 
that that language may fairly lay claim to an original right in 
it. The radical sense seems to be simply provisions, what is 

* This derivation of campana is supported by the Albanian kemboig, 
koumboig, I ring, resound, sound ; kambane, kembone, koumbone, a cattle- 
or church-bell. Halm. 


provided beforehand, from wara, foresight, caution, warning ; 
warata, to beware, to make provision ; wara-mies, a supple- 
mental man, a man provided to take the place of another ; 
wara-huonet, a barn, a provision-house ; t oka-war a, provision 
for the future (taka, after) ; warustaa, to provide one with 
necessaries, to fit one out, to arm. Hence Lapp warjo, arms ; 
G. wehr ; waret, warjet, to keep, to guard. 

Lapp welkes, white; welkotet, to become white, to grow 
pale, Finn walkia, white, walawa, whitish, explain G. welken, 
E. welk, welewe, to wither, fade, decay : 
" The which was whilome grene gras, 

Is welewedhay as tyme now is." Gower in Halliwell. 
So in Latin, pallescunt frondes, they wither. 

Finn ivako, Lith. wag a, wag as, a furrow, give a most satis- 
factory explanation of E. wake, the furrow -like track left by 
an object moving through the water, for which however it is 
remarkable that the Finn has a distinct word, wana, trans- 
lated ' f iirchen-ahnliche spur ' by those who had no thought of 
the connection of the English word with the Finn wako. 

From Finn salata, to hide, keep secret, the equivalent 
apparently of Lat. celare, has been shown the origin of an 
adverb salaan, corresponding exactly to the Lat. clam. The 
opposite palam seems also to have its analogue and expla- 
nation in Lapp palen, the locative of pale, a time or turn 
(vicis) . Akta palen, once ; tann palen, at that time ; tai pali 
(in the plural), those times, formerly; peiwe palen, in the 
day-time ; mo palen, in my presence ; weres almai palen, in 
the presence of witnesses (weres alma, literally a man uncon- 
nected by blood, a witness). The ultimate root seems to be 
the Finn palaan, pallata, to turn, return, to roll. 

Among the agreements pointed out by Professor Key in 
the Paper above alluded to, is Lat. ccecus with Finn sokia, 
blind ; which is supported by the number of cases in which 
we have seen a Finn initial s correspond to a Latin c. Now 
sokia in Finn appears to be derived from sakaan, sa'ata, to 
mix, to trouble, to make thick ; sakia, thick, turbid ; sekainen, 
sekawa, mixtus, promiscuus, confusus, perturbatus, haud 
clarus, e. c. aqua, intricatus, obscuratus, e. c. oculus, seu visus. 


Hence soka, what troubles or obscures, as a mote in the eye, 
dregs or sediment in water; sokainen, turbid, impure; so- 
kenen, soeta, to become turbid, to become blind; sokaan, 
soata, to make water foul, turbid ; sokaistus, making turbid, 
blinding. In the same way the G. trube is used of any defect 
of brightness or transparency and also of sight : trubes wasser, 
trubes wetter, and trube augen. We speak of a dull glass, 
dull weather, and dull of sight. 

But possibly the same Finn root may give the derivative 
also of Lat. secale, rye, which is spoken of by Pliny as a fertile 
but inferior grain, hardly eatable by itself, tantum ad ar- 
cendam famem utile, which it was usual to mix with another 
grain, admiscetur huic far, ut mitiget amaritudinem ejus. 
Now Finn sekuli, sekali, signify any kind of mixed food, 
though the former is chiefly applied to a mixture of barley 
and oats, the latter to one of greens and pease. Thus Lat. 
secale would be equivalent to G. mengkorn, Sw. bland-korn. 


[Read January the 25th.'] 

THE special phenomenon, the consideration of which led to 
the writing of the following paper, though not the only sub- 
ject treated in it, is the insertion of certain mutes in Greek 
and other languages into certain pairs of liquids. This, 
though the bare fact is one with which every scholar is 
familiar, has perhaps never yet been sufficiently accounted 
for. Matthiae says, speaking of fippporov, e^pa^evrj, &c. : 
" These are probably not mere poetic licences, but relics of 
old forms." Like Pott before them, Jelf and Latham simply 
apply the epithet " euphonic" to the intruding mute. Don- 
aldson predicates of the Greek ear " a particular aversion to 
the immediate concurrence of jj,\, pp, &c." 

In order however to arrive at a just solution of the 


problem, it is necessary carefully to investigate the character 
and mode of formation of the letters concerned, and especially 
of the so-called liquids, which, except by the Sanskrit gram- 
marians, have rarely been satisfactorily dealt with in the 
classification of the alphabet. Not only do we find one 
liquid, which plays a somewhat conspicuous part in Greek, 
and still more in our own language, commonly ignored; 
but even when the list of liquids is complete, a distinction 
of considerable importance that subsists between certain 
of them is often altogether overlooked. In Dr. Latham's 
works I have not found any allusion to it. Professor Key, 
in his admirable Essay on the Alphabet, affords us only a 
rapid glimpse of it. The distinction in question is however 
pointed out in Heyse's German Grammar, to which I have 
this moment referred, where he describes r and / as mund- 
laute, and m and n as nasenlaute (vol. i. p. 326). 

The liquid above alluded to, as commonly excluded from 
the list and that even by Heyse, although it is quite as 
important an element in the German language as in Greek 
or English is of course the ng of king, song, rung ; Uingen, 
meinung ; and the 7 of 7^09, dyKaXvj, &c. Of this Dr. 
Latham observes : ' ' The simple sound is related to n and g in 
a manner that has not yet been determined." (Eng. Lang, 
first edition, p. 110.) This relation then it is important for 
our present purpose that we endeavour to determine. But 
Dr. Latham's later researches afford no assistance. In 1855 
he affirms : ' ' Ng is no true consonant, but a vowel of a 
peculiar character, i . e. a nasal vowel, formed by the passage 
of air through the nostrils instead of the lips." (Handbook, 
2nd edition, p. 144.) Of the argument that seems to be 
implied in this last clause, it is not difficult to dispose. 
It seems tolerably plain, that with equally good reason m and 
n may be described as " vowels of peculiar character, i. e, 
nasal vowels, formed by the passage of air through the 
nostrils instead of the lips." Word for word, and letter for 
letter, the statement contained in the latter clause will hold 
good "mutato nomine," and therefore the same inference 
may be drawn, if the reasoning is conclusive. It is not 

c 2 

20 11. F. WEYMOUTH, ESQ., 

necessary for me, so far as my present purpose is concerned, 
to give definitions of a vowel and of a consonant respectively ; 
but it would be a singular definition indeed that would include 
ng in the list of vowels, and not embrace n and m also. 
These three consonants possess just this one striking feature 
in common, that when we pronounce them, the breath passes 
not through the lips, but through the nostrils. We will how- 
ever approach them from another quarter. 

In many, perhaps most, languages of civilized nations, 
there are, as in English, just six explosive consonants ; that is 
to say, consonants for the articulation of which all exit of the 
breath is restrained by a complete stoppage of the orifice 
of the mouth, preparatory to a sudden outburst. These are 
the two classes of mutes which we usually call the tenues and 
the mediae ; in Greek TT, K, T, and & 7, S. All the other con- 
sonants are continuous the liquids included. 

Again, the terms tenues and mediae are commonly applied 
only to the six mutes just mentioned. But the difference 
that subsists between them is found also to distinguish certain 
other pairs of sounds, as the English s and z, or the French 
ch and/, which are equivalent to the middle consonant sounds 
in lashing and measure. Now if those physiologists are right 
who attribute this difference to the relaxation of the vocal 
chords of the larynx when IT, K, T, are sounded, and the 
tension, and therefore vibration, of these same chords when 
the media and similar consonants are pronounced, so that 
with these latter there is a more perfect sound ; perhaps the 
names surd and sonant, adopted in some of our Sanskrit 
grammars, best express this distinction. In this sense the 
liquids are all sonant. In this they agree with the medial 
mutes; in being continuous, not explosive, they differ from 

Can the comparison be carried further ? Yes, if we exclude 
the mundlaute r and /, and confine our consideration to the 
three remaining liquids m, ng, and n. These, so far as the 
mouth alone is concerned, might be termed explosives, and as 
sonant explosives they would identify themselves with /3, 7, 8, 
which they closely resemble. Thus in sounding both b and 


m, the tongue lies passive, and the lips are tightly closed, so 
that no breath escapes thence. Comparing the final con- 
sonants of rug and rung, we find in each that, while the mouth 
is open, the body of the tongue is pressed against the palate, 
and thus the orifice of the mouth is completely stopped. So 
is it with d and n, to sound both of which the tip of the 
tongue is pressed against the palate, and, though the mouth 
is open, no exit is afforded for the breath. Thus these letters 
pair off most amicably, the difference in each case being the 

In the English language the liquids m, n, and ng, and in 
other languages these same sounds or such modifications of 
them as occur, alone are sounded by the aid of the nasal 
cavity. All others, including the remaining liquids r and /, 
are what Heyse calls mouth-sounds. More accurately thus : 
in pronouncing m, n, and ng y the pendulous portion of the 
velum palati is lowered, so that the breath passes through the 
nose instead of through the mouth. In sounding all the other 
letters, vowels included, this soft palate is raised so as to touch 
the back of the pharynx, and thus the nasal cavity is entirely 
closed. Yet not entirely in the case of those persons, either 
on this or on the other side of the Atlantic, who speak with 
what is not unaptly termed a " nasal twang." 

To distinguish the nasal from the non-nasal letters, a 
simple but decisive experiment is to hold, while sounding any 
vowel or consonant, a small looking-glass (or the blade of a 
penknife, or any similar object presenting a polished surface, 
and cold) horizontally against the upper lip, with the bright 
surface upwards; this surface will then be dulled by the 
breath only when m, n, and ng are produced, or when there 
is the "nasal twang*." While trying this experiment, we 
cannot fail to perceive how, the moment the velum palati is 

* This suggestion has already been made by the present writer in a few 
observations on a part of this subject that have appeared in the Adver- 
saria of the Cambridge Journal of Classical and Sacred Philology, No. 6, 
p. 333. But the presumed interest of the subject as a whole to the general 
philologer, seemed to warrant its somewhat fuller treatment in a more 
appropriate place. 


lowered, b changes into m, d into n, and g (hard) into ng ; or, 
when we raise it, m is turned into b, n into d, and ng into g. 

But the fact that these consonants are related in some such 
way is sufficiently established by the experience of any one 
who is suffering from a bad cold in the head, such experience 
not being pleasant, but profitable nevertheless to the phi- 
lologer. At such a time made becomes bade; tongue, tug; 
pain, paid; and so forth. In the Welsh and Irish languages 
moreover this affinity of b, d, g (hard), to m, n, ng, is fully 
recognized, the change from the former to the latter being 
termed " aspiration " in the grammars ; but I have nowhere 
met with a satisfactory attempt to explain the exact nature of 
this affinity. 

We have now, I hope, succeeded in fixing the relation of 
ng to g ; showing it at the same time to be fully coordinate 
with n and m, not less a liquid than either of these, nor on 
the other hand, as it has been I think inaccurately described, 
" a more complete nasal." 

As to r and /, which differ so materially from m, n, and ng, 
it seems to be not a happy arrangement by which, in the clas- 
sification of the alphabet, these are all herded together. It 
would seem far preferable that these two non-nasals should 
retain the name by which they were known to the Greek 
grammarians, of semivowels (rjfjLlffxova) . This name we now 
commonly apply only to w and y, but in fact r and / (and 
indeed the sibilants also) approach quite as nearly as do they 
to the nature of vowels : they can just as readily be sounded 
by themselves, and with just as little use of the more active 
organs of speech; and they as readily combine with other 
consonants to form what almost seems a single articulation ; 
so that if, notwithstanding the presence of the w or the y in 
the spoken words dwell and thwack, or duke and newt, we 
may yet consider them as all but biliterals (disregarding 
vowels of course in the use of this term), so we may regard 
trap and drill, or gleam and flat, notwithstanding the pre- 
sence of the / or the r. 

At length therefore we are in a position to consider why 
" euphony" changes /-te^Xerat into pe/Apterai, contracts dvepos 


into dvSpos, and so forth. In the transition from the p to 
the A,, and from the v to the p, besides the other changes in 
which the lips and tongue are concerned, the velum palati 
must be raised to close the nasal orifice. In fact, therefore, 
the process is divided in such instances as these before us, and 
the transition made by halves, the orifice being closed first, 
and the other changes effected afterwards. But thus, we find 
at the intermediate stage of the transition, the m has become 
b, the n, d, and the ng, g ; and this constitutes the so-called 
euphonic insertion of the mute*. 

It cannot fail to be observed, that as yet it has been as- 
sumed that the Greek /3 and 8 were equivalent to our b and d-, 
although those letters are pronounced by the modern Greeks 
much more like our v and sonant th-, and it is at least probable, 
as is Matthise's opinion, that they have preserved the true 
ancient pronunciation of these letters. But if we adopt this 
supposition, and these consonants classed by the gram- 
marians with the a<t>cova were thus, to use Plato's expression, 
(pcwrjevra fjiev ov, ov fjuivroi, 76 a<f>0o<y<ya, being continuous in- 
stead of explosive ; then, inasmuch as the orifice of the mouth 
will not be quite closed in pronouncing them, the resemblance 
that they bear to m and n respectively becomes somewhat less 
marked ; yet the difference is but slight, and they will still, as 
to the precise mode of their articulation, occupy an intermediate 
place between m or n and the succeeding non-nasal consonant. 

The individual phenomena upon which light seems to be 
thrown by the foregoing remarks, are readily divisible into 
the following classes : 

I. Those in which the mute is inserted between a nasal 
liquid and a following consonant. Of this kind are the 
Greek words already discussed, with several others a^iftpo- 

* Since this paper was read to the Society, a friend has informed me 
that I have been anticipated in these views to what extent I am not 
aware by a German writer little known on our side of the Channel. In 
1838, H. E. BINDSEIL published at Hamburg the first, and as yet only, 
volume of his ' Abhandlungen zur allgemeinen vergleichenden Spraeklehre,' 
the first Part of which is specially devoted to a consideration of the Phy- 
siology of vocal sounds. I. Physiologic der Stimm-und Sprachlaute. II. 
Ueber die verschiedenen Bezeichnungsweisen des Genus in den Sprachen, 


TO?, crivSpos, fjiecrrjfju^pLa, &c. ; and the numerous analogous 
instances furnished by other ancient or modern European 
languages ; Lat. templum, Fr. viendrai, &c. Of similar words 
a large collection may be found in Pott. Of the g thus 
inserted, the French epingle is the only example I have met 
with. Derived from the Lat. spinula, it assumes the forms 
espinule, epinule, epinle. In the last of these the second syl- 
lable terminates with the well-known French nasal akin to 
the English and German ng, though weaker, and thus more 
resembling the Sanscrit anuswdra. From this sound to the / 
there is then a transition which is broken by the insertion of 
the g. The French ebranler, like the English ringlety the 
Germ.jiingling, &c., does not take the mute. 

II. Those in which a vowel intervenes, at least in the 
word as spoken, between the nasal liquid of the root and 
the sequent consonant, the euphonic mute being still inserted. 
Of this we see examples in the Eng. number, tumble, &c., 
where the b belongs not to the root. Also in cinder, gender, 
thunder, gander*, t-run-dle, &c. ; in which a radical n is sup- 
ported by its cognate but exotic d. And thus we may explain 
the difference in the pronunciation of the ng in such a pair of 
words as the English younger and its German equivalent 
jiinger. In the latter the ng represents one simple sound, 
and in this comparative, as in alter, starker, and the Engl. 
broader, wiser, Sec., we find nothing anomalous ; the regular 
comparatival termination being appended immediately to the 
root. But in the comparatives of English adjectives in -ng, 

* " With regard to the d in gander," writes Dr. Latham, " it is not easy 
to say whether it is inserted in one word or omitted in the other [gans]." 
(Handbook, 2nd edition, p. 214.) The analogy of the other similar words 
mentioned in the text gives a high degree of probability to the former 
supposition ; and this is confirmed by the great rarity, if not non-existence, 
of precisely analogous instances of an omitted d, and by the long list of 
cognate words in various Indo-European languages, in none of which a d 
is found, except where an r is affixed, as in the A.-S. gandra and the Engl. 
and Low Germ, gander. Eichhoff, Pott, and Dr. Latham himself, furnish 
the following : Sansc. hansa, hansi; Pers. kay; Greek x*l v > Lat. anser 
(and gan-lus}', Germ, gans and hahn ; O.H.G. kans-, M.H.G., a mascu- 
line form ganazzo ; Lith. zusis, and several others. 


we insert between the root and the termination -er, the hard g, 
which, as has been above shown, is the explosive mute akin to 
the nasal ng ; and the young-g-er, long-g-er, strong-g-er, which 
result, are precisely analogous formations to number and cinder. 

But as the ng in younger does not stand for the simple 
nasal liquid as in young, but for that liquid + its cognate mute ; 
if we search in the direction thus indicated, may we not find 
other instances besides the French epingle, in which the ear, 
if not the eye, can discover an insertion of the mute between 
two consonants, without an intervening vowel, just as in 
ajjifipoTos and dvSpos? Wrangler, pronounced wrang-g-ler, 
contains a hard g which does not belong either to the root 
wring, or to the termination ; so does hungry, if the Germans 
have preserved the earlier pronunciation of the noun hunger. 
Compare also the French Hongrois with the German Ungar. 
Examples of this kind, however, are not numerous ; and with 
most of them there is a prior form, in which, as in younger, a 
vowel sound is interposed between the nasal liquid and the 
next consonant. 

And here we may not unsuitably inquire what combi- 
nations of consonants there are, either with or without an 
intervening vowel, which thus invite the introduction of a 
mute. 1. Mp takes the inserted @ in apPpoTos, &c. 2. MX 
inserts yS in fjLeppXcoKa, dfj,(3\vva) for djjuakvvw, &c. 3. In 
the Latin templum, ml has taken p. 4. In J^a^dtv from the 
Hebr. Shimshon, and le^a, ^ has a TT inserted according 
to our pronunciation, but a b (English) according to the pro- 
nunciation of the modern Greeks, who would read these words 
as Sambzo n and Tem'bza. 5. Mt takes p in the Latin 
emptus, sumptus, &c. 6. In the Greek \d/jL/38a, derived from 
the Semitic name lamed, we find /3 inserted between /*, and S. 
7. Chaucer's Sompnour, dampnacioun, &c., show the p in- 
serted between two nasal liquids. 

jV gives us less variety. 1. Between v and p there is a 8 
iii dvSpos. 2. Our English verb to trundle shows the d be- 
tween n and /. 

Ng takes the additional sound of the hard g : 1. before r 
in Hongrois, younger, &c. ; 2. before / in wrangler. 


III. But again we find in many words in some modern 
languages, that a nasal liquid, which is not followed by any 
syllable, but is itself the final of the root, has its cognate mute 
appended. A familiar example is found in the English hound 
as compared with the Latin can, the Greek KW, and the 
Sanskrit swan. Such also are comb and lamb, though in 
these in modern English we do not sound the final consonant. 
It is indeed contended, on the authority of the old Sanskrit 
grammarians, that such forms as hound, kind, tendo, are older 
than the allied forms without the mute, and that the change 
which has occurred is one of subtraction rather than addition. 
To discuss this point now would be simply a digression : 
suffice to say therefore that there are forms in which beyond 
dispute the addition has taken place ; such as tyrant, ancient, 
Normandy, Germ, jemand, niemand, &c. 

A similar formation to that of comb from the biliteral root 
that may be traced in the Germ, kamm (primarily signifying 
a range of hills), the French cime, the Lat. cum-ulus, and the 
Greek KVTTTW with its numerous oifspring, is the Greek 
Tafias. At least it seems probable that the root of this 
word is rvfj,, as found in the Lat. tumeo, tumulus, &c. ; and 
that the verb TV</>O>, from which Liddell and Scott derive it, is 
rather to be regarded as an imitative word of separate origin. 
The roots TV^ and TV<(>, though resembling each other, as 
they express kindred notions, may yet be altogether inde- 
pendent of one another ; the latter with its continuous non- 
nasal consonant being well-fitted like multitudes of similar 
words (e.g. Oeco, Tpe%co, curro, A.-S. yrnan, Germ, laufen) 
to convey that idea of visible motion, namely of the rising 
smoke, which is not inherent in the former. 

Lastly under this head must be mentioned words termi- 
nating in -ng, as king and song, when pronounced as in some 
parts of England, kingg and songg. But in young and the 
Germ. Jung, when we trace them back to their origin, we find 
reason to believe that the process by which kingg and songg 
have assumed a guttural that does not belong to them, is in 
this word, in the later stages of its growth, reversed. Sup- 
posing the radical form to be the Sanskrit yuvan or the Lat. 


juven, the middle consonant first becomes fully vocalized in 
such a form as the Lith. jaunas and in the Lat. junior for 
juvenior ; a guttural is then added in the Gothic juggs (pro- 
nounced junggs, the doubled g being = yy in Greek), the 
O. H. Gr.junk and the M. H. Gc.junc (theEngl. younker being 
evidently derived from some such form), and this guttural is 
again dropped in K.-^.geong (?),Engl. young, and N. H. Q.jung. 

IV. A fourth class of facts are supplied by some of the 
dialects of Western Africa, in which a great number of words 
may be found to begin with a suppressed vowel sound fol- 
lowed by a nasal liquid and a mute. In these it seems most 
probable that of the two consonants only one is radical and 
the other euphonic. The combinations are those of m with b 
or p ', of n with t, d, or its compound j ; and of ng with g or k ; 
as exhibited in the Dualla 'mbenffa, dove; 'mpimba, nose; 
'ndabo, house; 'ngodi, girdle or } nggodi, as the compiler of 
the fragmentary Isubu grammar writes such words, the ng 
being sounded as in the English younger. The nasal liquid is 
in this class of words always followed by its cognate surd or 
sonant explosive mute, each being distinctly pronounced 
(continuous mutes are apparently unknown to these languages 
except in imported words) . 

V. It is doubtless owing to the close affinity between the 
labial pair of these letters, that in the etymology of the lan- 
guages just alluded to, the Dualla and the Isubu, these letters 
have one function so much in common. In six out of seven 
classes of plural nouns, the plural prefix begins with an m or 
a b. And compare the Kafir plural prefixes aba and ama. 

But in like manner numerous instances may be adduced of 
the interchange of these letters, and not of the labial pair 
alone, but of the palatals and gutturals also. 

1 . M and b QIC p are apparently thus interchanged in the 
following words : Germ. bad, Engl. bath, Sanskr. mid, Lat. ma- 
deo, Lith. maudau : Germ, burg, Engl. borough, Sanskr. mur, 
Lat. murus : Germ, weib, Engl. womb, Germ, wamme ; also in 
husband for house-man-, and more clearly in certain cases 
where an r or / follows, as /SXlrrco from /jue\i, &c. (See Eich- 
hoff, and Key on the Alphabet.) 


2. N is interchanged with the palatal mutes in the Sanskr. 
navan, Lat. novem, Lith. dewyni, Russ. dewiat' ; in the Greek 
roots fjuad and pev, &c. (See Key.) 

8. The consonantal sound with which our word tongue 
closes, and which we have in the Germ, zunge, is not found 
in the cognate verbs, the Engl. lick, Germ, lecken, O. H. G. 
lekon, Goth, laiffon, Greek Xe/^o), Sanskr. lih. And as in the 
Sanskrit form of this root, the nasal is substituted by the 
softened guttural h, so in Gothic we ftndjuhiza as equivalent to 
the O. H. Gr.junffiro andN. H. Q.junger. Compare also the 
French join-d-re with the Greek root %evy, and the I^ai.jugum. 

But perhaps in most or all of these cases, if the interchange 
is real, and the words in question are not derivatives of roots 
of independent origin simulating affinity, such interchange is 
indirect and may be referred to euphonic causes. Thus the 
root jug being strengthened, as is the case in several words 
shortly to be alluded to, by the insertion of the allied nasal, 
becomes jung (the verb being doubtless pronounced jung-go), 
and the newcomer now ousts the original guttural mute to 
form the French join-t*, &c. So in the root lih, lick, Xe% 
&c., the same nasal being inserted gives the various forms 
lingua, Celt, dingua, Goth, tuggo (pronounced tung-go), 
O. H. G. zunka, Swedish tunga, N. H. G. zunge, Low Germ. 
tunge, Engl. tongue -, the original guttural being quite lost in 
the last three or four of these. Very similarly may we trace 
the growth of the German forms menge and Pfingsten from 
primitives in which this nasal is not found. They do, however, 
contain another nasal for which ng has been substituted. The 
former of these words is from the O. H. G. managi, allied to 
the Engl. many, &c. Here the vowel of the second syllable 
being dropped, euphony required an exchange of nasal liquids, 

and finally the guttural was lost from pronunciation. So in 


* It may be observed in passing, that this French anuswara may re- 
present as a final any one of the nasal liquids : thus it is substituted for 
the true labial m in chambre from camera, champ from campus, impur from 
impurus ; it stands for n in chanvre from cannabis, chanter from cantare, 
bonfvom bonus; and for ng in point from punctum, e'teindre from extinguo, 
plaindre from plango, and so forth. 


Pfingsten from the Greek Treyr^Kocrrrj, syncope produced some 
such form as 'jrev-Kocrrr], necessarily modified into peng-koste ; 
then the k is dropped, and the Low German pingsten, M. H. G. 
pfinyesten, and N. H. G. Pfingsten result. 

VI. Familiar to every Greek scholar are the numerous 
cases in which the mute is the radical, and the liquid the 
auxiliary ; such as p,avOdv(o, \av6dvw, irevOo^, &c. And as the 
6 here represents in some sort its kinsman S, which we have 
generally hitherto found combining with the n; so instead of 
the combination /u-/3 we have //,</> in pi^a and p^K^aXeo?, 
dashing, from the root piir. M/3 occurs however in Xapftdvco, 
ddfjufio? (and the Latin plumbum), and Tv^irayov, pronounced 
by the modern Greeks teem b anon, with TT = the English b ; in 
all of which the //, does not belong to the root. The inserted 
ng we find in \ay%dva) from Xa^, and Tvy%dvco from TV% ; as 
well as in the Lat. jungo and Goth, tuggo, which have been 
already discussed. 

But can this insertion be anyway accounted for ? I think 
so, if we bear in mind that in the class of Greek verbs just 
mentioned, the short form is used chiefly and almost exclu- 
sively in the 2nd Aorists, tenses which represent the action 
of the verb at once in its completeness ; the form with the 
liquid belongs only to the imperfect tenses, which represent 
the action as prolonged or habitual. It is that the sound may 
answer to the sense, that to express the idea in the latter 
form, the sound of the word is prolonged by the strengthening 
of the already continuous mute by another continuous con- 
sonant, its cognate liquid. Similarly, it is that the mind may 
dwell on the notion which the word conveys to it, that the 
sound is thus strengthened in /3ev#o<?, Tv^iravov, plumbum, 
&c. And may we not thus account for the fact that the par- 
ticipial termination t or d, the simple explosive mute, is so 
extensively used in the Indo-European languages for the 
perfect tense, and the form in nt or nd for the imperfect? 
Let us compare the two following lists. 

1. Perfect participles. Sanskr. dpt-a(s), Gr. /&a>T-o9, Lat. 
lect-us, Fr. convert, Germ, geliebt, O. H. G. giladot, Dutch 
gedrukt,K.-S.gelufod, Lith. let-as, Engl. loved and learnt. 


2. Imperfect participles. Sanskr. pachant (in declension), 
Gr. TUTrrovT-o? (pronounced teep'tondos by the modern 
Greeks with T = the English d], Lat. amant-is and amand- 
us, Fr. allant, Germ, liebend, Low Germ, lewent (in such a 
phrase as dat Lewent = Engl. " infinitive in -ing," living), 
Goth, stigands, Du. woonende, A.-S. tellende, Lith. lejand; 
Engl. telling, or in the midland counties tellingg. 
The prolonged sound of the termination in words of the latter 
class, as contrasted with the rapidity with which that of the 
others is dismissed, seems to render such forms very appro- 
priate for their office of expressing an action as still continued 
and incomplete. 

VII. The explanation above offered of the affinity of m 
with b will fully account also for the jj, substituted for (B in 
e/36yLtvo5 and <re/tvo9. The roots being epe/3 and cre/3, and the 
Hebrew language proves this in one case, and the Sanskrit in 
the other, the termination -vo? is to be appended. But the 
v is a nasal liquid, that is, it is sounded with the velum palati 
lowered. This may be lowered therefore after the /3 is pro- 
nounced ; but it is much easier to effect this change in the 
position of the organs at the opportunity which the preceding 
vowel affords, and to sound both the consonants that intervene 
between the e and o with the organs as much as possible in 
the same position, that is, with the nasal cavity open for them 
both; the first consonant being assimilated to the second, 
according to the rule of the Greek language, rather than the 
second to the first; and hence epe/Avos, o-efivos. A like 
change is effected by the letter n in the Swedish hamn and 
its derivatives and compounds. Here the termination -n is 
appended to a root ending in v or /, the root being doubtless 
found in the Danish hav, German haff, and Swedish haf; 
whence are derived havn in Danish, hafen in German, and 
haven in English. But contact with n has in the Swedish 
word changed the labial mute into the labial nasal liquid. 

VIII. There are, however, some instances of a mute inserted 
where the first consonant is not a nasal liquid, but a sibilant. 

The first is the adjective eV0Xo?, if at least we adopt the 
opinion of Jelf, Donaldson, and others, that the Doric eo-\o9 


is the original form. Then the 6 is thus derived. I assume 
its common pronunciation among ourselves to be the true 
one, being that of our surd th, as it is pronounced by the 
modern Greeks. It may then be remarked, that the formation 
of the three sounds of s, th, and /, requires the tongue to be 
pressed against the teeth for the first, to touch the anterior 
part of the palate for the second, and to touch the palate 
again further back for the third; so that in the transition 
from s to /, the tongue passes by, if not actually through, the 
very position that is necessary to pronounce the th. Hermann 
however identifies this word, and I think rightly, with the 
German edel-, and if this be allowed, so that e#X is the root, 
like the Anglo-Saxon seftel (csthel), and the Doric form stands 
for e'0\o? as cio? for 0eo9, the problem to solve will be to 
account for the a- prefixed to the 9 of the root. This is not 
easy ; though we may at least assert that the strengthening 
of the by the cognate semivowel c is analogous to that of 
the mute by an inserted nasal liquid in Tv^iravwv, Tvy%dvc0, 
&c. Whether in io-Q/jios, l^da-OXfj, pacOaKk, ae6p,a, (TVTTT)- 
ofjuea-Oa, &c., the a or the 6 is the radical, is perhaps not 
readily determined. 

A second case is that of Meo-rpai//-, as the name appears in 
Manetho's fragments apud Syncellum, though the LXX. write 
the name Meo-patju, or -iv without the T. This T may have 
been derived directly from the Hebrew tsade of the original 
word, so that Meo-rpatja i s but varied by metathesis for 
Mera-pat/jt,, which to Greek organs of speech would be an 
impossible form. If on the other hand, as seems more pro- 
bable, the T is simply euphonic, its introduction may be 
explained just as that of the 6 in eo-0Xo? : it serves in pre- 
cisely the same way as a stepping-stone from the sibilant to 
the succeeding consonant. A parallel case is the "Eo-Spa? of 
the LXX. for the Hebrew Ezra, where however the z is zayin, 
not tsade. 

From this point of view let us examine the Latin castrum. 
The root I believe to be the biliteral cas*, found also in casa 

* Dawson and Rushton, in their Terminational Dictionary, divide the 
word ca-stra. I venture to think this a mistake. They err in the other 


and the Croatic kuzha, and easy to be identified by the aid of 
Grimm's Law with the Germ, haus, A.-S. hus, Engl. house, 
&c. Is then -trum the termination ? I think not : I believe 
-trum, wherever its force can be distinctly seen, marks the 
instrument, as is laid down by Professor Key in his Latin 
Grammar. I would therefore hazard the conjecture, that 
cas-lum or cas-ulum was the original form, signifying pri- 
marily a little house, i. e. a tent or hut, that element in short 
of which a camp will be composed ; and then coming, through 
the associated military ideas, to signify a fortified dwelling, 
and hence a fort. Suppose this so. We know that no word 
either in Greek or Latin begins with si, and that this was 
scarcely a tolerable combination to the Greek or Latin mouth. 
A t was therefore inserted, on the principles above explained, 
stl being (in both these languages) a possible group of con- 
sonants, yet not a favourite combination, and the / was 
therefore changed into r to facilitate pronunciation, as in the 
very similar old French forms apostre from apostolus, epistre 
from epistola, and numerous other examples. (See Key on the 
Alphabet, p. 73.) 


[Read April the llt/i.'] 

GULL, a dupe ; to gull, to deceive, to make a dupe of. A 
metaphor taken from the utter helplessness of a young bird, 
still provincially called a gull. Wilbraharn (Cheshire Glos- 
sary) says that all nestling birds in quite an unfledged state 
are called ' naked gulls/ doubtless from the yellow tint of the 
naked skin about the beak and other parts ; Icel. gulr, Dan. 

direction in the case of astrum, of which they make ast the root and rum 
the termination; though the Germ, stern, the English star, the Greek 
dcrrfip, the Sanskrit tar a, &c. all prove that the r in this case belongs to the 


guul, yellow. In Surrey the term is commonly applied to a 
gosling. In the same way the Fr. bejaune, the proper meaning 
of which is a young bird (yellow-beak), is used in the sense 
of a novice or simpleton; bejaunage, bejaunise, simplicity, 
inexperience, doltishness (Cotgr.). Another variation of the 
metaphor designates the dupe as a pigeon, originally signi- 
fying a young bird, from Lat. pipio, ' a young pipping or 
chirping bird, a squab' (Andrews). Hence It. pippione, pass- 
ing into piccione (as sappia and saccia from sapere, abbia and 
aggia from habere), applied to the young pigeon in the same 
way as fowl to cocks and hens, or bird in sportsmen's language 
to the partridge. ' Pippione, a pigeon, a silly gull ; pippionare, 
to pigeon, to gull one, to make one swallow a gudgeon/ 
(Florio.) Again, the Fr. niais, a nestling, is taken as the 
type of simplicity or folly. ' Niais, a nestling; hence a 
youngling, novice, ninny, a simple, witless and inexperienced 
gull/ (Cotgr.) 

BEZONIAN. The Fr. bejaune, mentioned in the last article, is 
I doubt not the origin of the ' Bezonian ' of our dramatists, 
commonly supposed to be derived from It. bisogno, want, 
bisoanoso, necessitous, making the term equivalent to 'poor 
devil/ But this is not the sense of the Sp. bisono, from whence 
doubtless the expression immediately comes, that term being 
applied to a raw recruit, novice, tyro, simpleton, f incongru, 
bejaune, sot, niais/ (Nunez.) The term bisogni was also ap- 
plied in Italian to new-levied soldiers, and in the long Italian 
wars of the middle ages, when French, Spanish and Italians 
were mixed up together, any piece of military slang would 
pass with the utmost facility from one language to the other. 
The sound of the French j, being foreign to the two other 
languages, would naturally be represented by a z, as in the 
Piedmontese biso from bijou, a jewel. The Italian, unskilled 
in French, says zoli forjoli, zour for jour. 

GOBLIN. The Goblin, under one name or another, was a 
superstition very widely spread over Europe in less instructed 
times. It was generally conceived as a supernatural being of 
small size, but of great strength, dwelling under ground, in 
mounds or desert places, not generally ill-disposed towards 


man, and in some cases domesticating himself with him and 
rendering him service. Hence the frequent addition of a 
familiar appellation, as in Hob-goblin, Robin Goodfellow, Hob 
Thrush (Cotgr. in v. Lutin; Hob-drudge?). It was known 
in Germany by the name of Kobold, and was supposed par- 
ticularly to frequent mines, where it is capriciously favourable 
or mischievous. The miners, says Adelung, who have always 
much to do with the Kobold, call him Berg-geist, Berg- 
mdnnchen (which may be translated ' mine-ghost, mine-dwarf), 
Matthew Kobalein. From the prevalence of the superstition 
among this peculiar class has arisen the name of the mineral 
cobalt, the value of which has only been discovered in modern 
times, being formerly only known as an incumbrance among 
valuable ores attributed to the ill-offices of the Kobold, 
whence the name is said to have arisen. 

There can be no doubt that the name Kobold is identical 
with the Fr. gobelin, the habits of which are mentioned by 
Ordericus Vitalis, as quoted by Adelung: " Daemon enim 
quern de Dianse fano expulit adhuc in eadem urbe degit et in 
variis frequenter formis apparens neminem Isedit. Hunc 
vulgus gobelinum appellat." He is known in Breton by the 
name of gobilin, and is there supposed to engage in household 
drudgery, to curry the horses of a night, for instance, like 
Milton's Lubber-fiend. 

It is among the Celts probably that the origin as well of 
the name as of the superstition itself is to be looked for. The 
name in Welsh is coblyn, signifying in the first instance a 
knocker, from cobio, to knock, to thump ; cobiwr, a knocker, a 
pecker; coblyn y coed, a woodpecker. The origin of the 
appellation seems to be indicated in a passage in which there 
is no reference to the name goblin, and the writer of which 
had probably never thought of any connexion between that 
word and the superstition he is describing. "People will 
laugh at us Cardiganshire miners," says a correspondent, 
quoted in 'Bridge's Guide to Llandudno/ "who maintain the 
existence of knockers in mines, a kind of goodnatured im- 
palpable people, not to be seen, but heard, and who seem to us 
to work in the mines. The miners have a notion that these 


knockers, or little people, as we call them" (compare G. berg- 
mdnnchen) " are of their own tribe and profession, and are a 
harmless people who mean well." 

It will be observed that the Kobold in Germany is pecu- 
liarly a miner's superstition, while Cardiganshire has been a 
mining district from the earliest period. 

GAZETTE. Commonly derived from gazzetta, the name of a 
small Venetian coin supposed to be the price of the original 
newspaper. But the value of the gazzetta was so small (' not 
worth a farthing of ours/ Florio), that it never could have been 
the price either of a written or printed sheet. Schmeller was 
nearer the mark when he derived the word from gazzetta, the 
diminutive of gazza, a magpie, supposing that the image of 
that bird may have been impressed upon the earlier news- 
papers as the emblem of talkativeness. But without evidence 
of the supposed practice, a guess of this kind is worthless. 
Moreover, in the present instance the supposition is wholly 
unnecessary. The magpie is called gazza in Italian, as chat- 
ter-pie in English, from a widely-spread root representing a 
chattering noise, which is exemplified in E. chat, chatter; 
Hung, csatora (cs = English ch), noise, racket, csatordzni, to 
make a noise, chatter, talk much, csacsogni, to chatter or 
prattle, csacsogdny, a chatter-box, magpie, jackdaw; o\.gadac, 
to talk, gadu-gadu, chit-chat, tittle-tattle; Fr. gazouiller, 
to twitter, to murmur ; It. gazzerare, gazzolare, gazzogliare, 
gazzettare, to chatter as a pie or jay, to prate (Florio). 
Hence gazzetta, gazzette, 'all manner of idle chattings or 
vain prattlings, but now generally used for running reports, 
daily news, intelligences and advertisements as are daily 
invented and written unto foreign nations, viz. from Venice, 
Rome, and Amsterdam/ (Florio.) 

The primitive meaning of the word then is simply chit-chat, 
the appropriateness of which may be illustrated from a late 
Number of the Quarterly Review on Advertisements : "At the 
same time, the public journals, it is clear, had not performed 
that part of their office which was really more acceptable to 
the country reader than any other the retailing of the political 
and social chit-chat of the day." (No. 193. p. 204.) 



BOWELS. Lat. botulus, a sausage ; It. budelle ; Venet. buele-, 
O. Fr. boel; Bret, bouzellen, plur. bouzellou or bouellen, bou- 
ellou, bowels. Perhaps named on account of the (SopftopwyiJios 
or rumbling sound which takes place in them, from Bret. 
bouda, to hum, to murmur, the equivalent of E. buzz. The 
W. poten, the belly, a pudding, is probably the same word, 
and may be illustrated by the Finn potina, gemurmel, a rum- 
bling or murmuring, from potista, rauce ebullio ut puls 
fervida, mussito, dumpf tonen, murmeln. In like manner, in 
Icelandic the belly is termed bumbr, from bumla, to resound ; 
Gr. fto/j,/3v\i,ac0, to rumble, ventris murmur edo; fio/jiftvXr}, 
a narrow-mouthed guggling vessel. Probably guts, the proper 
English designation of the bowels, is derived from another 
imitation of the internal rumbling exhibited in Icel. gutl, by 
which is represented the agitation of liquids in a vessel ; ' at 
gutla, agitare liquida ut bilbiant/ to guggle. 
His guttes begonne to gothelen 
Like two gredy sowes. P.P. 

Plat. D. guddern is applied to the rattling sound of things 
falling in abundance, as apples from a tree, water pouring from 
a roof. The W. and Gael, bru, the belly, seem in like 
manner connected with It. bruire, to rumble; il venire mi 
bruisce, my guts rumble (Altieri) . So also Pol. brzuch, the 
belly, and brzeczec, to hum, to buzz; Russ. briucho, belly, 
and briuzchat, to grumble. 

To BUCK. A mode of preparation for washing formerly in 
universal use, by soaking the linen in a solution of wood 
ashes. The word was very generally spread. In G. it is 
beuchen, buchen, buchen, buken-, Sw. byka; Dan. byge; Fr. 
buquer, buer; It. bucatare. The derivation has been much 
discussed. The more plausible suggestions are 1. Dan. bog- 
aske, the ashes of beech wood, chiefly employed in making 
potash ; but the practice of bucking would have arisen long 
before any particular kind of wood was employed in procuring 
a supply of ashes. 2. It. bucata, buck-ashes, supposed to be 
derived from buca, a hole, because the ashes are strained 
through a pierced dish, whence the ashes for bucking, or the 
act of bucking itself, or the linen operated on, are called 


colada in Spanish,, from colar, to strain. But the analogy 
fails, because bucare does not appear ever to have been used 
in the sense of straining or filtering. 

The true derivation is the Gael, bog, moist, soft, and as a 
verb, to steep, to soak, to soften; Bret, bouk, soft, tender, 
whence boukaat, to soften, doubtless originally to soak. In 
the same way It. molle signifies both moist and soft, and the 
Lat. mollire, to soften, is identical with Fr. mouiller, to wet. 

The frequent interchange of b and m (as in W. baban, 
maban, a baby) leads us to identify the Celtic root with the 
Slavonic mok, wet, appearing in Eng. muck, meek, and Lat. 
macero, as mentioned in a former paper. Hence Russ. mokro, 
wet, moknut, to become wet, mochit, to wet, to soak ; Bohem. 
mok, a steep for flax; Pol. moczyc (mochits), to soak foul 
linen before washing. In Lat. imbuere, to soak, the root has 
lost the final guttural, as in Fr. buee for buquee. 

HOST, an army. This is one of the words, with respect to 
which little is gained by simply mentioning the origin with- 
out sufficient illustration to explain the mode in which it came 
to acquire the actual signification. 

In the troubled times following the breaking up of the 
Roman empire, the first duty of the subject was to follow his 
lord into the field when called on by proclamation to march 
against the enemy. The demand for military service was 
expressed by the term 'bannire in hostem/ to order out 
against the enemy, as in an edict of Charlemagne quoted by 
Muratori, Diss. 26 : ' Quicunque liber homo in hostem ban- 
nitus fuerit et venire contempserit, plenum heribannum compo- 
nat/ i. e. as it is explained, ' 60 solidos solvat.' The term hostis 
then, which primarily indicated the enemy against whom the 
expedition was to be made, was compendiously used for the 
military service itself, and is frequently taken as synonymous 
with ' hostilis expeditio/ or f exercitalis expeditio/ and is then 
used as a feminine noun. A supplication is addressed to Charle- 
magne, ' ne episcopi deinceps sicut hactenus vexentur hostibus 
(i.e. with demands of military service) sed quando nos in hostem 
pergimus' (which may be translated either, when we march 
against the enemy, or, when we proceed on military duty or 


join the ranks), 'ipsi propriis resideant in parochiis/ The 
same immunity is expressed in a charter of A.D. 965 : ' Nee 
ab hominibus ipsius ecclesiae, hostilis expeditio requiratur/ 
' Hostem facere' was to perform military service. In a law 
of Lothaire a certain fine is imposed on those who, having the 
means, neglect ' hostem bene facere/ while those are excused 
' qui propter nimiam paupertatem neque per se hostem facere, 
neque adjutorium praestare possunt;' and the same sense is 
expressed in contemporary documents, ' qui in exercitalem ire 
possunt expeditionem/ In like manner in Italian, 'Boglio 
fare la hoste sopra Palestrina/ Fragm. Hist. Rom. in Muratori. 
'Bandire hoste/ to proclaim war (Florio). The term would 
easily pass from signifying military service to the body of 
men engaged in such service, or to signify an army, and 
thence any numerous assemblage. 

TOURNAMENT. Commonly explained from the combatants 
having to turn back their horses after each tilt to make a 
fresh charge, ' quia scilicet equos celeriter in orbem circum- 
versant' (Skinner). But probably the signification has been 
attained by a somewhat different track. The peculiarity of a 
tournament was not so much the wheeling of the horses, 
which no doubt is one signification of Fr. tournoyer, but the 
fighting within a railed-off field, or lists, a ' champ clos/ as it 
was called in Fr. Now another meaning of Fr. tournoyer , as 
of It. tornear, is to surround or fence round; torneamento, a 
fence, hedge, enclosure ; and hence probably it was that the 
term torneo or torneamento was applied to a combat within 
lists. An old Italian chronicler in Muratori (vol. iii.), speak- 
ing of the Black Prince at the battle of Crecy, says, ' Fece 
attorniare soa huoste con pali di fierro moito spessi ficcati in 
terra. Quesso attorniamento fu fatto alia rotonna a modo di 
fierro da cavallo/ 

TRADE. This is one of those cases, several of which have 
been previously pointed out, where a modern worch has been 
formed from the coalescence of two others originally distinct, 
but resembling each other in sound, and of similar meaning in 
certain applications. 

From Lat. tractare^.to handle, transact, discuss, treat, was 


formed It. trattare, to treat; Fr. tr -alter, to treat, handle, 
deal in or meddle with, debate, contract with (Cotgr.) ; Sp. 
tratar, to handle, treat on a subject, to discuss, to manage, to 
traffic, to trade (Newman and Baretti). Hence trato, treat- 
ment, intercourse, trade, traffic, commerce ; Fr. traite, a 
draught, course, trace, proceeding, also a transportation, out- 
ward vent or shipping over (Cotgr.) . ( La traite des noirs/ the 
slave trade. At the same time, from a totally different action, 
expressed by A.-S. tredan, to tread, was formed A.-S. trod, a 
path, track, course; in O. E. trade, trode, troad. 

Wyth wynd at wylle the trad held thai, 

And in England com rycht swyth. Wyntoun. 

They say they con to heaven the highway, 
But by my soul I dare undersay, 
They never set foot in that same troad, 
But balk the right way and stray en abroad. 

Spenser, Shep. Cal. 

So ' trade wind,' a wind preserving a certain course. 

The word was then metaphorically used in the sense of 
course or habit of action : 

Tho would I seek for queen-apples unripe 
To give my Rosalind, and in summer shade 
Dight gaudy girlonds was my common trade 
To crown her golden locks. Shep. Cal. 

It seems then to have been applied to any special course or 
mode of occupation by which a man earned his living, and 
thus came to signify handicraft or mercantile business, as 
distinguished from agricultural labour, the common lot of the 
mass in less advanced times. It now became confounded 
with Sp. trato and Fr. traite, and attracted to itself the signi- 
fication of commerce or traffic properly belonging to the 
Roman derivation. 




[Read May the 23rd.'] 

The following vocabulary is one taken by Dr. H. Sandwith 
from a Kurd of the Zaza tribe, one of the rudest of the whole 
Kurd family, and one for which we have no philological 


head ...... sere-min. 

eyes ........ tchim-ewm. 

eyebrows .... burue-wm. 

nose ........ zmje-min. 

moustache . . simile-min. 

beard ...... ardishe-mm. 

tongue ...... zoane-mm. 

teeth ...... dildone-m'w. 

ears ........ gushe-?mVz. 

fingers ...... ingishte-mm. 

the back . 

hair ....... pore-?mw. 

cold ........ serdo. 

hot ........ auroghermo. 

sun ........ rojshwesho. 

moon ...... hashme. 

star ........ sterrai. 

mountain. . . . khoo. 

sea ........ aho. 

valley ...... * derei. 

eggs ........ hoiki. 

a fowl ...... kerghi. 

welcome .... tebexairomfc. 

come ...... beiri. 

stay ........ roshe. 







bread noan. 

water awe. 

child katchimo. 

virgin keinima. 

orphan lajekima. 

morning .... shaurow. 

tree dori. 

iron asin. 

hare aurish. 

greyhound . . taji. 

pig khooz. 

earth ert. 

fire adir. 

stone ...... see. 

silver sem. 

strength .... kote. 

sword shim shir. 

a fox krevesh. 

stag kive. 

partridge. . . . zaraj. 

milk shut. 

horse istor. 

mare mahiiie. 

grapes eshkijshi. 

a house .... ke. 

green kesk. 

' crimson .... soor. 

black siah. 

white supeo. 

sleep ...... rausume. 

go ; shoori. 


The meaning of the termination -min has been explained 
by Pott and Rodiger in their Kurdische Studien. It is the 
possessive pronoun of the first person =my=meus = 6/409, &c.; 
so that sere-mm = cwput-meum (or mei), and pie-mm = pater- 
meus (or mei). 

So little was the Zaza who supplied Dr. Sandwith with the 
list under notice able to conceive a hand or father, except so 
far as they were related to himself, or something else, and so 
essentially concrete rather than abstract were his notions, that 
he combined the pronoun with the substantive whenever he had 
a part of the human body or a degree of consanguinity to name. 
It is difficult to say how far this amalgamation is natural to 
the uncultivated understanding, i. e. it is difficult to say so on 
a priori grounds. That the condition of a person applied to 
for the purpose of making a glossary out of his communi- 
cations is different from that under which we maintain our 
ordinary conversation, is evident. Ordinary conversation 
gives us a certain number of words, and a context as well. A 
glossary gives us words only, and disappoints the speaker who 
is familiar with contexts. 

If this be true, imperfect contexts, like the combinations 
pie-min, &c. should be no uncommon occurrences. Nor are 
they so. They are pre-eminently common in the American 
languages. Thus in Mr. Wallace's vocabularies from River 
Uapes the list runs thus : 


head (my) .... m-bida .... tcho-kereu, . . . wo-dusia. 

mouth (my) . . m'-numa . . tcho-ia. wo-nunia. 

&c. &c. &c. &c. 

similar illustrations being found in almost every American 

In his Appendix to Macgillivray's Voyage of the Rattle- 
snake, the present writer pointed out instances of this amal- 
gamation in the languages of the Louisiade. He now adds, 
that he has also found it in some of the samples of the 
ordinary Gipsy language of England, as he has taken it from 
the mouth of English gipsies. 


He considers it to be a personal rather than a philological 
characteristic, certain individuals having a minimum amount 
of abstracting power, and such individuals being inordinately 
common amongst the American Indians. 



[Read May the 3th.~] 

If we inquire for the older etymologies which have been 
given of YITTLO^, we gather from Eustathius, that some gram- 
marians derived it from tfSc0 : against these the learned bishop 
urges the objection, that they disregarded the mutes in /carTj- 
TribcovTo and in eiri r rfma (j)dp/jLa/ca *irda-cre. Fol. 566, 40 
(edit, rom.) : ical a-rjiJLetaHrcu, &>9 tyCkovTai rb rJTriov. ov yap 
\eyei, KaOrjTnowvro, d\\d -^tXw? KaryTTibcovTO. Srj\ov o avrb 
real ev rot? ef?}? e/c rov eirl T ^iria <f)dpfj,a/ca Trdaraew rv Se 
76 fJieO 1 'Ofjujpov rtve? eSdcrvvov avrb, etc rov ijBa) irapdyovre?. 
Another explanation, generally adopted by modern lexico- 
graphers, is furnished by the Etymologicum Magnum : IJTTW 
Trporepov eica\e2ro 6 'A<7/cX777rt69* rj diro rwv rpoTrcov, T) 

, /cal r^9 TCOV '^eipwv rJTriorrjTos' a> KOI <yvvai/ca 
e^ ^5 avrq> <yevecrdai, 'Ia<rova, ITava- 
Ae;rtft)v ev uTro/^v^yLtart Au/to^povo?. TJITIO^ crrj/juawei, 
Kvpto)<; TOV \o<yia/jLov. Tiapd TO eVa) TO Xe^yaj, eVto? /cal IJTTIO?, 
b ev \6y<a Trdvra Troi&v, /cal /JLTJ jrddei. e/c //-eTaX?^i|rea)9 B /cal 
b Sid \byov Trpocrrjvr)*; /cal irpao^' /cal i?7Tta)TaT09, o eV Xo7Ot9 
irpabraTos /cal 770-1)^09. Supposing this derivation to be true, 
it would be strange, that while etVetv and e7ro9 show every- 
where an initial digamma in Homer, no trace of it should be 
preserved in 7/77709. On the contrary, the absence of it is 
evident in verses like A. 830 : 

vi vBan Xta/aoJ, eVl 6" iJTria <f)dpfjia/ca irdarae. @. 40 : 
ede\w Se Tot rJ7TtO9 elvai. 

This paper is sent simultaneously, in German, to Kuhn's Zeitschrift. 


If the Greeks had ever felt that any connexion existed 
between rjmo? and eliretv, we should have expected that 
the digamma would have alike remained in both words. 
Nor can I believe, that in such a case, Homer would have 
formed the adjective rjTnoSwpos, or even later poets the epithet 
rjTno^eip, the literal rendering of which would be, "with 
whose hands one can speak." Benfey's derivation (Wurzel- 
lexicon, ii. 356) from the Sanskrit vap (to cut, to shave), may 
be conformable with the ideas of India, where lovers scratch 
and bite each other, but it has not been handed down to us 
that the Greeks manifested their affection in a similar manner. 
Ebel, in Kuhn's Zeitschrift, iv. 447, not less arbitrarily com- 
pares rJTTios with the Latin plus. This word is pronounced 
piho in Umbrian and Volscan, and still more fully, piihio, in 
the Oscan, which latter is very far from iJTrio*;*. 

Homer employs TJ-TUO? as an attribute of persons, with the 
meaning of kind, affable, complying; and of things, in the 
sense of soothing, congenial, useful. It occurs in the following 
passages, @. 40. X. 184: 

Odpcrei,, TpiToyeveia, <f>L\ov re'tcos' ov vv TI 
TrpoQpovi, fj,v6eo]j,ai,' e'#eXft> Be TOL rJTrto? elvai. 

"I wiU comply with your wishes." ^. 281 : 
rolov <yap tfXeo9 e<r6\ov d e rru)\ea' 

, 6 <T(j)a)'lv fj,a\,a TroXXa/a? vypbv 
eo-cra9 vSart, 

fl. 770 : e/cvpbs 8e irarrjp a>? ^Trto? aleL 

O. 775 : ov yap Tt? JJLOL er aXXo? evl Tpoirj evpelrj 

iJTrio? ovSe (^Xo?, 7raj/re9 Be pe TTzfypiK 
ft. 47 : TraTep e<70\bv aTrooXefra, 09 TTOT' ev 

TOia-Sea-aw /BaalXeve, irarrjp 8 1 a>9 ^?rto 
13. 230, 234 = 6.8, 12: 

fjitj Tt9 eri Trpo(f)pa)v dyavbs Kal rpirios 
(3a(n\ev<s 9 ^Be cfrpecrlv 

* Freund (Lex. s. v.), and Mommsen (Unt. Dial. p. 287) say that Cicero 
wrote piius instead of plus. Both copied this false statement out of For- 
cellini, without taking the trouble to verify what really stands in Quin- 
tilian, who only mentions aiio and Maiia. 


dXX alel %a\e7r6$ r etrj ical alcrvka 
0)9 ouTt9 fjue/jLvijTcu 'OSixra-?^ Qeioio 
\a&v, ol(rw avaaa-e, Trarrjp 8' o>9 777^09 rjev. 
K. 337 : & Kip/cr), 7T&)9 yap /u,e /eeXeat trot tfiriov elvaf 
" To comply with your wishes." X. 441 : 

TO) vvv fJHJtrore /cal av ryvvcu/ci Trep r)irio<s elvcu. 
V. 314 : TOVTO S' eycov eu oltf, QTL poi, irdpo<s 


f . 139 : ou jap er aXXov 

^Trtov o>Se dya/cra /cicro^a^ OTTTTOV 

ouS' et /cev Trar/309 al 


o. 152 : ?7 7ap e/tto^e irarr^p 0)9 rJ 

o. 490 : eVet avSpo9 Scb/jbar a^ltceo TroXXa fjLoyij<ra$ 

fjiriov, 69 ST; rot Trape^et @pw<rfo re Troatv re. 

A. 218 : aura/3 eVet l'8ev eX09j o^' eyu,7re<76 Tnicpos otcrro9, 


A. 515 : IrjTpbs yap dvrjp TroXXwv aVraf^O9 aXXo>v 

[tou9 r' e/crdjjiveiVj eVt r' ^Trta <f>dpfjLa/ca Trdcr<rei,v.~] 

A. 830 : /J,7)pov S' e/cra/jC oia-rov, air avrov S* al/u,a 

vt J* vSart Xtapo), eVt 8' ^Trta ^dppa/ca irda-are. 

A. 361 : otSa 7a/o W9 rot OvfLos evl a-rrjBea-o-i, <f>l\oi<ri,v 
fyiria Sijvea olSe. 

Compare Hesiod, Th. 236 : 

avrdp (Nrjpea) /ca\eov<ri, yepoyra 

ovveica vrj/Aeprtf*; re /cal 777^095 ouSe 0e/j,i,<rrecov 

\ij6erai, dXXa St/caia /cal rJTria Sijvea olSev. 
v. 327 : 

T^Xe/^d^o) Se /ce /tv^ov 670) /cat ^rept, <f>altjv 

rfrfiov, ei cr^col'v KpaBlrj aSot 

n. 73 : *ra%a Ac 

7r\r)crei,av veicvoov, ei /JLOI /cpelcw ' 


" Had been kind towards me/' v. 405. o. 39 : 
auro5 Be Trpwria-ra <rv/3a)Tr)v ela- 
05 rot, vwv eVioupo5 5 O/JLWS Se rot 
vralSd re <rov 

o. 557 : 

e<r0Xo5 ecbv evlavev, dvd/creo-w ^ 
Z. 251 : eV#a ot rjirioSmpos zvavrvr) r/Xu#e 

The post-Homeric writers do not differ in the application 
of 777^05, except that they use it more freely as an epithet of 
things. Thus Hesiod, Op. 787 : 

aXX' eptyovs rdftvew /ecu 7ra>ea 
CTTJKOV r afjifyiftakelv rroipvri'iov v\rciov 

"A day suitable for." Soph. Phil. 691 : 

05 rav Oep/jiordrav ai/j,a$a /dfjiao/juevav e\Kecov 

Karevvdaeiev, etc. 

seems to me to be derived from a verb, just as 
is from d^co, dpiaos from dp/ceco, d<T7rd<rio<> from dcrir ' 
/cXo?rt05 from K\e7rra) } //,etXi^i05 from fj,et,\ta'a'c0 ) &<j)d<yi,os from 
a-(f>d^a). As #7*05 agrees in every point with the Sanskrit 
yajya (sacrificio colendus), so does ^7r^o5 correspond with the 
Sanskrit dpya, of which I shall treat presently. The root of 
both words is dp, to obtain, to acquire, which in Sanskrit 
appears in this form, but in Latin as ap. "H7rto5 might be 
explained as obtainable, accessible, easy to be got at, from 
which the meaning of kind would develope itself, just as in 
ei;7rp6<7oSo5 ; but I prefer to take another way. The original 
meaning of apiscor is not I get, but / tie for myself. In the 
primaeval state of civilization, when cattle* formed the only 
property, a man acquired it by tying up under his own roof 
cows and horses which he had either found in a wild state, or 
taken in incursions into the enemy^s territory. This meaning 
of apiscor rests upon the following facts. In the first instance 
we have aptus, which very commonly signifies joined, connected 
with (apta et connexa, apta et coh&rentia, Cic.), and the verb 
apere, to tie, is recorded by Festus and Servius. Festus apud 
* Pecus itself means alligatum. 


Paul. Diac. p. 16 : Apex, qui est sacerdotum insigne, dictus est 
ab eo, quod comprehendere antiqui vinculo apere dicebant. Ser- 
vius ad Virg. TEn. x. 270 : Apere veteres ritu flamiimm alligare 
dicebant, unde apicem dictum volunt*. Taking this meaning 
of the root ap as my basis, I believe that ^Trto? signified ori- 
ginally connected, connected by the ties of kindred or society, 
and that its usual meaning sprang from that source. I may 
remind my hearers, that the English kind owes its meaning to 
a similar process. 

This etymology is supported by two words which occur 
frequently in the Vaidic Sanskrit : d'pya, kindred and akin, 
and dpi, akin. I give a few instances. 

1. d'pya, kindred, relationship. 

Rv. i. 105, 13 : Agne tava tyad ukthyam deveshv asty apyam. 
"O Agni, thy relationship to the gods is worthy of being 

viii. 10, 3 : Yayor asti pra nah sakhyam deveshv adhy apyam. 
" Whose friendship to us, whose relationship to the gods, is 

viii. 27, 10: Astihivahsajatyamri9adaso devaso asty apyam. 
" O gods, destroyers of our enemies, you sprang from the same 
parents and family." 

2. d'pya, a relation. 

Rv. vii. 15, 1 : Upasadyaya milhusha asye juhuta havis, 

Yo no nedishtham apyam. 

" Pour the ghee into the mouth of the revered liberal Agni, 
who is our nearest relation." 

vii. 32, 19 : Nahi tvad anyan maghavan na apyam vasyo 
asti pita cana. 

" For no other relation, not even our father, is more liberal 
to us, than thou, O Indra." 

viii. 86, 7 : Ma na indra para vrinag,bhava nah sadhamadyah. 
Tvam na uti, tvam in na apyam, ma na indra para vrinak. 

* Compare also Paulus Diac. exc. : ape apud antiques dicebatur pro- 
hibe, compesce. 


" Do not repel us, O Indra, but partake of our rejoicings ; 
thou art our help and friend : O Indra, do not repel us." 

3. dpi, akin, related. 

iv. 25, 6 : Nasushver apir na sakha na jamir dushpravyo 
Vahanted avacas. 

" Indra is neither a relation, nor friend, nor brother, to a 
man who does not sacrifice to him ; he hears not, but destroys, 
a man who does not praise him." 

iv. 41, 2 : Indra ha yo varuna cakra api devau martah 

sakhyaya prayasvan, 

, Sa hanti vritra samitheshu yatrun. 

"The mortal who makes Indra and Varuna his friends by 
offering oblations, destroys in the battle all enemies." 

vi. 45, 17 : Yo grinatam id asitha apir uti 9ivah sakha 

Sa tvam na indra mrilaya. 

" O Indra, who provest thyself a near relation and true friend 
to all who praise thee, prosper us." 

If we except the neuter gender, which is peculiar to the 
Sanskrit, as for instance also in mitra, friend, vritra, enemy, it 
is clear that the above-mentioned dpya, a relation, agrees in 
every respect with TJ 


[Read May the 9^.] 

The terminations of the Welsh comparative and superlative 
ending in ach and af agree with the same in the Armorican, 
formed by och and , for which latter the ancient language 
shows af. We have, for instance, in Welsh : 

ffwenn, white, gwennach, whiter, gwennaf, whitest. 
In Armorican gwenn, gwennoch, gwenna(f.} 


These terminations have been rightly compared with the 
Latin ior (ios), and imus in such forms as minimus, infimus. 

The Welsh has, besides, a peculiar degree of comparison, 
which, according to the native grammarians, expresses either 
equality or admiration, and is translated in English by as, so, 
or how with the positive. This degree is formed from the 
positive by affixing the syllable ed, observing the same rules 
as in the derivation of the comparative and superlative. We 
have therefore, for instance : 


glan, pure .... glanach glanaf glaned. 

hardd, handsome harddach .... harddaf hardded. 

main, thin .... meinach meinaf meined. * , 

crwn, round .... crynach crynaf cryned. 

tlawd, poor tlotach tlotaf tloted. 

gwlyb, wet .... gwlypach gwlypaf gwlyped. 

rhad, cheap .... rhatach rhataf. rhated. 

teg, fine tecach tecaf teced. 

In construction, the particles cyn or can (as, so), and in 
South Wales mor (as, so) are frequently, but not necessarily, 
placed before it : daed, or cyn (can) ddaed, or mor ddaed, as 
good. A few examples, extracted from the grammars of 
Owen Pughe and Rowland, will serve to illustrate the appli- 
cation of this form. 


" Cued ganddo ei bleser, fel na ddaw." His pleasure is so 
dear to him, that he will not come. " Y mae cyn ddoethed, 
fel y gwyr y cwbl." He is so wise that he knows the whole. 
" Dos ymaith (cyn) gynted ag y gelli." Go away as soon as 
you can. "Rhedodd cyn gyflymed, fel na allodd ei ddal." 
He ran so fast that he could not be stopped. " Y mae Arthur 
cyn hardded a Dafydd." Arthur is as handsome as David. 
" Am dy laned 
Bardd tuchaned 
A griddfaned 

Gwrdd ofynion. W. Lleyn. 

For thou art so beauteous, let a bard murmur, and let him loudly 
sigh his ardent wishes. 


Fy march melyngan 
Cyfred a gwylan. Taliesin. 
My steed of yellow- white, as swift as a sea-mew. 

Drwg yw yn dryced an buchedd. G. ab Gwrgeneu. 
Evil it is to us that so evil our life. 
Arien deced, 
Eirian drefred, 
Arwydd codded, 
Am dy giried, 

Er dy garu. Gro. Dhu. 

As the hoar so fair, of splendid state, the token of affliction, for 
thy pleasure, on account of loving thee." 


" Wyned yw'r eira ! Cyn wyned yVr eira ! " How white is 
the snow ! " Duw anwyl, fyred einioes ! " My beloved Lord, 
how short is life ! 

" Merch brenin dwyrain a ddaeth i Frefi, 

Wrth glywed daed tynged Dewi. G. Brycheiniawg. 
A daughter of the king of the east did come to Brevi, by hearing 
how good the destiny of Dewi. 

Goddefwn, gwylwn gwaeled arnan 

Gwyth gyman. LI. P. Moch. 

Let us be patient, let us bewail how wretched upon us the contact 
of wrath. 


Dielwed fydd dyn y dydd y ganer. G. ab yr Ynad Coch. 
Let us consider how helpless is man the day he is born." 

The second category appears to me quite identical with the 
first, and only qualified by the rhetorical accent which accom- 
panies the sentence. The idea conveyed by saying " So short 
is life ! " or, " Life is so short ! " is much the same with 
" How short is life \" There is no reason to assume that the 
affix ed has in one case a demonstrative, in the other a 
relative or interrogative meaning. 

The ancient Welsh preserves et instead of ed as the termi- 
nation of the equal. I copy a passage in Zeuss's Celtic Gram- 


mar, i. 307 : " Notanda est post enumeratas terminationes 
gradationis Cambrica terminatio sequalitatis ET, addita adjec- 
tivis, quibus prsefigitur compositione part, ky, kyn, subditurque 
prsep. a } ac (cum) : niuer kyhardet a hwnnw (congregatio 
seque splendens cum hac), Mab. i. 16: gwas kynuonhe- 
dicket athi (puer seque nobilis ac tu), i. 264: achyntristet 
oedynt ac agheu (et seque tristes erunt cum morte), i. 36." 

A grammatical form so commonly used in one branch of 
the Celtic dialects, though not found in the others, cannot be 
without its parallel in the wider range of the Indo-European 
languages. I believe the Welsh et corresponds to the San- 
skrit vat, with the loss of the initial v, in the same manner as 
in oen, pi. wyn (lamb), compared with Lat. ovis, Sanskrit avi, 
ci pi. cwn (dog), KIKOV, Sanskrit cvan; hun (sleep), Sanskrit 
svapna. The Sanskrit vat* very commonly forms adverbs, 
expressing a similarity or likeness, as well from adjectives as 
substantives ; I give a few instances : " Sa 9rigala atmanam 
mritavat sandar9ya sthitas." The jackal pretended to be dead ; 
literally, showing himself as (if) dead. Bigveda, i. 124, 9. 

"Tah pratnavan navyasir nunam asme revad uchantu 
sudina ushasas." As in old times may the brilliant dawn 
appear again today with her glorious light. Rv. ii. 17, 1. 

" Tad asmai navyam Angirasvad arcata." Sing to him this 
new song, as Angiras did before you ; literally, like Angiras. 
Ev. 1. 31, 17. 

" Manushvad agne, Angirasvad angiras, Yaydtivat sadane 
pwrvavac chuce. 

Acha yahy, a vaha daivyam janam." Brilliant Agni, as 
thou earnest to Manus, as to Angiras, to Yayati, to our ances- 
tors, come to the place of sacrifice, and bring with thee the 

These passages, the number of which could be greatly 
increased, may suffice to show, that the two affixes et and 
vat, though not entirely agreeing in their application, still 
bear a certain likeness which proves them to be of the same 

* This suffix always throws the accent on to the last syllable. 



[Read May the 23rd.] 

Professor Key, in his Paper on the preposition avd, in the 
Society's Transactions for 1855 (p. 9), and again in his Paper 
on eV4 in the same volume (p. 93), tries to prove the affinity 
of ad and avd, Bva) and vev&>, by the analogy of similar 
consonantal changes in Welsh. He says : " In Welsh the 
interchange becomes in some cases a law of the language, so 
that an initial d is sure under certain circumstances to take 
the form of an n. Thus, though dant means ' tooth/ and 
dysgu 'learning/ yet for ' seven teeth/ f my learning/ the 
phrases are saith nant, fy nysgu." For my part I know no 
instance where d passes into n otherwise than by assimilation. 
We find indeed that the Old-Italian dialects change d into n, 
but only after a preceding n. The Umbrian substitutes 
regularly nn for nd in the middle of words, and writes for 
instance pihaner for piandi, pane for quande* ; the Oscan has 
upsannam for operandam, and Plautus, by birth an Umbrian, 
says in the well-known line of the Miles Gloriosus, 

" dispennite hominem divorsum et distennite," 

using dispennite and distennite for dispendite and distendite. 
Thus, for "to grunt," one finds grunnire as frequently as 
grundire. In these cases the cause by which the change is 
produced is clear, though the mode of assimilation differs 
from the usual one. 

As to the Welsh change of an initial mute letter into a 
nasal, we have to observe that it takes place only in certain 
combinations. Dant (tooth) can never become nant when it 
stands alone, but it may perhaps be allowable to say saith nant 
(seven teeth), though saith dant is now alone usual. But C G, 
PB, TD, are respectively changed into NGH NG, MH M, 

* The Umbrian, like the oldest Latin, does not express a double con- 
sonant in writing (compare Aufrecht and Kirchhoff, Umbrische Sprach- 
denkmdkr, i. pp. 70, 87). 

E 2 


NH N, if preceded by the possessive pronoun fy (my), the 
preposition yn (in), and certain numerals. These numerals are 
pump or pum (five), saith (seven), wyth (eight), naw (nine), deg 
(ten), ugain (twenty) and its compounds, can (a hundred). It 
would be a grammatical blunder to say chwech niwrnod (six 
days), or pedwar mwystfil (four animals) instead of chwech 
diwrnod and pedwar bwystfil. Only three words undergo 
usually a change after these, namely, blwydd or blynedd 
(year), and diwrnod (day). The simple reason why the 
above-mentioned numerals only, and no others, have this 
influence, is, because they alone ended originally with an n. 


pump pancan 

saith saptan sibun septyni. 

wyth ashtan asztuni. 

nau navan. niun dewyni. 

deg da?an taihun 

On the other hand we have : 

dau dvi tvai du. 

tri tri thri trys. 

pedwar catvar fidvor keturi. 

chwech .... shash saihs szeszi. 

Can, a corruption of cant, which still exists and agrees with 
the Irish cet, has exceeded these limits, and produces the 
change in consequence of its present final n, while un (one) is 
prevented from exercising a similar influence because it ori- 
ginally terminated with a vowel (uno). The preposition yn 
agrees with eV, Latin in, Gothic in, Oscan and Umbrian 
en, Lithuanian in, and belongs to the same category as 
pump, &c. 

In composition, a corresponding nasal must be substituted 
for a mute, if a word is preceded by the negative particle an, 
which corresponds to the Greek av, Umbrian and Oscan an, 
Sanskrit an, Gothic un, Latin in. Thus we have anghadarn 
(powerless) for an -f cadarn, anmhech (sinless) for an +pech, 
anneffro (not awake) for an + deffro. The same takes place 


after another particle, cy, %uv, cum-, we find, therefore, 
cynghas (mutual hate) from cy + cas, cyngofal (mutual care) 
from cy -\-gofal, cymhorth (mutual aid) from cy-porth, cy- 
mrawdd (discourse) from cy-brawdd, cynhebygu (to compare) 
from cy + tebygu, cynefod (custom) from cy + defod. 

For all these cases it is evident, that the n coming in con- 
tact with the following mute, had the power to assimilate it, 
though in course of time the cause might disappear and the 
effect alone remain. An ordinary Welshman in saying deg 
mlynedd is as little able to account for the transmutation of 
the b into m, as any unschooled man in England to explain the 
transition of the ou in mouse into the i of mice. Certain gram- 
matical processes are conventionally continued for centuries, 
when the power that first put them in operation has long 
vanished, and they appear then to the untutored eye as arbi- 
trary, or are falsely attributed to euphony. 

Having shown that nasalization took place only where a 
preceding word ended with an n, we are naturally led to suppose 
that fy (my) also must have been originally fy n, though this 
form is no longer to be discovered even in the oldest literary 
monuments of the Kelt. But we must recollect, that our my, 
thy, are a similar corruption of the Anglo-Saxon min, \m, 
and that the Gothic mein, as well as the Lithuanian mdnas, 
have an n in the possessive pronoun. 

This may suffice to show that the transition in Welsh of 
mutes into nasals is based on the same principle as that by 
which the Latin distendite is changed into distennite, and 
cannot be employed as an analogy for totally different cases. 
As long as it remains unproved, that d standing by itself, and 
not in contact with other consonants, can pass into n, the 
comparison of avd and ad, Sva> and vevco must be considered 
as problematical. This proof would be given, if the Lithuanian 
dewyni (nine) and debesis (heaven) were really simple trans- 
mutations of the Sanskrit navan and nabhas, Latin novem and 
nebula -, but I need not dwell on these words, as the true 
explanation of them has been already advanced by Professor 
Ahrens in the Rheinisches Museum, 1843, pp. 169, 170, where 
he shows that the oldest forms of navan and nabhas were 


dnavan and dnabhas (Greek i/6<o?, loBv<f>^) , of which some 
languages preserved the nasal, the Lithuanian the lingual 

[Read June the 13^.] 

Dr. Ebel proposes, in Kuhn's Zeitschrift, iv. 320, a new and 
very ingenious derivation of actutum. This adverb occurs 
frequently in the older Latin literature, especially in Plautus, 
very rarely in the classical period, and has the meaning of 
" quickly, shortly, instantly." Comparing it with the German 
augenblicklich, to which the English phrase " in the twinkling 
of an eye/' the French en un din d'ceil, St. Paul's eV pnrfj 
o<$6a\ij,ov, and the Sanskrit nimesha might have been 
added, Dr. Ebel thinks, that actutum is a compound of ac 
and tutum. He takes the latter word as the past participle 
of tueri, and recognizes in the former the shorter and older 
form of oculus. That such a shorter form has really ex- 
isted, is proved by the Greek coi/r, which appears with a 
short vowel in compounds like aWoty, olvoty. Again, if the 
Sanskrit akshi (eye) can be compared with these words, I 
would suggest that even this word exists in a monosyllabic 
form in the Vaidic an-aksh, eyeless, blind. Nor does the a 
in the supposed ac, as compared with the o in oculus, con- 
stitute a real difficulty, for the a appears in the Lithuanian 
akis and the old Prussian ackis, and there is no want of 
instances in which an original a coexisted with an e, i, or o. 
Thus we find gressus gradior, fessus fatiscor, ferctum farctum, 
pignus pang ere, ovis avilla, foveo favilla, fovea favissa. So 
far, therefore, we must allow that the proposed derivation, if 
not true, claims the right of being possible. 

But is there really any necessity to go beyond the actual 
state of the Latin, and to give up the usual explanation of 


actutum*? I think not. Scholars may differ as to the mode 
of derivation, but I doubt whether many will be inclined to 
separate actutum from actus. Passing over those authors 
who in full earnest explained our adverb as a compound of 
actu and turn, I quote a more reasonable explanation proposed 
by Lindemann (De Adverbio Latino Specimen iv., Zittaviae, 
1827, p. 17) : "Actutum quid sit, nondum recte explicatum 
legi, descendere videtur ab antique verbo actuere, quod eodem 
modo ab subst. actus efformatum fuit, ut statuere a statu. 
Sit igitur actuere in actu ponere, quemadmodum statuere 
statui reponere, statum alicui rei dare. Unde participium 
actutus in actu positus, ad actum emotus, exercitus. Ergo 
actutum significant cum actu multo, non segniter, celeriter, 
thatig, rasch, actutum redi, kehre rasch zuriick, kehre eilig 
zuriick." But are we to suppose also verbs like astuere, 
cornuere, nasuere, in order to explain astutus, cornutus, 
nasutus ? 

Actus signifies not only action, act, acting, but occurs also 
sometimes in the sense of motion, movement, activity. Lucan 

Pilaque contorsit violent! spiritus actu. 

Fertur in abruptum magno mons inprobus actu. 

Pocula quae facili vilis rota finxerat actu. 
Lucretius, iii. 186 

At quod mobile tantopere est, constare rutundis 
Perquam seminibus debet perquamque minutis, 
Momine uti parvo possint impulsa moveri. 
Namque movetur aqua, et tantillo momine flutat, 
Quippe volubilibus parvisque creata figuris. 
At contra mellis constantior est natura, 
Et pigri latices magis, et cunctantior actus. 
Two derivatives of actus show the same meaning, actuarius 
in actuarium navigium, a fast-sailing ship (compare celox), and 
actuosus (but this only metaphorically). Seneca says, " Noster 
animus in motu est, eo mobilior et actuosior, quo vehementior 
fuerit," and Cicero de Oratore, iii. 26, which passage must be 


read with the context, " quam leniter, quam remisse, quam 
non actuose." That agere itself implied sometimes a rapid 
motion, is shown by age, agite, " be on the move/' and agilis*. 
This is the point from which we must proceed, in order to 
explain actutum. The Romans formed from actus an adjective 
actutus, meaning "endowed with movement, being on the 
move, full of activity," so that for instance " ite actutum in 
frundiferos locos" would be translated literally "go in-a- 
state- of-lively-activity into leafy places." The neuter alone 
is now preserved, the adjective being lost in the same way as 
is the case with temere. The transition from the notion of 
activity into that of speed is simple and common enough. 
We find an analogy in "quick quickly, alive lively," life 
presupposing a superior degree of activity. 

I add a few words on the formation of actutum. It agrees 
entirely with cinctutus and versutus, which are derived from the 
substantives cinctus and versus, the formation from the latter 
having taken place at a time when it still had its original 
meaning of "turning." The same affix appears in astutus, 
cornutus, nasutus, verutus, from astu, cornu, nasus (us-f), veru. 
In all these forms I consider the utus as a contraction of 
u-itus, and compare them with the two adjectives fortu-ltus 
and gratu-itus. The two vowels u + i, that is, the u of the 
base and the i of the affix, coalesce into u just as in manus for 
manu-is, equitatu (dat.) for equitatu-i. The same affix appears 
in auritus, crinitus, Ignitus, pellitus, turritus, mellitus, for 
auri-itus, crini-itus and so on, and has the meaning of " pos- 
sessed of, endowed with." That this itus stands in a near 
connexion with the tus (itus) of the past participle need 
hardly be stated. 

* The best translation of agilis in German would be " ruhrig." 
t This form has not yet come to light, but must be inferred from nasutus. 
From nasus, nasi we should have nasitus, just as galentus comes from 
galerum, and amtus from avus. If this supposition be true, we should have 
for * nose ' five different forms in Latin, nasus, -i, and nasum, naris, nasus, 
-us, and lastly a monosyllabic form nas, seen in nasturtium. Compare Varro 
apud Nonium, p. 12, " nasturcium nonne vides ab eo dici, quod nasum 
torqueat, vestispicam, quod vestem speciat?" And Virgil, Moretum, 84, 
" quaeque trahunt acri vultus nasturcia morsu." 



[Read May the 9M.] 

The present paper is a supplement to two well-known con- 
tributions to American philology by the late A. Gallatin. The 
first was published in the second volume of the Archseologia 
Americana, and gives a systematic view of the languages 
spoken within the then boundaries of the United States ; these 
being the Eiver Sabine and the Rocky Mountains, Texas being 
then Mexican, and, a fortiori, New Mexico and California; 
Oregon, also, being common property between the Americans 
and ourselves. The second is a commentary, in the second 
volume of the Transactions of the American Ethnological 
Society, upon the multifarious mass of philological data col- 
lected by Mr. Hale, during the United States Exploring Expe- 
dition, to which he acted as official and professional philologue ; 
only, however, so far as they applied to the American parts of 
Oregon. The groups of this latter paper the paper of the 
Transactions as opposed to that of the Archaeologia so far 
as they are separate from those of the former, are 

1. The Kitunaha 

2. The Tsihaili-Selish. 

3. The Sahaptin. 

4. The Waiilatpu. 

5. TheTsinuk or Chinook. 

6. The Kalapuya. 

7. The Jakon. 

8. The Lutuami. 

9. The Shasti. 

10. The Palaik. 

11. The Shoshoni or Snake 


To which add the Arrapaho, a language of Kansas, concerning 
which information had been obtained since 1828, the date of 
the first paper. Of course, some of these families extended 
beyond the frontiers of the United States, so that any notice 
of them as American carried with it so much information 
respecting them to the investigators of the philology of the 
Canadas, the Hudson's Bay Territory, or Mexico. 

Again three languages, the Eskimo, and Kenai, and Ta- 
kulli, though not spoken within the limits of the United 
States, were illustrated. Hence, upon more than one of the 



groups of the papers in question there still remains something 
to be said; however much the special and proper subject of 
the present dissertation may be the languages that lay beyond 
the pale of Gallatin's researches. 

The first groups of tongues thus noticed for the second 
time are 


II. THE Sioux. I have little to say respecting these families 
except that they appear to belong to some higher class, a class 
which, without being raised to any inordinate value, may even- 
tually include not only these two now distinct families, but also 
the Catawba, Woccoon, Cherokee, Choctah, and (perhaps) Caddo 
groups, perhaps also the Pawni and its ally the Riccaree. 

III. THE ALGONKIN GROUP. The present form of this 
group differs from that which appears in the Archseologia Ame- 
ricana, by exhibiting larger dimensions. Nothing that was 
then placed within has since been subtracted from it ; indeed, 
subtractions from any class of Gallatur's making are well-nigh 
impossible. In respect to additions, the case stands differently. 

Additions of no slight importance have been made to the 
Algonkin group. The earliest was that of 

The Bethuck. The Bethuck is the native language of New- 
foundland. In 1846, the collation of a Bethuck vocabulary 
enabled me to state that the language of the extinct, or 
doubtfully extant, aborigines of that island was akin to those 
of the ordinary American Indians rather than to the Eskimo ; 
further investigation showing that, of the ordinary American 
languages, it was Algonkin rather than aught else. 

A sample of the evidence of this is to be found in the fol- 
lowing table ; a table formed, not upon the collation of the 
whole MS., but only upon the more important words con- 
tained in it. 

English, son. 
Bethuck, mageraguis. 
Cree, equssis. 

Oiibbeway, ninqwisis 1 

J J y . *> =mvson. 
, negwis. j 

Ottawa, kwis. 

Micmac, unquece. 
Passamaquoddy, n'kos. 
Narragansetts, nummuckiese 

my son. 

Delaware, quissau = his son. 
Miami, akwissima. 



Miami, ungwissah. 
Shawnoe, koisso. 
Sack & Fox, nekwessa. 
Menomeni, nekeesh. 

English, girl. 
Bethuck, woaseesh. 
Cree, squaisis. 
Ojibbeway, ekwaizais. 
Ottawa, aquesens. 
Old Algonkin, ickwessen. 
Sheshatapoosh, squashish. 
Passamaquoddy, pelsquasis. 
Narragansetts, squasese. 
Montaug, squasses. 
Sack & Fox, skwessah. 
Cree, awdsis = child. 
Sheshatapoosh, awash = child. 

English, mouth. 
Bethuck, mamadthun. 
Nanticoke, mettoon. 
Massachusetts, muttoon. 
Narragansetts, wuttoon. 
Penobscot, madoon. 
Acadcan, met on. 
Micmac, toon. 
Abenaki, ootoon. 

English, nose. 
Bethuck, gheen. 
Miami, keouane. 

English, teeth. 
Bethuck, bocbodza. 
Micmac, neebeet. 
Abenaki, neebeet. 

English, hand. 
Bethuck, maemed. 
Micmac, paeteen. 
Abenaki, mpateen. 

English, ear. 
Bethuck, mootchiman. 
Micmac, mootooween. 
Abenaki, nootawee. 

English, smoke. 
Bethuck, bassdik. 
Abenaki, ettoodake. 

English, oil. 
Bethuck, emet. 
Micmac, memaye. 
Abenaki, pemmee. 

English, sun. 
Bethuck, keuse. 
Cree, &c., kisis. 
Abenaki, kesus. 
Mohican, kesogh. 
Delaware, gishukh. 
Illinois, kisipol. 
Shawnoe, kesathwa. 
Sack & Fox, kejessoah. 
Menomeni, kaysho. 
Passamaquoddy, kisos = moon. 
Abenaki, kisus = moon. 
Illinois, Arm's = moon. 
Cree, kesecow = day. 
Ojibbeway, kijik= day and light. 
Ottawa, kijik = ditto. 
Abenaki, kiseoukou = ditto. 
Delaware, gieshku = ditto. 
Illinois, kisik = ditto. 
Shawnoe, keeshqua = ditto. 
Sack & Fox, keeshekeh = ditto. 

English, fire. 

Bethuck, boobeeshawt. 

Cree, esquitti, scoutay. 

Ojibbeway, ishkodai, skootae. 

Ottawa, ashkote. 

Old Algonkin, skootay. 



Sheshatapoosh, schootay. 
Passamaquoddy, skeet. 
Abenaki, skoutai. 
Massachusetts, squitta. 
Narragansetts, squtta. 

English) white. 
Bethuck, wobee. 
Cree, wabisca. 

, wapishkawo. 

Ojibbeway, wawbishkaw. 

, wawbizze. 

Old Algonkin, wabi. 
Sheshatapoosh, wahpou. 
Micmac, ouabeg, wabeck. 
Mountaineer, wapsiou. 
Passamaquoddy, wapiyo. 
Abenaki, wanbighenour. 

, wanbegan. 

Massachusetts, wompi. 
Narragansetts, wompesu. 
Mohican, waupaaeek. 
Montaug, wampayo. 
Delaware, wape, wapsu, wapsit, 
Nanticoke, wauppauyu. 
Miami, wapekinggek. 
Shawnoe, opee. 
Sack & Fox, wapeskayah. 
Menomeni, waubish keewah. 

English, black. 
Bethuck, mandzey. 
Ojibbeway, mukkudaiwa. 
Ottawa r maekateh. 
Narragansetts, mowesu. 
Massachusetts, mooi. 

English, house. 
Bethuck, meeootick. 
Narragansetts, wetu. 

English, shoe. 
Bethuck, mosen. 
Abenaki, mkessen. 

English, snow. 
Bethuck, kaasussabook. 
Cree, sasagun=.}\&i\. 
Ojibbeway, saisaigan. 
Sheshatapoosh, shashaygan. 

English, speak. 
Bethuck, ieroothack. 
Taculli, yaltuck. 
Cree, athemetakcouse. 
Wyandot, atakea. 

English, yes. 
Bethuck, yeathun. 
Cree, ahhah. 
Passamaquoddy, netek. 

English, no. 
Bethuck, newin. 
Cree, namaw. 
Ojibbeway, kawine. 
Ottawa, kauween. 

English, hatchet. 
Bethuck, dthoonanyen. 
Taculli, thynle. 

English, knife. 
Bethuck, eewaeen. 
Micmac, uagan. 

English, bad.. 
Bethuck, muddy. 
Cree, my a t on. 
Ojibbeway, monadud. 

, mudji. 

Ottawa, matche. 


Montaug, muttadeeaco. 
Delaware, makhtitsu. 
Nanticoke, mattik. 
Sack & Fox, motchie. 

-, matchathie. 

Micmac, matoualkr. 
Massachusetts, matche. 
Narragansetts, matchit. 
Mohican, matchit. 
Montaug, mattateayah. 

The Shyenne. A second addition of the Algonkin class was 
that of the Shyenne language a language suspected to be 
Algonkin at the publication of the Archseologia Americana. 
In a treaty made between the United States and the Shyenne 
Indians in 1825, the names of the chiefs who signed were 
Sioux, or significant in the Sioux language. It was not 
unreasonable to consider this as primd-facie evidence of the 
Shyenne tongue itself being Sioux. Nevertheless, there were 
some decided statements in the way of external evidence in 
another direction. There was the special evidence of a gen- 
tleman well-acquainted with the fact, that the names of the 
treaty, so significant in the Sioux language, were only trans- 
lations from the proper Shyenne, there having been no Shy- 
enne interpreter at the drawing-up of the document. What 
then was the true Shyenne ? A vocabulary of Lieut. Abert's 
settled this. The numerals of this were published earlier than 
the other words, and on these the present writer remarked 
that they were Algonkin (Report of the British Association 
for the Advancement of Science, 1847, Transactions of the 
Sections, p. 123). Meanwhile, the full vocabulary, which was 
in the hands of Gallatin, and collated by him, gave the con- 
templated result : " Out of forty-seven Shyenne words for 
which we have equivalents in other languages, there are 
thirteen which are indubitably Algonkin, and twenty-five 
which have affinities more or less remote with some of the 
languages of that family." (Transactions of the American 
Ethnological Society, vol. ii. p. cxi. 1848.) 

The Blackfoot. In the same volume (p. cxiii), and by the 
same author, we find a table showing the Blackfoot to be 
Algonkin; a fact that must now be generally recognized, 
having been confirmed by later data. The probability of this 
affinity was surmised in a paper in the 28th Number of the 
Proceedings of the present Society. 


The Arrapaho. This is the name of a tribe in Kansas ; 
occupant of a district in immediate contact with the Shyenne 

But the Shyennes are no indigence to Kansas. Neither 
are the Arrapahos. The so-called Fall Indians, of whose lan- 
guage we have long had a very short trader's vocabulary in 
Umfreville, are named from their occupancy which is on the 
Falls of the Saskatshewan. The Nehethewa, or Crees, of 
their neighbourhood call them so ; so that it is a Cree term 
of which the English is a translation. Another name (English 
also) is Big-belly, in French Gros ventre. This has given rise 
to some confusion. Gros-ventre is a name also given to the 
Minetari of the Yellow-stone River; whence the name 
Minetari itself has, most improperly, been applied (though 
not, perhaps, very often or by good authorities) to the Fall 

The Minetari Gros-ventres belong to the Sioux family. 
Not so the Gros-ventres of the Falls. Adelung remarked 
that some of their words had an affinity with the Algonkin, 
or as he called it, the Chippeway-Delaware, family, e.g. the 
names for tobacco, arrow, four, and ten. 

Umfreville's vocabulary was too short for anything but the 
most general purposes and the most cautious of suggestions. 
It was, however, for a long time the only one known. The 
next to it, in the order of time, was one in MS., belonging to 
Gallatin, but which was seen by Dr. Prichard and collated 
by the present writer, his remarks upon it being published 
in the 134th Number of the Proceedings of this Society. 
They were simply to the effect that the language had certain 
miscellaneous affinities. An Arrapaho vocabulary in School- 
craft tells us something more than this ; viz. not only that it 
is, decidedly, the same language as the Fall Indian of Umfre- 
ville, but that it has definite and preponderating affinities with 
the Shyenne, and, through it, with the great Algonkin class 
in general. 


scalp mithash metake. 

tongue nathun vetunno. 



tooth veathtah veisike. 

beard vasesanon meatsa. 

hand mahchetun maharts. 

blood bahe mahe. 

sinew anita antikah. 

heart battah estah. 

mouth nettee marthe. 

girl issaha xsa. 

husband nash nah. 

son naah nah. 

daughter nahtahnah nahtch. 

one chassa nuke. 

two neis neguth. 

three nas nahe. 

four yeane nave. 

five yorthun noane. 

six nitahter nahsato. 

seven nisorter nisoto. 

eight nahsorter . nahnoto. 

nine siautah soto. 

ten mahtahtah mahtoto. 


man enanetah .... enainneew, Menom. &c. 

father, my . . nasonnah .... nosaw, Miami. 

mother, my . . nanah nekeah, Menom. 

husband, my . nash nah, Shyenne. 

son, my .... naah nah, Shyenne. 

.... inkvfithah, Shawnee. 

daughter, my . nahtahnah .... netawnah, Miami, 

brother, my . . nasisthsah .... nesawsah, Miami, 

sister, my. . . . naecahtaiah . . nekoshaymank, Menom. 

Indian enenitah ah wainhukai, Delaware. 

eye mishishi maishkayshaik, Menom. 

mouth netti may tone, Menom. 

tongue nathun wilano, Delaware. 

tooth veathtah .... wi pit, Delaware. 

beard vasesanon .... witonahi, Delaware. 

back nerkorhah pawkawmema, Miami. 

hand machetun .... olatshi, Shawnee. 



foot nauthauitah . . ozit, Delaware. 

bone hahunnah .... ohkonne, Menom. 

heart battah may tali, Menom. 

blood bahe mainhki, Menom. 

sinew anita ohtah, Menom. 

flesh wonnunyah . . weensama, Miami. 

skin tahyatch .... xais, Delaware. 

town haitan otainahe, Delaware. 

door tichunwa .... kwawntame, Miami. 

sun nishi-ish .... kayshoh, Menom. 

star ahthah allangwh, Delaware. 

day ishi kishko, Delaware. 

autumn .... tahuni tahkoxko, Delaware. 

wind assissi kaishxing, Delaware. 

fire ishshitta .... ishkotawi, Menom. 

water nutch nape, Miami. 

ice wahhu mainquom, Menom. 

mountain .... ahhi wahchiwi, Shawnee. 

hot hastah ksita, Shawnee. 

he enun enaw, Miami. 

waynanh, Menom. 

that (in) ... hinnah aynaih, Menom. 

who unnahah .... ahwahnay, Menom. 

no chinnani .... kawn, Menom. 

eat mennisi mitishin, Menom. 

drink bannah maynaan, Menom. 

kill nauaiut o*A-nainhnay, Menom. 

Fitzhugh Sound forms in -SKUM. There is still a possible 
addition to the Algonkin group ; though it is probable that it 
cannot be added to it without raising the value of the class. 
The exact value and interpretation of the following fact has 
yet to be made out. I lay it, however, before the reader. 
The language for the parts about Fitzhugh Sound seems to 
belong to a class which will appear in the sequel under the 
name Hailtsa or Haeetsuk. The numerals, however, have 
this peculiarity, viz. they end in the syllable -kum. And 
this is what, in one specimen, at least, two of the Blackfoot 
terms do. 


English, two. || English, three. 

Fitzhugh Sound, mal-skum. Fitzhugh Sound, uta-skum. 

Hailtsuk, maluk. \ Hailtsuk, yutuk. 

Blackfoot, nartoke-skum. \ Blackfoot, nahoke-skum. 

What, however, if this syllable -skum be other than true 
Blackfoot ; i. e. what if the numerals were taken from the 
mouth of a Hailtsa Indian ? The possibility of this must be 
borne in mind. With this remark upon the similarity of 
ending between one specimen of Blackfoot numerals and the 
Hailtsa dialect of Fitzhugh Sound, we may take leave of the 
Algonkin class of tongues and pass on to 

IV. THE ATHABASKAN GROUP. The vast size of the area 
over which the Athabaskan tongues have spread themselves, 
has commanded less attention than it deserves. It should 
command attention if it were only for the fact of its touching 
both the Oceans the Atlantic on the one side, the Pacific 
on the other. But this is not all. With the exception of the 
Eskimo, the Athabaskan forms of speech are the most north- 
ern of the New World ; nay, as the Eskimos are, by 110 means, 
universally recognized as American, the Athabaskan area is, 
in the eyes of many, absolutely and actually the most northern 
portion of America the most northern portion of America 
considered ethnologically or philologically, the Eskimo coun- 
try being considered Asiatic. To say that the Athabaskan 
area extends from ocean to ocean, is to say that, as a matter 
of course, it extends to both sides of the Rocky Mountains. 
It is also to say that the Athabaskan family is common to 
both British and Russian America. 

For the northern Athabaskans, the main body of the family, 
the philological details were, until lately, eminently scanty 
and insufficient. There was, indeed, an imperfect substi- 
tute for them in the statements of several highly trust- 
worthy authors as to certain tribes who spoke a language 
allied to the Chepewyan, and as to others who did not; 
statements which, on the whole, have been shown to be 
correct ; statements, however, which required the confirmation 
of vocabularies. These have now been procured; if not to 
the full extent of all the details of the family, to an extent 



quite sufficient for the purposes of the philologue. They show 
that the most western branch of the stock, the Chepewyan 
proper, or the language of what Dobbs called the Northern 
Indians, is closely akin to that of the Dog-ribs, the Hare 
(or Slave) and the Beaver Indians, and that the Dahodinni, 
called from their warlike habits the Mauvais Monde, are but 
slightly separated from them. Farther west a change takes 
place, but not one of much importance. Interpreters are 
understood with greater difficulty, but still understood. 

The Sikani and Sussi tongues are known by specimens of 
considerable length and value, and these languages, lying as 
far south as the drainage of the Saskatshewan, and as far west 
as the Rocky Mountains, are, and have been for some years, 
known as Athabaskan. 

Then came the Takulli of New Caledonia, of whose lan- 
guage there was an old sample procured by Harmon. This 
was the Nagail, or Chin Indian of Mackenzie, or nearly so. 
Now, Nagail I hold to be the same word as Takull-i, whilst 
Chin is Tshin = Dinne = Tnai = Atna = Knai = Man. The 
Takulli division falls into no less than eleven (?) minor sec- 
tions ; all of which but one end in this root, viz. -tin. 

1. The Tau-/m, or Talko-m. 

(?) 2. The Tsilko-tfm or Chilko-fm/ perhaps the same word 
in a different dialect. 

3. The Nasko-/m. 8. The Natliau-tfm. 

4. The Thetlio-tfw. 9. The Nikozliau-#w. 

5. The Tsatsno-/m. 10. The Tatshiau-m, and 

6. The Nulaau-^w. 11. The Babin Indians. 

7. The Ntaauo-lm. 

Sir John Richardson, from vocabularies procured by him 
during his last expedition, the value of which is greatly en- 
hanced by his ethnological chapter on the characteristics of the 
populations which supplied them, has shown, what was before 
but suspected, that the Loucheux Indians of Mackenzie River 
are Athabaskan ; a most important addition to our knowledge. 
Now, the Loucheux are a tribe known under many names ; 
under that of the Quarrellers, under that of the Squinters, 
under that of the Thycothe and Digothi. Sir John Richard- 


son calls them Kutshin, a name which we shall find in several 
compounds, just as we found the root -tin in the several sec- 
tions of the Takulli, and as we shall find its modified form 
dinni among the eastern Athabascans. The particular tribes of 
the Kutshin division, occupants of either the eastern frontier 
of Russian America, or the north-western parts of the Hudson's 
Bay Territory, are (according to the same authority) as follows : 

1. The Axtez-kutshi = Hard people. 

2. The Tskm-kutshi = Water people. 

3. The T&tzei-kutshi = Rampart people; falling into four 


4. The Teystse-kutshi People of the shelter. 

5. The Vania-kutshi = People of the lakes. 

6. The Neyetse-kutshi = People of the open country. 

7. The Tlagga-silla = Little dogs. 

This brings us to the Kenay. Word for word Kenay is 
Knai = Tnai, a modified form of the now familiar root t-n = 
man, a root which has yet to appear and reappear under 
various new, and sometimes unfamiliar and unexpected, forms. 
A Kenay vocabulary has long been known. It appears in 
Lisianisky tabulated with the Kadiak, Sitkan, and Unalaskan 
of the Aleutian Islands. It was supplied by the occupants 
of Cook's Inlet. Were these Athabaskan? The present 
writer owes to Mr. Isbister the suggestion that they were 
Loucheux, and to the same authority he was indebted for the 
use of a very short Loucheux vocabulary. Having compared 
this with Lisiansky's, he placed both languages in the same 
category rightly in respect to the main point, wrongly in 
respect to a subordinate. He determined the place of the 
Loucheux (Kutshin as he would now call them) by that of the 
Kenay, and made both Kolush. He wjould now reverse the 
process and make both Athabaskan, as Sir John Richardson 
has also suggested. 

To proceed three vocabularies in Baer's Beitrdge are in 
the same category with the Kenay, viz. 

1. The Atna. This is our old friend t-n again, the form 
Tnai and others occurring. It deserves notice, because, 


unless noticed, it may create confusion. As more populations 
than one may call themselves man, a word like Atna may 
appear and re-appear as often, as there is a dialect which so 
renders the Latin word homo. Hence, there may not only be 
more Atnas than one, but there actually are more than one. 
This is a point to which we shall again revert. At present it 
is enough that the Atnas under notice are occupants of the 
mouth of the Copper River, Indians of Russian America and 

2. The Koltshani. As t-n = man, so does k-ltsh = stranger, 
guest, enemy, friend ; and mutatis mutandis, the criticism that 
applied to Atna applies to words like Koltshan, Golzan, and 
Kolush. There may be more than one population so called. 

3. The Ugalents or Ugalyackh-mutsi. This is the name of 
a few families near Mount St. Elias. Now 

The Atna at the mouth of the Copper River, the Koltshani 
higher up the stream, and the Ugalents, are all held by the 
present writer to be Athabaskan not, indeed, so decidedly as 
the Beaver Indians, the Dog-ribs, or the Proper Chepewyans, 
but still Athabaskan. .They are not Eskimo, though they 
have Eskimo affinities. They are not Kolush, though they 
have Kolush affinities. They are by no means isolated, and 
as little are they to be made into a class by themselves. At 
the same time, it should be added that by including these we 
raise the value of the class. 

For all the languages hitherto mentioned we have spe- 
cimens. For some, however, of the populations whose names 
appear in the maps, within the Athabaskan area, we have yet 
to satisfy ourselves with the testimony of writers, or to rely 
on inference. In some cases, too, we have the same popu- 
lation under different names. This is the case when we have 
a native designation .as well as a French or English one 
e. a. Loucheux, Squinters, Kutshin. This, too, is the case 
when we have, besides the native name (or instead of it), the 
name by which a tribe is called by its neighbours. Without 
giving any minute criticism, I will briefly state that all the 
Indians of the Athabaskan area whose names end in -dinni 
are Athabaskan; viz. 


1. The See-issaw-dinni = Rising- sun-mew. 

2. The Tau-tsawot-e?mm = Birch-rind-mew. 

3. The Thlingeha-dwm = Dog-rib-mew. 

4. The Etsh-tawut-fi?mra = Thickwood-mew. 

5. The Ambah-tawut-6/mm = Mountain-sheep-men. 

6. The Tsillaw-awdut-e?mm = Bushwood-mew. 

Lastly Carriers, Slave-Indians, Yellow-knives, Copper- 
Indians, and Strong-bows are synonyms for some of the tribes 
already mentioned. The Hare- Indians are called Kancho. 
The Nehanni and some other populations of less importance 
are also, to almost a certainty, Athabaskan. 

If we compare the Athabaskan with the tongues in its 
neighbourhood, we shall find that it is broadly and definitely 
separated from them in proportion as we move from west to 
east. In Russian America, the Eskimo, Sitkan, and Atha- 
baskan tongues graduate into each other. In the same parts 
the Athabaskan forms of speech differ most from each other. 
On the other hand, to the east of the Rocky Mountains, the 
Dog-rib, the Hare, and the Chepewyan are cut off by lines 
equally trenchant from the Eskimo to the north, and from 
the Algonkin to the south. I infer from this that the 
diffusion of the language over those parts is comparatively 
recent ; in other words, that the Athabaskan family has moved 
from west to east rather than from east to west. 

Of the proper Athabaskan, i. e. of the Athabaskan in the 
original sense of the word, the southern boundary, beginning 
at Fort Churchill, on Hudson's Bay, follows (there or there- 
abouts) the course of the Mississippi ; to the north of which lie 
the Chepewyans who are Athabaskan, to the south of which 
lie the Crees, or Knistenaux, who are Algonkin. Westward 
come the Blackfeet (Algonkin) and the Sussees (Athabaskan), 
the former to the north, the latter to the south, until the 
Rocky Mountains are reached. The Takulli succeed occu- 
pants of New Caledonia ; to the south of whom lie Kutani 
and Atnas. The Takulli area nowhere touches the ocean, 
from which its western frontier is separated to the south of 
55 north latitude by some unplaced languages ; to the north 
of 55, by the Sitkan but only as far as the Rocky Moun- 


tains ; unless, indeed, some faint Algonkin characteristics lead 
future inquirers to extend the Algonkin area westwards, 
which is not improbable. The value of the class, however, if 
this be done, will have to be raised. 

The most southern of the Athabaskans are the Sussees, in 
north latitude 51 there or thereabouts. But the Sussees, 
far south as they lie, are only the most southern of the Atha- 
baskans en masse. There are outliers of the stock as far south 
as the southern parts of Oregon. More than this, there are 
Athabaskans in California, New Mexico, and Sonora. 

Few discoveries respecting the distribution of languages are 
more interesting than one made by Mr. Hale, to the effect 
that the Umkwa, Kwaliokwa, and Tlatskanai dialects of a 
district so far south as the mouth of the River Columbia, 
and the upper portion of the Umkwa river (further south 
still) were outlying members of the Athabaskan stock, a 
stock pre-eminently northern not to say Arctic in its main 

Yet the dialects just named were shown, by a subsequent 
discovery of Professor Turner' s, to be only penultimate ramifi- 
cations of their stock ; inasmuch as further south and further 
south still, in California, New Mexico, Sonora, and even Chi- 
huhua, as far south as 30 north latitude, Athabaskan forms 
of speech were to be found ; the Navaho of Utah and New 
Mexico, the Jecorilla of New Mexico, and the Apatch of 
New Mexico, California, and Sonora, being Athabaskan. The 
Hoopah of California is also Athabaskan. 

The first of the populations to the south of the Athabaskan 
area, who, lying on, or to the west of, the Rocky Mountains, 
are other than Algonkin, are 

V. THE KITUNAHA. The Kitunaha, Cutani, Cootanie or 
Flatbow area is long rather than broad, and it follows the line 
of the Rocky Mountains between 52 and 48 north latitude. 
How definitely it is divided by the main ridge from that of 
the Blackfoots I am unable to say, but as a general rule, the 
Kutani lie west, the Blackfoots east ; the former being Indians 
of New Caledonia and Oregon, the latter of the Hudson's Bay 
Territory and the United States. On the west the Kutani 


country is bounded by that of the Shuhap and Selish Atnas, 
on the north by the Sussee, Sikanni, and Nagail Athabaskans, 
on the south (I think) by some of the Upsaroka or Crow 
tribes. All these relations are remarkable, and so is the geo- 
graphical position of the area. It is in a mountain-range ; and, 
as such, in a district likely to be an ancient occupancy. The 
languages with which the Kutani lies in contact are referable 
to four different families the Athabaskan, the Atna, the 
Algonkin, and the Sioux ; the last two of which, the Black- 
foot (Algonkin) and the Crow (Sioux), are both extreme 
forms, i. e. forms sufficiently unlike the other members of 
these respective groups to have had their true position long 
overlooked; forms, too, sufficiently peculiar to justify the 
philologue in raising them to the rank of separate divisions. 
It suffices, however, for the present to say, that the Kutani 
language is bounded by four tongues differing in respect to 
the class to which they belong and from each other, and 
different from the Kutani itself. 

The Kutani, then, differs notably from the tongues with 
which it is in geographical contact ; though, like all the lan- 
guages of America, it has numerous miscellaneous affinities. 
In respect to its phonesis it agrees with the North Oregon 
languages. The similarity in name to the Loucheux, whom 
Richardson calls Kutshin, deserves notice. Upon the whole, 
few languages deserve attention more than the one under 

VI. THE ATNA GROUP. West of the Kutanis and south of 
the Takulli Athabaskans lie the northernmost members of a 
great family which extends as far south as the Sahaptin 
frontier, the Sahaptin being a family of Southern, or Ame- 
rican, Oregon. Such being the case, the great group now 
under notice came under the cognizance of the two American 
philologues, whose important labours have already been 
noticed, by whom it has been denominated Tsihaili- Selish. 
It contains the Shushwap, Selish, Skitsnish (or Cosur d'Alene) 
Piskwans, Nusdalum, Kawitchen, Skwali, Chechili, Kowelits, 
and Nsietshawus forms of speech. 

In regard to the Atna I have a statement of my own to 


correct, or at any rate to modify. In a paper, read before the 
Ethnological Society, on the Languages of the Oregon Ter- 
ritory (Dec. 11, 1844), I pronounced that an Atna vocabulary 
found in Mackenzie's Travels, though different from the Atna 
of the Copper River, belonged to the same group. The group, 
however, to which the Atna of the Copper River belongs is 
the Athabaskan. 

The Tsihaili-Selish languages reach the sea in the parts to 
the south of the mouth of Frazer's Eiver, i. e. the parts 
opposite Vancouver's Island; perhaps they touch it further 
to the north also ; perhaps, too, some of the Takulli forms of 
the speech further north still reach the sea. The current 
statements, however, are to the effect, that to the south of the 
parts opposite Sitka, and to the north of the parts opposite 
Vancouver's Island, the two families in question are separated 
from the Pacific by a narrow strip of separate languages 
separate and but imperfectly known. These are, beginning 
from the north 

Skittegats, Massetts, Kumshahas, and Kyganie of Queen 
Charlotte's Islands and the Prince of Wales Archipelago. 
Its area lies immediately to that of the south of the so-called 
Kolush languages. 

VIII. THE CHEMMESYAN. Spoken along the sea-coast and 
islands of north latitude 55. 

IX. THE BILLECHULA. Spoken at the mouth of Salmon 
River ; a language to which I have shown, elsewhere, that a 
vocabulary from Mackenzie's Travels of the dialect spoken at 
Friendly Village was referable. 

X. THE HAILTSA. The Hailtsa contains the dialects of the 
sea-coast between Hawkesbury Island and Broughton's Ar- 
chipelago, also those of the northern part of Vancouver's 

In Gallatin, the Chemmesyan, Billechula, and Hailtsa are 
all thrown in a group called Naas. The Billechula numerals 
are, certainly, the same as the Hailtsa ; the remainder of the 
vocabulary being unlike, though not altogether destitute of 
coincidences. The Chemmesyan is more outlying still. I 


do not, however, in thus separating these three languages, 
absolutely deny the validity of the Naas family. I only 
imagine that if it really contain languages so different as the 
Chemmesyan and Hailtsa, it may also contain the Haidah 
and other groups, e.g. the one that comes next, or 

XI. THE WAKASH of Quadra and Vancouver's Island. 

South of the Wakash area come, over and above the 
southern members of the Atna family and the Oregon outliers 
of the Athabaskan, the following groups, of value hitherto 

A. The Tshinuk, or Chinuk; 

B. The Kalapuya; 

C. The Jakon; all agreeing in the harshness of their 
phonesis, and (so doing) contrasted with 

D. The Sahaptin, and 

E. The Shoshoni. 

The Sahaptin is separated by Gallatin from the Waiilatpu 
containing the Cayiis or Molele form of speech. The present 
writer throws them both into the same group. The numerals, 
the words wherein it must be admitted that the two languages 
agree the most closely, are in 


one naks na. 

two lapit lepl-in. 

three mitat mat-nin. 

six oi-lak noi-na. 

seven oi-napt noi-lip. 

eight oi-matat noi-mat. 

The meaning of the oi and noi in these words requires 
investigation. It is not five', the Sahaptin and Cayiis for 
five being pakhat (S.) and tawit (C.). Nor yet is it hand (as 
the word for five often is), the word for hand being epih and 
apah. It ought, however, theoretically to be something of 
the kind, inasmuch as 

Oi-lak and noi-na = ? + 1. 
Oi-napt and noi-lip = ? + 2. 
Oi-matat and woi-mat = ? + 3. 


Of the Shoshoni more will be said in the sequel. At pre- 
sent it is enough to state that the Shoshoni and Sahaptin 
languages are as remarkable for the apparent ease and sim- 
plicity of their phonesis as the Jakon, Kalapuya, and Tshinuk 
are for the opposite qualities. It may also be added that the 
Shoshoni tongues will often be called by the more general 
name of Paduca. 

South of the Caytis, Waiilatpu, and Wihinast, or Western 
Shoshonis, come the languages which are common to Oregon 


For three of these we have vocabularies (Mr. Hale's) : 


There may be other forms of speech common to the two 
countries, but these three are the only ones known to us by 
specimens. The Lutuami, Shasti, and Palaik are thrown by 
Gallatin into three separate classes. They are, without doubt, 
mutually unintelligible. Nevertheless they cannot be very 
widely separated. 

Man = in Lutuami hishu-atsus, in Palaik = yatui. Qu. atsus=yatui. 
Woman = Lutuami tar-itsi, Palaik = umtew-itsen Qu. itsi = itsen. 

In Palaik, Son = yau-itsa, Daughter = lumau-itsa. 
Head = Palaik lah. In Lutuami lak=hair. Qu. mak = head in 

Shasti, makh = hair, Shasti. 
Ear = Lutuami mumoutsh, Palaik ku-mumuats. 
Mouth = au Shasti, ap Palaik. 
Tooth = itsau Shasti, itsi Palaik. 
Sun = tsoare Shasti, tsul Palaik = sun and moon. In Lutuami tsol 

= star. 
Fire = Shasti ima = Palaik mails. The termination -/- common in 

Palaik, ipili = tongue, kelala = shoes, usehela = sky, &c. 
Water = Shasti atsa, Palaik as. 
Snow = Lutuami kais, Shasti kae. 
Earth = Lutuami kaela, Palaik kela, Shasti tarak. This is the second 

time we have had a Shasti r for a Palaik I tsoare = tsul. 
Sear = tokunks Lutuami, lokhoa Palaik. 
Bird= Lutuami lalak, Shasti tararakh. 
1= Lutuami no. Qu. is this the n in n-as = head and n-ap = hand, 

for which latter word the Shasti is ap-ka ? 



one tshiamu umis. 

two hoka kaki. 

Neither are there wanting affinities to the Sahaptin and 
Cayiis languages, allied to each other. Thus 

Ear = mumutsh Lutuami = ku-mumuats Palaik = mutsaui Sahaptin. 

tsak Shasti = taksh Cayiis. 

Mouth = shum Lutuami = shum-kaksh Cayiis = him Sahaptin. 
Tongue =pawus Lutuami =pawish Sahaptin =push Cayiis. 
Tooth = tut Lutuami = tit Sahaptin. 
Foot = akwes Shasti = akhua Sahaptin. 
Blood = ahati Palaik = kiket Sahaptin. 
Fire = loloks Lutuami = ihiksha Sahaptin. 
One = natshik Lutuami = naks Sahaptin = na Cayiis. 
Two = lapit Lutuami = lapit Sahaptin = leptin Cayiis. 

The Lutuami seems somewhat the most Sahaptin of the 
three, and this is what we expect from its geographical position, 
it being conterminous with the Molele (or Cayiis) and the 
allied Waiilatpu. It is also conterminous with the Wihinast 
Shoshoni, or Paduca, as is the Palaik. Both Palaik and Lu- 
tuami (along with the Shasti) have Shoshoni affinities. 


nose moui=iami, Palaik. 

mouth timpa = shum, Lutuami. 

ear inaka=isak, Shasti. 

sun tava=sapas, Lutuami. 

water pa=ampo, Lutuami. 

I ni=no, Lutuami. 

thou i=i> Lutuami. 

he 00= hot, Lutuami. 

one * . . shimutsi=tshiamuu, Shasti ; umis, Palaik. 

The chief language in contact with the Shasti is the in- 
trusive Athabaskan of the Umkwa and Tlatskanai tribes. 
Hence the nearest languages with which it should be com- 
pared are the Jakon and Kalapuya, from which it is geogra- 
phically separated. For this reason we do not expect any 
great amount of coincidence. We find however the following 


head tkhlokia=lah, Palaik. 

star tkhlalt=tshol, Lutuami. 

night kaehe=apkha, Shasti. 

blood pouts =poits, Lutuami. 

one khum=tshiamu, Palaik. 

Of three languages spoken in the north of California and 
mentioned in Schooler aft, by name, though not given in 
specimens, (1) the Watsahewa, (2) the Howtetech, and (3) 
the Nabiltse, the first is said to be that of the Shasti 

Of the Howtetech I can say nothing ; 

The Nabiltse is, probably, the language of the Tototune ; at 
least Rogue's River is its locality, and the Rascal Indians is 
an English name for the Tototune. 

South of the Shasti and Lutuami areas we find 



The latter vocabulary is short, and taken from a Seragoin 
Indian, i. e. from an Indian to whom it was not the native 
tongue. We are warned of this the inference being that the 
Tahlewah vocabulary is less trustworthy than the others. 


man ahwunsh pohlusarih. 

boy anak'hocha kerrhn. 

girl yehnipahoitch. ..... kernihl. 

Indian ahrah astowah. 

head akhoutshhoutsh .... astintah. 

beard merruhw semerrhperrh. 

neck sihn schdbiti. 

face ahve wetawaluh. 

tongue upri so'h. 

teeth wu'h shti. 

foot fissi stah. 

one issah titskoh. 

two achhok kitchnik. 

three. . . keurakh . . kltchnah. 



four peehs tshahanik. 

Jive tirahho schwallah. 

ten trail swellah. 

The junction of the Rivers Klamatl and Trinity gives us 
the locality for 

spek itself is spoken at the junction, but its dialects of the 
Weyot and Wishosk extend far into Humboldt County, where 
they are, probably, the prevailing forms of speech, being used 
on the Mad River, and the parts about Cape Mendocino. 

The Weyot and Wishosk are mere dialects of the same 
language. From the Weitspek they diifer much more than 
they do from each other. It is in the names of the parts of 
the body where the chief resemblances lie. 

V. THE MENDOCINO (?) GROUP. This is the name sug- 
gested for the Choweshak, Batemdaikai, Kulanapo, Yukai, and 
Khwaklamayu forms of speech collectively. 

1, 2. The Choweshak and Batemdaikai are spoken on Eel 
River, and in the direction of the southern branches of the 
Weitspek group, with which they have affinities. 

3, 4, 5. The Kulanapo is spoken about Clear Lake, the 
Yukai on Russian River. These forms of speech, closely allied 
to each other, are also allied to the so-called Northern Indians 
of Baer's Beitrage, Northern meaning to the north of the 
settlement of Ross. The particular tribe of which we have 
a vocabulary called themselves Khwakhlamayu. 


head khommo kaiyah. 

hair shuka musuh. 

eye iiu ui. 

ear shuma shimah. 

nose pla labahbo. 

mouth aa katsideh. 

tooth oo yaoh. 

tongue aba bal. 

hand psha biyah. 

foot sakki kahmah. - 

sun ada . . . lah. 



moon kalazha luelah. 

star kamoi uiyahhoh. 

fire okho . . k'hoh. 

water aka k'hah. 

one ku khahlih. 

two koo kots. 

three subo homeka. 

four mura dol. 

five tysha lehmah. 

six lara tsadi. 

The following shows the difference between the Weitspek. 

and Kulanapo ; one belonging to the northern, the other to 
the southern division of their respective groups. 


man pagehk kaah. 

woman wintsuk dah. 

boy hohksh kahwih. 

girl wai inuksh dahhats. 

head tegueh kaiyah. 

hair leptaitl musuh. 

ear spehguh shimah. 

eye mylih ui. 

nose metpi labahbo. 

mouth mihlutl katsedeh. 

tongue mehpl'h bal. 

teeth. merpetl yadh. 

beard mehperch katsutsu. 

arm mehsheh' tsuah. 

hand tsewush biyyah. 

foot metske kahmah. 

blood happ'l bahlaik. 

sun wanoushleh lah. 

moon ketnewahr luelah. 

star haugets uiyahoh. 

day tehnep dahmul. 

dark ketutski petih. 

fire mets k'hoh. 

water paha k'hah. 

I . . nek . . hah. 



thou kehl ma. 

one spinekoh k'hahlih. 

two nuehr kots. 

three naksa homeka. 

four tohhunne dol. 

Jive mahrotum lehmah. 

six hohtcho tsadi. 

seven tchewurr kulahots. 

eight k'hehwuh kokodohl. 

nine kerr hadarolshum. 

ten^ wert'hlehwerh hadorutlek. 

In the Kulanapo language yacal ma napo = all the cities. 
Here napo = Napa, the name of one of the counties to the north 
of the Bay of San Francisco and to the south of Clear Lake. 

We may now turn to the drainage of the Sacramento and 
the parts south of the Shasti area. Here we shall find three 
vocabularies, of which the chief is called 

VI. THE COPEH. How far this will eventually turn out to 
be a convenient name for the group (or how far the group 
itself will be real), is uncertain. A vocabulary in Gallatin from 
the Upper Sacramento, and one from -Mag Readings (in the 
south of Shasti county) in Schoolcraft, belong to the group. 

Mag Readings is on the upper third of the Sacramento 
there or thereabouts. 






pehtluk . . . 

. . . winnoke .... 


muhlteh. . . 

. . . dokke 





. . . tomi 

. . tomoi. 

e y e 


. . . chuti 

. tumut. 


kiunik . . . 

. . . 

. . tsono. 

mouth ... 


. kal. 



. . . shi , 





sahlah . . . 

. keole. 



. . . shim 

. . tsemut(/ 



. . . mat 

. . ktamoso. 

blood . 

sahk . 

. . . chedik 




sun sunh tuku sas. 

wind toudi kleyhi 

rain yohro luhollo 

snow yohl yola 

fire poh pau po. 

water mehm mem mem. 

earth kirrh kosh 

In the paper of No. 134 the import of a slight amount of 
likeness between the Upper Sacramento vocabulary and the 
Jakon is overvalued. The real preponderance of the affinities 
of the group taken in mass is that which its geographical 
position induces us to expect a priori. With the Shasti, &c. 
the Copeh has the following words in common : 


head buhk uiak, S. 

hair teih tiyi, P. 

teeth siih itsa, P. 

ear maht mu-mutsh, L. 

eye sah asu, P. 

foot mat pats, L. 

sun sunh tsul, P. 

thou mih mai, S. 

and, probably, others. 

The Copeh is spoken at the head of Putos Creek. 

Observe that the Copeh for water is mem, as it is in the 
languages of the next group, which we may provisionally 

VII. THE PUJUNI. Concerning this we have a notice in 
Hale, based upon information given by Captain Suter to 
Mr. Dana. It was to the effect that, about eighty or a hun- 
dred miles from its mouth, the river Sacramento formed a 
division between two languages, one using momi, the other 
kik = water. 

The Pujuni, &c. say momi ; as did the speakers of the Copeh. 

For the group we have the (a) Pujuni, (b) Secumne, and 
(c) Tsamak specimens of Hale, as also the Cushna vocabulary, 
from the county Yuba, of Schoolcraft ; the Cushna numerals, 


as well as other words, being nearly the same as the Se- 
cumne, e.g. 


one wikte wikte-m. 

two pen pani-m. 

three sapui sapui-z. 

four tsi tsui-wz. 

Jive mauk marku-m (mahkum?). 

So are several other words besides, as 

head tsol chole. 

hair ono ono. 

ear bono bono. 

eye il bin. 

*un oko okpi. 

VIII. THE MOQUELUMNE GROUP. Male's vocabulary of 
the Talatui belongs to the group for which the name Moque- 
lumne is proposed, a Moquelumne Hill (in Calaveras county) 
and a Moquelumne Kiver being found within the area over 
which the languages belonging to it are spoken. Again, the 
names of the tribes that speak them end largely in -mne, 
Chupumne, &c. As far south as Tuol-wmwe county the language 
belongs to this division, as may be seen from the following 
table ; the Talatui being from Hale, the Tuolumne from School- 
craft ; the Tuolumne Indians being on the Tuolumne River, 
and Cornelius being their great chief, with six subordinates 
under him, each at the head of a different ranchora containing 
from fifty to two hundred individuals. Of these six members 
of what we may call the Cornelian captaincy, five speak the 
language represented by the vocabulary : viz. 

1. The Mumaltachi. 

2. The Mullateco. 

3. The Apangasi. 

4. The Lapappu. 

5. The Siyante or Typoxi. 

The sixth band is that of the Aplaches (? Apaches), under 
Hawhaw, residing further in the mountains. 




head hownah tiket. 

hair esok munu. 

ear tolko alok. 

eye hunteh wilai. 

nose nito uk (?). 

mouth ahwiik hube (?). 

sky wutsha wituk. 

sun heamhah hi. 

day hemaah hiumu. 

night kowwillah kawil. 

darkness .... pozattah hunaba. 

fire wiikah wike. 

water kikah kik. 

stone lowwak sawa. 

As far west as the sea-coast languages of the Moquelumne 
group are spoken. Thus 

A short vocabulary of the San Rafael is Moquelumne. 
So are the Sonoma dialects, as represented by the Tshokoyem 
vocabulary and the Chocouyem and Yonkiousme Paternosters. 
So is the Olamentke of Kostromitonov in Baer's Beitrage. 
So much for the forms of speech to the north of the Gulf 
of San Francisco. On the south the philology is somewhat 
more obscure. The Paternosters for the Mission de Santa 
Clara and the Vallee de los Tulares of Mofras seem to belong 
to the same language. Then there is, in the same author, one 
of the Langue Guiloco de la Mission de San Francisco. These 
I make Moquelumne provisionally. I also make a provisional 
division for a vocabulary called 

IX. THE COSTANO. The tribes under the supervision of 
the Mission of Dolores were five in number; the Ahwastes, 
the Olhones, or Costanos of the coast, the Romonans, the 
Tulomos, and the Altatmos. The vocabulary of which the 
following is an extract was taken from Pedro Alcantara, who 
was a boy when the Mission was founded, A.D. 1776. He 
was of the Romonan tribe. 


man imhen t&i-esse. 

woman ratichma . . . kuleh-esse. 



boy shimsmuk yokeh (small}. 

girl katra koyah. 

head illc moloh. 

ear tuorus ahlohk. 

eye rehin , shut. 

nose us huk. 

mouth werper lapgup. 

tongue tassek lehntip. 

tooth sift kuht. 

neck lani helekke. 

foot kolo koyok. 

blood payan kichawh. 

sky reneme lihlih. 

sun ishmen hih. 

moon kolma pululuk. 

star agweh hittish. 

day puhe (light) hiahnah. 

night moor (dark) kawul. 

fire roretaon wikih. 

water sir kihk. 

river orush polali. 

stone erek lepeh. 

I kahnah kahni. 

thou mene mih. 

he wahche ikkoh. 

they nekumsah mukkam. 

all kete mukkam. 

who mato mahnti. 

eat ahmush . . yohlomusih. 

drink owahto ushu. 

run akamtoha hihchiah. 

see atempimah ellih. 

This shows that it differs notably from the Tshokoyem; 
the personal pronouns, however, being alike. Again, the word 
for man = l-aman-tiya in the San Rafael. On the other hand, 
it has certain Cushna affinities. 

Upon the whole, however, the affinities seem to run in the 
direction of the languages of the next group, especially in that 
of the Ruslen ; 

H 2 


I=kah-nah, Cost. =ka= mine, Ruslen. 
Thou=me-ne, Cost. =me= thine, Ruslen. 
Sun=-ishmen, Cost.=ishmen=light, Ruslen. 
TPater=sii, Cost.=ziy, Ruslen. 
(?) Boy=shmishmuk ) Cost. = enshinsh, Ruslen. 
(?) Girl=katra, Cost.=kaana, Ruslen. 

Lest these last three coincidences seem far-fetched, it 
should be remembered that the phonesis in these languages is 
very difficult, and that the Ruslen orthography is Spanish, the 
Costano being English. Add to this, there is every appear- 
ance, in the San Miguel and other vocabularies, of the r being 
something more than the r in brand, &c., every appearance of 
its being some guttural or palatal, which may, by a variation 
of orthography, be spelt by /. 

Finally, I remark that the -ma in the Costano ratich-ma = 
woman, is, probably, the -me in the Soledad mue (=man) and 
shurish-me ( = woman), and the amk (ank) of the Ruslen muguy- 
amk (=man) and latrayam-ank (= woman) ; (?) latraya 
ratich. Nevertheless, for the present I place the Costano by 
itself, as a transitional form of speech to the languages spoken 
north, east, and south of the Bay of San Francisco. 

X. THE MARIPOSA LANGUAGES. In the north of Mariposa 
county, and not far south of the Tuolumne area, the language 
seems changed, and the Coconoons is spoken by some bands 
on the Mercede River, under a chief named Nuella. They 
are said to be the remnants of three distinct bands, each with 
its own distinct language. 


head , oto utno. 

hair tolus celis. 

ear took took. 

nose thedick tuneck. 

mouth sammack shemmak. 

tongue talcotch talkat. 

tooth talee talee. 

sun suyou oop. 

moon offaum taahmemna. 

star tchietas . . sahel. 



day hial tahoh *. 

fire sottol ossel. 

water illeck illick. 

XI. THE SALINAS GROUP. This is a name which I propose 
.for a group of considerable compass, and one which contains 
more than one mutually unintelligible form of speech. It is 
taken from the river Salinas, the drainage of w r hich lies in the 
counties of Monterey and San Luis Obispo. The southern 
boundary of Santa Cruz lies but a little to the north of its 

The Gioloco may possibly belong to this group, notwith- 
standing its reference to the Mission of San Francisco. The 
alia, and mut- (in mut-ryocuse), may = the ahay and i-mit-a 
(sky) of the Eslen. 

The Ruslen has already been mentioned, and that in respect 
to its relations to the Costano. It belongs to this group. 

So does the Soledad of Mofras ; which, though it differs 
from that of Hale in the last half of the numerals, seems to 
represent the same language. 

So do the Eslen and Carmel forms of speech ; allied to one 
another somewhat more closely than to the Buslen and So- 

So do the San Antonio and San Miguel forms of speech. 

The Ruslen, Eslen, San Antonio, and San Miguel are, pro- 
bably, four mutually unintelligible languages. 

The Salinas languages are succeeded to the south by the 
forms of speech of 

XII. THE SANTA BARBARA GROUP containing the Santa 
Barbara, Santa Inez, and San Luis Obispo languages. 

XIII. THE CAPISTRANO GROUP. Capistrano is a name 
suggested by that of the Mission of San Juan Capistrano. 
The group, I think, falls into two divisions : 

1. The Proper Capistrano, or Netela, of San Luis Rey and 
San Juan Capistrano. 

2. Tlie San Gabriel, or Kij, of San Gabriel and San Fer- 

* Same word as taech = light in Coconoons; in Pima /a*. 



XIV. THE YUMA LANGUAGES. At the junction of the Gila 
and Colorado stands Fort Yuma, in the district of the Yuma 
Indians. They occupy each side of the Colorado, both above 
and below its junction with the Gila. How far they extend 
northwards is unknown, probably more than 100 miles. They 
are also called Cuchans, and are a fierce predatory nation, 
encroaching equally on tribes of their own language and on 

From these Yuma Indians I take the name for the group 
now under notice. It contains, besides the Yuma Proper, the 
Dieguno of San Diego and the Coco-maricopa. 

The Coco-maricopa Indians are joint-occupants of certain 
villages on the Gila; the population with which they are 
associated being Pima. Alike in other respects, the Pima and 
Coco-maricopa Indians differ in language, as may be seen 
from the following table, confirmatory of the testimony of 
numerous trustworthy authorities to the same effect. 




man.. . huth.. . epatsh . . apatch . . . 

1 epatch. 

woman . . hahri. . . . sinyak .......... seniact .... sun. 

Indian . . huup. . . . metepaie ......... - ...... - * 

fecoutsucherowo "") 

head ---- mouk . . -| and ^ - ...... estar, 

i^umwelthoocouo. . J 

hair . . ptmuk . . eetche .......... - ...... hiletar. 

ear .... ptnahauk . smythl .......... - ...... - 

nose . . tahnk. . . . - ............ - ...... hu. 

mouth. . chinits . . ...... ...... ...... ah. 

tongue . neuen. . . . epulche .......... - ...... 

tooth . . ptahan . . aredoche ........ - ...... - 

beard . . chinyo . . yahboineh ........ - ...... - 

hand . . mahahtk. . eesalche ......... issalis ...... selh. 

foot . . tetaght . . emetchslipaslapya. . ametche. . . . hamulyay. 

sky ---- ptchuwik . ammai .......... - ...... - 

tun ---- tahs .... nyatch .......... - ...... - 

moon . . mahsa. . . . huthlya .......... - ...... - 

star . . uon ...... klupwalaie ........ - ...... - 

snoiv , . chiah .... halup .......... - ...... - 




fire. . . . tahi . . 

water . . suutik. . 

I. ahan . . 

he .... yeutah 

one .... yumako 

two. . , . kuak . . 

three . . vaik 


aawoh house 

aha haache .... kha. 

nyat nyah. 


sin. ... sandek , . 

havick haveka . . 

hamuk hamoka . . 

four . . kiik chapop champapa 

five .... puitas. . . . scrap sarap .... 






San Diego lies in 32| north latitude, a point at which the 
philology diverges in one direction into Old California, in 
another into Sonora. I first follow it in the direction of 


San Diego, as has just been stated, lies in 32 J north 
latitude. Now it is stated in the Mithridates that the most 
northern of the Proper Old Californian tongues, the Cochimi, 
is spoken as far north as 33. If so, the Dieguno may be Old 
Californian as well as New ; which I think it is ; believing, at 
the same time, that Cochimi and Cuchan are the same words. 

Again, in the following Paternoster the word for sky = 
ammai in the Cuchan vocabulary. 


father sky 

Pennayu make^amba yaa ambayujui miya mo ; 

name men confess and love all 

Buhu mombojua tamma gkomenda hi nogodono demuejueg gkajim ; 

and sky earth 

Pennayula bogodono gkajim, gui hi ambayujup maba yaa keammete 

decuinyi mo puegin ; 

sky earth 

Yaa m blihula mujua ambayup mo dedahijua, amet e no guilugui 
hi pagkajim ; 

this day day 

Tamada yaa ibo tejueg quiluguiqui pe,mijich e mbu ibo yanno 
puegin ; 


and man evil 

Guihi tamraa yaa gambuegjula k^pujui ambinyijua pennayala 

dedaudugujua, giulugui pagkajim ; 

and although and 

Guihi yaa tagamuegla hui ambinyijua hi doomo puhuegjua, he 

doomo pogonunyim ; 

and earth bless 

Tagamuegjua guihi usimahel keammet e decuinyimo, guihi yaa hui 

ambinyi yaa gambuegpea pagkaudugum. 

Lastly, in 33 north latitude, the language of San Luis El 
Rey, which is Yuma, is succeeded by that of San Luis Obispo, 
which is Capistrano. 

I conclude, then, that the Yuma language belongs to the 
southern parts of New and the northern parts of Old Cali- 

Of recent notices of any of the languages of Old California, 
eo nomine, I know none. In the Mithridates the information 
is pre-eminently scanty. 

According to the only work which I have examined at first- 
hand, the Nachrichten von der Americanisehen Halbinsel 
Kalifornien (Mannheim, 1772; in the Mithridates, 1773), the 
anonymous author of which was a Jesuit missionary in the 
middle parts of the Peninsula, the languages of Old California 

1. The Waikur, spoken in several dialects, 

2. The Ushiti. 

3. The Layamon. 

4. The Cochimi, north, and 

5. The Pericu, at the southern extremity of the peninsula. 

6. A probably new form of speech used by some tribes 
visited by Linck. 

This is what we learn from what we call the Mannheim ac- 
count ; the way in which the author expresses himself being 
not exactly in the form just exhibited, but to the effect that, 
besides the Waikur with its dialects, there were five others. 

The Waikur Proper, the language which the author under 


notice was most especially engaged on, and which he says that 
he knew sufficiently for his purposes as a missionary, is the 
language of the middle part of the peninsula. How far the 
Utshiti and Layamon were dialects of it, how far they were 
separate substantive languages, is not very clearly expressed. 
The writer had Utshis, and Utshipujes, and Atschimes in his 
mission, " thoroughly distinct tribes lauter verschiedene 
Volcklein." Nevertheless he always speaks as if the Waikur 
tongue was sufficient for his purposes. On the other hand, 
the Utshiti is especially mentioned as a separate language. 
Adelung makes it a form of the Waikur ; as he does the Laya- 
mon, and also the Cora and Aripe. Then there comes a popu- 
lation called Ika, probably the Picos or Ficos of Bagert, 
another authority for these parts. Are these, the sixth popu- 
lation of the Mannheim account, the unknown tribes visited by 
Linck ? I think not. They are mentioned in another part 
of the book as known. 

To the names already mentioned 

1. Ika, 3. Utshipuje, 

2. Utshi, 4. Atschime, 


5. Paurus, 9. Mitsheriku-tamais, 

6. Teakwas, 10. Mitsheriku-tearus, 

7. Teengiiabebes, 11. Mitsheriku-ruanajeres, 

8. Anguk wares, 

and you have a list of the tribes with which a missionary for 
those parts of California where the Waikur languages pre- 
vailed, came in contact. Altogether they gave no more than 
some 500 individuals, so miserably scanty was the popu- 

The occupancies of these lay chiefly within the Cochimi 
area, which reached as far south as the parts about Loretto in 
26 north latitude ; the Loretto language being the Layamon. 
This at least is the inference from the very short table of the 
Mithridates, which, however little it may tell us in other 
respects, at least informs us that the San Xavier, San Borgia, 


and Loretto forms of speech were nearer akin to each other 
than to the Waikur. 


sky .... ambayujub. . ambeink .... . . . . terereka-datemba. 

earth . . amet amate-guang. . . . . . datemba. 

fire usi ussi 

man .... tamma .... tama tamma . . ti. 

father . . kakka .... iham keneda.. . 

son uisaham .... . . . . tshanu. 

The short compositions of Hervas (given in the Mithridates) 
show the same. 

THE WAIKUR. This is the language of what I have called 
the Mannheim account, namely the anonymous work of a Jesuit 
missionary of the Waikur country published at Mannheim. 

It gives us the following specimens Waikur and German : 

Kepe-dare tekerekadatembi dai ; 
unser Vater gebogene Erd du bist ; 

ei-ri akatuike-pu-me ; 

dichodas erkennen alle werden ; 

tshakarrake-pu-me ti tschie ; 

loben alle werden Lent und\ 

ecun gracia-ri acume care tekerekadatembi tschie ; 
dein gratia o dass haben werden wir gebogene Erd und ; 

eiri jebarrakemi ti pu jaupe datemba 

dir o dass gehorsamen werden Menschen alle heer Erd, 

pae ei jebarrakere ae'na kea ; 
wie dir gehorsamen droben seynd ; 

kepecun bu. kepe ken jatupe untairi ; 
unser Speis uns gebe dieser tag ; 

cate kuitscharake tei tschie kepecun atacamara 
uns verzehe du und unser Boses ; 

pae kuitscharrakere cate tschie cavape atukiara keperujake ; 
wie versehen wir auch die Boses uns thun ; 

cate tikakamba tei tschie; 
uns helfe du und; 

cuvumera cate ue atukiara; 

wollen werden Nicht wir etwas Boses ; 

kepe kakunja pe atacara tschie. Amen. 
uns beschutze von Bo'sen und. Amen. 



The compound tekereka-datembi = bent land '= sky = heaven. 
To this very periphrastic Paternoster we may add the fol- 
lowing fragments of the Waikur conjugation : 






Tucava j 



Tutau , . . .. . 

> amukmrikeri = 
Cate j 

Tucava J 

Amukirime = 
Amukiri tei=lude. 
Amukiri tu=ludite. 







ego ludo. 
tu ludis. 
ille ludit. 
nos ludimus. 
vos luditis. 
illi ludunt. 
'ego lusi. 
tu lusisti. 
ille lusit. 
nos lusimus. 
vos lusistis. 
_illi luserunt. 


1 wish I had not played. 

Thou $r. 


We #-c. 


Of the Pericu spoken at the south extremity of the penin- 
sula, I know no specimens. 

We now turn to that part of the Yuma area which lies along 
the course of the Gila, and more especially the parts along 
the Cocomaricopa villages, of which one portion of the occu- 
pants speak a language belonging to the Yuma, the other one 
belonging to the Pima class. 

This latter leads us to the languages of the northern pro- 
vinces of Mexico 


For these two provinces, the languages for which we have 
specimens fall into five divisions : 







That the Pima group contains the Pima Proper, the Opata, 
and the Eudeve, may be seen from the Mithridates. That the 
language of the Papagos, or Papago-cotam, is also Pima, rests 
upon good external evidence. Whether the speech of the 
Ciris, and population of the island of Tiburon and the parts 
opposite, be also Pima, is at present uncertain ; though not 
likely to be so long, inasmuch as I believe that Mr. Bartlett, 
the Boundary Commissioner, is about to publish samples, not 
only of this, but of the other languages of Sonora. 

West of the Pima lies the Tarahumara, and south of it the 
Hiaqui, succeeded by the Tubar and Cora of Sinaloa. 

The following Paternosters of these four languages may be 
compared with the Opata dialect of the Pima. The words 
that, by appearing in more than one of them, command our 
attention and suggest the likelihood of a closer relationship 
than is indicated in the Mithridates, or elsewhere, are in 


Tamo mas teguiacachigua. cacame ; 

Amo tegua santo a ; 

Amo reino tame macte ; 

Hinadeia iguati terepa ania teguiacachiveri ; 

Chiama tamo guaco veu tamo mac ; 

Guatame neavere tamo cai naideni aca api tame neavere tomo opagua; 

Gua cai tame taotitudare ; 

Cai naideni chiguadu rApita cachia. 


Itom-achai t eve-capo catecame ; 
Che-chevasu yoyorvva ; 
Itou piepsana em yaorahua ; 

Em harepo in buyapo annua amante (tevecapol} vecapo annua beni ; 
Machuveio/w-buareu yem itom amica-itom ; 

Esoc alulutiria ca-aljiton-anecau itepo soc alulutiria ebeni itom ve- 
herim ; 


Caitom butia huenacuchi cativiri betana ; 
Aman itom-jeretua. 


Ite-canar <?rmuicarichua catemat ; 

Imit teffmu&rac milituraba teochiqualac ; 

Imit huegmica carinite bacachin-assifaguin ; 

Imit avamunarir echu nanagualac imo cuigan amo nachic tegmue- 

caricheri ; 

Ite cokuatarit, essemer taniguarit, iabbe ite micam ; 
Ite tatacoli ikiri atzomua ikirirain ite bacachin cale kuegma nanegua 

cantem ; 

Caisa ite nosam bacatatacoli ; 
Bacachin ackiro muetzerac ite. 


Tami nono, mamu regui guami gatiki ; 

Tami noineruje mu regua ; 

Telimea rekijena ; 

Tami neguaruje mu jelaliki henna, guetshiki, mapu hatschibe re- 

guega guami ; 
Tami nututuge hipeba ; 
Tami guecanje tami guikeliki, matame hatschibe reguega tami gue- 

canje putse tami guikejameke ; 
Ke ta tami satuje ; 
Telegatigemeke mechka hula. Amen. 


Ta yaoppe tapahoa pethebe ; 

Cherihuaca eiia teaguarira ; 

Chemeahuabeni tahemi (to us) eiia chianaca ; 

Cheaquasteni eiia jevira iye (as) chianacatapoan tup up tapahoa ; 

Eii ta hamuit (bread) eu te huima tahetze rej rujeve ihic (to-day} 

ta taa ; 

Huatauniraca ta xanacan tetup itcahmo tatahuatauni titaxanacante ; 
Ta vaehre teatcai havobereni xanacat hetze huabachreaca tecai tahemi 

rutahuaga tehai eu ene. 

With these end our data, but not our lists of dialects ; the 


names Maya, Guazave, Heria, Sicuraba, Xixime, Topia, Te- 
peguana, and Acaxee all being, either in Hervas, or elsewhere, 
as applied to the different forms of speech of Sonora and Si- 
naloa ; to which may be added the Tahu, the Pacasca, and the 
Acasca, which is probably the same word as Acaxee, as Huimi 
is the same as Yuma, and Zaque as Hiaqui. Of the Guazave 
a particular dialect is named as the Ahome. Add also the 
Zoe and Huitcole, probably the same as the Huite. 

That some of these unrepresented forms of speech belong to 
the same class with the Pima, Hiaqui, &c., is nearly certain. 
How many, however, do so is another question ; it may be 
that all are in the same predicament ; it may be only a few. 

The languages of 


These are 




The last will be considered at once, and dismissed. More 
has been written on the Otomi than any other language of 
these parts ; the proper Mexican not excepted. It was ob- 
served by Naxera that it was monosyllabic rather than 
poly synthetic, as so many of the American languages are, 
with somewhat doubtful propriety, denominated. A Mexican 
language, with a Chinese characteristic, could scarcely fail to 
suggest comparisons. Hence, the first operation on the 
Otomi was to disconnect it from the languages of the New, 
and to connect it with those of the Old World. With his ac- 
customed caution, Gallatin satisfies himself with stating what 
others have said, his own opinion evidently being that 
the relation to the Chinese was one of analogy rather than 

Doubtless this is the sounder view ; and one confirmed by 
three series of comparisons made by the present writer. 

The first shows that the Otomi, as compared with the mono- 
syllabic languages of Asia, en masse, has several words in com- 
mon. But the second qualifies our inferences, by showing that 
the Maya, a language more distant from China than the Otomi, 


and, by no means inordinately monosyllabic in its structure, 
has, there or thereabouts, as many. The third forbids any 
separation of the Otomi from the other languages of America, 
by showing that it has the ordinary amount of miscellaneous 

In respect to the Chinese, &c., the real question is not 
whether it has 50 many affinities with the Otomi, but whether 
it has more affinities with the Otomi than with the Maya or 
any other American language ; a matter which we must not 
investigate without remembering that some difference in 
favour of the Otomi is to be expected, inasmuch as two lan- 
guages with short or monosyllabic words will, from the very 
fact of the shortness and simplicity of their constituent 
elements, have more words alike than two polysyllabic forms 
of speech. 

The fact, however, which most affects the place of the 
Otomi language is the monosyllabic character of other Ame- 
rican languages, e. g. the Athabaskan and the Attacapa. 

As these are likely to be the subject of some future investi- 
gation, I lay the Otomi, for the present, out of consideration ; 
limiting myself to the expression of an opinion, to the effect 
that its philological affinities are not very different from what 
its geographical position suggests. 

Of the Pirinda and Tarasca we have grammars, or rather 
grammatical sketches ; abstracts of which, by Gallatin, may 
be found in his Notes on the Semi- civilized Nations of 
Mexico, Yucatan, and Central America, in the first volume of 
the Transactions of the American Ethnological Society. The 
following are from the Mithridates. 


Cabutumtaki ke exjechori pininte ; 
Niboteachatii tucathi nitubuteallu ; 
Tantoki hacacovi nitubutea pininte ; 
Tarejoki nirihonta manicatii ninujami propininte ; 
Boturimegui dammuce tupacovi chii ; 

Exgemundicovi boturichochii, kicatii pracavovi ku^entumundijo bo- 
turichochijo ; 


Niantexechichovi rumkuejentuvi innivochochii ; 

Moripachitovi cuinenzimo tegui. 



Tata uchaveri tukire hacahini avandaro ; 

Santo arikeve tucheveti hacangurikua ; 

Wetzin andarenoni tucheveti irecheekua ; 

Ukuareve tucheveti wekua iskire avandaro, na humengaca istu umen- 

gave ixu excherendo. 

Huchaeveri curinda hanganari pakua intzcutzini yaru ; 
Santzin wepovacheras huchaeveri hatzingakuareta, izki huchanac 

wepocacuvanita haca huchaveri hatzingakuaechani ; 
Ca hastzin teruhtazema teruniguta perakua himbo ; 
Evapentztatzini yaru catzingurita himbo. Isevengua. 

It now becomes convenient to turn to the parts to the east 
of California, viz. 


In Utah the philology is simple, all its forms of speech 

1 . Athabaskan ; 

2. Paduca; or 

3. Pueblo. 

1. The Navaho, along with the Jecorilla of New Mexico, 
the Hoopah of California, and Apatch of California, New 
Mexico and Sonora, is Athabaskan. 


man tennai ailee. 

woman estsonnee eetzan. 

head (my). . . . Awtzeetsin seezee. 

hair (my) .... hutzee seesga. 

face (my) .... Awnnee streenee. 

ear (my) .... hutjala seetza. 

eye (my) .... Awnnah sleeda. 

nose (my) .... Awtchih seetzee. 

mouth (my) . . huzzai : sheeda. 

tongue (my) . . huttso sheedare. 

tooth (my) . . hurgo sheego. 



sky eeyah eah. 

sun chokonoi skeemai. 

moon klaihonoi clanai. 

star sonh suns. 

day cheen-ffo eeska. 

night klai-^o cla. 

light hoascen-^o skee. 

rain ....:. . . naheltinh nagostee. 

snow yas zalis. 

hail neelo heeloah. 

fire konh kou. 

water tonh toah. 

stone tsai zeyzay. 

one tlahee tahse. 

two nahkee nahkee. 

three tanh. . tail. 

2. The Utah with its allied dialects is Paduca, i. e. a mem- 
ber of the class to which the Shoshoni, Wihinast, and Cu- 
manch languages belong. 

3. The Moqui is one of the languages of 


The comparative civilization of the Pueblo Indians has 
always attracted the attention of the ethnologist. Until lately, 
however, he had but a minimum amount of trustworthy infor- 
mation concerning either their habits or their language. He 
has now a fair amount of data for both. For philological pur- 
poses he has vocabularies for six (probably for all) of them. 

Of the Pueblo languages two belong to the drainage of the 
Rio Colorado and four to that of the Rio Grande. Of these 
two divisions the former lies the farthest west, and, of the two 
Colorado Pueblos, the most western is that of 

The Moqui. The Moqui vocabulary was procured by Lieut. 
Simpson from a Moqui Indian who happened to be at Chelly. 

The Zuni country lies in 35 north latitude, to the south 
and east of the Moqui, and is probably divided by the Sierra 
de Zuni from 


The Acoma, or Laguna, the most southern of the Pueblos 
of the Rio Grande. North of the Acoma area lies that of 

The Jemez, on the San Josef. 

The two that still stand over lie on the main stream of the 
Rio Grande itself. They are 

The Tesuque; and 

The Taos or Picuri. The northern boundaries of the Te- 
suque seem to be the southern ones of Taos. Connect these 
Pueblos with the town of Taos, and the Tesuque with Santa 
Fe, and the ordinary maps give us the geography. 

The philological affinities of the Pueblo languages scarcely 
coincide with the geographical relations. The Moqui lies far 
west. Laying this then out of the question, the three that, 
in their outward signs, most strike the eye in tables, as 
agreeing with each other, are the Laguna, the Jemez, and the 
Tesuque. The other two that thus outwardly agree are the 
Taos and the Zuni, two that are not in the most immediate 
geographical juxtaposition. 

What is meant by the " outward signs that most strike the 
eye on tables" ? This is shown in the following tables : 


head oshoqui/wee pto. 

hair tiysJiwee po. 

ear , . . lahjotinnee oyez. 

eye ton&hwee tzie. 

nose nohahhunee heu. 

mouth aihvfsihtinnee so. 

tongue houinnee . . . hae. 

tooth oaknahwee nmai. 

The following are some of the most patent miscellaneous 
affinities : 

English, sun. 
Tesuque, pah. 
Jemez, pah. 

English, moon. 
Tesuque, poyye. 

Jemez, pahah. 
Taos, pannah. 
Moqui, muyah. 
English, man. 
Tesuque, say en. 
Jemez, tahhanenah. 



English, woman. 
Tesuque, ker. 
Zuni, ocare. 
English, wife. 
Tesuque, naveso. 
Jemez, neohoy. 
English, boy. 
Tesuque, onue. 
Jemez, annoh. 
English, forehead. 
Tesuque, siccovah. 
Laguua, cophay. 
English, face. 
Tesuque, chaay. 
Laguna, kowah. 
English, eye. 
Tesuque, chay. 
Jemez, saech. 
English, teeth. 
Tesuque, muah. 
Taos, moen-nahenhay . 
Moqui, moah= mouth. 
English, chin. 
Tesuque, shabbok. 
Taos, claybonhai. 

English, hand. 
Tesuque, mah. 
Jemez, mahtish. 
Moqui, moktay. 
Moqui, mahlatz finger. 
English, breast. 
Tesuque, peah. 
Laguna, quaist-pay. 
Taos, pahahkaynaynemay . 
Jemez, pay-lu. 
Utah, pay. 
English, deer. 
Tesuque, pahye. 
Jemez, pahah. 
English, rattlesnake. 
Tesuque, pay y oh. 
Taos, pihoown. 
English, cat. 
Tesuque, musah. 
Laguna, mus. 
Taos, museenah. 
Jemez, moonsah. 
Zuni, musah. 
English, fire. 
Tesuque, tah. 
Jemez, twaah. 

The Moqui, which is not to be separated from the other 
Pueblo languages, has, out of twenty-one words compared, 
eight coinciding with the Utah. 

Neither are there wanting words common to the Pueblo 
languages and those of the Athabaskan Navahos, Jecorillas 
and Apatches. 

English, deer. 
Navaho, payer. 
Jecorilla, pay ah. 
Jemez, pahah. 
English, cat. 
Navaho, muse. 
Jecorilla, muss ah. 

Tesuque, musah. 
Laguna, &c.*, mus. 

English, earth. 
Navaho, ne. 
Jecorilla, nay. 
Tesuque, nah. 

* The Utah is musah. 


English, man. 
Navaho, tennay. 
Jecorilla, tinlay. 
Tesuque, say en. 

English, mouth. 
Navaho, hu-zzay. 
Jecorilla, hu-zzy. 
Tesuque, sho. 

Jemez, tahhanenah. 

Of these the first two may be borrowed. In 


the languages are Arapaho, and Shyenne, already noticed ; 
and Cumanch, which is Paduca. 

For the Kioway we want specimens. In 


they are Sioux, already noticed, and Pawni, allied to the Ric- 
caree. Kanzas leads us to 


It is convenient in a notice of the languages of the State 
of Texas to bear in mind its early, as well as its present 
relations to the United States. In a country where the spread 
of the population from the other portions of the Union has 
been so rapid, and where the occupancy is so complete, we are 
prepared to expect but a small proportion of aborigines. And 
such, upon the whole, is the case. The displacement of the 
Indian tribes of Texas has been great. Even, however, when 
Mexican, Texas was not in the category of the older and more 
original portions of Mexico. It was not brought under the 
regime of the missionaries, as we may see by turning to that 
portion of the Mithridates which treats of the parts west of 
the Mississippi. The references here are to Dupratz, to Lewis 
and Clarke, to Charlevoix, to French and English writers 
rather than to the great authority for the other parts of 
Spanish America Hervas. And the information is less pre- 
cise and complete. All this is because Texas in the earlier 
part of its history was, in respect to its exploration and de- 
scription, a part of Louisiana (and, as such, French) rather 
than a part of Mexico, and (as such) Spanish. 

The notices of Texas, in the Mithridates, taken along with 


our subsequent data, are to the effect that (a) the Caddo, (b) 
the Adaize or Adahi, (c) the Attakapa, and (d) the Choktah 
are the prevailing languages of Texas ; to which may be added 
a few others of minor importance. 

The details as to the distribution of the subordinate forms 
of speech over these four leading languages are as follows : 

a. The Nandakoes, Nabadaches, Alich (or Eyish), and Ini 
or Tachi are expressly stated to be Caddo ; and, as it is from 
the name of the last of these that the word Texas is derived, 
we have satisfactory evidence that some members, at least, of 
the Caddo family are truly and originally Texian. 

b. The Yatassi, Natchitoches, Adaize (or Adaye), Nacog- 
doches, and Keyes, belong to the Caddo confederacy, but with- 
out speaking the Caddo language. 

c. The Carancouas, the Attacapas, the Apelusas, the Mayes 
speak dialects of the same language. 

d. The Tunicas speak the same language as the Cheetahs. 
Concerning the philology of the Washas, the Bedies, the 

Acossesaws, and the Cances, no statements are made. 

It is obvious that the information supplied by the Mithri- 
dates is measured by the extent of our knowledge of the four 
languages to which it refers. 

Of these, the Choktah, which Adelung calls the Mobilian, 
is the only one for which the Mithridates itself supplies, or 
could supply, specimens ; the other three being unrepresented 
by any sample whatever. Hence, to say that the Tachi was 
Caddo, that the Yatassi was Adahi, or that the Carancoua 
was Attacapa, was to give an instance, in the way of explana- 
tion, of the obscurum per obscurius. Since the publication of 
the Mithridates, however, we have got samples of all three 
Caddo, Adahi, and Attacapa so that our standards of com- 
parison are improved. They are to be found in a tabulated 
form, and in a form convenient for collation and comparison 
in both of Gallatin's papers. They were all collected before 
the annexation of Texas, and they appear in the papers just 
referred to as Louisiana, rather than truly Texian, languages ; 
being common to the two areas. 

Of the works and papers written upon Texas since it 


became a field of observation for English and American, as 
opposed to French and Spanish observers, the two on which 
the present writer, when he treated of the subject in his work 
on the Varieties of Mankind, most especially, and perhaps ex- 
clusively relied, were the well-known work of Kennedy on 
Texas, and a MS. with which he was favoured by Mr. Bol- 
laert, specially limited to the ethnology of the State. Of this 
MS. a short abstract is to be found in the Report of the 
British Association for the Advancement of Science for the 
year 1846, made by Mr. Bollaert himself. 

The later the notice of Texas the greater the prominence 
given to a tribe of which nothing is said in the Mithridates, 
viz. the Cumanch. As late as 1844 we had nothing beyond the 
numerals and a most scanty MS. list of words to tell us what 
the Cumanch language really was. These, however, were 
sufficient to show that its affinities were of a somewhat 
remarkable kind, viz. with the Shoshoni, or Snake, tongues of 
the southern parts of Oregon*. In Mr. Bollaerfs notice the 
Cumanches are divided into three sections : (1) the Cumanch 
or Jetan, (2) the Lemparack, and (3) the Tenuha, and a list of 
no less than thirty-five other tribes follows this division, some 
of these being said to be wholly extinct, some partially so ; 
some to be more or less Cumanch, some to be other than Cu- 

The tendency of the Mithridates is to give prominence to 
the Caddo, Attacapa, and Adahi tongues, and to incline the 
investigator, when dealing with the other forms of speech, to 
ask how far they are connected with one of these three. The 
tendency of the writers last-named is to give prominence to 
the Cumanch, and to suggest the question : How far is this 
(or that) form of speech Cumanch or other than Cumanch ? 

Working with the Mithridates, the MS. of Mr. Bollaert, 
and Mr. Kennedy's volume on Texas before me, I find that the 
list of Texian Indians which these authorities justified me in 
publishing in 1848, contained (1) Coshattas, (2) Towiachs, 
Towakenos, Towecas, and Wacos, (3) Lipans or Sipans, (4) 

* " On the Languages of the Oregon Territory." By R. G. Latham, 
M.D. Read before the Ethnological Society, Dec. 1844.- Note. 


Aliche or Eyish, (5) Acossesaws, (6) Navaosos, (7) Mayes, 
(8) Cances, (9) Toncahuas, (10) Tuhuktukis, (11) Unataquas 
or Anadarcos, (12) Mascovie, (13) Tawanis or lonis, (14)' 
Wico, ? Waco, (15) Avoyelles, (16) Washitas, (17) Ketclii, 
(18) Xaramenes, (19) Caicaches, (20) Bidias, (21) Caddo, (22) 
Attacapa, (23) Adahi ; besides the Carankahuas (of which the 
Cokes are made a branch) classed with the Attacapa, and not 
including certain Cherokees, Choctahs, Chikkasahs, and Sioux. 

A Washita vocabulary, which will be referred to in the 
sequel, concludes the list of Texian languages known by spe- 

At present, then, the chief question respecting the philo- 
logy of Texas is one of distribution. Given as centres to cer- 
tain groups 

1. The Choctah, 

2. The Caddo, 

3. The Adahi, 

4. The Attakapa, 

5. The Cumanch, and 

6. The Washita languages, 

how do we arrange the tribes just enumerated ? Two works 
help us here : 1. A Letter from the Ex-president Burnett to 
Schoolcraft on the Indians of Texas. Date 1847. 2. A Sta- 
tistical Notice of the same by Jesse Stem. Date 1851. 
Stem's statistics run thus : 


Towacarros . . 


114 >293 


38 J 


161 -| 

Andarcos . . . 

202 V476 






.... 100 

Limns . . 


Comanches 20,000 

giving us several of the names that have already appeared 


giving also great prominence to the Cumanches numerically 
at least. 

In Mr. Burnett's Letter the term Caddo is prominent ; but 
whether it denote the Caddo language, or merely the Caddo 
confederation, is uncertain. Neither can I find from the con- 
text whether the statements respecting the Indians of the 
Caddo connexion (for this is what we must call it at present) 
are made on the personal authority of the writer, or whether 
they are taken, either directly or indirectly, from the Mithri- 
dates. The term that Burnett uses is stock, his statement 
being that the Waco, the Tawacani, the Towiash, the Aynic, 
the San Pedro Indians, the Nabaducho, and the Nacodocheets 
are all both Texian in origin and Caddo in stock. 

His other tribes are 

1 . The Ketchi : a small tribe on Trinity River, hated by 
the Cumanches as sorcerers, and, perhaps, the same as 

2. The Hitchi, once a distinct tribe, now assimilated with 
their neighbours. 

3. The Tonkaways, a separate tribe, of which, however, the 
distinctive characters are not stated. 

Whatever may be the exact details of the languages, dia- 
lects, and subdialects of Texas, the general outline is simple. 

The Choctah forms of speech are anything but native. 
They are of foreign origin and recent introduction. So are 
certain Sioux and other dialects spoken within the Texian 

The Cumanch is in the same predicament; though not, 
perhaps, so decidedly. It belongs to the Paduca class, and its 
affinities are with the Shoshoni and Wihinast of Oregon. 

The Caddo Proper is said to be intrusive, having been intro- 
duced so late as 1819 from the parts between the great Eaft 
and the Natchitoches or Red River. I hold, however, that 
some Caddo forms of speech must be indigenous. 

The Witchita is probably one of these : 


head cundp etskase. 

hair beunno deodske. 

eye nockkochun kidahkuck. 



nose sol dutstistoe. 

mouth nowoese hawkoo. 

tongue ockkotunna hutskee. 

tooth ockkodeta awk. 

one whiste cherche. 

two .'. bit mitch. 

three dowoh daub. 

four peaweh dawquats. 

Jive dissickka esquats. 

six dunkkee kehass. 

seven bissickka keopits. 

eight dowsickka keotope. 

nine pewesickka sherchekeeite. 

ten binnah skedorash. 

The Adahi has already been noticed as being a compara- 
tively isolated language, but, nevertheless, a language with 
numerous miscellaneous affinities. 

The Attacapa is one of the pauro-syllabic languages of 
America, by which I mean languages that, if not monosyllabic 
after the fashion of the languages of south-eastern Asia, have 
the appearance of being so. They form a remarkable class, 
but it is doubtful whether they form a natural one, i. e. whether 
they are more closely connected with each other in the other 
elements of philological affinity than they are with the 
tongues not so characterized. They deserve, however, what 
cannot be given in the present paper, a special consideration. 
For the north-eastern districts of Mexico, New Leon, Ta- 
maulipas, &c., i. e. for the parts between the Rio Grande and 
Tampico, no language is known to us by specimens. It is 
only known that the Cumanch dips deeply into Mexico. So 
does the Apatsh. 

A tribe, lately mentioned, that of the Lipans, is, perhaps, 
Apatsh. Burnett states that they agree with the Mescalero 
and Seratics of the parts about the Paso del Norte. For these, 
however, we still want vocabularies iis nominibm. 

Be the Lipan affinities what they may, it is clear that both 
the Cumanch and Apatsh languages belong to a class foreign 
to a great part of the areas over which they are spread 


foreign, and (as such) intrusive intrusive, and (as such) de- 
veloped at the expense of some native language. 

That the original area of the latter is that of the Navahos, 
Jecorillas, Hoopahs, Umkwas, Tlatskanai, and that these oc- 
cupy the parts between the Algonkin and Eskimo frontiers 
parts as far north as the Arctic circle has already been stated. 
No repetition, however, is superfluous that gives definitude 
and familiarity to the very remarkable phenomena connected 
with the geographical distribution of the Athabaskans. 

Neither are the details of the Paduca area the area of the 
Wihinast, Shoshoni, Utah, and Cumanch forms of speech 
without interest. To the north of California, the Wihinast, 
or Western Shoshonis, are separated from the Pacific by a 
thin strip of Jacon and Kalapuya country, being succeeded in 
the direction of Utah by the Shoshonis Proper. Then follow 
the Bonaks and Sampiches ; the Shoshoni affinities of which 
need not be doubted, though the evidence of them is still 
capable of improvement. The Utah of the parts about Lake 
Utah is known to us by a vocabulary ; and known to be Cu- 
manch or Shoshoni call it which you will. I call them all 
Paduca, from a population so named by Pike. 

Now, out of twenty-one words common to the Utah and 
Moqui, eight are alike. 

Again, the Shoshoni and Sahaptin have several words in 
common, and those out of short vocabularies. 

Thirdly, the Shoshoni and Wihinast, though spoken within 
(comparatively) narrow limits, differ from each other more 
than the several forms of the Cumanch, though spread over a 
vast tract of land. 

The inference from this is, that the Paduca forms of 
South Oregon and Utah are in situ ; those of New Mexico, 
Texas, and New Leon, &c. being intrusive. In respect to 
these, I imagine that a line drawn from the south-eastern 
corner of the Utah Lake to the source of the Red or Salt Fork 
branch of the River Arkansas, would pass through a country 
nearly, if not wholly, Paduca; a country which would lie 
partly in Utah, partly in New Mexico, and partly in Kansas. 
It would cross the Rocky Mountains, or the watershed be- 


tween the drainages of the Colorado and the Missouri. It 
would lie along a high and barren country. It would have 
on its west the Navaho, Moqui, and Apatsh areas ; on its east 
certain Sioux tribes, and (further south) the Arapahos and 
Shyennes. It would begin in California and end in the parts 
about Tampico. 


The Cumanches, on the very verge, or within the tropics, 
vex by their predatory inroads the Mexican states of Zacatecas 
and Durango. Along with the Lipans they are the sparse 
occupants of the Bolson de Mapimi. Along with the Apaches 
they plunder the traders and travellers of Chihuhua. 

For the parts about Tampico the language belongs to the 
Huasteca branch of 

THE MAYA. The Maya succeeds the languages just enu- 
merated on the east. On the west, the Otomi, Pirinda, and 
Tarasca are succeeded by 

THE MEXICAN PROPER. But the Maya and Mexican 
Proper are languages of such importance, that the present 
paper will merely notify their presence in Mexico and Cen- 
tral America. 

The languages that, from their comparative obscurity, 
claim the attention of the investigator, are those which are 
other than Maya and other than Mexican Proper. 

Of these, the first succeeds the Huasteca of Huastecapan, 
or the parts about Tampico ; which it separates, or helps to 
separate, from the northern branches of the Maya Proper, 

THE TOTONACA of Vera Cruz, of which the following is the 
Paternoster ; the German being that of the Mithridates. 


Unser Vater o im Himmel steht 
Quintlatcane nac tiayan huil ; 

gemacht hoch werde dein Nahme 
Tacollalihuacahuanli 6 mi maocxot ; 

komme dein (m'cA?) 

Niquiminanin 6 mintacacchi 


gethan werde dein Wille 
Tacholahuanla 6 min pahuat 

wie wie im Himmel 

Cholei ix cacnitiet chalchix nac tiayan ; 

unser Brot, 
O quin chouhcan lacalliya 

uns gib heute 
niquilaixquiuh yanohue ; 

uns vergib unsre Sunde 

Caquilamatzancaniuh quintacallitcan 

wie wir vergeben 

Chonlei 6 quitnan lamatzancaniyauh 

unsern 'Schuldigern 
6 quintalac allaniyan ; 

Und nicht uns lasse 

Ca ala quilamactaxtoyauh 

damit wir stehen in Versuchuny 
Nali yojauh naca liyogai 

gethan werde 
Chontacholacahuanla . 

The same from Hervas. 

Kintaccan 6 natiayan huill ; 
Tacotllali huacahuanla o min paxca maocxot 
Camill omintagchi, 
Tacholaca huanla ixcacgnitiet ot 

skiniau chon cholacan ocnatiayan ; 
Alyanohue nikila ixkiu ki lacali chaocan ; 
Kilamatzancaniau kintacagllitcan 

Kintalacatlanian ochonkinan iclamatzan 

Caniau kintalacatlanian ; 
Nikilamapotaxtou ala nicliyolau 
lacotlanacatalit nikilamapotexto 
lamatzon lacacoltana. 


Cross the watershed from Vera Paz to Oaxaca, and you 
come to the area of 

THE MIXTECA. In the ordinary maps, Tepezcolula, on the 
boundaries of Oaxaca and Puebla, is the locality for its chief 
dialect, of which there are several. 



Dzutundoo, zo dzicani andihui ; 

Naca cuneihuando sasanine ; 

Nakisi santoniisini ; 

Nacahui nuuiiaihui saha yocuhui mini dzahuatnaha yocuhui andihui ; 

Dzitandoo yutnaa yutnaa tasinisindo hiutni ; 

Dzandooni cuachisindo dzaguatnaha yodzandoondoondi hindo suhani 

sindoo ; 

Huasi kihui fiahani nucuitandodzondo kuachi ; 
Tahui nahani ndihindo sahafiavvhuaka dzahua ; 

The Mixteca succeeds the Mexican Proper, itself being other 
than Mexican, just as the Totonaca succeeded the Huasteca, 
which was Maya, the Totonaca being other than Maya. 

The Mixteca is the language of Northern, 

THE ZAPOTECA that of Southern, Oaxaca. 

Hervas writes, that the Zapoteca, Mazateca, Chinanteca, 
and Mixe were allied. The Mixe locality is the district around 

South of the areas of the three languages just enumerated 
comes the main division of the Maya the Maya of Guatemala 
and Yucatan, as opposed to the Huasteca of the parts about 
Tampico. This, however, we pass over sicco pede, for 


Limiting ourselves to the districts that undeniably belong 
to those two States, we have samples of four dialects of 

THE LENCA language ; these being from the four Pueblos 
of Guajiquiro, Opatoro, Intibuca, and Sirmlaton, those of the 
last being shorter and less complete than the others. They 
are quite recent, and are to be found only in the Spanish edi- 
tion of Mr. Squier^s Notes on Central America. The English 
is without them. 


man taho amashe. 

woman . . move napu. 

boy .... ........ guagua hua. 

head. . . . toro tohoro cagasi. 



ear .... yang yan yangaga. 

eye .... saing saringla saring. 

nose .... napse napseh nepton. 

mouth . . ingh ambeingh .... ingori. 

tongue . . nafel navel napel. 

teeth . . nagha neas nigh. 

neck . . . ampshala .... ampshala .... cange. 

arm .... kenin kenin kening. 

fingers . . lasel gualalasel .... 

foot .... guagi quagi guaskaring. 

blood . . uahug uah quch. 

sun .... gasi gashi gashi. 

star .... siri siri 

fire .... uga 'ua 

water . . guass uash gnash. 

stone . . caa caa tupan. 

tree .... ili ili ill. 

one .... ita ita itaska. 

two ... naa 

three . . lagua 

four .... aria 

five .... saihe saihe 

six huie hue 

seven . . huis-ca 

eight . . teef-ca 

nine .... kaiapa 

ten .... isis issis 

As Mr. Squier is the sole authority for the Lenca of San 
Salvador and Honduras, so he is for 


Limiting ourselves to the undoubtedly Nicaraguan area, 
and taking no note of the Mexican Proper of more than 
one interesting Mexican settlement, the three forms of speech 
for which we have specimens are 



3. THE WULWA, of the Chontal district. 

And now we pass to the Debateable Ground. The language of 



gives us a fourth form of speech ; at least (I think) as different 
from the Choretega, Nagranda, Wulwa, and Lenca, as they 
are from each other. This is 

THE WAIKNA of the Indians of the coast, and, probably, of 
several allied tribes inland. 

Of the Waikna, Wulwa, Nagranda, and Choretega, sam- 
ples may be found either in Squier's Nicaragua, or vol. iii. of 
the Transactions of the American Ethnological Society. 


man rahpa miho. 

woman rapa-ku w-ahseyomo. 

boy sai-ka w-asome. 

girl sai-kee f . . . w-aheyum. 

child chichi w-aneyame. 

father ana goo-ha. 

mother .... autu goo-mo. 

husband .... a'mbin 'mhohue. 

wife a'guyu wume. 

son sacul-e w-asomeyamo. 

daughter . . . saicul-a w-asayme. 

r a'cu eoochemo. 

head < ,. 

\ edi 

hair tu'su membe. 

face enu grote. 

forehead .... guitu goola. 

ear nau nuhme. 

eye setu nahte. 

nose ta'co mungoo. 

mouth dahnu nunsu. 

tongue duhu greuhe. 

tooth semu nahe. 

foot naku graho. 

sky dehmalu nekupe. 

sun ahca numbu. 

star ucu nuete. 

fire ahku nahu. 

water eeia . . nimbu. 

fesee nugo. 

stone ^ 

I esenu. 


I ic-u saho. 

thou ic-a sumusheta. 

he ic-a 

we hechel-u semehmu. 

ye hechel-a 

they icanu 

this ca-la 

For the Waikna there are other materials. The Wulwa 
specimens are few. Hence it may be doubtful whether the 
real difference between it and the Waikna be so great as the 
following table suggests. 


man all waikna. 

woman .... y-all mairen. 

son pau-ni-ma lupia- waikna. 

daughter. . . . pau-co-ma lupia-mairen. 

head tunni let. 

" eye minik-taka nakro. 

nose magni-tak kamka. 

mouth dinibas . . bila. 

blood anassca tala. 

all duwawa . semehmu. 

drink mahuia bo-prima. 

run dagalnu bo-tupu. 

leap masiga bo-ora. 

f aiyu pa-ya. 

90 lieu 

sing nagamo pa-coondamu. 

ami pa-yacope. 


The following is from a vocubulary of Dr. Karl Scherzers 
of the languages of the Blanco, Valient e, and Talamenca Indians 
of Costa Rica, occupants of the parts between the River Zent 
and the Boca del Toro. We may call it a specimen of 

THE TALAMENCA. It seems to be, there or thereabouts, as 
different from the preceding languages as they are from each 







1 ENGL. 






. tshuko. 



water . . 

. ditzita. 

mouth, .... 


one .... 

. . e-tawa. 

beard .... 
neck-joint ? . 

-karku mezili. 
sa-fra- tzin-sek . 

four .... 
five .... 
six .... 
seven . . 

. ske-tewa. 
. . si-tawa. 
. . si-wo-ske-le. 
. si-wo-vfora.. 

finger .... 


eight. . . . 

. . *-M7o-magnana 





moon. . 


The same volume of the Transactions of the American 
Ethnological Society that supplies us with Mr. Squier's voca- 
bularies for Nicaragua supplies us with Dr. Seeman's for 

These being for 

The Cholo is the same as Dr. Cullen's Yule, and also the 
same as Cunacuna and Darien of Balbi and the Mithridates. 


.... conj ungo. 

.... poquah. 

.... pauquah. 

paque-cua pake-quah. 

atale eterrah. 

ner-cua indricah. 

cugle coogolah. 

vau-agua paukopah. 

nine paque-haguc pakekopah. 

ten ambegui anivego. 

It is also the same as some short specimens of the Mithri- 
dates; where 


one quensa-cua 

two vo-cua .... 

three paa-cua . . 




seven . . . 

eight . . . 


water dulah. 



wife (woman) = poonah. 

mother = naunah. 

The Cholo leads us into South America, where, for the 
present, we leave it. 


I will now add two notes, which may possibly save some future 
investigator an unremunerative search. 

First, concerning a language called Mocorosi. In Julg, this is 
made a language of Mexico. It is really the Moxa of South Ame- 
rica under an altered name. 


/ nuti nuti. 

thou piti piti. 

he '. . . ema ema. 

this maca , . . maca. 

that maena . t niaena. 

that you .... maro maro. 

the esu esu. 

my nuyee nuyee. 

thy piyee piyee. 

his mayee mayee. 

one eto eto. 

two api. api. 

three mopo mopo. 

This is from an Arte y vocabulario de la Lengua Mocorosi, com- 
puesto por un padre de la compania de Jesus missionero de la Pro- 
vincias de los Moxos dedicado a la Serenissima Reyna de los Angeles, 
siempre Virgen Maria, Patrona de estas Missiones ; en Madrid, ano 
de 1699. 

A Lima edition A.D. 1701 differs from this in omitting the name 
Mokorosi, and being dedicated to a different patron. In other 
respects the two works agree verbatim et literatim. 

Secondly, in respect to a language called Timuacuana. For this 



we have a Catechismo y examen para los que comulgan ex lengua 
Castellana y Timuquana, por el Padre Fr. Francisco Pareja; and 
y Padre de la Provincia de Santa Elena de la Florida, &c. Mexico, 

Also, the following numerals in Balbi, perhaps, taken from the 
above : 



one. . . . 


three . . nahapumina. 
four . . nacheketamima. 
Jive . . namaruama. 

six. . 
seven , 
nine , 


napikinahuma . 




[Read June the 27/A.] 

The Sanskrit ksh, a combination of k+s, is usually repre- 
sented in Greek and Latin by f x y <TK sc, KT ct. Some 
cognate words will exemplify this : 



aksha (axle) afo>v . 

dakshina (right) Sef to? , 

shash, Zend, khsvas* . . Fef . , 

kshura (razor) (vpov . 

makshu (quickly) .... . 

maksha (fly) . 

kshap (night, literally she") / 
who covers) J 






* Both a corruption of kshvaksh. The Greek form with the digamma 
occurs in the ' Tabulae Heracleenses.' 

f Mvm ought not as it generally is to be compared with these words, 
before it is shown that or <nc can be dropped between two vowels iji 
Greek, as is apparently, but only apparently, the case in Latin. I believe 
that /iula stands for /xvVta, and that this little animal, as well as p.vs, mus 
received its name from its propensity for stealing. 



riksha (bear) ap/cro? . . ursus (for urcsus) . 

, 7 , /,. i N , ftexere (basilicam, 

takshan (faber) TGKTWV . .4 

\_ naves). 

kshan (to kill) KTAN . . 

kshi (to kill) KTI . . . 

kshinumas /crtVu/u-e?. 

naksha-tra (star)* .... vv/cr . . . noct. 

Wherever the Sanskrit ksh agrees with KT in Greek, we must 
consider the latter as the older form, and the ksh as a cor- 
ruption, because s being weaker than /, can never, unless 
influenced by a subsequent mute, turn into the stronger 
sound. Sanskrit is as little able as Latin to bear kt at the 
beginning of words, but Sanskrit shows an additional weak- 
ness in never suffering it to stand at the end of roots. Forms 
like fleet, nect, pect, plect, TCKT, are impossible in Sanskrit. 
But all three languages have often transformed kt into some 
softer sound, and in many cases we are hardly able to trace 
the original form. I should not venture at present to prove 
the maxim, though I believe it will be confirmed hereafter, 
that every ksh, %, and x, found in the radical part of words, 
arose from kt. It may be interesting to show the different 
organic transmutations which kt might undergo in Greek and 

1. The k might be dropped. Compare KTVTTO^ and TVTTTW. 
kt } t. 

* This term rendered literally signifies ' watcher of the night,' from 
naksha for nakta (night) and tra (protector). The latter stands RV. I. 100, 
7. As naksha does not occur separately, it is probahle that the change 
took place in order to avoid the cacophony of two ? s in two adjoining 
unaccented syllables. Bopp and Benfey (S.V.) derive ndkshatra from 
naksh, without stating the meaning of that root. The native grammarians, 
as in most cases where a derivation does not lie on the surface, indulge in 
all kinds of absurdities. Yaska, one of the oldest, derives it from a verb 
naksh ' to go.' But as most things might be called from the same activity, 
and locomotion is not a very striking feature of the stars, and as naksh never 
signifies simply 'to go,' this etymology seems to be arbitrary. According to 
other grammarians, the stars are the imperishable, eternal, from na-{ kshar 
or na+kshi. This shows more sense, but less knowledge of grammar. 


2. The t is changed into s. Compare TeKTwv and texo, 
fixum for fie -turn, noxa for noc-ta. kt, ks. 

3. The initial k might be changed into p. 

4. The f and x might either be changed into GK, sc, or lose 
the initial guttural and appear as a and s. Compare 1, f /(/>o? 
and ova'009, maksha and musca. 2, %vv and <TVJ>, Zovvufo? 
and AtoWcro?, /mWu/u and <7tW9, Sextius and Sestius, mixtus 
and mistus, the Oscan Santia and Xantias. kt = &s, sA: ; 
kt=ks, s. 

Having laid down these rules, I shall proceed more safely 
in tracing the origin of sons at present, and of one or two 
Latin words hereafter. I readily believe that Festus is right 
in explaining sons by nocens. Qu. xiv. 1, 22, " Sons nocens, 
ut ex contrario insons innocens." Qu. xiii. 27, 24, " Son- 
ticum morbum in xii. significare ait Aelius Stilo certum cum 
justa causa, quem non nulli putant esse, qui noceat, quod 
sontes significat esse nocentes. Naevius ait : sonticam esse 
oportet causam, quam ob rem perdas mulierem." But the 
analogy between nocens, noxius, and sons, appears to me far 
more intimate than the Latin grammarians are aware of. As 
" nocere alicui " is nothing else but ' ' neci esse alicui," to be 
the cause of destruction, of death, to somebody, so sons sig- 
nified originally " destroying, killing," and, as every destroyer 
is held to account by the laws of society, passed from thence 
easily into the usual meaning of ' ' guilty." The original sig- 
nification appears clearly in sonticus morbus, a deadly disease, 
that is, a disease which either causes or threatens death. 
Compare Gellius, xx. 1, 27, "Ceteroquin morbum vehemen- 
tiorem, vim graviter nocendi habentem, legum istarum scrip- 
tores alio in loco non per se morbum, sed morbum sonticum 
appellant." An attack of such a disease excused a soldier 
from appearing at the appointed day of a levy, and stopped 
all farther proceedings in a lawsuit. Hence, or as I am more 
inclined to believe, from the fact that death and murder inspire 
the human mind with the greatest awe in any state of society, 
we find sonticus, but very rarely, in the sense of " extreme, 

* We have an analogy in the use of " deadly," for " extremely, exceed- 


I consider sons as one of those participles a small number 
of which remain in every language which have passed into 
adjectives and substantives, and are apparently unconnected 
with any primitive verb. Thus in Latin dent (edent), font 
(%eovr, or rather an obsolete %vvr, according to Pott*), fre- 
quent, clement, in Greek d/covr, e/covr (=Sanskr. ucant, willing, 
Pott), ryepovT, Spdfcovr. The verb to which sons belongs, is 
the Greek KTAN, in that shorter form KTA, which appears in 
the aorist e/crav, so that sont agrees in every respect with KTOVT 
(/cra?t)- KTAN, when turned into Latin, could after what 
I have previously said only become xan or scan, and if we 
suppose it took the first form at a time when the Latin could 
bear an x at the beginning of words, it was necessary at a later 
period to give up the guttural. In the same manner we find 
that the Greek crtvis is derived from KTI, a third form in which 
our verb appears. For /crav and KTI, we find in Sanskrit kshan 
and kshi. In Icelandic we have the verb KTA as ska. Com- 
pare Edda, 1110. 

Mjb'k er osviSr ef harm enn sparir 
f janda inn F o L K s K A ; 

' ' he is very unwise, if he any longer spares the man-hurting 
enemy." The neuter skae, hurt, occurs frequently. I find, 
for instance, a ship called, in the Fagrskinna, p. 21, bldmoerar 
skae, " the hurter of the blue plain." 

ingly," in some provincial dialects, as for instance, " a deadly lively child," 
for "a very lively child." The Dialect and Folk-lore of Northamptonshire, 
by Thomas Sternberg, p. 29 : John Noakes and Mary Styles, by Charles 
Clark, p. 38. 

* Kuhn in his Journal, iii. 399, proposes a new, but by no means supe- 
rior, derivation from the Sanskrit dhavant, currens, lavans, abluens. 

f With regard to the o, compare dos from dare, cos from *care, the par- 
ticiple of which we have in catus. 



[Read April the 25f&.] 

[Mr. JAMES YATES communicated to the Society " An Essay 
on the Irregularities in Homer's Versification." 

Mr. Yates stated that his Essay had been written many 
years ago, and had been originally communicated to a private 
Society bearing the same name as our own. As the essay ex- 
tended to a considerable length, Mr. Yates read only parts of 
it, omitting, besides other portions of it, very many lines 
cited from Homer in proof of his positions. 

Professor Maiden has kindly prepared the MS. for the 
press, retaining in general Mr. Yates' s words, and adding 
notes of his own, with the consent of the author.] 

In the poems attributed to Homer we perpetually find 
combinations of letters, which contradict the established rules 
of prosody. Among the ancients these irregularities seem to 
have excited little attention ; but by modern critics they have 
been placed among the most curious subjects of classical 
investigation. It will be the object of the following essay to 
explain the circumstances in which they occur, and the causes 
to which they are to be ascribed. 

Respecting these irregularities, it may be remarked as a 
universal principle, that they consist not in the excess, but in 
the deficiency of letters. For in every instance, the prosody 
may be made regular by the insertion of one or two additional 
letters. It will be proved, that, in many cases, such letters 
were originally in the text; but that, in others, the time 
which would have been occupied in their enunciation, was 
filled up in a different manner. 

As it would be impossible to ascertain in what instances 
letters have been omitted out of the original text, until we 
have determined what latitude was taken in deviating from 
the general rules of prosody, it appears proper to attend, first, 
to those cases in which the time was completed without the 
use of letters subsequently expunged. 

I. In the first place, the time was often occupied by a 


pause depending upon the sense. Any one who recites poetry 
uses the liberty of making such pauses longer or shorter, 
and more or less distinct, according to his own taste and 
choice. Examples of this license affecting the versification 
present themselves with a variety of circumstances. A short 
vowel at the end of a word is not cut off, and a diphthong or 
long vowel retains its time, although the word following begins 
with another vowel; and where a consonant intervenes, a 
syllable naturally short is used as long. Thus in 
II. vi. v. 46. Zwypei, 'Arpeo? vie. 

viii. v. 120. Tibv virepOv/jLov Syftafov, 'Hwo7n}a. 

ib. v. 105. *A\X' ay ejjLwv o^ecov eiriftya-eo, 6<f)pa iSvjai. 

ib. v. 158. AvTt? av ICOXJJLOV eVl Se TpcSe? re /cai'l&KTcop. 

The license occurs, whether the syllable be the first, second, 
or third of a foot"* ; and whether there be, or be not, a caesura. 
The principle here stated is an obvious one, arising from the 
nature of sound, which necessarily occupies time, and of lan- 
guage, which requires that the time, usually given to sound, 
be occupied at intervals by pauses. The effect of the pause 
has been recognized by some of the most distinguished writers 
upon this subject f- 

II. Besides the pause, the time necessary to complete the 
metre was, in many cases, filled up by lengthening the sound 
of syllables naturally short, or retaining short vowels, which 

* In the first syllable of a foot more especially, the principle to be 
mentioned next comes into play likewise. ED. 

f Eandem, nisi majorem, efficaciam habet interpunctio, quse brevem 
syllabam excipit. Cujus generis longe plurima Ilias et Odyssea exempla 
suppeditant, et tempus, sive moram, quse syllabse deerat, pausa explent, 

ut etiamnum musici nostri facere consueverunt Neque autem hanc 

numero ipsi insitam vim veteres prorsus effugisse statuendum est. Jam 
Eustathius ad II. . 265, p. 645, vulgatam scripturam, peveos, d\K^s re 
\dda>p,aL, eo nomine defendit, scribens : TO Se ptveos eVreiWt fiTpikcos evravda 
rrjv \r)yovo~av dia TTJV dpearKOVcrav 'Aptorap^o) TfXeiav o"Tiyp.r)V, Kal TO eV avTrj 
ovTO) %poviov Kal (TTa.o-ip.ov T^S (pavrjs Xdyo) KOIVIJS o-vXXaftrjS. Profecto 
autem in aprico est, earn ob rem nonnunquam syllabas brevissimas produci, 
quibus alio modo jus illud vix concedendum esset; idque, quo gravior, 
quse a tergo quasi instat, interpunctio est, eo lubentius admitti posse. 
Spitzner, De Versu Grsecorum Heroico, p. 20. 


according to the usual practice would be cut off. This prin- 
ciple may be deduced, almost as obviously as the last, from 
the nature of speech and of verse. Since metre consists in 
the succession of long and short sounds arranged in a certain 
order, any one who recites verses, will, through the force of 
habit, become disposed to enunciate long and short syllables 
in their proper metrical places according to the prescribed 
arrangement, although their times are not represented by the 
letters before his eyes. Yielding to this propensity, he will 
supply the deficiencies of the metre by dwelling upon those 
syllables, the shortening or elision of which would interrupt 
its regularity. 

This prolongation, or retention, of short syllables in accom- 
modation to the metre may take place, whether those syl- 
lables are the final or initial, or, in some cases, even medial 
syllables of a word. Hence this mode of supplying metrical 
deficiencies is far more frequent than that already described, 
which can be employed only in final syllables : and although 
the admission of this license is restricted by a regard to the 
necessity of determining the metre by a succession of syllables 
which are of the proper length in their own nature, yet the 
instances of its adoption are much more numerous than the 
cases in which the metre is completed by the intervention of 
a pause. The following passages selected from the 8th book 
of the Iliad are examples : 

v. 13. "H fjLLV e\a)i> pty\o) e? | Tdprapov. v. 25. 7re\pl piov \ 
Ovkvjji'jTOLo. v. 66. a|efero | lepov rfpap. v. 229. Tli} e/Sav. 
v. 248. re|/co9 e\d\(f>oio ra^eirj^. v. 262. 6ov\piv e7ri\eip,evoi 
a\Kijv. v. 267. <7a,K\i T\a\/jiQ)Vi,dSao. v. 290. Bv\a) iTrfrovs. 
w. 300 & 309. aJTro vev\pri^iv t'aXXev. v. 324. B^ice % 4\irl 
vev\py. v. 359. (f>8lft\vo<} ev \ TrarplSi, yairj. v. 392. "Uprj \ Se 
liao\Ti/i. v. 473. 7ro\e\fjLov d7ro\7rava-eTcu. v. 474. \7rplv 
wp\0au. v. 517. Atjl <j>i\oi | dyy6\\6vTO)V. 

The cases referable to the principle here described, consist, 
first, of diphthongs or long vowels, which retain their time, 
instead of becoming short, before another vowel ; secondly, of 
short vowels (a, e, i, o), which in the same situations do not 
suffer elision; and, thirdly, of short vowels, followed by a 



single consonant either at the end of the same word or at the 
beginning of the next, which supply the place of long syl- 
lables. In most, if not in all cases of this third class, the 
syllable was made long by dwelling, not upon the vowel, but 
upon the consonant. This may be inferred from the analogy 
of those syllables, which are not final, but in which the 
same prolongation occurs. In these the consonant is usually 
written double, as in 'A^AXeu?, 'OSuo-o-ei;?, OTTTTT?, OTTL. In 
the few cases where it is not so doubled, such as o/3/jtyu,o? 
"Apr)?, we may infer from analogy, that it was pronounced, 
though not written, double* ; and by extending the analogy 
to the final syllables, which were lengthened upon the same 
principle, we may conclude, that in them also the consonant 
is to be pronounced twice, though written only once. A 
circumstance which confirms this doctrine is, that the pro- 
longation of short syllables, whether at the end of words, or 
not at the end, commonly took place before the consonants, 
which were most readily dwelt upon, or doubled, in pronun- 
ciation, namely, the four liquids, X, //,, v, p, and the letter cr. 
This tends to prove that the prolongation depended upon the 
consonant rather than upon the vowel. According to this 
view the instances of short vowels used as long, cited from the 
8th Iliad, ought to be pronounced as follows : ireplp piov 
OV\V/JLTTOIO, re/co? <ie\d<f>oio Ta^6/7;9, Oovpw 
dirov vevpr)(])Wy eirlv vevpy, <f)0l/jievo<> aev TrarplBi, 

iy irplv v&pQai, ; and perhaps o-d/ceir Te\afjico- 

It has been usual with the authors, who have treated of 
these irregularities, to say that they were occasioned by the 
caesura. A long section in Spitzner's Treatise is entitled De 

* That the consonant was not always doubled appears from instances 
in which a short vowel was changed into a diphthong or a corresponding 
long vowel, as in ov\6p.evos for 6\6p.evos, Ov\vp.iroio from "OXvpfros, and the 
adjectives r}vep,6eis, TjpaOoeis, from the nouns avep-os, a^aOos. There is no 
reason why a vowel should not be lengthened before a consonant, as well as 
before another vowel, as in e'iapos from the nominative cap, and the fami- 
liar forms xpv<reioy, xdXKetos. ED. 

f Hardly Au< </>i'Aot, since the aspirate consonants are never doubled. 
Here, more probably, the final vowel was lengthened, Atf $i\ot. ED. 


syllabis in vocabulorum fine caesurce vi productis* ; and Her- 
mann, in a disquisition De productionibus ob c<BSuram-\, says, 
"His constat nihil apud Homerum atque Hesiodum esse 

The force of the csesura appears to be assumed by these 
authors. They offer no evidence whatsoever in proof of its 
reality. It is true, that there very frequently is a csesura 
where this prolongation occurs. In other words, it is a fact, 
that the prolonged final syllable of a word is seldom the final 
syllable of a foot. This arises from two causes : first, because, 
as the first syllable of every foot is long, the reciter habitually 
expects a long syllable at the regular intervals, and therefore 
is ready to make a syllable long in that place, even though it 
be not long in itself; but as the feet may be dactyls or 
spondees, there is no habit of recitation which leads to the 
lengthening of the second syllable of a foot, and consequently 
the lengthening of a syllable in that place is comparatively 
rare : and, secondly, because, in the structure of hexameter 
verse, it was a general rule to avoid the well-known want of 
euphony, which results from feet ending at the end of a word. 
The authors, who assign the csesura as the cause of the pro- 
longation of short syllables, do not offer any reason why the 
csesura should have this effect ; nor is it possible to prove that 
there was any connexion of the one circumstance with the 
other, except the coincidence arising from the structure of the 
verse, which has been pointed out. Examples of prolongation 
without csesura are necessarily uncommon, but they are not 
unknown. We find the following instances of short syllables 
used as long: TroXXa \ Xto-cro/u-ej/^, II. e. 358. TroXXa | pva-- 
rd^ea-KVy co. 755. /SXoo-upcoTTY? | e'crre^avfOTO, X. 36. irplv | 
eX0etv, v. 172, %. 156. irplv \ ovrda-cu, TT. 322; and the 
instances in which a short vowel at the end of a word retains 
its quantity, instead of suffering elision, before a word begin- 
ning with a vowel, as in 6. 66, aegero \ lepov 17 pap, are sub- 
stantially of the same nature. 

* De Versu Grsecorum Heroico, inaxime Homerico, cap. ii. sec. i. pp. 
f Orphica, pp. 69/-720. 


While most authors have ascribed the irregularities in 
question to the caesura, some have attributed them to the 
Ictus metricus or Arsis. This account, though it has the 
advantage of being applicable to the initial and medial syl- 
lables of words, as well as to final syllables, does not on the 
whole appear more satisfactory than that which has been 
considered. There is a confusion in the use of the terms Ictus 
and Arsis. Some authors use these words as synonymous, 
while others employ them in distinct senses. According to 
some, Arsis was the raising of the voice ; according to others, 
the raising of the hand or foot in beating time. Bockh 
indeed has proved, that, in the language of the ancient Greek 
writers on music and metre, Arsis meant the raising of the 
hand or foot ; Thesis, the putting it down ; and consequently 
that the Thesis coincided with the elevation of the voice, 
which Bentley, and the modern writers who have followed 
him, call Arsis*. The learned editor of MorelFs Thesaurus, 
who favours the doctrine now under review, has collected the 
principal definitions of Arsis -f, from which one thing at least 
is evident, that the meaning of the term is unsettled, and 
consequently that any theory, which attributes metrical irre- 
gularities to arsis, must be obscure. Professor Dunbar of 
Edinburgh says rightly, that " in hexameter verse the ictus, 
or arsis (using the words as synonymous) is always upon the 
first syllable of the foot J :" but, if so, no theory of arsis will 
explain irregularities in the second and third syllables. It 
appears also, that the writers who speak of arsis, consider it 
as something which affects either the accent of syllables, 
giving them a higher tone on the musical scale, or the loud- 
ness and strength of the voice in uttering them, rather than 
the time occupied in their enunciation ; so that after all, the 

* Priscian, in writing upon accents, not upon metre, applies the terms 
to the voice, and says that the syllables of a word, up to the accented 
syllable inclusively, are in arsi, and the remaining syllables in thesi. This 
use has misled modem metrical writers. ED. 

f See Maltby's valuable " Observations, " prefixed to Morell's The- 
saurus, cap. iii. 2. 

J Prosodia Graeca, p. 24. 


doctrine of the arsis does not reach the case to which it is 
intended to apply, and which has to do, not with the accent 
of syllables, nor with their loudness, but with the slowness or 
rapidity of their enunciation. 

If the opinions of those authors, who attribute the various 
usages in question either to caesura, or to arsis, be unsa- 
tisfactory, there is the greater reason to believe, that Homer 
did not acknowledge the formal restraints of inviolable rules, 
but chose occasionally to employ combinations of sounds, 
which, though offensive if too frequent, give an agreeable 
variety to the versification of a long poem, when admitted in 
moderate proportion, and which require from the reader 
slight and appropriate modifications, which are easy and 
natural to him, because coinciding with the general strain of 
the metre. And if it be admitted, that habit thus operated 
in preserving the regularity of Homer's verses, when they 
were uttered aloud, it is evident that, since the first syllable 
of every foot was long, and since the reciter would conse- 
quently be more disposed and prepared to supply any deficiency 
in the first syllable than in the second or third, the same 
principle which explains these irregularities in all their variety, 
shows also why they were admitted most frequently at the 
beginning of the foot. 

III. Having ascertained what licenses the poet himself 
used in constructing his verses, we may now proceed to 
determine what irregularities have arisen from the omission 
of letters originally belonging to the words which he em- 
ployed. The letters so omitted were principally, if not solely, 
two, F, called Van, and S, called San or Sigma. These were 
not mere breathings, nor arbitrary and occasional modifi- 
cations of the words to which they belonged, but constituent 
parts of them, which in the early stages of the language were 
uttered as distinctly, as fully, and as constantly, as the other 
letters in the same words. 

The existence of F, as a letter of the primitive Greek 
alphabet, appears from the testimony of Dionysius of Hali- 
carnassus, and many others of the ancient critics and gram- 
marians. By most of them it is called the Molic Digamma, 


because its form was that of a double gamma, and because it 
continued in use among those who employed the ^olic dialect, 
after it was rejected by the other Greeks. The fact testified 
by these authors is confirmed by the use of this letter among 
the Greeks to the latest times as a numeral standing for 6, 
and by the existence of the same letter, occupying the same 
place, in alphabets, which had the same origin with the Greek, 
and which exhibit a general correspondence with it. To 
these evidences, which were long esteemed sufficient by the 
ablest judges, we may now add the actual appearance of this 
letter in ancient inscriptions. 

The sagacity of Bentley long ago assigned a place to F in 
particular words, because he observed its effect in rectifying 
the prosody of Homer, and because he noticed the existence of 
letters having the same sound (V in Latin, and W in English) 
in corresponding Latin and English words. It is a remark- 
able coincidence of fact with theory, and a singularly strong 
confirmation of the general doctrine of this great critic, that 
the digamma has been found in coins and marbles at the 
commencement of the very same words to which he had pre- 
fixed it in his copy of Homer. Since his time also, the very 
curious and important inquiry into the analogy of languages 
has been pushed much further, and has furnished decisive 
proofs of the accuracy of his conclusions in many instances. 
The languages which are more particularly allied to the Greek, 
and which are consequently subservient to the illustration of 
this subject, are the Latin, and the other ancient languages of 
Italy; the Mceso- Gothic, and other languages of the Teutonic 
stock, including our own Saxon; the Sanscrit; and the 
ancient Persic. 

With respect to the sound of F in those Homeric words, 
from which it has been excluded, nothing has yet been brought 
forward sufficient to shake the opinion originally advanced by 
Bentley, that it was that of the English W*. The argument 
founded upon the fancied harshness of this sound, though 

* That this was the opinion of Bentley appears from Clarke's note on 
IliadTr. 172. 


principally relied on in opposition to the common opinion*, 
appears very futile. The later Greeks did esteem it an 
offensive sound, and therefore rejected it. But their prede- 
\ cessors, we may- be assured, perceived no more coarseness in 
\it than the modern English do, when they employ it in 
sXoeaking their mother tongue. If scholars of the present day 
are unable to endure the insertion of W before epyov and 
othsr Homeric words, it is because they are unaccustomed to 
it in those situations. They never complain of the frequent 
occurrence of this sound in the lines of Milton, Pope, or 
Spenser; and would exclaim with vehemence against any 
proposal to improve their euphony by expunging W, or sub- 
stituting F or V in its place. Even in Homer they are habitu- 
ated to the sound in certain words, such as vlos and pzpavia,, 
and in them they perceive no harshness. The arguments 
which conspire to show that F was pronounced like the 
English W, have been so often stated, that it is unnecessary 
to repeat them here. The reader is referred to 'Foster on 
Accent and Quantity/ pp. 127-130; Burgess, e Adnotatio in 
Dawesii Miscellanea Critica/ p. 422 ; and the Critical Review 
for February 1817, pp. 112, 113f. 

The general considerations which have been now stated, 
respecting the restoration of the digamma to Homer's poems, 
will be best illustrated by a review of the principal words 
which began with this letter in his time, and must begin with 
it now, in order that his prosody may exhibit its original 
degree of regularity. 

"Ao-n;, a City. 

The use of aa-rv in Homer requires the introduction of an 
initial consonant. It is true that a great number of the 
instances in which it occurs, afford no evidence upon the sub- 
ject, inasmuch as it is found at the beginning of a line, or 

* Marsh's Horse Pelasgicae, chap. iv. 5. 

t There is no more harshness in an initial F in Greek, than in an initial 
W in English. The difficulty in pronunciation occurs where the f would 
appear between two vowels, as in the augmented tenses of verbs beginning 
with an F, especially if the second syllable is short, as in the aorists efiftov, 
(fadov, and where there is a reduplication, as in fffouca, Fffopya. ED. 


preceded by a long syllable, or by v e^eX/euart/cov. With 
these perhaps may be included forty-four passages in which 
it is preceded by the prepositions ire pi, TTOTL, Trporl, and irpo, 
in which the final short vowel is not subject to elision. But 
besides these, there are more than forty which favour the 
admission of an initial consonant, and only seven which are 
against it. In some of the verses which oppose the insertion 
of an initial consonant, it may be introduced by very slight 
alterations of the text. Thus, in II. 7. 140, for 'Avfy>6? re 
Trporepoto Kal do-reos, we may read with Heyne, ai/S/oo? re 
Trporepov KOI Fdareos. In or. 207, e acreo?, and ft). 320, 
vTrep aoreo?, we may substitute e/c and Bid, the latter word 
being supported by the evidence of numerous MSS. now 
extant, and by the testimony of scholia, respecting those of 
ancient times. There remain X. 732, d/ji(f>laravTo Sr) aa-rv; 
a. 274, ego/jiev, d<rrv B Trvpyoi; Od. p. 25, eicdOev Be roi 
do-rv, and in II. o. 455, there is an elision before the proper 
name, 'A<JTWO&>. 

The inscription FAST appears on a coin published by 
Goltz*. To what place this inscription referred is uncertain. 
Havercamp supposes it to stand for Fatrru/^vaV, meaning the 
inhabitants of "Aorupt9, a city in Bceotia, mentioned by Ste- 
phanus Byzantinusf. But there can be no doubt, that what- 
ever city might be indicated by this inscription, its name was 
derived from the noun, the primitive form of which is the 
subject of this investigation. 

"h.<TTv or Fao-, is the Sanskrit vdstu, site of a habitation ; 
and probably contains the same root as words of a similar 
meaning, which ran through the ancient Teutonic languages, 
appearing, for example, in Suio-Gothic, as FASTE, a citadel 
or fortification (Hire's Glossar. Suio-Goth. vol. i. p. 437), 
and represented in our own language by FASTNESS. The 

* Graeciae ejusque Insularum Numismata, Tab. xvii. (quoted by Haver- 
camp, Sylloge Scriptorum, p. 275). 

f " In nummis Bceoticis apud Eckhel. Doctr. Num. ii. p. 196 legitur : 
Kvfapa et Fatrr. quod supplendum esse videtur Faorv (aoru)." Savelsberg 
De Digammo, cap. ii. p. 5. The same writer gives Fao-rv/^iSdirios on the 
authority of Ulrich, Iter Graccum, vol. i. p. 247. ED. 


French verb BATIR, anciently B ASTIR, with its derivatives 
Bastile, Bastion, &c., may be of the same origin. 

, a Word-, EtTrov, I said-, *O^r, "CWa, Voice. 
An inscription, discovered in Elis, and brought to England 
by Sir W. Gell, contains the word FEHO2, word, thus 
written with the digamma"^. Dawes has filled more than 
three pages of his ' Miscellanea Critica' with those instances 
of the occurrence of this word in Homer, in which it is pre- 
ceded by a short syllable ending in a consonant, and that 
syllable is made long. Spitzner admits, that the number of 
verses which seem to require the introduction of the digamma 
before eVo?, is almost infinite. He nevertheless contends that 
a large proportion oppose its use. His collection, which is 
copious, and apparently complete, contains twenty-six exam- 
ples from the Iliad, in which the words eVo? and etTrov, in 
different positions, seem to refuse the digamma, besides rj. 68, 
and many other lines, in which the phrase o<j)p ewco occurs. 
In no less than twelve of these instances, the dative plural in 
the form eVeeo-crt is preceded by an elision; and the lines 
may be corrected by the substitution of FeTrecrcrt, with the pre- 
ceding vowel not elided : e. g. in e. 893, for Bd/jLvrj/ju e-Treecrcrt, 
we may read SdfjLvrjfjii Peireo-ai,. In some other instances the 
genuine form may be restored by alterations equally slight. 
For example, in /3. 213 and 342, for o? p eirea and yap p 
eTreeo-o-', we may read o? FeTrea, and yap /eTreeoV : and in 
i). 375, for KOI Be roS' elir eleven, simply KOI roSe FeLTre^evai. 
These, and some other corrections, are no greater than those 
which a modern critic conceives himself entitled to make in 
conformity with any established principle of language; and 
the reverse alterations are exactly such as the ancient editors 
of Homer would introduce upon their own authority, in order 
to correct the irregularities arising from the omission of the 
digamma. The remaining instances, the correction of which 
is less easy, do not present an amount of evidence sufficient 
to counterbalance, or even to throw doubt upon, the evidence 
for the existence of the digamma in this family of words ; 
* Bockh. Corp. Inscrip. t.i. p. 26, n. 11. 


especially as some at least of the lines are liable to the sus- 
picion of interpolation. 

The insertion of F explains also the frequent occurrence of 
the second aorist with the syllabic augment, ee^Trov, eeuTre, 
and in compounds, as fjuereeiTre, Trpocreetire', and the occur- 
rence of compounds, in which the short vowel of a preposition 
is not elided, as aTroetTrwv, Siaenrefjuev. 

Heyne has maintained that in compounds the F was often 

omitted. But the instances which favour this opinion are 

neither numerous, nor very decisive. In II. a. 555, we find 

NOv 8' atVo>9 SeiSoi/ca Kara cfrpeva, fitf ere 

In the expression al'o^/m TrdpeiTrcov (i. e. TrapFeisrrwV) II. . 
62, 77. 121), the verb is followed by an accusative of the thing 
spoken. It is used without any case after it in II. \. 792, 
o. 404, 

ei K,kv ol <rvv Bai/jLon 0v/j,bv opivco 

And in the only other passage, where it occurs, II. f. 337, 
Nv Se fj,6 TrapFeiTrova aXo^o? ^akaKoicri FeTreao-w 
" lp/jLr)cr* e? TroXe/xov, 

/z-e may be considered as governed by wp^ce. The genuine 
reading, therefore, of II. a. 555, may be fjurj Trap Ferny, without 
the accusative of the person. 

The passages which remain, are 

II. K. 425, evSovv, r) dirdvevde ; Siewe pot,, o(/>pa Saeico, 
II. T. 75, fjifjvw aTreiTTovros fie^/aOvfiov 
Od. a. 91, fJLvr]<TTripe(T(Tiv aTrenre^ev. 
In the last passage we may read pvrja-Tripecrcr 
With regard to II. T. 75, according to Bentley, Heyne, and 
Knight, the whole verse is spurious ; and certainly it may be 
spared with advantage rather than injury to the passage*. 

"O'v/r, the voice, being in all probability of the same origin 
as eVo?, word, had, like it, the digamma. In thirteen places 

* It is doubtful whether it is worth while to try to correct the parti- 
cular line, II. K. 425. Some critics, not without reason, believe this whole 
book to be of later date than the main bulk of the Iliad. ED. 


where it occurs in Homer, the circumstances are such as to 
afford no evidence either for or against the introduction of 
this additional letter. But thirteen others require its ad- 
mission to make the prosody regular ; and three only would 
require alteration. By the usual substitution of aa for K, the K 
itself taking the place of TT, Foo-aa was produced, which also 
the prosody of Homer requires. 

That the words belonging to this root originally began with 
F, and that this sound was a component part of the root, and 
that the p sound at the end of the root varied to k, is manifest, 
from a comparison of the cognate tongues. In Latin we have 
vox, voc-is, voice, voca-re, to call, in Sanskrit, vdch, voice, 
vach, to talk. The Moeso-Gothic vopsan, to cry aloud, to 
shout, and the English whoop, probably represent the same 
root with the final p. 

"Epyov, Work. 

The Elean Inscription, to which reference has been made 
under this last head, contains the word written FAPFON. 
The expression is, At 8e n Seot, aire Feno?, cure Fdpyov : " If 
there be occasion for anything, either to be said or done/' 
This combination was frequent in Greek, like the phrase, 
" Aut dicto aut facto" in Latin, and "Rath und That" in 
German, or "word or deed" in English. Hermann, in a 
note to one of the Homeric Hymns (Hymn, ad Ven. v. 86, 
p. 92) states the various prosodiacal circumstances in which 
this root (epy) is found, and refers to more than sixty passages 
in the Iliad and Odyssey, where the measure is rendered 
complete by the insertion of the digamma before it. The 
form Fapyov for Fepyov in the Elean Inscription, is like other 
antiquated forms, which the grammarians call ^Eolic, such as 
al and alVe just before, for el and etre. 

The insertion of the digamma before this word is coun- 
tenanced by strong analogies in the cognate languages; in 
Moeso-Gothic, Waurk or Waurg, work, and the verb Waurk- 
jan, to work; in the Suio-Gothic, Werk, opus (Ihre, Gloss. S. 
Goth. p. 1096, t. ii.); Alemannic, Werch-, Anglo-Saxon 
Weorc, and the verb wyrcan. Hence the modern German 


and Dutch Werk, and the English Work, and the Lowland- 
Scotch Wark. 

"Er7?, Fellow-citizen. 

In the Elean Inscription we further find FETA2, meaning 
a private citizen (Bockh. Corp. Ins. p. 31), as opposed to a 
magistrate, reXeo-ra, or to a small community or village, 
Sa/j,o<?. As this inscription throws light on the sense, as well 
as the form, of this word, it may be worth while with the help 
of it to examine the several instances of the word in Homer. 

II. f. 239. When Hector comes from the field of battle to 
the Scsean Gates, the wives and daughters of the Trojans 
flock around him, 

Wipopevai TratSa? re, /caa-iyvvjTOVs re, /era? re, 

11 inquiring after their sons, brothers, fellow-citizens, and 


II. ij. 295. Hector advises Ajax to relinquish the contest, 

so as to gratiiy all the Greeks, but especially his own fellow- 

citizens, and his particular friends : 

f fl? crv T evcftprfvys irdvra^ Trapa v^vcrlv 'A^atou?, 
2)ou? re fj,d\ia-Ta fera?, Kal eraipovs, 01 rot eaatv. 

This passage illustrates the difference between fer^? and 
eralpos. Hesychius and Apollonius, in the usual vague 
manner of the ancient lexicographers, represent these words 
as synonymous ; and one of the ancient scholiasts uses their 
assumed identity of signification as a reason for marking the 
verse as spurious (Heyne ad loc.) But we cannot have a 
more decisive proof of its genuineness, than that it contains a 
word used in its exact and proper sense, and in its ancient 
form, the meaning of which was generally forgotten in the 
time of the grammarians. The difference between the two 
words is this : Ferca denotes those joined by citizenship 
eralpoi, those joined by familiarity and friendship. 

II. t. 460. ^H yu-ev 7ro\\a Feral Kal dvetyiol d/jL<f>l$ eovre?. 

Here Phoenix speaks of his fellow-citizens, Ferai, as opposed 
to his cousins or distant relatives, dv\fnol. 


II. TT. 456. ev6a (re rap^vaovai fcaa-iyvrjTol re Ferai re. 
The corpse of Sarpedon was to be conveyed to Lycia, his 
native land, where his brethren and fellow-citizens were to 
raise a tomb over him. 

II. f. 262. w? TVW} KeKfiTjrcas, a^vvwv <jolvi Ferrjcri,. 
i. e. " as thou art wearied, defending thy fellow -citizens" 
The words are addressed to Hector by his mother, on his 
return from the battle. 

In the Odyssey the word occurs only twice, Od. 8. 3, and 
16; where we are told, that the neighbours and fellow- 
citizens of Menelaus 

Tetrove? r)8e Ferai MeveXaou Kv^akifAoio, 
were feasting in his palace on occasion of the marriages of his 
son and daughter. The Scholia published by Mai (p. 120, 
ed. Buttmann) give the true explanation : "Ercu Se, ol etc TT;? 
TroXeo)?, ol a-vvr)6ei<;. 

"Eros, Year. 

This word is found written with the digamma, not only in 
the Elean Inscription, but also in the Heraclean Tablets, and 
in the Orchomenian Inscription now in the British Museum 
among the Elgin Marbles*. It requires an initial consonant 
to complete the prosody in Homer. 

Et'/eo<n, Twenty. 

The Orchomenian Marble presents this word written 
FIKATI. In the Heraclean Tablets it is written LEIKATI, 
the digamma having here the form E, which has been con- 
founded with Sf- 

In eleven passages the metre of Homer requires that ei/coai 
should begin with a consonant : viz. II. j3. 510, 748; v. 260; 
o. 678; ^. 264; G>. 765; Od. ft. 212; 8. 669; i. 209, 241 ; 
K. 208. 

A slight alteration is required in the following passages : 
viz. the omission of a redundant re in II. t. 379; %. 349; 

* On this inscription, see Bockh, Corp. Inscr., vol. i, p. 740. 
f Mazocchi, Tab. Heracl. Napoli, 1754, fol. 


Od. TT. 249 ; of xpvcrov for ^pva-oto in II. X. 25, and of r(\6ov 
for rjKvOov in Od. <ir. 206, r. 484, <. 208, w. 322 ; and the 
omission of K in Od. e. 34. 

The digamma at the beginning of this numeral is probably 
the v of Svo. In some languages the first letter d or t is 
retained : thus MOBSO- Gothic twaimtig ; German zwanzig ; 
Anglo-Saxon and Dutch twentig. In others the initial con- 
sonant is dropped : thus Sanskrit vinqati ; Latin viginti y 
from which come the French, Italian, Spanish, &c. ; and 
Greek h/cari, Pei/cart,, fei/coai,. 

The latter part of this word (tiff in the Teutonic languages; 
cat in the Sanskrit; gint in the Latin; and KOLT or KOO- in the 
Greek) seems to be a modification of daqan, decem, Se/ea, ten ; 
so that twain-tiff, &c. signify two tens. 

, Elis. 

The inhabitants of Elis are called in the before-mentioned 
Inscription, brought from that country, FAAEIOI ; and 
various coins are represented by Goltz (pi. 35, 36), Pellerin 
(Recueil de Medailles, torn. i. pi. 10), and Combe (Hunt. 
Mus. No. 21, 22), as having the inscription FAAEION more 
or less curtailed. Mr. Spencer Stanhope, in his splendid 
work on 'Olympia' (London, 1824, folio), has published 
eleven coins with the first two letters only, FA, and two with 
the entire word FAAEION. There can be little doubt, that 
the name of this country was still used in its ancient form 
when Homer wrote. 

Ot/eo?, House. 

A brazen tablet, discovered near the site of the ancient 
Petilia, contains the word FOIKIAN, very distinctly written* : 
and in an inscription from Orchomenus in Bo3otia, published 
by Leake in the Classical Journal, we find FTKIAS, v being 
substituted for 01, as in numerous examples of the same 
dialect f. 

* Marsh's Horae Pelasgicse, pp. 60-62; Bockh, Corp. Inscr. n. 4, p. 11. 

f Class. Journal, vol. xiii. p. 332. Bockh, Staats-haushaltung, vol. ii. 
p. 398, and Corp. Inscr. i. n. 1562. According to Bockh, this inscription 
is as late as Alexander the Great. 


The rules of prosody require that the digamma be restored 
to ot/co? and its derivatives throughout the poems attributed to 
Homer. For although a certain number of passages may 
require amendment in order to admit it, they are very few 
compared with those passages in which the prosody is ren- 
dered perfect by the insertion of it. 

The root exhibits a corresponding form in the cognate 
languages : Sanskrit viq ; Latin vicus, a village ; Anglo-Saxon 
wick, a house or castle ; Armorican gwic, villa (see Ihre, Gloss. 
Suio-Goth. Prooem. p. xi.). 

v Apve9, Lambs. 

A Ta//,/a? or treasurer of Orchomenus is mentioned in an 
inscription from Boeotia, among the Elgin Marbles, by the 
name FAPNHN (Walpole's Memoirs relating to European 
and Asiatic Turkey, p. 474*), and Greek proper names in -o>v, 
-wvo?, were sometimes taken from the names of animals, as 
Av/ccov, -ft>vo9, and ^dpvwv may have been derived from the 
noun signifying a lamb. 

Previously to the discovery of this inscription, Heyne had 
placed this noun (of which the nominative singular is not 
found, but the other cases are apvo?, dpvi, &c.) in his cata- 
logue of digammated words, although the metrical evidence 
is not by any means so decisive in this as in most other 
instances f. 

"Aryvu/u, I break. 

The occurrence of F on ancient coins, and in inscriptions, 
in some of the same words, to which it had been ascribed with 
a view to the correction of the irregularities in the versifi- 
cation of Homer, presents the strongest confirmation of the 
hypothesis which could have been desired, and justifies the 
insertion of that letter at the commencement of other words, 
where the prosody requires it, and where the addition is sup- 
ported by the existence of the same or equivalent letters in 
corresponding words in any of the cognate languages. 

* Corp. Inscr. i. n. 1569. 

f Cf. Sanskrit urnd, wool, from the root vr. The affinity has been 
noticed by Pott, Etym. Forsch. ED. 


Proceeding in alphabetical order, we shall first consider 
the evidence, that ayvv/jii, I break, was written Fayvv/ju. 
There are eight passages in Homer, which have such words 
as ap/j,ara and av%eva before the verb, without the loss of 
their final short vowel, and which therefore require an initial 
consonant; to which may be added fifteen instances of the 
first aorist eafa, or second aorist passive edyrjv, formed with 
the syllabic instead of the temporal augment. Two passages 
require alteration, viz. II. i/r. 392, "iTTTmov Be ol ^f e Oea vyov, 
and Od. T. 539, Hdat> KCUT au^era? ^fe, for which we may 
read iracri tear av^eva Page, or av^ev e/af e. 

Kauafa-i? (/caffafat?), which occurs in Hesiod, 'Works 
and Days, 7 w. 664, 691, contains a remarkable remnant of 
the general use of the digamma. But for the existence of 
this letter at the beginning of the simple verb, we should 
have met with /earafat?, not /cavdgais. Compare the San- 
skrit bhanj, to bv*eak. 

o?, Slender, Narrow. 

This is a word of rare occurrence ; but as it is applied to a 
variety of objects, we are the better able to determine what it 
means. It is used to describe a narrow passage (Od. K. 90), 
a thin delicate hand (II. e. 425), the weak slender legs of 
Vulcan (II. a-. 411, v. 37) ; and wolves are represented (II. 
TT. 161) lapping water, ryKaxro-ya-w dpatyon, with long slender 
tongues. In this, and in one other passage (Od. K. 90), it is 
preceded by the paragogic v, and consequently these passages 
decide nothing. In the other three passages, however, the 
prosody is incomplete without an initial consonant. These 
are, II. e. 425, %etpa aparfv, and II. cr. 411, i|r. 37, VTTO e 

*Eiap, Spring. 

The lines, Od. r. 519, and II. 6. 307, indicate that Homer 
used this word with an initial consonant ; and that the con- 
sonant was F, we may conclude from the fact, that the same 
word exists in this form in the languages of two countries so 
remote as Latium and Sweden. Ver is the corresponding 


word in Latin; war in Suio-Gothic (Hire, v. ii. p. 1082). All 
the intermediate nations have described this season by words 
expressive of other ideas, such as Printemps, Fruhjahr, Spring. 
Terentianus Maurus quotes cap, rjp, among the words to 
which the digamma was prefixed in the ^Eolic dialect*. 

"ESvov, Wedding-gift. 

This* word, which Homer uses only in the plural, denoted 
the presents given by the bridegroom to the bride before 
marriage (see the Scholia on II. i. 146). In the present text 
of Homer we find both eSva and eeSva, for which Heyne pro- 
poses FeSva and eFebva. Such a variety of form, which is 
found also in other digammated words (FeiKoa-i, cPel/coai,, &c.), 
is agreeable to the genius of the older Greek language. 
Hence the adjective dvefeSvos, without wedding-gifts (II. i. 
146, 288, v. 366f). 

, To see. 

This verb, with modifications expressing the ideas of 
knowing, seeming, appearing, occurs continually, and with 
comparatively few exceptions requires an initial consonant to 
complete the prosody. That the lost consonant was F may 
be concluded from the parallel forms in various other lan- 
guages, such as vid-ere in Latin; vid, know, in Sanskrit; 
wizzan, to know, Alemannic; wissen in German; and the 
Anglo-Saxon forms represented in modern English by the 
verbs wit, wot, wist, and the noun wit, and adjective wise. 
In Suio-Gothic we find sam-wete, con-scientia, o-v 

"Evvu/u, / clothe. 

This word was Pevvv^i, and hence Fel/jua, a garment, and 
j^, raiment, and numerous other forms. II. e. 905, must 

* " loiies dicunt prjp ;" Varro, De Ling. Lat. vi. p. 192. 

f In the ordinary texts of Homer this word appears as dvdcdvos, i. e. 
dvdffdvos. Both ova and ai>e are possible forms of the negative in com- 
position, which commonly appears as av or a, but which is certainly akin 
to the preposition avev. See Buttm. Ausf. Gr. Sprachlehre, vol. ii. 
p. 466. ED. 



have been ^apievra Se Fei^ara Fecrcre, and ty. 67, rola Trepl 
%poi' Fei/j,ara Fearo. 

Analogous to the Greek FecrOtfs is the Latin vestis, and 
Sanskrit vas, to wear, as clothes. From, the same root we 
find in Suio-Gothic wad and wast ; in Moeso-Gothic wastjom, 
clothes; in Anglo-Saxon, vaeda, a garment; and we have in 
English a curious remnant of the same root in weed, used 
now only in two phrases, a palmer's weeds, and a widow's 

'E/ccbv, Willing. 

This is an ancient participle, and connected with efcrjn, by 
the will of, which appears to be the dative of an obsolete 
noun. The words were Feicwv and FetcijTi, and hence are 
formed aFefccw, aFeK^n, afe/eao//-evo9. The root FeK is found 
in Sanskrit with no material difference of sound, as vaq, to 
desire, to will, whence vaqa, wish, will. 

IV. It has, I believe, been universally supposed by the 
authors, who have recommended the insertion of the digamma 
in the Homeric poems, that this is the only letter which has 
been omitted at the commencement of certain words. But, 
as the digamma has been expunged from one set of words, so 
the letter sigma has been taken from the commencement of 
others. If, on the one hand, it is an ascertained fact, that 
F existed at the beginning of some Greek words, which were 
afterwards always used without it, it is no less certain that 
this was the case with the letter 2j also ; since there are words 
used in both forms, such as 0-1)9 and 9, a-vcpopfio? and v<f>op/36<?, 
which show the transition from the complete to the abbre- 
viated state ; and since numerous words beginning with the 
aspirate in Greek begin with s in the cognate languages, 
such as ef, sex, Sanskr. shash ; eirra, septem, Sanskr. saptan ; 
oX/eo9, sulcus*. The evidence for the insertion of 2 in the one 

* The list of words, which in the later Greek began only with the rough 
breathing, but which in Latin began with s, may be easily augmented. 
But the traces of the passage of o- into the rough breathing in Greek itself 
have been less often observed. "Ia-rrjp.1 must have been originally a-i-a-Trj-fjn, 
formed with the usual and regular reduplication from the root ora, as 
from So, and corresponded in form, as well as in meaning, with 


set of words, is, generally speaking, as copious and decisive as 
the evidence for the insertion of F in the other set ; and we 
are able to determine by circumstances in each case, whether 
the one letter or the other ought to be supplied. 

o?, Father-in-law. 

A line, which has made a principal figure in controversies 
respecting the digamma, is one addressed by Helen to Priam, 
11.7. 172: 

AtSo?o9 re JJLOI ecrGi, (f)i\e etcvpe, Sewos re. 
The digamma has been prefixed to eicvpe, but this does not 
remove all difficulty. It appears strange that the analogy of 
socer in Latin should not have suggested the insertion of <r 
instead of f. The lengthening of the preceding syllable then 
takes place without any difficulty ; since, as was stated in the 
second section, no letter is more frequently prolonged or 
doubled to accommodate the quantity than a. f E/eiyjo?, and 
the feminine etcvpr), occur only thrice besides in the Iliad (in 
%. 451, and o>. 770), in which passages the cr may be inserted 
without any further change*. 

the Latin causative verb sisto. So the perfect eo-rrjica. must have been 
originally o-e-o-Trj-Ka. From the second aorist evxov, and other forms, we 
conclude that the root of theverb e^co was primitively trex (cf. Sanskrit sah.). 
It appears with the aspiration in the future e<, and other forms, in which 
the final x is modified. But the Greek law of euphony, which forbids the 
same syllable to begin and end with an aspirate letter, or two consecutive 
syllables to begin with aspirates, leaves the present ex * w ith the smooth 
breathing, so that the original a- is not represented. In like manner, to-^w, 
the strengthened form of e^eo, has taken the place of a-i-o^-a), which ori- 
ginally stood to o-e^G) in the same relation as p.i-p.v-a> to pevm ; that is, it 
was formed by reduplication, like from the root yev, and 7rt-7rr-o> 
from Trer. There are vestiges of such archaic forms in Homer. The 
defective metre of II. X. 36, 

rrj 8' em [lev Topya) J3\0(rvp>7ris (TT(f)dvo)TO, 

which has been noticed above in II. may be restored by reading with the 
old reduplication of the perfect, jSXoo-vpwTri? o-ea-TfffxivwTo. The hiatus in 
Od. t. 122, our' apa Troip-vrjo-iv Karai*cr^rai, our' aporoifriv, will be removed 
if we read Karao~to-^erat. So in II. e. 90, our' apa epKca ur^ei dXeadcoz/ 
epidyXew, we should probably read our' apa fep<ea o-icrxei. ED. 

* The German Schwieger, used in the compounds Schwieger-vater and 


a9, Far ; "Etfao-ro?, Each. 

Festus, a grammatical writer of the fourth century, informs 
us that Valgius derived the Latin secus, otherwise, from the 
Greek e/ea?"*. This derivation seems highly probable, although 
Festus quotes the remark for the purpose of refuting it. The 
aspirate of e/ea? may be considered as a remnant of the initial 
5. With this restoration the Greek and Latin words are 
almost the same in sound ; and the sense of the Latin word 
is obviously deducible from the primary acceptation in Greek. 
With etcds are to be associated eicaOev, and the derivatives 
'E/earo?, ''Ei/cdrrj, ercdepyos, eKr)(36\os, e/ear^/QeXeT^j which are 
titles or epithets of Apollo and Artemis. 

"E/eao-ro?, e/cdrepos, and e/cdrepOe, are perhaps to be referred 
to the same rootf. 

There are many passages in Homer, in which 

Schwieger-mutter, father-in-law and mother-in-law, and likewise Schwa- 
ger t brother-in-law, strongly suggest the conjecture, that the Greek 
fKvpos originally had both the Vau and the Sigma, and was vfeKvpos, so 
that the e in the preceding <t\e was lengthened simply by position, $i\e 
o-ffKvpe. The same conclusion might be drawn, though less certainly, 
from the appearance of o in the Latin socer. The o is not merely sub- 
stituted for e, but represents fe. The Sanskrit pvafura is conclusive. It 
is noticed by Bopp, &c. The combination a-f at the beginning of 
words appears to have been not unusual. It is commonly admitted that 
the adjective r)8vs began with a consonant; but the comparison of the 
Latin words suavis, suadeo, and the English sweet, shows that it probably 
began with the two consonants erf. A similar conclusion is drawn from a 
comparison of eQos, wont, and kindred words, with the Latin forms suesco, 
suetus, &c. The pronoun ov, ot, e, which has the old accusative cr<e, and 
the plural creels, &c., and which corresponds to the Latin sui, se, with the 
possessive pronoun os, corresponding to the Latin suus, must have had 
originally the forms crfe, &c. ; and where the o- was retained in later Greek 
the f passed into (p. In words of this class in Homer, although in some 
passages there are indications that both consonants were preserved, as in 
vfexvpe in II. y. 172, yet, more usually, one consonant (probably the <r) 
seems to have been dropped, and one (the f) retained alone. Cf. Donald- 
son, New Cratylus, p. 120. ED. 

* So Donaldson, New Cratylus, p. 356. See also Transactions of the 
Philological Society, 1854, p. 167. ED. 

f These latter words are apparently derived from the Sanskrit numeral 
eka, one. ED. 


requires an initial consonant; but many also, in which it 
seems to reject it. In several of the latter kind the preceding 
word ends in s, as in II. o. 288, fjuaXa eXirero Ovfjubs e/cda-rov. 
It is worth considering, whether, in the older language, when 
one word ended in s, and the next began with the same letter, 
the former s might not be dropped in pronunciation, as in the 
old Latin. We have sanctu' Serapis (Lucilius), and perhaps 
we ought to have OV^JLO o-e/caarov. 

Ov, ol, e, Him ; 09, His. 

The advocates of the digamma have found no words more 
perplexing than the personal and possessive pronouns of the 
third person. They agree, however, in believing that the 
digamma belonged to their primitive form. But the insertion 
of f is little adapted for removing the difficulty of those pas- 
sages, in which a short vowel at the end of the preceding 
word is not only not cut off, but even takes the place of a 
long syllable ; as in the following instances : 

II. e. 343. 'H 8e fjieya id%ov(ra aTrb eo /cd/3{3a\v viov. 

. 62. -- 6 S' dirb edev wa-aTO %e(,pt. 

p. 205. - aTrb edev TJKG ^apa^e. 

v. 163. 'A<77Ti / 8a ravpefyv vykW airo eo, Sela-e Se 

v. 261. Hvj\LSrj^ Be cra/co? yitev airo eo xei 
_ o 8^2 "1 

QQ ' >"Ht8ee pavTocrvvas, ovSe 01/5 TratSa? eacr/ce. 

A. OoU._J 

It is more probable that the pronoun began with a cr, which 
in these passages was doubled in pronunciation (see above, 
II. p. 122* ). 

In not a few of the passages, in which the pronoun appa- 
rently rejects an initial consonant, the preceding word ends 
in 5, as in II. X. 403 : 

' dpa elire Trpo? ov fjieya\rJ70pa Qvpov, 

* It has been suggested in a preceding note, that the original form of 
this pronoun was a-fe, o-fto, &c. This assumption satisfies the require- 
ments of these passages. Elsewhere one consonant only is required j and 
it is likely that the o- was dropped, and the forms remained Fe, foi, 
&c. ED. 


a line which recurs often ; and in a. 609, 

Zeu? Be 7T/JO? ov Xe^o? r\'C 'OXu/ATUo? da-repO'jnJT'rjs. 
It is possible that in these instances the text is not corrupted, 
as has been often supposed, but that the final 5 was dropped 
in pronunciation before the initial sigma, and that we ought 
to pronounce irpo vov Xe^o?, &c. 

The insertion of <r receives abundant confirmation from the 
cognate languages. We have in Sanskrit, sva ; in Latin, sui, 
se, sum, &c. ; in Alemannic and Suio-Gothic, sin; and in 
Mceso-Gothic, sein, his, whence the German sein ; in German 
also, sie, she, and sie, they, &c. 


Whether f 'Hprj was ever used by Homer with an initial 
consonant, as Heyne supposes, appears to me to be doubtful ; 
but, if it was, the circumstances tend to show, that the initial 
consonant was not F, but %. The cases which require an 
initial consonant to complete the prosody, and which have 
induced Heyne and others to assume the reading ^rjprj, are 
twenty-nine in number, but consist altogether of the recur- 
rence of one combination of words, viz. ITOTVUI "Hpy. The 
evidence is certainly much weaker than if we found several 
phrases with the hiatus. 

But of the passages in which "Hprj apparently rejects an 
initial consonant, thirty-four (which are about three-fifths of 
the whole number) present before "Hprj a word ending in 9, 
as in the frequently recurring phrase Qea Xeu/eeoXevo? " 
It is possible that we ought to pronounce Oea \evicu>\evo 
upon the principle already indicated*. 

* Where there is evidence that a word, which in the later Greek began 
with an aspirated vowel, began in the earlier Greek with a consonant, the 
aspiration gives ground for assuming, that the lost consonant was <r rather 
than f. The instances in which an initial f is represented by a rough 
breathing, are comparatively few ; while those in which the breathing is 
manifestly the substitute for an original o-, are many. It deserves, how- 
ever, to be considered, whether the rough breathing itself, the H of the 
ancient alphabet, had not sometimes the power of a consonant. There is 
another most important consideration, which has been overlooked by the 



[Read June the 13^.] 

Amongst the abstracts in ia, io- (ium), we find several 
derived from a past participle. Thus, exercitium from exercito-, 
nuptiae from nupto-, argutiae, minutiae, from arguto-, minuto-, 
controversies from controverso-, inscitia from in-scito-, impolitia 
from im-polito-, comitium, exitium, initium, from comito-, exito-, 
inito-. Lastly, I mention lotium from loto-, as bearing the 
greatest resemblance with our word. As lotus is a contraction 
of lav-i-tus from lavere, we may perhaps suppose that otio- 
comes from a participle oto-=avito- y this being derived from 
a verb avere. Everybody knows, that the verbs terminating 
with a v undergo a strong syncope in the past participle, as 
jiito-j cauto-, fauto-, foto-, moto-, for juv-i-to-, cav-i-to-, fav- 
i-to-, fov-i-to-, mov-i-to-. I hope that a better etymology will 
be proposed hereafter; in the mean time I offer it as my 
conjecture, that otium meant originally " enjoyment, hap- 
piness," and owes its origin to the verb avere, " to be happy," 
of which the imperative ave, aveto, avete, and the infinitive 
have alone been preserved*. The fact that avere has an 
initial h in some inscriptions and manuscripts is, I conceive, 
no objection to my proposed derivation. I tried to show 
elsewhere (Aufrecht und Kuhn, Zeitschrift. I. p. 358) that 
the English rest, and German rast (Gothic rimis) come from 
a root ram, which in Sanskrit signifies both ' to be happy' and 
< to rest/ 

scholars who have been most intent upon replacing the digamma in the 
Homeric text ; and the same remark will apply to the restitution of the 
initial sigma. As the consonants were confessedly lost in the progress of 
the language, and as such a change in pronunciation could not have taken 
place suddenly, there must have been a period and state of transition ; and 
this time of transition may have been earlier for some words, and later for 
others. It is possible, therefore, and not at all improbable, that, when 
the Homeric poems were composed, some at least of the words, which 
anciently began with f or <r, were in a state of transition, and may have 
been used by the poet, sometimes as beginning with a consonant, some- 
times with a vowel. ED. 

* Compare also Gellius xix. 7, 9, (Laevius) ' avens ' posuit pro 'libens.' 


I know only of one previous explanation which deserves any 
attention. Graff, Wellmann, and Doderlein connected otium 
with the Gothic auty ' deserted/ autyida ' a desert/ Solitude 
might be an apt name for leisure, if au]> signified solitary, but 
neither the Gothic nor the other Teutonic languages exhibit 
it in any other meaning than that of ( deserted, waste, barren, 

[Head November the Sth.'] 

The Latin abstracts ending in tia, tie-, tio- (Hum), must 
not be confounded with those in ia, ie-, io-, derived from 
nouns the crude form of which has the termination to or t, as 
for instance angustia from angusto-, scientia from sclent-, 
septimontium from septimont-, silentium from silent-. In Latin 
the former are never derived from a verbal root. Professor 
Bopp says indeed, in his Comparative Grammar, 844 : " We 
find in Latin, together with i-tio, also i-tiu-m in the com- 
pound in-i-tiu-m, which agrees in its suffix with the noun- 
derivative servi-tium." But it need hardly be stated that 
this comparison is wrong. While servitium contains the ter- 
mination tio-, we have to derive initium, as well as eaciiium, 
comitium, from the participles inito-, eoeito-, comito-, by means 
of the suffix io, unless indeed words like exercitium, lotium, 
nuptiae, argutiae, inscitia, are to be divided into exerci-tium y 
lo-tium, &c. I intend at present to offer a conjecture as to the 
origin of the first-mentioned abstracts in tia, tie-, tio-. Most 
of them are derived from adjectives, only a few from substan- 
tives. I know of the following : 

I. A-declension. 

Amicitia, inimicitia, avaritia, blanditia, canitia, duritia, 
justitia, injustitia, laetitia, lautitia, malitia, moestitia, mollitia, 
munditia, immunditia, notitia, pigritia, planitia, primitiae, 


pudicitia, impudicitia, impuritia, saevitia, scabritia, segnitia, 
spurcitia, stultitia, tristitia, vafritia. Nequitia comes from 
nequam, the latter being derived, according to Ritschl, from 
an adjective nequus for ne-cequus. Pueritia is the only in- 
stance of tia being connected with a substantive ; for lanitia, 
lanities, lanitium, seem to be more rightly spelt with a c, all 
three being derivatives from the adjective lanicius. 

II. ^-declension. 

Most of the words above mentioned belong also to this 
declension. They are, nequities, amarities, amicities, avarities, 
blandities, calvities, canities, durities, mollities, mundities, no- 
tifies, planities, saevities, scabrities, segnities, spurcities, 
tardities, tristities, vastities. Only imbalnities and pullities 
are derived from other nouns. 

III. ^-declension. 

Calvitium from calvus. All the others are derived from 
substantives, namely, famulitium, servitium, conservitium, 
sodalitium. Ostium is so concrete in its meaning, that I 
hesitate to derive it immediately from os. Gurgustium is 
quite obscure. It is probable that convitium also belongs to 
this class, and stands for convoc-i-tium. This derivation has 
been very ably defended by Fleckeisen in the Rheinisches 
Museum, 1853. p. 221 seq. 

Bopp, in his Comparative Grammar, 846, considers tia, 
and tio- as lengthened forms of ti. But ti, Greek ai (com- 
pare men-ti-, fl^-o-t-?), form only primary derivatives, and I 
know of not a single instance where -ti- is attached to another 
noun. For sementis is not derived from semen, but both are 
independently formed from the root se ; or, to express the 
fact more precisely, the one word shows the shorter suffix 
men for ment, the other the enlarged form menti ; and each 
stands in the same relation to the other as momen does to 
momentum. I divide, therefore, sementis thus, se-ment-i-s, 
not semen-ti-s. Pott, in his Etymol. Forsch. ii. 494, con- 
siders tio, tia, as increased forms of io, ia, without accounting 
in any other way for the existence of the t. 


The two principal suffixes for forming secondary abstracts 
in Sanskrit are td (fern.) and tva (neut.). The former is 
represented in Latin by ta mjuven-ta, senec-ta, olivi-ta, and 
TT) in Greek in ape-rrf, and appears frequently in Gothic in 
the shape of tha, for instance, diupi-tha, depth ; hauhi-tha, 
height; garaihti-tha } justice. The other suffix, tva, very 
frequently forms abstracts from adjectives and substantives in 
Sanskrit, as for instance, mahat-tva, greatness, from mahat, 
great; sakhi-tva, friendship, from sakhi, friend. It appears 
in Slavonic in the form of s-tvo, as apostol'-stvo, the mission of 
apostles, from apostol' ; mnoz'-stvo, multitude, from mnog'*. 
I believe that the Latin tia, tio, agree completely with the 
Sanskrit tva, on the assumption that the v was first vocalized 
(tua, tuo], and the u at a later time weakened to i. We know 
that the Latin i stands frequently for an older u. Even the 
oldest Latin knows only tibi, but it is certain that this pro* 
noun is weakened from tubi, Sanskr. tubhyam. The Umbrian 
has, in the accusative of the same pronoun, tiom, which, when 
we compare it with the Sanskrit tvdm, we are sure can only 
stand for tuo-m. This explanation of the abstracts in tia, tio, 
removes all other difficulties, and restores to the Latin a 
suffix, traces of which remain in all the Indo-European 


[Read February the 8th.~] 

The idioms of a language are its strength. Far more cha- 
racteristic than its words, even in their earlier and simpler 
forms, the Idiom individualizes a language, and marks it out 
from the group to which it belongs. When we find the 
same words (and often the same grammatical inflections) 
in Sanskrit and in Greek, in German and in English, we are 

* Schleicher in Aufrecht and Kuhn's Zeitschrift, i. 142. 


convinced that these languages had a common origin. When 
we find that each language has notwithstanding a perfectly 
distinct manner of expressing the same ideas, we see in this 
fact evidence of different culture, different associations, 
different pursuits. 

The Idiom may be said generally to consist, not in the 
peculiarity of the words employed, but in the peculiarity of 
their collocation. Each word by itself may be literally trans- 
lateable in another tongue, but not in the peculiar position 
in which it is placed. We must resort to other words, or to 
a different collocation, to express the same thought. Take, 
for instance, the German phrase ' das Seinige zu Rathe halten/ 
and the corresponding English phrase 'to husband one's 
means :' the one is not a rendering of the other, but the sub- 
stitution merely of a phrase of similar meaning to convey the 
same idea. Each word in the German has its English equi- 
valent, and yet a literal translation would be out of the 
question. So common an expression as ' Es thut mir leid/ 
must be expressed by ' I am sorry for it ; ' not, ' it does me 
woe/ Even e II a raison 7 must be Englished by ' He is right;' 
for although Dryden did venture on the Gallicism ' he has 
reason, 3 his authority was not sufficient to make it current. 
Such transplanted idioms seem to have a natural tendency to 
die out. Berners, in his translation of Froissart's Crony cle, 
renders the French 'se battre a F entrance/ by ' fight at 
utteraunce;' and Shakspere also writes 'to the utterance/ 
Modern English refuses to recognize the stranger except in 
its native garb. 

There are idioms no doubt which are identical in two or more 
languages. These are probably, in some instances, a common 
inheritance derived originally from the same parent. Some- 
times, again, they may very nearly approach, but a single 
word in the one language will refuse to surrender to the 
other. Thus the French phrase ' II y a tout lieu de croire/ 
may be rendered in English ( There is much room to suppose/ 
or ' there is every reason to suppose/ In the one case we slightly 
modify the expression; in the other we change the figure. 
There is this difference between idioms and words. Foreign 


words are admitted and naturalized. In a language like our 
own, their foreign birth is no bar to their reception ; but the 
language will not bow its neck to a foreign yoke. It is one 
thing to welcome strangers ; they may increase our wealth. It 
is another to submit to their dictation ; this is to resign our 

Examples might readily be multiplied; but my object 
in this paper is rather to direct attention to the idioms 
of our own language, and more especially to those (many of 
them now obsolete) which are to be found in our earlier writers. 
Under this head I shall also include certain peculiarities of 
construction which I do not remember to have seen noticed 
elsewhere. These must be regarded only as first-fruits. There 
is a large harvest still to be gathered. Our dictionaries and 
grammars have done but little for us here. Richardson's Dic- 
tionary, useful as it is in many respects, is in this extremely 
deficient. One looks almost in vain in his pages for idiomatic 
usages, and it is to be regretted that such idiomatic expres- 
sions as he does give, are not classed separately, as in our 
Latin and Greek Lexicons, instead of being mixed up with 
the ordinary usages of the word. So again with regard to 
constructions, you will not learn from him that Hooker writes, 
' Drawn from those beaten paths wherewith they have been 
inured/ or that Latimer tells us ' not to flatter with anybody/ 
or that Roger Ascham speaks of ' chaunging a good [word] 
with a worse/ Nor do I know of any grammar that at all 
supplies the deficiency. We have numbers of books on the 
study of words, and the changes through which words have 
passed, what have fallen into desuetude, what still survive, 
what additions have been made from time to time to our 
existing stock. But nothing I believe has yet been done to 
illustrate the idioms* of our language, or to classify its 
constructions. And yet, important as the study of words 
is, that of idioms is certainly not less important. We admit 
this readily enough in our study of foreign languages. One 
of the first things in such a study is to notice the idioms 

* Perhaps I ought to except Dr. Roget's useful Thesaurus, but this 
does not profess to travel beyond modern usage. 


of the language. We know well enough how our first essays 
at composition in Greek, or Latin, or French, engage us in 
a perpetual hunt after phrases. It is wonderful with what 
zest these are seized on and treasured up, and with what 
ingenuity we torture ourselves that they may figure in our 
translations. But it is remarkable that the same solicitude 
ahout the idioms of our own language is never instilled into 
us, and we are expected to write as we speak, by a kind of 
natural gift*. 

The collection of idioms and constructions which follows, is, 
as I have said, only a fragmentary contribution to a know- 
ledge of this subject. It is taken chiefly from Piers Plough- 
man, Wiclif, Gower, and Chaucer. In a few instances I 
have traced the idiom down to a later time. I hope in a 
future paper to continue my investigation at least through 
the writers of the Elizabethan period, if not to our own. 


Blame. ' To fall in blame, set in blame/ 
(1.) Forthy men shulden nothing hide 

That mighte fall in blame of pride. 
i. e. that might be censured as pride. 

Gowerf, C. A. vol. i. p. 145. 
And again, 

So might thou lightly fall in blame. Ib. p. 229. 

With this we may compare the French 'tomber en faute/ 
The Germans say, ' die Schuld f allt auf mich/ and we now 
say, 'the blame falls or rests on me/ or f l am to blame/ 
This however is somewhat different from Gower's phrase, 
especially in the first instance given above. 

(2.) And thei have self ben thilke same 

That setten most the world in blame. Gower. 

* Coleridge somewhere gives as a test of a good style, that it should be 
untranslateable in other and simpler words of the same language without 
loss of sense or dignity. He has also remarked on the great excellence of 
a thoroughly idiomatic style. It seems probable that the more idiomatic 
a style is, the less translateable it will be in other words. 

t The references are to Dr. Pauli's edition of Gower, which will shortly 
be published. 


Cf. the Latin expression ' Ponere aliquem in culpa/ Cic. pro 
Cluent. c. 45. We should now say, ' find fatdt with/ When 
this latter idiom first arose I cannot say, but in Heb. viii. 8, 
fjuefjL^ofjLevo^ avrols is translated ' finding fault with them.' 
This is clearly idiomatic, and very different from the expression 
1 to find fault in a person/ which in the Bible is merely a 
literal rendering of the Hebrew or Greek words. 

Boot. ' To do boot of a thing/ &c., z. e. ( to remedy, make 
amends/ &c. 

And ye that may do boot 

Of al my languor with your wordes glad. 

Chaucer*, vol. iv. p. 182. 

There can no wight thereof do bote. Gower, Prol. C. A. 
So also, ' to have bote' (Chaucer) ; and ' to find bote 5 (Gower). 
Shakspere's ' Make boot upon the summer's velvet buds/ is a 
different kind of expression. ' To do boot' is to do service, 
confer a benefit, &c. 'To make boot' is to reap a benefit, 
to gain in any way, as by spoil or pillage. Hence the word 
booty. The original meaning of the word is simply that of 
addition. This appears in the phrase 'to boot,' which we 
still use. On the different usage of the verbs 'make' and 
'do' in the composition of phrases, I shall say something 
further on. 

Cast. 'To cast one's cheer, look, wits/ &c. The phrase 
'to cast one's cheer/ in the sense of 'to cast one's eye/ 'turn 
one's face/ &c., is of frequent occurrence. 

Up to the heven he caste his chere. Gower, i. 143. 

His chere aweiward from me caste. Ib. p. 46. 

She caste on me no goodly chere. Ib. 

In the following passage the expression seems to be used 
rather differently : 

This Acteon, as he wel might 

Above all other caste his chere 

And used it from yere to yere 

With houndes and with grete homes 

* The edition of Chaucer referred to is that in Bell's Annotated Edition 
of the English Poets. 


Among the wodes and the thornes 
To make his hunting, &c. Ib. 53. 

Here it seems equivalent to ' carried his head/ 
Again, ' to cast one's look ' : 

And as he caste his loke 

Into the well, and hede toke. Ib. p. 120. 

We still say f to cast a look/ Also 

Here wittes thereupon they caste 
And ben appointed atte laste. p. 1 14. 

i. e. they reflected upon it and at last came to an agreement. 

Other phrases are : 

And such a loue on her he caste 
That he her wedded ate laste. Ib. 125. 

So that upon his trecherie 

A lesinge in his herte he caste. Ib. 137. 

Chaucer uses the expression ' to cast off the heart * in the 
sense ' to give up/ f to despair/ 

Cast off thine herte, for all her wordes white. 

Vol. iv. p. 168. 

Hand.(l.) 'To take or have on hand, upon hand,' &c. 
Gower writes 

And thus the whele is all miswent 

The which fortune hath upon honde. Vol. i. p. 130. 

Which every kinde hath upon honde. Ib. p. 42. 
Chaucer : 

Such maner wordes hadde we on honde. Vol. ii. p. 56. 
It is curious in what a general way this phrase is made use of 
by Gower. He writes 

For who that hath humblesse on honde. Vol. i. p. 153. 
And even 

Though I sikenesse have upon honde. Prol. p. 5. 
(2.) Similarly ' to take on honde/ which however does not 
correspond so much to the modern ' take in hand/ as to the 
phrase ' to take to :' 



And thus they casten care awey 

And token lustes upon honde. Gower, i. 126. 

i. e. took to, or engaged in, tourneys, &c. 

Tho toJce he lesinge upon honde. Ih. 214. 

Thou must humhlesse take on honde. Ib. 145. 
i. e. have recourse to lying, to humility, &c. 

(3.) And to add an instance without the verbs ( have ' or 

Tho was ther gret merveile on honde. Ib. p. 151. 
(4.) The very curious expression * to bear on honde ' = ' to 
insist upon/ ' persuade/ Tyrwhitt explains it, ' to accuse 
falsely, to persuade falsely/ but I think incorrectly. The false- 
ness of the persuasion is only an accidental, not a necessary, 
idea : 

And bare on honde it was no wit 

Ne time for to speke as tho. Gower, ii. 2. 

i. e. ' would have it.' 

I wis a wyf if that she can hir good 

Shal beren him on hond the cow is wood. Chaucer, ii. 5 1 . 
i. e. shall make him believe anything, however extravagant. 

Bar I styf min housebondes on honde 
That thus they sayde. Ib. 56. 
I bare him on honde he had enchanted me. Ib. 62. 
So also, with a sort oiprcegnans locutio : 

Ye wise wyves that can understonde 

Thus schulde ye speke, and bere hem wrong on honde, 

For half so boldely can there no man 

Swere and lye as a womman can. Id. ii. 51. 

i. e. make them out to be, insist upon it that they are, in the 

(5.) 'To be brought to honde' = ' to be brought down (sc. 
under the hand), subdued/ &c. 

But yet he was nought of such might 
The strength of love to withstonde 
That he ne was so brought to honde 


That malgre wher he wol or no 

This yonge wife he loveth so, &c. Gower, i. p. 68. 


(.) . . your bewte may not strecche 

To make amendis of so cruel a dede. Chaucer, v. 67. 
(6.) . . and assaieth 

His God which made him nothing straunge. 

Gower, i. 140. 

i. e. who did not turn away as a stranger from his request, 
but listened to and granted it. 

The persoun of the toun, for sche was feir, 

In purpos was to maken hir his heir, 

Bothe of his catel and his mesuage, 

And straunge made it of hir mariage. Chaucer, i. 222. 

i. e. made it a matter of difficulty to obtain her in marriage. 

He made it strange, and swore so God him save, 
Lesse than a thousand pound he wold not have. 

Id. ii. 242, 
Similarly, ' to make wise,' 

(c.) Oure counseil was not longe for to seeke ; 
Us thought it was not worth to make it wys. 
And graunted him withoute more avys. Id. i. 108. 

i. e. to make it a matter of wisdom or deliberation. (Tyrwhitt.) 

(d.) And he shoulde eke here truth alowe 

With al his herte and make him chere. Gower, ii. 8. 
He maketh the messanger no chere. Id. i. 193. 
(e.) And swore, if she him daunger make. Ib. 195. 
It is a matter of some interest to compare the usages of 
this verb with those of the verb * to do '. We shall find that 
the modern idiom differs widely in some of these from the 
ancient. I have already observed on the difference of meaning 
between such phrases as ( to do boot ' and ' to make boot/ and 
I will now supply some further illustrations. Thus Chaucer 

The sely wydow, and her doughtres tuo, 
Herden these hennys crie and maken wo. iii. 233. 

M 2 



Witness on Jobe,whom we dedefnl wo. Chaucer, ii. 96. 

Feyne wold she wote al hole your thoughte 
And why you do here al this wo. vi. 69. 

We see that in the above examples, f to make wo ' is 
intransitive, ' to do wo ' is transitive ; that in the former the 
action rests with the agent, in the latter it passes over to 
an object. The same kind of distinction holds in some 
phrases still in use. Thus we say ' to make mischief/ and ' to 
do mischief/ but with this difference, that in the first phrase 
the subject or maker is the more prominent; in the other, 
the object or the person injured. So again we say ' to make 
good ' i. e. to repair some injury which the subject has in- 
flicted and for which the subject is responsible ; ' to do good ' 
i. e. to benefit some person who is the object of our regard. 
Yet on the other hand, ' to do well/ ' to do ill/ ' to do evil/ 
in all these forms of speech the action of the verb limits itself 
to the agent. 

The following are some of the principal idiomatic usages of 
the verb ' do/ which are deserving of notice : 
( 1 .) Falsnesse for fere thanne 

Fleigh to the ffreres, 

And Gyle dooth hym to go, 

A gast for to dye. Vision of Piers Ploughman, 1302. 
i. e. ' stirs himself,' ' sets himself to go.' 

(2.) Have mercy quod Mede 
Of men that it haunteth, 
And I shal covere your kirk 
Your cloistre do maken. Ib. 1473. 
We should now say ' have made. 3 

But natheles this marquys hath doon make 

Of gemmes, set in gold and in asure, 

Broches andrynges, for Grisilde's sake. Chaucer, ii. 132. 

(3.) Where we should now use the verb ' make* ; ' followed 

* Chaucer also uses the verb ' make ' in such construction : 
Sche made to clippe or schere his heres awey 
And made his foomen all his craft espien. iii. 189. 


by another verb, either with, or without, the prepositions ' to/ 
'for to': 

Thou schalt no more, thurgh thy flaterye 

Do me to synge and wynke with myn ye. 

Chaucer, iii. 234. 

And doo that I my shippe to haven wynne. Id. vi. 180. 

Which is the way to doon you to be trewe ? Ib. 189. 

Men wolde say that we were theves stronge 

And for our tresour doon us for to honge. Id. iii. 84. 

And did him plainly for to wite. Gower, ii. 4. 

i. e. ' made him to know/ or as we should probably express it, 

' gave him to understand/ 

As yet for aught that is befalle- 

May no man do my chekes rede. Gower, ii. 7. 

i. e. make them blush. 

So that his loking dooth myn herte colde. 

Chaucer, viii. 52. 

So also followed by the transitive verb, where we should now 
use the passsive : 

. . what a peyne 

Al sodeynly about myn herte 

Ther com at ones, and how smerte 

In creping softe ! as who should stele 

Or do me robbe of al myn hele. Chaucer, vi. 62. 

The modern idiom would be 'make me, cause me to be 
robbed/ &c. 

(4.) Where 'do' is equivalent to 'put/ This Professor 
Key considers the original meaning of the verb (Lat. Gram, 
p. 65 note) . 

Thei ben acombred with covitise 
Thei konne nought doon it from them. 

Piers PL Vision, 1. 852. 

How I may do lachesse away. Gower, ii. 4. 

Quod Pandarus, ( Be stil ! and let me slepe 
And do down* thin hood, thi nedis spedde be.' 

* Tyrwhitt reads doe on, whence don. 


(5.) With various nouns, either with (.) or without (b.) 
a personal object : 

(a.) But he and hise disciplis don many extortions to the pore 
puple. Wiclif, Three Treatises*, p. cxliv. 

Her men taken stimwhat sooth, and don dremyng to this treuthe. 
Ib. p. xvii. 

. . if that I may doon ease 

To thee, sir cook, &c. Chaucer, iii. 257, and often. 
Fain would I do you mirthe wiste I how. Id. i. 108. 
Ye done hym neither good ne gentilnesse. Id. v. 146. 
She doth her self a shame and hym a gyle. Ib. 143. 
And doo to me adversite and grame. Id. vi. 188. 
I do nofors the whether of the two. Id. ii. 86. 
(b.) Now wolden some men say parad venture 
That for my necgligence I do no cure 
To telle you the joye and tharray, &c. Chaucer, ii. 81. 
Ther was a knyght, that loved and dede his peyne 
To serven a lady in his beste wise. Ib. 227. 
For whiche cause the more wee doute 
To do a fault while sche is oute. Id. vi. 62. 
(6.) Absolutely: 

And everich had a chapelet on her hedde, 
Which did right well upon the shining here. Chaucer. 
i. e. sat, or looked well. 

(7.) Passive, with the meaning ' to be killed ' : 
And this thei seien is mortesied and patrimonye of Crist, that was 
doon on the cross. Wiclif, Three Treatises, xxviii. 

Tyrwhitt observes that Chaucer rarely uses 'do' as an 
auxiliary verb. He quotes, as illustrative of this usage, a 
passage in the Monkes Tale : 

His yonge sone that three yere was of age, 
Unto him said, fader, why do ye wepe ? 
Whan will the gailer bringen our potage ? 
Is there no morsel bred that ye do kepe? 1. 14742. 
It occurs again a little further on, 

And whan the woful fader did it sey. 

* The references are to Dr. Todd's edition, Dublin, 1851. 


He also notices that the transitive use is more common, 
as in v. 10074, Do stripen me, Faites me depouiller ; v. 10075, 
Do me drenche, Faites me noyer. But it occurs still more 
frequently, he says, to save the repetition of a verb*. 


We may not pynche at this lawe that God him silf ordeyned first, 
but if we putten blasfeme on God, that he ordeynede thaime 
foolily. Wiclif, Three Treatises, p. xxviii. 

But for I shewed you Arcyte, 

Al that men wolde to me wryte 

And was so bysy you to delyte, 

Myn honour saufe, meke, kynde and fre, 

Therefore ye put on me this wyte. Chaucer, vi. 188. 

And thinke ye that furthrid be your name 

To love a newe, and ben untrewe aye, 

And put yow in sclaunder now and blame. Ibid. 

There is no sleighte. . . . 
That he ne put it in assay 
As him belongeth for to done. Gower, i. 65. 

That he hath put all his assay 
To winne thing, &c. Ib. 68. 

Prise or Price : 

And to gret chepe is holden at litel prise. 

Chaucer, ii. 60. 

And evermore he hadde a sovereyn prys. Id. i. 79. 
i. e. was in high esteem. 

Tho was knighthood wpris by name. Gower, i. 6. 

In which he found so mochel grace 
That all his prise on her he laide 
In audience. Ib. 154. 

* Wiclif uses 'do/ 'make/ and 'give,' as synonymous in the following 

.... Crist seith hymself in the gospel, false Cristis and false prophetis 
shulen rise and shulen gyve grete syngnes and grete wondris, so that if it 
may be don, also the chosen ben sent into errours. Now oure faithful 
men done wondres whenne thei suffren persecutions; but thenne the 
kny?tis of this beemoth, that is Sathanas, shulun make wondris, $u whenne 
thei maken persecucion. Three Treatises, p. cxx. 



And be other men never so hooly and kepen Goddis lawe, lewde 
ydiotis thei ben clepyn, and lityl thei sett bi hem. Wielif, Three 
Treatises, p. cxlii. 

I sette nought an hawe 

Ofhi8 proverbe, ne of his old sawe. Chaucer, ii. 65. 
Of grete men, for of the smale 
As for to accompt he set no tale. Gower, i. 64. 
And o/"the counseil none accompte 
He sette whiche his fader taught. Id. ii. 37. 
A king whilom was yonge and wise 
The which set of his wit great prise. Gower, i. 145. 
And they have self ben thilke same 
That setten most the world in blame. Id. i. 63. 

. . and set your herte in ese. Chaucer, iv. 150. 
Whom folwest thou? where is thy herte ysette1I\>. 1 76'. 
Was none of hem so ware, that might 
Set eye, where that he becom*. Gower, i. 143. 
Of suche men as now aday 
This vice setten in assay. Gower, i. 229. 
For thilke shirte unto the bone 
His body set afire anone. Ib. 236. 
i. e. as we still say, set on fire. 

Among these other of slouthes kinde 
Which all labour set behinde. Id. ii. 38. 
we should say ' set aside/ 

Take : This verb is used in a variety of phrases : 
(1.) The thridde part of the chirche fijhtith her after Crist; 
and taketh ensaumple and wei of him to com to hevene as he cam. 
Wielif, Three Treatises, p. viii. 

But for thou canst not, as in this centre 
Wynne thy cost, tak her ensaumple of me. 

Chaucer, ii. 99. 
(2.) But Jewes ajenstoden hem fast, and hethen mens tooken 

* Compare with this, Chaucer, Leg. of Good Women (Leg. of Ariadne, 
ad. fin.} : 

Alas, wher shall I wretched wight become ? 


hem with wille, and receyveden the Hooli Ghoost. Wiclif, ut 

supra, p. ix. 

(3.) This fals Arcyte, of his newefanglenesse, 
For she to hym so lowly was and trewe 
Too ke lessedeynteforhir stedfasnesse. Chaucer, vi. 184. 

i. e. set less value upon it. 

(4.) Take keep = take heed : 

What shulde I take kepe hem for to plese. 

Chaucer, ii. 50. 

We loveth no man that takith kepe or charge 
When that we goon. Ib. p. 54. 

He bad hem of the stremes depe 
That they beware and take kepe. Gower, i. 233. 
Also construed with the preposition ' upon/ 
For it is good ye take kepe 
Upon a thing which is me tolde. Ib. 215. 

With the negative 

And take offonl delite no kepe. Ib. 56. 

This idiom is very common, both in Chaucer and Gower, 
perhaps even more so than the synonymous 'take heed/ 
Wiclif, so far as my observation has gone, does not use ' to 
take keep ' : but he has f to take tent,' which I have not met 
with in either of the others : 

God grant thise lordes grace to take tent thereto : to bisy hem 
for the cause of God more thenne for her owne. Three Treatises, 
p. cliii. 

The idiom which is common to all three writers ' take heed/ 
is the only one which still survives. We are, however, so rich 
in phrases of similar import that we need scarcely regret our 
loss. Thus we can say either ' to give, 3 or ' to take heed ' : 
' to pay attention ' : ' to take care ' ; not, however, ' to give 
care'; for this we must substitute f to give diligence/ or 
' to do one's diligence.' 

Tell, in the sense of to reckon, account, &c. : 

Crist telde not by siche abite. Wiclif. 
i.e. Christ made no account of, set no value on, such habit. 

Crist loved more ye treue prestis thenne thise worldly goods ; he 


[i. e. Antecrist] and hise tellen more bi strumpetis prestis, and more 
thei shal be sett by and worshiped, &c. Id. Three Treatises, p. cli. 

But by my fay ! I told of it no stoor. Chaucer, ii. 50. 
i. e. did not care about it. 

And I say forther more 

That I ne tell of laxatives no store. Id. iii. 225. 
. . but he was but seven yer old 

And therefor litel tale hath he told 

Ofeny drem, so holy was his hert. Ib. 224. 

They loved me so wel, by God above ! 

That I tolde no deynte of her lo\e. Id. ii. 50. 
Instead of to ' tell store of/ we now say ' to set store by * a 
thing. The word ' tell ' means properly ' to count ' or ' reckon/ 
Arnold connects it with the Greek reXo?, tax, toll, and reXetv. 
We have preserved this meaning of the root in the modern 
' tellers ' in parliamentary phrase, and in such expressions as 
1 it tells against him/ ' an argument, or a blow which tells/ 

Well: 'To be well/ &c., with the opposite 'to be wo/ 
'Well worth/ and 'wo worth/ 

My faire maid, well the be 

Of thin answere, and eke of the 

Me liketh well. Gower, i. 154. 

Wel were they that thider mighte winne ! 

Chaucer, vi. 63. 

Well worthe of this thinge grete clerkes 

That trete of this and other werkes.Ib. 195. 
In B ell's edition of Chaucer there is a strange note on this 
idiom. He says, " Worthe is a verb, of which grete clerkes 
is the subject. It often occurs in combination with wele and 
wo, and appears to mean to attribute. Thus the meaning of 
this sentence would be, 'Great scholars attribute a great 
value to this thing ' ; and the exclamation, ' Woe worth the 
day/ would mean ' May evil be attributed to this day/ Thus 
the worth of a thing means the value attributed to it." It 
is scarcely necessary to observe that the word ' worthe ' here 
is only the Anglo-Saxon weoriSe, from the verb weorSan 
(Germ, werderi), ' to become ' -, and ' well worthe/ ' wo worthe ' 


is ' well become/ or ' well be ' ; 'wo become/ or ' wo be.' 
Hence we find Gower writes : 

Hereof was Poliphemus wo. i. 163. 
. . and to beware also 

Of the perill er him be woo, Ib. 78. 
And Chaucer, 

And if so be my ladye it refuse 

For lack of ornate speech, I would be woe 

That I presume to her to writen so. iv. 130. 

Wo worthe the faire gemme vertules ! 

Wo worthe that herbe also that doth no boot ! 

Wo worthe that bewte also that is rowtheless ! 

Wo worthe that wight that tret ech undur foot. v. 67. 

I. Words construed with the preposition ' of/ 

A. Verbs. 

(1.) Thank, of a thing, where we should now use for (Fr. 
remercier de) : 

And she ayen, in, right goodly manere 
Thanketh her of her most frendely chere. 

Chaucer, iv. 252. 

They may now, God be thanked of his lone, 
Maken her jubile, and walk alloone 
* * * # 

Save that to Crist I sayd an orisoun 
Thankyng him of my revelacioun. Id. ii. 109, 110. 
So also the same construction occurs twice in the Preces de 
Chauceres at the end of the Canterbury Tales, vol. iv. p. 102. 4. 
The person thanked is governed by the preposition 'to/ 

And to our hihe goddes thanke we 
O/"honoures that our eldres with us left. Id. iii. 194. 
(2.) Pray, beseech, &c. of a thing : 

So longe preyeden they the king of grace 
Til he his lif hath graunted in the place. 

Chaucer, ii. 751. 
And him o/lordschip and of mercy prayde. Id. i. 147. 


Besechyny her of mercy and of grace, 

As she that is my lady soveraine. Id. iv. 207. 

Pray, beseech, &c. to a person : 

And mekely she to the sergeant preyde, Id. iii. 141. 

(3.) Reprove, upbraid. &c. : 

And therefore, sir, syth that I yow nought greve 
Of my poverb no more me repreve. 
Now, sir, o/elde ye repreve me, &c. Ib. 85. 

(Cf. ii. 77.) 

Soo pleyne she was, and did her fulle myghte, 
That she nyl hiden noothyng from her knyghte, 
Lest he q/any untrouthe her upbreyde. Id. vi. 184. 

. . and whenne synne regneth among greet men : and thei dreden 
of worldli harme : thei doren not snybbe men of this synne : lest her 
order leese worldli help. Wiclif, Three Treatises, p. xxxvi. 

(4.) Hearken : 

. . and this wey is cleped penitence. Of which men schulden 
gladly herken and enquere with al here herte, to wyte what is peni- 
tence, &c. Chaucer, C. T., Persones Tale. 

Perhaps, however, in this instance, the verb ' herken/ by a 
kind of attraction, takes the same government as the verb 
1 enquere.' 

(5.) Espy: 

And as God wolde, he gan so faste ryde 
That no wight of his countenance espyed. 

Chaucer, T. & C. vi. 6. 

(6.) Know: 

And sythen that I knewe of loves peyne. Id. i. 146. 
The same construction as of the verb ' to wite.' 
Fuljitel woot Arcite of his felawe. 

Id. i. 137, and frequently. 

(7.) He/p o/=cure of: 

Ther nas quyksylver, litarge, ne brimstone 

Ne oynement that wolde dense and byte, 

That him might helpen q/his whelkes white 

Ne o/Hhe knobbes sittyng on his cheekes. Id. i. 103. 


(8.) Give: 

He gaf nat of that text a pulled hen 
That seith that hunters been noon holy men. Id. i. 85. 
Here, f of' is probably = f for.' And, ' he gaf not of/ to be 
explained by, ' he would not have given for/ 

(9.) Pass, &c. : 

. . whose hevenly figured face 
So pleasaunt was, and her wele shape person 
That of beauty she past hem everichon. Id. iv. 242. 

That, as me thought, of goodlihede 

They passeden alle, and womanhede. Id. vi. 48. 

For al the worlde, so hadde she 

Surmountede hem alle of beaute 

Of manere and of comelynesse 

Of stature and of so wel sette gladnesse, &c. Ib. 162. 

[But also ' pass in ' : 

That no wight passed hir in hardynesse 

Ne in lynage, ne in other gentilnesse. Id. iii. 195.] 

(10.) Remember: 

And for to doon his observaunce to May 
Remembryng of the poynt of his desire, 
He on his courser, stertyng as the fire 
Is riden into feeldes him to pleye. Id. i. 136. 

[Another construction, 

This noble wyf Prudence remembered hire upon the sentens of 
Ovide, in his book that cleped is the Remedy of Love, &c. Chaucer, 
iii. 130. 

On Dorigen remembreth atte lest. Id. ii. 254.] 

(11.) Hope: 

For sothly, whil contricioun lastith, man may ever hope of for- 
givenes. Chaucer, iv. 30. 

(12.) Cease: 

Sche never cessed, as I writen finde, 
O/*hire prayer and God to love and drede. 

Chaucer, iii. 10. 


(13.) Teach: 

Tho gan sche him ful besily to preche 
Of Cristes come, and of his peynes teche. Chaucer, p. 1 7. 
(14.) This Preposition is also used after verbs in expressing 
not immediate dependence, but a more general relation to the 
action contained in the verb. Thus, 

Whanne that Anelyda, this wofull quene, 
Hathq/'herhande written in this wise. Chaucer, vi. 191. 

Right ofhir honde a letter maked she. Chaucer. 
And again, Tr. & C. b. ii. 1005 (Tyrwhitt). Fr. de sa main. 
We should now say, * with her hand/ 

(15.) In the partitive construction. In the case of certain 
verbs the action of the verb is extended, not to the whole, 
but only to a part of its object. The preposition 'of in 
these instances expresses the same relation as is expressed 
in Greek by the genitive case. But this usage is more limited 
in modern, than in early, English. Chaucer not only says 

Or gifus ofjoure braune, if ye have eny, ii. 105 ; 
(with which we may compare ' Give us of your oil, for our 
lamps are gone out/ Matt. xxv. 8, Auth. Vers.), but 
Of smale houndes hadde sche, that sche fedde 
With rested fleissh and my Ik and wastel breed. 

Prol, C. T. i. 83. 

Perhaps to the same principle may be referred the con- 
struction : 

It snewed in his hous of mete and drynk, 
Ofalle deyntees that men cowde thynke. Id. i. 92. 
and that in the following passages : 

And then seyde Gamelyn, ' So mot I wel fare 

I have nought yet halvendel sold up my ware.' 

Tho seyde the champioun, ' So brouk I my swere, 

He is a fool that thereof buyeth, thou selleth it so deere. 

Id. i. 248. 

And in the schippe me drewe on heye 
And seyden alle that I wolde deye ; 
And leyde me long down by the maste, 
And of hire clothes on me caste. Chaucer, vi. 89. 


Wiclif also employs the partitive construction : 

Crist parted with folke of goodis that he had. Three Treatises, 

' To eat of/ 'to take of/ (found in Chaucer, vol. iv. 32), and 
* to give of/ all occur in the authorised version of the Bible, 
and with these verbs the construction is still not obsolete. 

(16.) A sort of pleonastic use : 

And thus though I that lawe obeie 

Of which that kinges hen put under. Gower, i. 117. 

(17.) Where e of, 3 = ( concerning 9 : 
For thi I lere yow, lordes, 
Leveth swiche werkes 
To writen in wyndowes 
O/youre wel deedes. Piers PI. Vision, 1493. 

Passives. The construction with the preposition 'of after 
passive verbs and participles is extremely common. This 
construction has been retained to a very considerable extent 
in the authorised translation of the Scriptures, but modern 
English has almost entirely discarded it, substituting mostly 
the preposition ' by' for the preposition 'of'*) : 
Thus in delyt he liveth and hath don yore 
Biloved and dred, thurgh favour of fortune, 
Bothe of his lordes and of his comune. Chaucer, ii. 127. 

[But in iii. 154 we have, 

. . that he be bilovid with his subgites and with his neighebours.] 
Al was this land fulfilled o/fayrie. Ib. 73. 

Though here doubtless < fulfilled ' = < filled fall.' Cf. vi. 181 

et al 

Who that holt him payd of his povert, 

I holde him riche, al had he nought a schert. Ib. 84. 

Cf. Wiclif (Three Treatises, cxxxviii.) : 

Antecrist holdeth hym a pay^ed of this : and punysheth hem not 

* The French carefully distinguish as to the use of de or par after 
passive verbs. 


[But also with the preposition ' on ' : 

. . and therfor Cristis apostlis and other disciplis longe after hem, 
weren not bisie aboute dymes : but helden hem payed on litil that 
the puple $af hem redily. Ib. p. xxviii.j 

. . false freris that blynden myche puple bi colour of hir clothes ; 
the wheche were never grounded of God, &c. Wiclif, ut supra, cxlii. 
. . for ellis eche pope were blessed ; aljif he were falsly chosun of 
fends. Ib. xxi. 

And that was proved well by night 
Whilome of the maidens five. Gower, ii. 10. 
Now sith that I have told yow of whiche folke ye schul be coun- 
selled, now wil I telle yow which counseil ye ought eschewe. Chau- 
cer, iii. 144. 

[Just before, however, we have the same word construed with 
the preposition ' by ' : 

And werke nought alwey in every need by oon counseilour alloone ; 
for som tyme byhoveth it be counselled by mony.] 

Tho they were served of messes two or thre, 
Than sayde Gamelyn, * How serve ye me ? ' 

Chaucer, i. 253. 

In this instance we should now employ the preposition ' with ' 
instead of ( by/ The preposition ' of seems, however, to con- 
vey somewhat of a partitive meaning, = ' they were served 
with some of/ &c. : 

The following passage illustrates more than one of the 
constructions already noticed : 

Then am I fed of thai they faste, 

And laugh of* that I se hem loure, 

And thus of that they brewe soure 

I drinke swete, and am wel esed 

Of that I wote they ben disesed. Gower, i. 167, 168. 

B. Adjectives. 

The adjective is frequently followed by the preposition 
'of/ and a noun, when the noun serves still further to 

* Cf. Chaucer, i. 136 : 

And fyrv Phebus riseth up so bright 
That all the orient lauyheth of the light. 


define and limit the meaning of the adjective. Generally, 
the noun thus dependent on the adjective will be a thing, 
more rarely a per 'son. 
(1.) Clean: 

A good wyf that is dene of werk and thought. 

Chaucer, iii. 241. 

Chaucer uses both constructions, ' clean o/ J and ' clean in ' : 
Thise manner wymmen, that observe chastite, 
Muste be dene in herte as wel as in body. . . . 
and it bihoveth that sche be holy in herte, and dene o/"body. 
Persones Tale, iv. 90. 

But ' clean of/ is also used as = ' pure from.' 

For be we never so vicious withinne, 

We schuln be holde wys and dene q/synne. Id. ii. 77. 
(2.) Large: 

She loveyde as man may do hys brother, 

Of which love she was wounder large. Id. vi. 164. 
[Wiclif uses this adjective with the preposition *to' before 
the personal object : 

Crist and hise apostlis weren large to the puple. Three Treatises, 
p. cxliv.] 

(3.) Rich: 

But riche he was of holy thought and werk. 

Chaucer, i. 97. 

O Salomon, fulfilled of sapience 
.... and richest o/"richesse. Gower. 

"(4.) Big:- 

The mellere was a stout carle for the nones, 
Ful big he was q/'braun and eek of bones. 

Chaucer, i. 100. 
(5.) High: 

So high he set him selfe above 
Of stature and of beaute bothe 
That him thought alle women lothe. Gower, i. 118. 

(6.) Worthy: 

This knight Brauchus was of his honde 
The worthiest of all that londe. Gower. 



And all they were so worthy of hir honde 
In hir time that none might hem withstond. 

Chaucer, iv. 255. 

(7.) Wise: 

And oon of hem, that wisest was of lore 

* * * * 

He to themarquys sayd, as ye schulnhiere. Id. ii. 127- 

(8.) Most: 

He was a jangler, and a golyardeys, 

And that was most o/synne and harlotries. Id. i. 100. 

i. e. one who was most given to sin, &c. 
(9.) Least: 

And were it the foulest cherl, or the foulest womman that lyveth, 
and lest of value, &c. Id. iv. 1 7. 

Chaucer has a vast number of such constructions; such as 
'gentle of kinde' (iv. 244), ' daungerous of speche/ 'expert 
o/lawe' (i. 101), &c. 

(10.) Negative adjectives : 

This Galathe, saith the poete 

Above all other was unmete 

O/"beaute, that men thanne knewe. Gower, i. 163. 

' Nay/ quod the fox, * but God give him meschaunce 
That is so undiscret of governaunce.' Chaucer, iii. 234. 
(11.) Gower uses the phrase 'to be glad of a person* : we 
now say only ' to be glad of a thing. 3 Just as we have seen 
before he speaks of e setting the heart upon a person/ whereas 
we apply the expression only to a neuter object : 
They toke her into felaship, 
As they that weren of her glade. i. 184. 

An instance of this construction likewise occurs in the autho- 
rized version of the Bible (Isaiah, xxxix. 1, 2) : 

" At that time Merodach-Baladan, the sou of Baladan, king of 
Babylon, sent letters and a present to Hezekiah, for he had heard 
that he had been sick and was recovered. And Hezekiah was glad 
of them, and showed them the house of his precious things, the silver 
and the gold," &c. 

It is true no personal object is mentioned, but messengers 


are implied in the verb ' sent/ and that Hezekiah was glad of 
the messengers, not of the letters, is clear from the following 
' and showed them/ &c. 

II. Words construed with the preposition ' to ' or ' unto/ 

(1.) Verbs implying obedience, &c. : 

No man may serve to two lordis. Wiclif. 

. . for siche semen not to Crist ; but seruen to her womb. Id. 
Three Treatises, p. cxxii. 

But men most nede unto her lust obeye. 

Chaucer, ii. 140. 

.... and ben redy to obeye to alle youre commandements. Id. 
iii. 179. 

And she to his bidding obeid*. Gower, i. 128. 
But also without the preposition. 

And hire obeie, and folwe hire wille in al. Chau. ii. 227. 
(2.) Command, &c. 

. . but ilche man myte ylyche comaunde\ to other. Wiclif, 
Three Treatises, p. xcv. 

. . but Crist bad to the poor man : let ye dede birye the dede. 
Ib. p. cxlix. 

With the verb 'enjoin/ Wiclif uses the preposition, not 
before the personal object, but before the thing enjoined : 
. . thei enjoy nen hem to brede and watur and to go barefote. Ib. 

* We have a lingering remnant of this construction in the authorized 
version of Acts, vii. 39 : " . . to whom our fathers would not obey." So 
also in Rom. vi. 16 : " his servants ye are to whom ye obey." 

f There are a few instances in the authorized version of the Bible where 
the verb ' command ' is thus followed by the preposition ' to ' : 

Then an herald cried aloud, To you it is commanded, O people, nations, 
and languages. Dan. iii. 4. 

And unto the married I command. 1 Cor. vii. 10. 

It is to be observed, however, that here (and the same holds good of the 
examples cited under the verb 'to obey') the pronoun precedes the verb, 
and the preposition seems to be demanded by the emphasis, as will be 
evident to any one who will try to read these passages without the pre- 

N 2 


(3.) Injury, harm, &c. : 

TOUCH. To speke or jangle in any wise 

That toucheth to my ladies name. Gower, i. 177. 

ANNOY. Salamon saith, that right as motthes in schepes flies 
annoyeth the clothes, and the smale wormes to the tre, right so 
annoyeth sorwe to the herte. Chaucer, iii. 131. 

TRESPASS. Al be it so that of your pryde and heigh presumpcioun 
and folye. ... ye have mysbore yow, and trespassed unto me, &c. 

And, again 

. . that God of his endeles mercye woll at the tyme of our 
deyinge forgive us our giltes, that we have trespassed to him in this 
wretchid world. Ib. p. 181. 

(4.) Profit: 

For it spedith to thee that one of thi membres perische than all 
thi body go into helle. Wiclif, Transl. of Matt. v. 

(5.) Accord: 

And after this, thou schalt considere the thinges that accorden to 
that purpos for to do by thy counseil, if resoun accorde thereto, 
and eek if thy might may accorde thereto, and if the more part and 
the better part of thy counseillours accorde thereto or noon. Chau- 
eer, iii. 146, 7. 

Wiclif also uses this construction. Elsewhere Chaucer 
prefers the preposition ' with'*. 

The wise Plato saith, as ye may rede, 

The word mot neede accorde with the deede, 

If men schul telle properly a thing 

The word mot corde with the thing werkyng. iii. 243. 

(6.) To forgive, &c. 

Forsothe $if $ee schulen/o^yw to men here synnes. Wiclif f. 

* So also in Wiclif we find the verb ' assent ' (as well as the noun), fol- 
lowed by the preposition * with ' : 

For fals mayntenyng makith eretikis, and so assente with siche falshed 
bryngith inne ofte eresies, and Crist wote not assente with thes : for thei 
may not be sothe. Three Treatises, &c. p. xxiv. 

t So in 2 Cor. ii. 10 (authorized version) : 

To whom yeforyive anything, I forgive also; 

where the same remark holds with regard to the emphatic position of the 
pronoun as above, p. 169, note f- 


. . the whiche shul not spare to the folk of God. Wielif, Three 
Treatises, p. cxxii. 

(7.) To have need : 

hole men han no nede to a leche. Wiclif. 
And if thou to us have neede > thou schalt fynde us prest. 

Chaucer, i. 264. 

for youre fadir woot what is nede to you. Wiclif. 

(8.) The following constructions with this preposition are 
also worth notice : 

(#.) And if there be to this matere 

Some goodly tale for to here. Gower, ii. 3. 
where ' to ' = ' in reference to/ 
(6.) Whilom Eneas 

Whom Anchises to sonne hadde. Gower, ii. 4. 

And he a lusty maide 
To doughter hadde. Ib. 43. 

. . whilom I was one 

That to my fader hadde a kinge. Ib. 48. 

We can still say 'to have to wife/ but not 'to son/ 'to 
daughter, 7 &c. 

(c.) This proude king let make a statu of gold, 
Sixty cubites long, and seven in brede, 
To which ymage bo the yonge and olde 
Comaunded he to love and have in drede. 

Chaucer, iii. 1 92. 

Here, either c to love' = ' to have love'; or, 'to* may be 
used in an indefinite way for ' with regard to * ; or possibly 
for ' loue ' we ought to read ' boue/ L e. ' bow/ 

(9.) After adjectives and nouns : 

The pover childe is bore as able 
To vertue as is the kinges sone. Gower, i. 269. 
Eke thou that art his sone art proud also, 
And knowest al this thing so verrayly, 
And art rebel to God and art his fo. Chaucer, iii. 194. 
. thei senden maundementis thikke aboute for covetise of ve- 


niaunce to curse and to putte out of chirche for rebelnesse to hem. 

Wiclif, Three Treatises, p. cxlv. 

A little further on we have ' rebelnes ageyns God/ 

III. Use of the preposition ( upon/ 
Gower uses this preposition in a very peculiar manner. 


For God 

.... hath set him but a little while 
That he shall regne upon depose. 

i. e. subject to deposition : or, so as to be deposed ; on such 

terms and conditions. 

And she upon childehood him tolde 

That Perse her litel hounde was dede. i. 219. 

i. e. as you would expect from a child, in her child-like way. 

Here it is not so easy to see the exact force of the preposition. 


. . so that upon his trecherie 

A lesinge in his herte he caste. i. 187. 

And similarly Chaucer, 

Ere ye doon eny execuccioun, 

Upon your ire for suspeccioun. iii. 245 . 

i. e. because you are angry, or, on the ground of your anger. 

In each of the last two instances the preposition will allow of 

a similar explanation, viz. : ( on the ground of.' 

[To be continued.] 



[Read November the Gth.'] 

Since my former communication on the connexion of the 
Finn and Lapp with the languages of Western Europe, I have 
had an opportunity of inspecting the grammars and voca- 
bularies of other members of the Ugrian race, reaching to 


the extremities of Siberia, and in all of them I find, to a 
greater or less extent, instances of a similar community of 
radical forms, strongly corroborating the evidence already 
produced of a primaeval connexion between the languages of 
the Ugrian and Indo-European classes. 

It might be suspected that the Lapp wuoksa, an ox, was 
merely a modification of the Swedish oxe, but the same name 
may be recognized in Ostiak uges, Syrianian 6s, Wotiak oj 
(French j), and even in the Turkish ogys. 

In like manner Finn porsas a pig, agreeing in so striking a 
manner with Lat. porcus, corresponds to Syr. pors, Ost. 
pur 'ash , Wot. parj (Fr.y), Samoiede pares, pores. 

The names of two of our wild berries are cranberry (i.e. 
crane-berry) and crow-berry, and it is remarkable that they 
are known by names having the same meaning among the 
Samoiedes, with whom the word for crane is kar, karra, or 
in another dialect haru, haro (which may perhaps be radically 
identical with crane, W. gar an), while the name of the cran- 
berry is kar an af or kara chober, harun ode -, chober or odea 
signifying a berry. The name of the crow-berry is warno, 
while warna is a crow, agreeing with Lith. warna, Servian 
wrana and W. bran. Esthon. warres, a crow, warresse mar- 
jad, crowberries. 

As the r of Lat. morus, a mulberry, changes to an I in the 
provincial Fr. molle (Vocabulaire de Berri) and E. mulberry, 
so we find Lapp muorje corresponding to Esthon. and Wot. 
muli, berry, fruit. In the latter language a berry is also de- 
signated by a name, bory, almost identical with the English 
word. In most of the Ugrian languages the word for grease 
is voi or vai, as Ost. voi, tallow, fat ; Finn and Esthon. voi, 
butter ; Lapp wuoi, butter, oil. The Albanian has a double 
form, voi or vai, and valj, by which the Finnish voi is con- 
nected with E. oil. 

Other instances in which the Albanian seems to connect 
Ugrian and Indo-European roots may be cited, as 

Esthon. HI, lillik ; Alban. Ijoulje, a flower ; Lat. lilium. 

Turk, bulbul, a nightingale ; Alban. bilj bilj ; Gr. 


Syrianian #0; (Fr, j), joy, gajma, I rejoice ; Alban. gazelim, 
gezim, joy, gezoig, I rejoice ; Lat. gaudeo. 

Several unsatisfactory explanations of the word king have 
been given from different Teutonic roots. An inspection of 
the Ugrian forms will convince us of the radical identity of 
the word with the Tartar chan, the g of E. king, or ig of G. 
konig being in all probability a suffix originally of diminutive 
signification. Thus we have Ostiak chon, emperor; Wot. 
kun, king, emperor; kunlen, queen; kunoka, lord, chief; 
Lapp konogas or konoges, king ; Lith. kuningas, a proprietor 
of a higher class (petty proprietors being addressed by the 
title pori), and especially the pastor of the parish; kuningene, 
the pastor's wife. 

The Greek apa, a prayer^ may perhaps admit of explanation 
from the Syrianian ara, a song ; Turum-ara, prayers, literally 
a God-song, from Turum, God. 

As examples of verbal agreement may be cited 

E. pot; Finnj0ta; Ost. put; Cheremiss. pat, a kettle. 

E. weather; Pol. wiatr; Ost. wot, wind. 

E. teat ; G. teitze ; Alb. tsitse ; Ost. tuti, the breast. 

E. name; Lapp namm, namma; Ost. nem; Wot. nim, 
name; nimo, celebrated. 

E. meed ; G. miethen, to hire ; Ost. mit ; Syr. Wot. med, 
reward, wages ; Ost. midaden ; Wot. medjalo, to hire ; Wot. 
medOj hired person, 

E. to bore; Lat. for are; Hung, furni, to bore, fur 6, a 
borer; Ost. por, par, a borer; ket-por, an awl (&/=hand); 
SamoieAe parti ; Finn pur as, a chisel, borer, scalprum fabrile, 
terebra sculptoria. 

E. cot; Esthon. koddo, a house; Lapp kate, a tent, house; 
Ost chot. 

E. must; Ost. most, in the same sense, tede most, one 
must eat. 

Sc. gang, to go ; Ost. jangam. 

E. nasty ; Syrian, njasti, sordes, njasties, sordidus. 

E. lip; Syrian. Ijijb. 

E. son; Syr. zon. 


E. lime, bird-lime; Esthon. limma, slime; Syr. lam, glue. 

E. latch, lace; Syr. latsj, laqueus. 

W. cogel, a distaff; Lapp kakkel; Syr. kozjalj. 

E. year; Icel. ar, harvest, produce of the earth, year; 
Syr. ar, autumn; Wot. ar, year, aran, harvest. 

Lat. multus; Wot. multes, much. 

E. kill, quell; Dan. quale, to choke; Esthon. kolima, to 
die; Finn kuolia, to die, kuolettaa, to kill; Wot. kulo, to 
die, kulem, dead, kulto, to kill ; Syrian, kula, I die ; kulj, a 
water-demon, always on the watch to drown his victims 
(a Nicker.) 

E. border, O.-E. brade, a brim, as brade-ful, brim-ful; 
Icel. bard, hatt-bard, ala pilei; Syrian, bord, ala, axilla, an 
edge, brim, as nyr-bord, nares, ala nasi ; Wot. bord, side. 

E. burn, a brook; G. brunn, a spring, well; Syrian, burnja, 
a well. 

E. sere, dry; to sear, to dry up; Hung, szdras, dry, szdrit, 
to become dry; Ost. sorom, dry, sorettem, to make dry, to 

E. brink; Pol. brzeg, edge, brink, shore; Bohem. breh-, 
Wot. bereg, shore. 

E. care; Lat. cur a-, Wot. kur, care, sorrow. 

E. sinew; Esthon. soon, sinew, vein; Wot. son, sinew, 
son-wir, vein (wir=s blood). 

But perhaps a more striking instance than any of these, is 
the occurrence of a word of so abstract a meaning as the 
relative such, as if transplanted from English, in the Wotiak 
sotsche. Nor is it a mere coincidence of sound, but the 
logical construction and manifold relations of the word are 
the same in both cases. The formation of such is apparent 
from the Goth, sva-leiks, O.H.G. solih, G. solcher, O.E. swilk, 
and with absorption of the I, swich, such, literally so-like. 
The Gr. T^/CO?, Lat. tails, O.E. thilk, Prov.E. thic, Lith. 
tokis, are equivalent forms with the dental instead of the 
sibilant modification of the demonstrative, and with a similar 
loss of the / in the two last forms. Corresponding to these is 
a series formed of the same termination, with a relative 
instead of a demonstrative particle ; Goth, hvileiks, Gr. 



7rr)\LKo<; (for KIJ\IKO<S), Lat. qualis, E. whilk, which, Lith. 
kokis. A fourth series must be considered as formed of the 
indefinite pronoun is, he, with the same termination, having 
the signification of ' ejus formse/ ' ejus speciei/ Of this series 
we can only point out Gr. fjXucos, of such a size, A.S. ilc, 
same. Now the Wotiak possesses analogues of each of the 
four series above mentioned. It has both forms of the demon- 
strative, ta, this, and so, that, which are confounded with each 
other in Sanskr. sa, sd, tat, Gr. 6, rj, TO (with an aspirate 
instead of an initial s], A.S. se, seo, that ; and from these 
are formed tatsche, the equivalent of Gr. TTJ\,IKO<; } Lat. talis, 
E. thilk, thic, Lith. tokis-, and sotsche, corresponding to Goth. 
svaleiks, G. solcher, E. swilk, such. The analogue of Goth. 
hvileiks, Lat. qualis, E. whilk, which, Lith. Mis, is the 
Wotiak ketsche, what, of what kind, and with a negative, 
no-ketsche, of no kind; and between ketsche and sotsche 
there is an indefinite etsche, of such a nature, him-like, cor- 
responding to Gr. r)\t,Kos and A.S. /c. The Wot. kyzi, 
manner, and adverbially in the sense of how, as, seems iden- 
tical with G. weise, E. guise, wise, Bret, giz, kiz; Wot. nokyzi, 
nohow, in nowise. 

I have said in accordance with the general doctrine that 
such is to be explained so-like, and in the same way is com- 
monly understood the termination in words like Goth, sama- 
leiks, Lat. similis, G. dhnlich, E. slovenly, but the Lapp 
enables us to take up the explanation at an earlier stage, and 
to treat the idea of likeness itself as a secondary formation. 
In that language the substantive lake, manner, custom, mode, 
is also used as a termination equivalent to E. ly, by which 
adjectives are converted into adverbs. Mann lakai? kutte 
lakai ? in what manner, how ? Arges laka, in a timid man- 
ner, timidly, from arges, timid; heimalaka, homely, more 
domestico, tanquam domi, from heima, home. From this 
element is formed the adjectival termination lakats, the exact 
equivalent of the Goth, leiks, Lat. Us, and G. lich. Thus 
from kalkos, slow, are formed kalkos laka, slowly, and kalkos 
lakats, slowish; from akta, one, aktalaka, in one and the 
same manner, aktalakats, sequalis, sirailis, of one form or 


mode of being (whence it would seem that the Sanskrit eka, 
one may constitute the first element of Lat. sequalis) ; to 
lakats, of your nature, tui similis, your like ; tann lakats (tan 
being the genitive of tat, that), literally, of that nature, talis ; 
mann lakats, of what nature, qualis. 

A nearer approach to the Lapp form was preserved in the 
O.E. termination lock, where we now use ly. We find in 
Layamon, kenlok, bold, from A.S. cen, keen; wod-lok, syno- 
nymous with wood, mad; worthlok, from worth, worthy; 
grislok, grisly, frightful. The same element is also employed 
in the formation of substantives, regularly in Icelandic, and 
in one or two scattered instances in English. Thus in the 
former language, from karg, obstinate, is formed karg leiki, 
obstinate condition or character, obstinacy ; from rosk, brave, 
roskleiki, bravery. In English we have wed-lock, wedded 
condition ; knowledge, formerly knowleche, the form or scheme 
of what is known, or condition of one knowing; and in A.S. 
reaf-lac, the condition designated by the term rob, robbery. 
The Esthonian form of the word luggu or lukku is explained 
state, manner, subject, condition, zustand, art, sache, be- 
schaffenheit. The Finnish form is lai, genus vel indoles rei, 
agendi modus, mos, giving rise to an adjectival termination 
lainen; pahan lainen, mali indolis; sen lainen, ejus generis, 
talis (the proper equivalent of E. such), minka lainen, of what 
nature, qualis, the equivalent of Lapp man lakats above 
mentioned. Here doubtless is shown the import of the ter- 
mination lei in G. einerlei, of one kind, alter -lei, of all kinds. 
The course of development in meaning is probably, look, 
countenance, appearance, form, mode of being ; Servian lik, 
vultus ; Pers. liqa, facies, vultus, forma (Dieffenbach) ; O.E. 
IcBche, liche, form; l&che, leche, look, countenance, gesture 
(Layamon) ; Lapp lake, custom, mode. Then with a prefix 
implying unity, community or identity of nature, Goth. 
galeiks, of common form, G. gleich, similar, like; Goth. 
samaleiks, of the same nature, Lat. similis; O.H.G. ana-lih, 
anagalih, A.S. anlic, G. ahnlich, of one form or nature, re- 
sembling, a meaning which has been transferred in most of 
the Teutonic dialects to the simple form like. 


The foregoing view of the original meaning of the adjec- 
tival termination lis, lich, ly, and of the mode in which the 
sense of like arises out of that of appearance, form, is corro- 
borated by several similar formations in the Finnish lan- 
guages. The substantive muoto is used in Lapp in the sense 
of face, appearance, form, image ; whence muotok, like ; attje 
muotok, like his father, having the appearance of his father ; 
muotolas, likeness. 

The meaning is extended in Finn to the mode or manner 
of doing anything, the word itself being probably identical 
with Lat. modus; niin muodoin, in that manner; monella 
muodolla, in many manners. From muoto is then formed 
the adjectival termination muotoinen, contracted into moinen, 
alicujus formse, gestaltet, ahnlich, equivalent to the Lapp 
lakats or Finn lainen, above mentioned; sen muotoinen, or 
sem moinen, of that nature, ejus generis, talis, as from lai, 
sen lainen, in the same sense ; isdnsa muotoinen, patri similis, 
from isa, isdn, father. So also from kuwa, form, figure, 
image, kuwainen, resembling ; from hahmo, form, appearance, 
hahmoinen, resembling. 

The Lapp has also wuoke, form, figure, appearance, manner 
(apparently from the same primitive root with Gr. 
seem, ei/ccov, an image, or with the digamma, Fei/ca), 
tan wuokai, in this manner, as tan lakai above mentioned. 
Hence wuokak, like, equal, and wuokok, or wuokasats, as an 
adjectival termination, equivalent to our ly ; piadnak, a dog, 
piadnak-wuokasats, or piadnak-lakats, dog-like ; akta-wuokok, 
or akta-lakats, uniformis, sequalis. 

In Esthonian also, the adjectival termination analogous to 
lis or ly, is formed from words signifying form, manner, sort, 
viz. kombe, and wiis, the latter identical with G. weise, E. wise. 
Sedda wisi, or sel wisil, or sel kombel, on this wise; latse 
wisil or latse kombel, in the manner of a child, child-like, 
childish. The employment of so many words, and especially 
of Esthon. wiis, signifying form or mode, in the formation 
of the adjectival suffix suggests an analogous explanation of 
the termination sam or some, in G. einsam, langsam, E. lone- 
some, gamesome. 


I endeavoured to show in a former paper that the meaning 
of E. wise, Bret, giz, kiz, manner, was derived from the sense 
of 'footsteps, traces, track/ which seems to be the original 
signification of the Breton word, the track or way to a place 
affording the most natural metaphor by which to express the 
mode of obtaining an end, or manner of doing anything. 
Now the Esthon. has sam, a step, corresponding probably to 
Alban. kame, a foot, kames, a foot passenger, W. cam, a foot- 
step (whence Fr. chemin, a way), as an initial s in the Finnish 
dialects often corresponds to a hard c in Latin ; and in Wotiak, 
where also the s of other Ugrian dialects is in other instances 
represented by aFr.y, we find, jam (Fr.y) signifying manner, 
way, closely approaching Fr. chemin, and through it uniting 
sam and cam above mentioned. 

The word, jam is then employed in composition in a mode 
exactly similar to Esthon. kombe or wisi, or to the Goth, leiks ; 
ta jamen, so ; muzon jamen, otherwise. If we consider the 
Ugrian element as identical with the Teutonic sam or some, 
it will explain in a satisfactory manner the force of that ter- 
mination in such examples as those above quoted ; einsam, 
one-wise, in the manner of one ; langsam, in long manner, 
slow ; gamesome, in the way of game. 


[Read December the 4<A.] 

IHRE somewhere observes that where a word is of so ancient 
a standing as to be common to several of the great ethno- 
graphic divisions, as Latin, Gothic, Celtic, it is in vain to 
search for the original derivation ; and doubtless in the vast 
extent of time that must have elapsed since those great 
branches of language can have separated from a common 
stock, there is ample opportunity for a particular form to 
have been lost in any given language, or to have become 



irrecognizable by continual change in sound and signification. 
But words are not, like material things, subject to the action 
of disintegrating forces, certain to produce their effect if a 
sufficient period of time be allowed. Their duration seems 
altogether matter of accident, and a root which was lost two 
thousand years ago, leaving perhaps a solitary derivative in 
Latin or Greek, may be preserved and largely developed in 
the uncultivated Finnish, Slavonic, or Celtic tongues. Thus 
we have shown that the purely Finn root mu (other), affords 
a satisfactory origin of the Lat. muto, Finn muuttaa, Esthon. 
muduma, to change. We have traced the Lat. macero, to soak, 
to a root mok or mak, signifying f wet/ which has a nume- 
rous progeny in the Slavonic and Gothic languages. In like 
manner the "W. coll, loss, damage, supplies us with precisely 
the form which is required to explain the Lat. incolumis, safe, 
without loss or damage. But as there is a deeply implanted 
prejudice against this kind of derivation, while the evidence 
in its favour gains rapidly in strength in proportion to the 
number of individual cases which are satisfactorily made 
out, I shall proceed in the present paper to the discussion of 
additional instances in which doubtful or wholly unsettled 
etymologies may be illustrated from the Finnish languages. 

To BORE, BURIN, BUR. The wide-spread range of the 
word bore was mentioned in a late paper (p. 174), in which it 
was identified with the Lat. for are \ Hung, furni, to bore, 
fur 6, borer ; Finn pur as , a chisel, terebra sculptoria ; pu- 
rastaa, to make holes with such a tool, scalpro terebro, 
sculpo ; and the same root was plainly recognized in several 
Siberian dialects. 

The Finn purr a, to bite, leaves little doubt of the primitive 
image from whence the expression is taken, the action of 
biting affording the most obvious analogy from whence to 
name the operation of a cutting instrument or the gradual 
working a hole in anything. The Icel. bit is used to signify 
the point or edge of a knife ; bitr, sharp, pointed. We speak 
in English of an edge which will not bite, and it is doubtless 
in the sense of the Icel. bit that the term centre-bit is applied 
to an instrument for boring. 


The analogy between the operation of a cutting instrument 
and the act of gnawing or biting leads to the application of 
the Finn puru, Esthon. purro, to anything comminuted by 
either kind of action, as puru, chewed food for infants, sahan 
puru, pu purro (pu = wood), sawdust, identical with O.H.G. 
urboro, uzboro, the gnawings as it were of the saw or borer. 
Probably also we may here have the origin of Fr. bourre, 
flocks or locks of wool, &c., used to stuff saddles, &c., also 
less properly (says Cotgrave), any such trash as chaff, shales, 
husks, &c. But if our theory be correct, the original appli- 
cation would be to sawdust or bran used for stuffing, then 
to flocks of wool, from their use in stuffing, and thence to 
down or nap in general. Hence Eng. bur, a flock of wool, 
then applied to the seed-vessels of certain plants which stick 
to one's clothes like a flock of wool. 

Again, from Finn purra, to bite, is derived purin, dens 
mordens vel caninus, a biter, giving a satisfactory explanation 
of burin, a graving tool, the tool with which the engraver 
bites into his copper-plate, a word which is commonly con- 
nected with the verb to bore. 

The Lapp equivalent is parret, to bite, and thence to eat ; 
whence parrets, an awl, a borer ; parremas, food ; parrestallet, 
to devour, eat greedily, vorare, cibo se ingurgitare. Hence 
we are led to Gr. fiopa, food, ftopos, voracious, fipoo), /Bpcoa-Ka), 
fiiftpcoa-KQ), to eat, from which it is difficult to separate Lat. 
voro, although the signification of the latter word is to swallow 
down, with apparent opposition to the notion of biting or 
chewing. But the notion of eating greedily, gulping down, 
might easily be expressed by a modification more or less 
distinct of the word for biting or eating, as in the Lapp 
example above quoted. Thus Lat. vorare and forare would 
be brought under the same root as slightly modified forms 
designating the act of biting applied to different purposes. 

AUGER. The consideration of the verb to bore naturally 
leads to that of auger, of which the explanation in Finn is 
singularly complete. It must be observed that auger is one 
of that numerous class of words which are used with and 
without an initial n, which may have been improperly added 


in the first case or omitted in the second. It was formerly 
written nauger in English, in A.-S. nafogar, Ober Deutsch 
nabeger, Pl.D. naviger, Du. neviger, eviger (KiL). Another 
A.-S. form was naf-bor, giving rise to G. nebber. The word is 
explained by the author of the Bremisch Worterbuch as sig- 
nifying such a tool as is used to bore the nave of a wheel. 
But the Finn establishes a connexion of a totally different 
kind with the nave of the wheel. In that language napa 
signifies navel, and thence the middle of anything, centre of 
a circle, axis of a wheel, anything which revolves, as from 
mere, the sea, meren-napa, a whirlpool; from rauta, iron, 
napa-rauta, the iron stem on which the upper millstone rests 
and turns; maan-napa, the axis of the earth; Lapp nape, 
navel, centre, axle. With kaira, a borer, the equivalent of 
A.-S. gar, the Finn forms napa-kaira, precisely corresponding 
to the common English name of the tool, a centre-bit ; the 
first element in the English, as in the Finnish word, indi- 
cating the nature of the action, namely the revolution of the 
tool round a fixed axis or centre. 

The root of the Finn kaira, a piercer, is preserved in the 
English expression of being gored by a bull, i. e. being torn 
or transfixed by his horns : 

Oh, be advised, thou knowest not what it is, 

With javelin's point a churlish swine to gore. Shakspeare. 

A.-S. gar, a javelin; gara, an angular point of land, a pro- 
montory, seem named from their pointed shape. Finn kairi, 
a gore in a garment, is a pointed piece of cloth let in to 
increase the width downwards. Perhaps Gr. <yapov, Lat. 
garum are originally so named from the pungent biting taste 
of the sauce which they designated. 

TURNIP. The force of the Finn napa seems also to explain 
Lat. napus, A.-S. nape, a turnip, as a root shaped like the 
nave of a wheel, spindle-shaped, or having an axis projecting 
out of the centre. The ordinary name of a turnip in Finn 
is nauris, whence nauriin napa, radix rapse perpendicularis ; 
napakka, a long tap-shaped turnip; napoan, nawota (exhi- 
biting the same relation between^ and v as in Lat. napus, 


Fr. navet), to cut off the roots of turnips. The syllable 
turn seems to have been added in English to express the 
same axial or spindle-shaped character which was conveyed 
by the Finn napa, but was no longer sensible to an English 
ear in the element ruep or nep. The Esthon. equivalent 
nabba, the navel or centre part of anything, is applied 
to other tap-rooted plants. Nabba-juur (juur=ioot)j tor- 
mentil or burnet. 

Amv. In the Finnish dialects liika, like, liig, signify excess, 
superfluity, unfitness; Finn liika, a tumour in the body, 
excrescence on a tree, the oblique cases of which are used in 
the sense of Lat. nimis ; liian suuri, too great (identical with 
Gr. Xtav) ; liian paljo (paljo = 7ro\vs, much) or Hiaksi f or 
liialta, too much ; teen liika, I do what is not fit ; liika-wieras, 
an uninvited guest ; liika-liha, proud flesh ; liika-aika, tempus 
vacuum, otiosum ; liika-nimi, cognomen ; liikenen, liieta, to re- 
main over. Lapp like-namm, a surname ; like-mana, a bastard ; 
likai } to boot. Esthon. liig-assi, ungerechte sache ; liig juus, 
false hair, perruque; liig kitsi, too tight; liig naene, con- 
cubine; liig-nimmi, zuname, surname, nickname; liig-te, by- 
way, wrong way ; liig-pajatus (pajatus = speech), lies, unpro- 
fitable talk. I have given so many examples of the force 
of this Finnish element on account of the light which it 
throws on several English etymologies where we have hitherto 
been either quite at a loss or troubled with a superfluity of 

NICKNAME. The present word is one of the same class 
with auger and nauger above mentioned, which vary with and 
without an initial n. In O.E. eke-name was current ; neke- 
name or ekename, agnomen (Prompt.) ; in Sw. oknamn, G. 
eich-name, ekel-name, as well as neckname. Here we have 
three plausible derivations : 1st, from Sw. oka, Eng. eke, to 
increase or lengthen out, as if the intention were to signify 
an additional name. But oka or eke are not used in compo- 
sition in the sense of additional in any other instance ; 2nd, 
from G. ekelj disgust, as a name given from dislike ; and 3rd, 
from necken, to tease or banter, a name given in banter or 
ridicule : 


" Susurro, a privy whisperer or secret carry-tale, that slaundreth, 
backbiteth and nicJceth one's name." Junius Nomenclator in Way's 

The last explanation would do very well if it stood by itself; 
but if such be the true origin of nickname, it is plain that in 
the formation of eke-name and ekel-name, the principle has 
been at work to which we have often alluded, in virtue of 
which, when some element of a compound word has lost its 
significance by lapse of time or introduction into a foreign 
language, a blind attempt is made to support the meaning 
which it seems to bear in the expression by such a modi- 
fication of the sound as may serve to give the significance 
required. And this I believe has really been the case with 
nickname itself. The derivation from the form corresponding 
to Lapp like-namm, signifying additional name, seems far 
more probable ; and in the case of G. zu-name, we see a word 
of this general significance also applied to a name given in 
ridicule. The interchange of an initial I and n is very 
common. The Lat. lympha and nympha seem both to have 
signified water, whence nymphtea, a water-lily, the Finn name 
of which has also a double form, lupukka and nupukka ; in the 
same language laskata and naskata, Sw. laska, to lash, to 
sew leather edge upon edge; Eng. level corresponds to Fr. 
niveau, It. nivello, livello ; Lapp lakkula, Sw. nyckel, a key. 

LEISURE. The compound liika-aika, vacant time, leisure, 
affords a plausible origin of Lat. licet, it is open to you to do 
so and so, you are permitted to do so. The significations of 
permission and leisure readily pass into each other, and as we 
suppose the Lat. licet to arise out of a word signifying leisure, 
so doubtless the word leisure itself, Fr. loisir, is derived from 
the infinitive licere, as plaisir from placere. The Provenyal 
equivalent lezer is rendered by Raynouard loisir, permission, 
moyen. ' A selat o per lezer/ in secret or with permission. 

The like development of meaning may be seen in the form- 
ation of G. mussen, Eng. must, from musse, leisure, where the 
verb must originally have had the signification of Lat. licere, 
and must thence have proceeded to imply ' necessity ' by a 
figure of speech, inasmuch as the permission given by a supe- 


rior is often only a civil cloaking of command. We have thus 
the same idea of necessity or compulsion expressed in Latin 
by opus, work, and in Teutonic by the exact opposite musse, 
leisure, opus est, you must; while in Latin itself the two 
opposites are expressed by the same word opera, in one appli- 
cation signifying work, and in another leisure. 

LIE. On a similar principle to that on which from Finn 
liika-aika has been deduced Lat. licet, the Esthon. liig-pajatus 
above mentioned may be regarded as the origin of Slavon. 
lugati, Goth, liugan, Ger. lugen, to lie, viz. by supposing in 
the one case the loss of the element signifying time, and in 
the other of that which signifies speech. 

The root we are considering gives rise to a considerable 
number of words in Lith. : lykus, excess, overplus ; likti, to 
remain over, to leave (the equivalent of Finn liikenen, liieta, 
to be superfluous, to remain over, and of Lat. linquere, relin- 
quere, relictum, to leave) ; lekas, over and above, odd ; lekani, 
daikti (daikti = things), relics. The numbers from ten to 
twenty, weno-lika, dwa-lika, &c., must be explained, one, two, 
&c. in excess (over ten). In the Slavonic the root gives Boh. 
licho, an odd number ; lichy, odd, mean, wrong, unjust : and 
probably lichwa, usury, to be compared with Esthon. liig-otus, 
usury, excessive interest, where the element signifying interest 
is lost, as in liika-aika and liig-pajatus above mentioned. 

LACK. To the same root must probably be referred Eng. 
lack, to want, although it seems to signify the reverse of excess ; 
but it must be remembered that opposites are often expressed 
by the same root, or slight modifications of the same. So in 
English, to cleave, is expressed the adherence of two things 
together, or the separation of one thing into two; the It. 
caldo, hot, is radically identical with G. kalt, cold, and in 
Lith. we have sziltas, warm, szaltas, cold; the Fr. blanc, 
white, is essentially the same word with E. black; the Lat. 
opera is used, as we have just observed, in the apparently 
opposite applications of work and leisure. Now what we lack 
may be considered as something in excess of what we have, 
and the Gr. \et7ro>, the correlative of the Lat. linquo, is also 
used in the sense of wanting, being deficient. The factitive 


form of the Lapp verb from like, excess, is likotet, which 
regularly should signify to cause to be superfluous, but is 
translated perdere posse, carere posse, opus non habere, to do 
without, a notion which might easily pass into that of being 
without, feeling the want of. 

LIND-WURM. Attempts have in vain been made to explain 
G. lind wurm, Icel. ling ormr, a dragon, from a Gothic source. 
The compound is in fact a mixture of Finnish and Teutonic. 
The Esthon. is lendew-maddo, a flying serpent, from lendama, 
to fly, lendwa, lendew, flying (whence lind, a bird), and maddo, 
a worm or snake. Finn, lentda, leta ; Bohem. letati, to fly. 

Mrjtccw. The name of the poppy, Gr. ^KWV, Esthon. mag- 
gona, Ober Deutsch magen, G. mohn, is explained by the 
Esthon. maggema, makkema, to sleep, the sleep-inducing 

NURUS. Lat. nurus, a daughter-in-law, young woman, 
married woman (Andrews). The original meaning is pro- 
bably young married woman. Esthon. noor, fresh, young; 
norik, young woman. 

ARBITER. The primary sense of arbiter is commonly given 
as an eye-witness, from whence that of an umpire or judge is 
supposed to be derived, as a witness specially called in for the 
purpose of declaring the event under trial. But there is no 
recognized derivation in Latin itself which would explain 
either of these meanings*. Now the Finn affords what is at 
least a very plausible explanation. 

There is a common tendency in an uninformed state of 
society to seek for the resolution of questions in which there 
is no means of direct knowledge, by casting of lots in some 
shape or another. Thus in Latin, sors is taken in the sense 
of an oracle, and sortilegus is a soothsayer, one who gives 
oracles or answers questions by the casting of lots. Alban. 
short, a lot, shortdr, a soothsayer, and this doubtless is the 
origin of our sorcerer, sorcery. One of the points upon 
which the cunning man of the present day is most frequently 

* Prof. Key, in vol. iv. of the Society's Proceedings, p. 94, derives 
arbiter from the old preposition ar, near (as seen in arvena=advena, &c.), 
and the root bit, go. 


consulted is the finding of lost property, and if a dispute upon 
such a question had arisen among a barbarous people, the 
most obvious means of settling it would be to refer it to one 
^who was supposed to have supernatural means of knowing the 
truth. Thus the wahr-sager, the truthsayer or soothsayer, 
would naturally be called in as an arbiter or doomsman. 
Now we find in Finn, arpa, a lot, symbol, divining rod, or any 
instrument of divination; arpa-mies (mes = man), sortium 
ductor, arbiter, hariolus ; arpelen, arwella, to decide by lot, 
to divine; arwata, conjicio, auguror, sestimo, arbitror; ar- 
waaja, arbiter in re censenda; arwelo, arbitrium, opinio, 
conjectura ; arwaus, conjectura, aestimatio arbitraria. It will 
be observed in how large a proportion of these cases the Lat. 
arbiter and its derivatives are used in explanation of the Finn 
words derived from arpa. 

QUISQUILI^E. The signification of quisquilics seems to be, 
light dry fragments of things, the small twigs and leaves that 
fall from trees; stipulae immixtse surculis et foliis aridis 
(Isidore in Forcellini). Hence rubbish, refuse. Langued. 
couscoulioUj husks of peas, beans, &c. The Gaelic equivalent 
gusgul is explained, refuse, filth, idle words, and we have the 
same metaphorical application in Latin : 

Quisquilias, volantis ^enti spolia, memoras. Csecilius in Forcellini. 

Now in Finn, light refuse matters of the foregoing description 
are designated from the rustling noise which they make. 
From kuhata, kuhista, to whisper, hum, rustle, is formed 
kuhu-ohrat, refuse barley; kuhuja, quisquiliae vel paleae quse 
motae leviter susurrant; from kahala, kahista, leviter crepo, 
movendo parum strideo nt gramen sub pedibus euntis vel 
arundo vento agitata, kahu, kahina, kahuja, refuse oats or 
barley, mere husks. 

Another modification of the same imitative root gives 
kulista, kulata, obscure sonare, whence kulu, kulina, hordeum 
vile, paleae; kulo, quisquiliae graminis vel gramen aridum 
tempore vernali in pratis. Here we have the element forming 
the latter half of the Gael, gusgul or its Lat. equivalent with 
the signification of the entire word. The syllable quis may 


be compared with the whis of whisper, O.-Sc. quhisper, or 
(with the ordinary interchange of p and qu) with It. pissi- 
pissij a whispering, buzzing, or humming noise. The original 
structure then of the word quisquilia would qualify it (by the^ 
repetition of the imitative element, as in susurro and nu- 
merous other instances) for the vivid expression of a whispering 
or rustling noise, such as is actually signified by It. bisbiglio 
or pispiglio, a form which differs only from quisquilice, as W. 
pump from Lat. quinque, or Oscan pitpit from quicquid. But 
in the process of logical development, the Latin word, like the 
Finnish equivalent above mentioned, has passed on to signify 
a rustling object, while the Italian one has been confined to 
the sound originally represented. 

ORTS. The word orts is used in Provincial English much 
in the sense of quisquilw in Latin, for scraps and remnants of 
fodder dropped by cattle, chips or odds and ends left by a person 
working : f quisquilise pabuli a pecoribus rejectee.' Finn Diet. 
The word is very widely spread through the Teutonic dia- 
lects. Ooraete, oorete, reliquise fastiditi pabuli sive cibi, esca 
superflua (Kil.) . Swiss urschi, ursi, remnants of food (Stal- 
der) ; Westerwald urze ; Prov Dan. orre, orred, orret, ovred, 
ovretj ort. 

Kilian's derivation from over-aete, as if the word signified 
what remained over after eating, is plausible in itself, and 
seems supported by forms like ooraetigh, fastidiens nimia 
satietate (Kil.) ; oorassen, to eat with disgust ; oorassiger, 
one who picks and chooses, does not eat all that is set before 
him ; ur'dssen, ur'dzen mit etwas, verurdssen etwas, to waste, 
to use wastefully, fastidiose cibum capere aliqua legendo, 
rejiciendo aliqua (Schmeller). It is probable, however, that 
the foregoing forms (like Fr. brinoter, to eat little and without 
appetite, from brin, a fragment. Patois de Braye.) are de- 
rived from the noun ooraete, urdss, urez, orts, remnants, the 
last syllable of which has been unconsciously assimilated to 
the verb essen, to eat, in consequence of the chief application 
of the word to remnants of food. The original meaning 
of the word seems to be far more general, as it appears in 
Gaelic ord, Irish orda, a fragment. The Lapp arates, reliquise 


cibi, prandii vel cceni, has a satisfactory native root in Esthon. 
warrisema, to rustle, to fall out, as ripe oats; warrid was 
herunter fallt, crumbs, droppings ; Finn warista, minutim et 
sparsim decide ut grana e spica vel folia arborum autumno ; 
waretj spicee, glumse, &c. in triturando decidentes, chaff, 
parallel with which may be mentioned the application of Eng. 
orts in America to the coarser siftings of flour, sharps, or 
pollards. It is remarkable that there is the same adoption of 
an initial w in Sc. worts ; ' Evenings worts are gude mornings 
fodderings' (Jamieson). 

ORDURE. Schmeller has already suggested that Fr. ordure 
may be derived from ort in the signification of refuse, rubbish. 
And such no doubt would be a most natural step in the deve- 
lopment of language. From the notion of fragments, rem- 
nants, offal of food, to that of rejection, refuse, rubbish, 
and ultimately, filth, is an easy transition. A child is said to 
ort his bread when he breaks it down into crumbs. The term 
is also applied to a cow that refuses or throws aside its pro- 
vender. It is hence metaphorically used to denote rejection 
in whatever sense : ' The lasses nowadays ort nane of God's 
creatures' (Jamieson). In the same way in Finn, runsu, 
quisquilise pabuli a pecoribus rejectee, orts, inde rejectaneum 
quid, purgamentum; runsimies, a scavenger, a remover of 

RUNCARE, to weed. The verb runsia, derived from the 
last-mentioned root, is explained e rejectanea vel purgamenta 
secerno, inde viliora qusevis rejicio/ describing exactly the 
object effected in weeding, viz. the removal of the worthless 
herbs from among the cultivated corn. Now although the 
word runsi itself does not appear to be used in Finn to desig- 
nate a weed, yet we find an exact synonym applied to that 
purpose. From rikkoa, to break or crumble, is formed rikko, 
a fragment, minutum quid rejectaneum, naucus, purgamentum, 
and hence (as a weed is the rubbish of a corn-field), rikka- 
ruoho (rwoAo=herb), a weed. Remembering then how often 
the hard c in Latin corresponds to an s in Finn, it will be no 
forced comparison if we indicate in the Finn runsia the ana- 
logue and explanation of the Latin runcare. 



[Read December the 18^.] 

The present paper is an attempt to reconcile the logical 
and etymological meanings of the word Distributed. 

Speaking roughly, distributed means universal : " a term is 
said be distributed when it is taken universally, so as to 
stand for everything it is capable of being applied to." 
Wfiately,\. 5. 

Speaking more closely, it means universal in one premiss ; 
it being a rule in the ordinary logic that no conclusion is 
possible unless one premiss be, either negatively or affirma- 
tively, universal. 

Assuredly there is no etymological connexion between the 
two words. Hence De Morgan writes : " By distributed is 
here meant universally spoken of. I do not use this term 
in the present work, because I do not see why, in any 
deducible meaning of the word distributed, it can be applied 
to universal as distinguished from particular." Formal Logic, 
chap. vii. 

Neither can it be so applied. It is nevertheless an accurate 

Let it mean related to more than one class, and the power 
of the prefix dis-, at least, becomes intelligible. 

For all the purposes of logic this is not enough ; inasmuch 
as the particular character of the relation (all-important in 
the structure of the syllogism) is not, at present, given. It 
is enough, however, to give import to the syllable dis-. 

In affirmative propositions this relation is connective on 
both sides, i. e. the middle term forms part of both the others. 
In negative propositions this relation is connective on one 
side, disjunctive on the other. 

In All men are mortal, 

All heroes are men, 

the middle term men forms a part of the class called mortal, 
by being connected with it in the way that certain contents 
are connected with the case that contains them ; whilst it also 

BY R. G. LATHAM, M.D. 191 

stands in connexion with the class of heroes in the way that 
cases are connected with their contents. In 
No man is perfect, 
Heroes are men, 

the same double relation occurs. The class man, however, 
though part of the class hero, is no part of the class perfect ; 
but, on the contrary, expressly excluded from it. Now this 
expression of exclusion constitutes a relation disjunctive 
indeed, but still a relation ; and this is all that is wanted to 
give an import to the prefix dis- in distributed. 

Wherever there is distribution there is inference, no matter 
whether the distributed term be universal or not. If the 
ordinary rules for the structure of the syllogism tell us the 
contrary to this, they only tell the truth, so far as certain 
assumptions on which they rest are legitimate. These limit us 
to the use of three terms expressive of quantity, all, none, 
and some ; and it is quite true that, with this limitation, uni- 
versality and distribution coincide. 

Say that Some Y is X, 

Some Z is \ y 

and the question will arise whether the Y that is X is also 
the Y that is Z. That some Y belongs to both classes is clear ; 
whether, however, it be the same Y is doubtful. Yet unless 
it be so, no conclusion can be drawn. And it may easily 
be different. Hence, as long as we use the word some, we 
have no assurance that there is any distribution of the middle 

Instead, however, of some write all, and it is obvious that 
some Y must be both X and Z ; and when such is the case 
Some X must be Z, and 
Some Z must be X. 

Universality, then, of the middle term in one premiss is, by 
no means, the direct condition that gives us an inference, but 
only a secondary one. The direct condition is the distri- 
bution. Of this, the universality of the middle term is only 
a sign, and it is the only sign we have, because all and some 
are the only words we have to choose from. If others were 
allowed, the appearance which the two words (distributed and 


universal] have of being synonymous would disappear. And 
so they do when we abandon the limitations imposed upon us 
by the words all and some. So they do in the numerically 
definite syllogism, exemplified in 

More than half Y is X, 
More than half Y is Z, 
Some Z is X. 

So, also, they do when it is assumed that the Y's which are 
X and the Y's which are Z are identical. 

The same Y is Z, 
Some Z is X. 

In each of these formulae there is distribution without 
universality, i. e. there is distribution with a quality other 
than that of universality as its criterion. The following 
extract not only explains this, but gives a fresh proof, if fresh 
proof be needed, that distributed and universal are used syno- 
nymously. The " comparison of each of the two terms must 
be equally with the whole, or with the same part of the third 
term ; and to secure this, j(l) either the middle term must be 
distributed in one premiss at least, or (2) the two terms must 
be compared with the same specified part of the middle, or 
(3), in the two premises taken together, the middle must be 
distributed, and something more, though not distributed in 
either singly." Thompson, Outline of the Laws of Thought, 39. 

Here distributed means universal Mr. Thompson's being 
the ordinary terminology. In the eyes of the present writer 
" distributed in one premiss " is a contradiction in terms. 

Of the two terms, distributed is the more general ; yet it 
is not the usual one. That it has been avoided by De Morgan 
has been shown. It may be added, that from the Port Royal 
Logic it is wholly excluded. 

The statement that, in negative propositions, the relation 
is connective on one side, and disjunctive on the other, 
requires further notice. It is by no means a matter of indif- 
ference on which side the connexion or disjunction lies. 

(a.) It is the class denoted by the major, of which the middle 
term of a negative syllogism is expressly stated to form no 

BY R. G. LATHAM, M.D. 193 

part, or from which it is disjoined, (b.) It is the class denoted 
by the minor, of which the same middle term is expressly 
stated to form part, or with which it is connected. 

No man is perfect 

here the proposition is a major, and the middle term man is 
expressly separated from the class perfect. 
All heroes are men 

here it is a minor, and the middle term man is expressly con- 
nected with class hero. 

A connective relation to the major, and a disjunctive 
relation to the minor are impossible in negative syllogisms. 
The exceptions to this are only apparent. The two most 
prominent are the formulae Camestres and Camenes, in both 
of which it is the minor premiss wherein the relation is 
disjunctive. But this is an accident ; an accident arising out 
of the fact of the major and minor being convertible. 

Bokardo is in a different predicament. Bokardo, along 
with Baroko, is the only formula containing a particular 
negative as a premiss. Now the particular negatives are, for 
so many of the purposes of logic, particular affirmatives, that 
they may be neglected for the present ; the object at present 
being to ascertain the rules for the structure of truly and 
unquestionably negative syllogisms. Of these we may pre- 
dicate that their minor proposition is always either actually 
affirmative or capable of becoming so by transposition. 

To go further into the relations between the middle term 
and the minor, would be to travel beyond the field under 
present notice; the immediate object of the present paper 
being to explain the import of the word distributed. That it 
may, both logically and etymologically, mean related to two 
classes is clear clear as a matter of fact. Whether, however, 
related to two classes be the meaning that the history of 
logic gives us, is a point upon which I abstain from giving an 
opinion. I only suggest that, in elementary treatises, the 
terms universal and distributed should be separated more 
widely than they are ; one series of remarks upon 

a. Distribution as a condition of inference, being followed 
by another on 


b. Universality of the middle term in one premiss as a 
sign of distribution. 

So much for the extent to which the present remarks 
suggest the purely practical question as to how the teaching 
of Aristotelian logic may be improved. There is another, 
however, beyond it ; one of a more theoretical, indeed of an 
eminejitly theoretical, nature. It raises doubts as to the 
propriety of the word all itself; doubts as to the propriety of 
the term universal. 

The existence of such a word as all in the premiss, although 
existing therein merely as a contrivance for reconciling the 
evidence of the distribution of the middle term with a certain 
amount of simplicity in the way of terminology, could scarcely 
fail, in conjunction with some of its other properties, to give 
it what is here considered an undue amount of importance. It 
made it look like the opposite to none. Yet this is what it is 
not. The opposite to none is not-none, or some ; the opposite 
to all is one. In one and all we have the highest and lowest 
numbers of the individuals that constitute a class. In none 
and some we have the difference between existence and non- 
existence. That all is a mere mode otf some, has been insisted 
on by many logicians, denied by few or none. Between all 
and some, there is, at best, but a difference of degree. Be- 
tween some and none, the difference is a difference of kind. 
Some may, by strengthening, be converted into all. No 
strengthening may obliterate the difference between all and 
not- all. From this it follows that the logic of none and some, 
the logic of connexion and disjunction (the logic of two signs), 
is much more widely different from the logic of part and 
whole (the logic of three signs) than is usually admitted ; the 
former being a logic of pure quality, the latter a logic of 
quality and quantity as well. 

Has the admixture done good? I doubt whether it has. 
The logic of pure and simple Quality would, undoubtedly, 
have given but little; nothing but negative conclusions on 
one side, and possible particulars on the other. Nevertheless 
it would have given a logic of the Possible and Impossible. 
Again, as at present constituted, the Quantitative logic, the 

BY R. G. LATHAM, M.D. 195 

logic of all and some, embraces either too much or too little. 
All is, as aforesaid, only a particular form of more than none. 
So is most. Now such syllogisms as 

Most men are fallible, 

Most men are rational, 

Some men are both frail and fallible ; 

Some frail things are fallible, 

are inadmissible in the Aristotelian paradigms. A claim, 
however, is set up for their admission. Grant it, and you 
may say instead of most 

Fifty-one per cent., &c. ; 

but this is only a particular instance. You may combine any 
two numbers in any way you like, provided only that the sum 
be greater than unity. Now this may be arithmetic, and it 
may be fact -, but it is scarcely formal logic ; at any rate it 
is anything but general. 

It is the logic of some and its modifications one, all, and 
anything between one and all, as opposed to the logic of the 
simple absolute some (some the opposite to none], and a little 
consideration will show that it is also the logic of the probable, 
with its modification the proven, (proven is probable, as all is 
some,) as opposed to the logic of the possible and impossible. 
Let, in such a pair of propositions as 

Some of the men of the brigade were brave, 
Some of the men of the brigade were killed, 
the number expressed by some, as well as the number of the 
men of the brigade, be known, and the question as to whether 

Some brave men were killed, 

is a problem in the doctrine of chances. One per cent, of 
each will make it very unlikely that the single brave man was 
also the single killed one. Forty-nine per cent, of each will 
make it highly probable that more than one good soldier met 
his fate. With fifty on one side, and fifty-one on the other, 
we have one at least., With all (either killed or brave], we 
have the same; and that without knowing any numbers 
at all. 



Juris Doctor, Member of the Friesic Chivalry. 
[Read December the 4^.] 

On reading the Rev. J. Davies's Paper " On the Races of 
Lancashire, as indicated by the Local Names and the Dialect 
of the County" (Phil. Soc. Trans. 1855, pp. 210-245), I was 
led to doubt whether all the words there indicated as Keltic, 
really have a Keltic origin, or whether the most part of them 
have not an Old-Friesic origin. I will give the results of my 
inquiry, by comparing some of these words with similar ones 
in the Old-Friesic, Dutch, and Flemish. 

I shall add a comparison between the Old-English and Old- 
Scotch words that I have found in a Paper by P. Hjort, 
" Om det engelske Konjugations-system," Kjobenhavn, 1843, 
and the same languages. This will prove that the greater 
part of these too have an Old-Friesic origin. 

I have little doubt, therefore, that in the dialects of the 
counties of England there remain many words of Old-Friesic 
origin which are considered as Keltic, or of which the true 
origin has not been shown by English authors. If I had the 
opportunity of examining all the glossaries of those dialects 
which are not to be got here, I would take upon myself to 
prove, that many of these words are to be found in Old-Friesic, 
Dutch, or Flemish; and this will confirm the thesis of the 
Rev. J. Davies, at the end of his Paper, where he says : 
" It is highly important for the purposes of English philology, 
that this (Old-Friesic) language should be more carefully 
studied by us, as it is, above all others, the fons et or iff o of 
our own." 



Compared with the Old-Friesic and with the Dutch. 
The following words are all names of towns, villages, ham- 
lets, lakes, &c. in the district of Friesland, taken from the 



work, " Oud en Nieuw Friesland, of Aardrykskundige Bes- 
chryving van die Provintie, byeen verzameld door Jhr. Mr. 
M. de Haan Hettema, Leeuwarden, 1840." It is very difficult 
to give the original meanings of these words, because they so 
often depend on the situation of the places, whether they 
are on heights, in low places, on rivers, or marshes; and 
there are no terms now extant in Friesic to explain them. 
Help, however, is to be obtained by referring to the languages 
related to the Friesic. 

From my comparison with the names of places given by 
the Rev. J. Davies, I expect it will be found that the situ- 
ations of his Lancashire places agree with those of the Friesic 
ones given by myself. But before giving the names of the 
places, I will give some of the words that form parts of those 
names, the meanings of which are known, and are to be 
found in the above-mentioned work : 

aard, eer, eerd; hill, hillock. 
bal, bel t bol; height, convex 

body, head. 
bird-, bank, border. 
bran, bron, brun; pointed, 


buurty buren; hamlet. 
corn, horn-, corner. 
deel; district. 
end-, end. 
eer, ir, ee; water. 
ga, gae ; village. 
gaast', heath. 
go -, district. 
gjum, gum, jum, um ; home, 

abode, village. 
ham, hem-, idem, 
hem', districtus. 
hal, hoi-, hill, height. 
herne, horne', corner, top, tip. 
hes, has; marsh. 

haule, hoole-, hill, hillock. 

kat ; dirt, mire, turfmoor. 

kerk', church. 

kol; cold. 

krim, krom; inflected. 

land-, district. 

lau, lee; smooth. 

mar, mor, mur; marsh. 

man, men ; common. 

meer; lake. 

pan, pen, pin; head or summit. 

piek, pike -, a pointed end. 

ryp ; way, road. 

scharn, schern; shred, part, 
corner, marshy ground. 

schet, schot; dirt, mire, turf- 

tan, ton, tun; environing, en- 

toet; mouth. 

terp ; height, hill. 



trop, troop; height. 

waerd, waert, werd, werth, 

wier ; hill, hillock. 
wir, wier-, sea-weed. 

ivin, wyn ; wrinkle (wynerts, 
a wrinkled currant) . 

wolde, woud, woude; wood, 


Pendle Hill. Fr. Pingmeer, a lake ; Pingjum, a village ; Pan- 
dregae, id. 

Eivington Pike. Fr. Piekmeer, a lake ; Pikmeer, id. 

Hentoe, Hentor. Fr. Hennaard, a village ; Henshuizen, a ham- 
let ; Henswoude, id. ; Hantum, a village. 

Sholver. Fr. Hallum, a village ; Hollum, id. ; Holwerd, id. 

Tandle Hills. Fr. Tania, a farm with right of voting ; Ton- 
nawerth, a village. 

Bryn. Fr. Brantgum, a village ; Brongergae, id. ; Bruindeer, 
a hamlet. 

Buersitt Hill. Fr. Burgwert, a village ; Burnm, id. 

Crimbles. Fr. Krinserarm, or Krimserarm, an inflected dam 
against the water ; Kromwal, a hamlet (krom, curved ; wal, 
shore) . 

Tooter Hill. Fr. Toetsmeer, a lake. 


IrkIrwell (Irkwell? the well of the Irk). Fr. Eernsum, a 

village ; Irnsum, id. ; Eernwonde, id. 
Medlock. Fr. Medemelaca, a town ; D. Medenblok, id. 
Ribble. Fr. Ryperkerk, a village ; Rypend, a hamlet. 
Colder. Fr. Kolderwolde, a village ; Koldum, id. ; Kollum, id. 
Lune. Fr. Terluine, a farm with right of voting; Luinjebird, 

a village. 

Wyre. Fr. Wierum, a village; Wirdum, id. 
Beal. Fr. Beuil, a viUage ; Balk, id. ; Belkum, id. 
Leven. Fr. Lauwers, a sea, a lake ; Leeuwarden, a town. 
Loud. Fr. Lioessens, a village ; Luds, a little lake. 
Rennet (Kunnet). Fr. Kuinder, a river. 
Morecambe. Fr. Morra, a village ; Marrum, id.; Marsum, id. 
Winander. Fr. Winerts, a little current ; Wynjeterp, a viUage. 


Manchester. Fr. Mantgum, a village. 

Catterall. Fr. Katlyk, a village ; Kattebuuren, a hamlet. 

Werneth. Fr. Warns, a village ; Warniahuizen, a hamlet. 

Carnforth. Fr. Cornjum, a village ; Cornwerd, id. 

Scotforth. Fr. Schoterland, a district in Friesland. 

Cinderland. Fr. Sindelra or Sondel, a village. 

Penketh. See Pendle Hill. 

Heskin, Hesketh. Fr. Haskerland, a district in Friesland; 

Hesensermeer, a lake. 
Sarneyford, Sharneyford. Fr. Scharnum, or Schernhemstra, 

or Scharnegoutum, a village ; Scharnebuuren, a hamlet ; 

De Scharren, id. 
Camel Hill. Fr. Kahool, a farm ; Koehool, a hamlet. 


addle } rotten. Fr. after \ D. etter (pus) ; Fr. aedel, dung-hole ; 

D. aal, dung- water. 

agog, eager, desirous. Fr. aegjen, eagjen, to aim at. 
boggart, an apparition, a hobgoblin. Fr. boghen, deceits 

(fraudes, deceptiones) . 

brawsen, stuffed with food, gorged. D. gebreeuwd, calked. 
brewis, a dish made of oat-cakes soaked in broth. Fr. bry, 

milk-porridge ; D. pap, id. 
brog, a bushy or swampy spot. D. broek, broekland, marshy 


bruit, to talk, to publish. D. verbreiden (to divulge). 
burleymon, a person appointed at courts-leet, to examine and 

to determine about disputed fences. Fr. buraldermon (judex 

vici) . 

ceckle, to speak insolently. D. kakelen, to chatter (garrire). 
cleawse, an enclosure, a field, a close. Fr. clowa (districtus) ; 

D. kluft, id. 

cock-boat, a small boat. D. kogge, koghschip (celox). 
cosy, comfortable, snug. Fr. kosya (ludere modo amatoris) ; 

D. liefkoosen', Fr. kos (pacificatio) . 
cratchinly, feebly, weakly. D. Kil. kraecke (domus ruinosa). 


crib, to steal, to filch a small part of anything. Fr. krabben, 

kribbelen (occulte auferre, furare); D. krabbelen, schrabben 

(colligere) . 
cuddle, to fondle, to embrace, to press to the bosom, to lie 

closely. D. kittelen, to tickle. 

dossuck, a dirty, slovenly woman. D. Kil. duyse (concubina) . 
dunder-head, a blockhead, a silly fellow. D. dun (tenuis, 

f arrant, decent, respectable, worthy. Fr. fara (agere) ; fera 

(administrare) ; fere (utilis) ; D. ervaren, expert, expe- 
fattle, to trifle about business. D. vaddich, vadsig (ignavus, 


garth, the belly-band of a horse. D. buik-gordel, id. 
goltch, to be gluttonous. D. gulzig (gulosus) ; Kil. golpe 

(gurges, vorago). 
gry, to be in an ague-fit. D. grysen (ringere, fremere) ; Kil. 

greesen (perterrefacere) . 
gullion, a soft, worthless fellow. Fr. gol, golle (mitis, bene- 

volus) ; sul, a very good-natured man ; sulachtig, simple, 


gyre, to purge. D. keeren (scopis purgare pavimentum). 
hawk, to cough, to bring up phlegm. D. hoesten (tussire). 
hopper, a receptacle for corn in a mill, a basket. Fr. opper, 

van hooi (meta foeni). 
howse, to stir up. D. husschen, hisschen, hitsen (accendere, 

inflammare) . 
huff, huff, to treat scornfully, to attack with scornful reproofs. 

Fr. schoff (opprobrium) ; schoffieren (afficere ignominia). 
hutch, to lift up the shoulders uneasily, to move the body with 

an uneasy motion. D. hutsen, hutselen, hotsen (quatere, 

concutere) . 

keen, to burn. D. kenen (regerminare) . 
lake, to idle, to play truant. Fr. loayckjen (sedere pigritise) ; 

loay (ignavus). 
lithe, v. to thicken broth or soup with meal. Fr. lithe, milk 

porridge ; D. pap, bry, id. 
lurch, to lurk, to lie hid. D. loeren (observare, insidiari). 


lutch, to pulsate strongly and painfully, as an angry tumour. 

D. klotsen, klutsen (quatere). 
mog, to move off, to depart quickly. D. moffelen, to remove 

secretly to some place. 

mulloch, dirt, rubbish. Fr. molde (humus) ; moude (pulvis) . 
natter, to gnaw, to nibble. D. knotten, to top (amputare). 
oandurth, afternoon. Fr. unden (post meridiem) . 
powse, powsement, dirt, refuse, offal. D. poesen, morsen, to 

dirt, to puddle. 
punse, to kick. Fr. bonsjen, bonsen, to throw ; D. bons, bounce, 

thump, hard blow. 

purr, to kick. Fr. porren, to thrust. 
reawt, a way, a route. Fr. reed, alley ; Kil. rafter, rauftere 

(materia trabis). 
reeack, to scream, to shriek. Fr. rogia, ruia (accusare); 

Kil. roken (instigare); roeck (dilator). 
rock, rocket, a frock. D. rok, coat. 
slat, to spill, to dash water about. Fr. slatten, id. ; D. slooten, 

to intersect with ditches, to dig ditches (purgare lamas). 
sow, the head. Kil. sop, tsop (supremum, summitas) . 
spree, a wild mischievous frolic. D. spreeuw, a jester, a 

scoffer; spreeuwen, to jest. 
tackle, to equip, to set in order, to take a person in hand 

with the intent to subdue him, or set him in order. Fr. 

optakelen, toetakelen (adornare, verberare) ; D. takelen, to 

whop, a smart, sharp blow. Fr. wepen (bellum) ; Kil. wapper 

(flagellum) . 
wyzles, the stalks of the potato-plant. D. vezels, fibres, strings. 

berm, barm, yeast. D. Kil. berm, barme (faex, spuma cere- 

visise) . 
cark, to be careful or anxious. Fr. karfesta, karena, karina 

(pcenitentia 40 dierum) ; karefester (cui posna inflicta est, 

jejunii 40 dierum, poenitentiarius) . 
drab, a prostitute, a vile, dirty woman. D. dribbe, a scold; 

dribben (mentiri, injuriare) . 


gabloc, an iron bar, a gavelock. Fr. gaffel, pitchfork (bidens, 

merga) . 
riddle, a coarse sieve. Fr. riddle (febris a terrore, sive horrore) ; 

D. redde, ryde, id. 
rhute, passion, a paroxysm of anger. Fr. rit (vexat) ; D. Kil. 

ryden (agitari ira, irasci) ; ritsch (catuliens) . 


beetneed, a helper, one applied to in distress. Fr. beta (re- 
par are) ; nede (periculum). 
bigg, to build. Fr. buwa-, boeghia (habitare). 
brattle, to spend money foolishly or ostentatiously, to squan- 
der. D. brassen (bacchari), to feast, to debauch. 
bryed, to spread abroad. Fr. breia (projicere) ; D. verbreiden, 

to spread. 

crib, a pen, a manger or rack. Fr. crib, id. ; D. kribbe, id. 
dateless, foolish, silly, weak in body and mind. D. Kil. doten, 

dutten (delirare, desipere) ; dotelore (mentis error, insania, 

delitium) . 
ding, to strike or knock about, to reiterate an accusation. Fr. 

thingia, id.; D. dingen (judicare) ; thinght (processus). 
dree, long, tedious, wearisome. Fr. dreeg iten, heavy food; 

dreeg wurk, heavy or hard work. 
fleet, to take the cream off the milk. D. vlooten ; vlieten de 

melk, to skim (cremorem lactis colligere, cremorem tollere) . 

Fr. flut, skim-milk (lac gelatum) ; vlotemelk, id. 
flooze, fleeze, small particles of wool or cotton. D. vlies, id. ; 

Fr. fluus (lana ex ove demta), fleece, flock, flue ; D. pluis, id. 
frist, trust, confidence. Fr. frithia (liberare, pacificare); friu- 

delf (maritus, amatus), wooer. 

gawster, to boast, to swagger. Fr. gysten (vefaemens). 
glendur, to stare, to look in amazement. Fr. gleon, gleaun, 

glandig (iratus, calidus) ; D. glinsteren (fulgere). 
haust, a cough. D. hoest, id. 
lit, a few, little. Fr. litje (parum) ; lits (parvus). 
menseful, decent, managing, thoughtful. D. meenen (arbi- 

trari, sentire) ; meening (sententia, mens, opinio, mente 



neb, an edge or rim, the peak of a bonnet, a piece broken off. 

Fr. neb (os) ; D. Kil. nebbe (rostrum navium). 
neeze, to sneeze. D. niesen, id. ; Fr. fniezen, id. 
snidge, a greedy sordid person. Fr. snoad, pauper; Kil. 

snoode (vilis, turpis) ; snodder, sordes. 
snite, to blow the nose. Fr. snuten, id. ; D. snuiten, id. 
steigh, a ladder, a stile. Fr. D. steiger, scaffold. 
swill, v. to wash or rinse vessels. D. dweilen, to clean with 

a clout ; dwell, towel, swab. 

sye, to drain milk through a sieve. D. zeeven (cribrare). 
syle, to rain continuously. Fr. syle (cataracta), sluice; D. 

sluis, id. 

tan, a twig. D. teen, twyg, id.; Fr. ft/ft (virga, vimen). 
teagle, a crane for winding-up goods. D. Kil. taeckel (re- 

mulcus) ; taekelen (subducere) . 
teend, to light a fire. D. Kil. teenen (irritare). 
tore, to labour hard for a living. Fr. toarnen (laborare) . 
wakes, the extremities of the lips, the corners of the mouth. 

Fr. weage (paries) ; Kil. weeg, id. 

fey, to do anything cleverly. D. Kil. vey (vigens, vegetus). 
spur, a prop in building. D. Kil. sparre (sudes). 


barkle, to stick to, to adhere ; trans, to cover over. D. Kil. 

barcke, bercke (cortex); barcken, bercken (arbores decorti- 

care) ; bergen (condere, abscondere) . 
cleg, a clever person, an adept. Fr. clewa, bycliwa (flores- 

cere, firmem sive fortem fieri); D. beklyven (coalescere, 

creel, a frame to wind yarn upon. D. Kil. kreelen, to bind ; 

Fr. kraga (boja, vinculum, quo collum circumdatur) . 
dab, a blow. D. douw, a push; Kil. dabben (palpare, subi- 

gere) ; Fr. tapa (capere). 
doage, wet, damp. Fr. douwe, id. ; D. daauw (ros) ; Kil. daeck, 

dake (nebula). 

faddle, nonsense, trifling. D. Kil. vaesen (farcire). 
fleak, to bask in the sun. Fr. blakerje ynne son, id. 


flit, to remove from one house to another. Fr. flet (mobilis) ; 

fletech (refugus). 
for elders, seniors, ancestors. Fr. eldra (seniores) ; D. voor- 

ouders, id. ; Fr. fordders, id. 
frum, tender, delicate, easily broken. Fr. frumdede (actio 

principalis) ; fremo (ntilis) . 
gain, gainer (a gainer way is a shorter way). D. Kil. gaen- 

eruen (haeredes accelerantes) ; Fr. gaelick (repentinns, in- 

tempestivus) ; galick (conveniens) . 
gar, to make, to do, to compel. D. Kil. gaerwen (prseparare, 

conficere) . 
gawby, a clownish simpleton. D. Kil. gabberen (nugari) ; 

gabber, gabbarus (homo insulsus). 
geek, a jest, a mocking sarcasm. D. Kil. gheck (jocus) ; 

gekken, to jest. 
hetter, keen, eager, as a dog in fighting. D. Kil. hetsen 

(incitare, instigare). 
hippin-stones, stones at the crossing of a stream. Fr. wippen 

(saltare) ; D. Kil. hippen, wippen (agitare, vibrare). 
kench, a twist, a strain. D. kinkhoorn (turbo, concha), a 

kind of shell like a paper case in the form of a cone. 
kick, fashion, mode. D. Kil. schick (apparatus) ; opschik, 

finery; Fr. schick jaen, to model. 
kipper, amorous, lascivious. D. Kil. kippen (pullulare) ; kip 

(pullities) . 
lam, to beat soundly, to chastise. Fr. lamma, lemma (debi- 

litare); lorn (debilis). 

lane, to conceal. Fr. leyna (mentiri) ; leynd (mendax). 
late, to seek. Fr. let ten (advertere animum, speculare ali- 

quem, vacare alicui rei) . 
lither, idle, lazy. D. Kil. lydden-tyd (homo ignavus, otiosus, 

tempus transigens ignave) ; lyden (tolerare). 
lurgy, idle. D. Kil. lor en (ignave aliquid agere) ; loeren (con- 

nivere); lurts (sinister). 

mood, satiated, filled to repletion. Fr. moed (satisfactio) . 
neeve, neyve, a fist. Fr. knevel (homo fortis) ; D. knevelen 

(manus vinculis illaqueare); knevelband (manicae, vincula 

manuum) . 


plucher, to pilfer, to steal slyly. D. Kil. pluysteren (diripere, 

spoliare) . 

scar, a steep bare rock. D. Kil. schaere (scopulus, rupes). 
so wl } whatever is eaten with bread. D. suyvel (lactantia). 
skellut, crooked, awry. Fr. schelf (quod non est rectus, pla- 

nus); D. schelferen (stringere, radere). 
skyme, skyoymej to look scornfully, to be cold and distant 

in manner, as a purse-proud parvenu to his old friends. 

D. Kil. schuymer (delator, musca). 
stood, the track of wheels. Fr. slat a (excitare incilia) ; slate, 

slaet, sleat (fossa). 
slunt, to be idle. D. slenderen, to loiter; Kil. sluns (homo 

ignavus) . 
sny, to turn up the nose in contempt, to affect dislike. D. Kil. 

snoecks (nasutulus) ; snoecks sien (argutis et acribus oculis 

intueri) ; snuytert (nasutus) . 
whack, a heavy blow. D. kwak, plump, sudden; kwakken, 

to throw, cast. 
whip off, to go off quickly. Fr. wippen (saltare) ; D. Kil. 

wippen, to hasten, to jump. 
whoave, to cover over, to overwhelm. Fr. wob (vestis) ; D. Kil. 

woack (amiculum ferale). 
yark, to strike hard. D. Kil. jacken (flagellare scutica) ;jacke 

(scutica) ; jackener (auriga) . 


(Davies, p. 277.) 
botch, to mend clumsily. D. Kil. boeten (emendare); Fr. 

beta (reparare). 
cant, to raise up a barrel, to set it on edge. D. kantelen, 

to overturn; kant, edge. 
frame, to set about a thing, to show capacity in beginning 

anything. Fr. framia (prodesse) ; fremo (utilis). 
fremd, strange, not belonging to the family. Fr. fraemd, id. ; 

D. vreemd, id. 

grit, sand. D. Kil. gries, greus (arena, glarea) . 
gull, a fool, one easily cheated. Fr. kul, id. ; D. sul, id. ; 

kullen, to fool. 


greel, to weep, to lament. Fr. greta (accusare) ; D. kryten, 

to lament. 
kittle, ticklish, difficult, uncertain. D. kittelen, to tickle; 

kitteloorig, kittelig, easily offended. 
mack, race, family, sort. Fr. meek (conventus matrimonialis) ; 

Fr. maga (cognatus). 
wad, a pledge, a forfeit. Fr. wed (impensa, noxa, cautio, 

promissio; laesio). 

P. HJORT. Om det engelske Konjugations-system. Kjoben- 
havn, 1843. Tillag (Side 79seq.). 


A. Verbs. 

claw, to stroke, to rub. Fr. clawa, to scratch, claw. 

dele, to divide. Fr. dela, id. 

deme, to judge. Fr. dema, id. 

foster, to nourish. D. voeden, voederen, id. 

hete, to be called. Fr. heten, id. 

kyke, to look stedfastly. D. kyken, to look. 

legge (hond upon him), to lay, &c. Fr. leggia hond up him, 

id. ; D. leg gen, to lay. 

ligge, to lie down. Fr. liggia, id. ; D. liggen, id. 
mene, to mean. Fr. mena, id. 
shifte, to divide. Fr. skifta, to separate. 
snibbed, reproved. Fr. snauwd, id. 
spille, to throw away. D. verspillen, to squander away. 
thole, to suffer. Fr. thola, id. ; D. dulden, id. 
uttre, to publish. Fr. utia, to utter; D. uitten, id. 
welde, to govern. Fr. walda, welda, id. 

B. Substantives. 

length and brede, breadth. D. lengte en breedte, id. 
/ee (all that lond and). Fr. al that lond and fia (omnes 

possessiones, omnes agri et omne pecus) . 
fostring, nutriment. D. voeder, voedering, id. 


hevedj head. Fr. hafed, id. 

knoppe, a button. D. knoop, knop, id. 

querne, a hand-mill. D. kwern, quern, quernmolen, id. 

unhele, misfortune. D. onheil, id. 

wanhope, despair. D. wanhoop, id. 

wantrust, distrust. D. wantrouwen, id. 

c. Other Words. 

deve, deaf. Er. daf, id. ; D. doof, id. 

owerthwart, across. D. overdwars, id. 

threttene, thirteen. Fr. threttine, id. 

thridde, third. Fr. thredde, id. 

wis^, certainly. D. zm, gewisselyk, id. 

whilke, way is he gone. Fr. hwelke wei is hi gongen, id. 


blere, to stay. Fr. bliva ; D. blyven, id. 

bollen, swollen. D. verbolgen, id. 

forlese, to lose entirely. Fr. forliesa, to lose. 

forlete, to quit. Fr. forlitta, id. 

Aa/fe, to go lamely. Fr. A/ta, id. 

knopped, buttoned. D. geknoopt, geknopt, id. 

rere, to raise. Fr. rera, to move ; D. roeren, id. 

welwilly, propitious. D. welwillend, id. 

wrote, to dig with the snout. D. wroeten, to turn up. 


A. Verbs. 

he bad, prayed. Fr. hi bad, id. 
bede, bide, to abide, remain. D. beiden, id. 
bygge, to build. Fr. buwia, bowa, id. 
bygginge, building. Fr. buwinge, id. 
drogh, drew. D. ro&, id. 
<scAe, to add. Fr. aca (augmentare) . 
feltred, felter' d, shaggy. D. viltig. 

grade, gredde, cried, wept. Fr. greta, to weep; D. kryten, id. 
haylse, salute. D. heil (salus). 
leke, lock, shut. Fr. luka (claudere). 


quadth, said. Fr. quath, id. 

rive, to tear. D. ryven, ryten, id. 

rope, to cry loud. Fr. hropa, to call. 

styfte, deal out, divide. Fr. skifta, to separate. 

spir, to ask, inquire. Fr. spera, to investigate ; D. speuren, id. 

spreden, spread. Fr. spreid, id. 

wete, wite, know, learn. Fr. wit an, to know > D. weeten, id. 

B. Substantives. 

ande, onde, breath, life. Fr. andema, ondema (anima, animus) . 

bane, death, misery. Fr. banthe (homicidium) ; bona (ho- 

barn, child. Fr. bern, id. 

brygge, bridge. Fr. bregge, id. 

egge, edge. Fr. eg, igge, id. 

ern, eagle. Fr. earne, id. 

gase, goose. Fr. gies (anseres) . 

get, goat. D. geit, id. 

glede, a burning coal. Fr. glede, glowing fire. 

hawe, churchyard. Fr. hof, id. 

leche, leech or physician. Fr. letza, id. 

make, mate, companion, D. makker, id. 

mawe, stomach. Fr. maga, id. 

meollen, mills. D. molens, id. 

nese, nose. D. neus, id. 

panne, paune, head, skull, brain-pan. Fr. breinpanne (cra- 
nium) ; D. hersenpan, id. 

punge, purse. Fr. ponge, id. 

ack- Fr. r^e, id. ; D. rug, id. 
e, cause, right. Fr. scheel (dissidium) . 

stede, place. Fr. stede, id. ; D. steed, id. 

gyrdyl-steed, the waist. D. #ordeZ steed, id. 

stubbe, stump, stake. Fr. stobbe, thump. 

sty, house, building. D. s&&, sfeed, stead. 

*^eme, swimming, qualm. Fr. swima, id. 

tale, talk, speech. D. taaj (lingua). 

tide, time. Fr. fwJ, id. ; D. tyd, id. 

wrethe, rage, harm, wrath. D. wreedheid, cruelty. 


c. Other Words and Combinations. 

ain, heyen, eyes. Fr. eaghen, eyes ; also, eren and eghen, Fr. 

ear en and eaghen, or ara and agha, ears and eyes. 
alond, ashore. Fr. a londe, id. 
blyde, blithe. Fr. blide } id. ; D. blyde, id. 
eiffhte, eghte, goods, property. Fr. ain, egin (proprium). 
ek, also. Fr. ek, id. 
ellis, else, otherwise. D. elders (alibi). 
fyle, vile, foul. D. vuil, id. 
hoi, whole, sound. Fr. heel (sanatus). 
godhede. D. goedheid, goodness. 
gowl, gules. Fr. giel, id. ; D. #ee/, id. 
foA, loth. D. laatdunkend, self-conceited. 
lite, lytte, little. Fr. litje-, litka (parvus). 
overtwert, overthwart. D. overdwars, id. 
recke, care. D. roeck (cura) ; Fr. rokolos (temerarius). . 
rightwise, righteous. Fr. riuchtfirdich (Justus). 
skere, shy re, sheer, clear. D. schieren (ornare). 
slike, such. D. zulke, id. 

store, loud, stark, stir. Fr. stoer, stor (magnus). 
thermyd, therewith. Fr. thermithe, id. 
tholmod, patient. Fr. thola (pati) ; D. dulden, id. ; Fr. mode 



A. Verbs. 

aynding, breathing. Fr. andema (spiritus, anima). 

big, to build. Fr. buwa, id. 

duller, move like the tide when it meets with resistance. D. 

bulderen, to bluster. 
dag, to clog, adhere. Fr. clay (argilla) ; D. klei, id. ; D. kleven 

(to cleave). 
clever, to climb. Fr. cliwa, klieuwen, id. ; D. klaveren, klau- 

teren, to clamber. 

deve, to deafen. Fr. daua, dawa, id. ; D. doven, verdoven, id. 
doop, to dip, to baptize. D. doopen, id. ; Fr. depa (to baptize) . 
dreip, to drop. Fr. drippa, id. ; D. druppen, id. 


dunder, to make a noise like thunder. D. donderen, to thun- 
der one about. 

dwine, to decay. D. kwynen, id. 

eak, to increase. Fr. aket (auctus) . 

forstaw, to understand. D. verstaan, id. 

ga, to go. Fr. #a, id. ; D.gaan, id. 

#arc^, to go, walk. Fr. ganga, id. 

gigle, kekle, to laugh. D. kakelen (insane loqui). 

#wop, to eat. D. knappen, id. 

gnidge, to pinch, to squeeze. D. knypen, id. 

Ame, to plunder, to ruin. D. verheeren, to waste. 

keik, to spy, peep. D. &/fa, id. 

&emp, to strive, contend. Fr. kempa, to fight; D. kampen, id. 

lak, to depreciate, vilify. Fr. leckia, id. ; D. Mew, id. 

layke, to sport. Fr. hlakia, laytse, to laugh. 

/05m (pediculos capere). D. luizen, id. 

:, to shut up, to inclose. Fr. Mc, to shut. 
, nurr, to snarl, as dogs. D. gnorren, knorren, id. 
nikker, nichar, to neigh like a horse. D. hinneken, id. 
reefe, to unravel, clear away. D. reeden, gereed maken, be- 

reiden, to prepare. 

schute, to push. Fr. scodda; D. schudden, to shake. 
sipe, to leak. D. zypen, to drip. 
smikker, to smile in a seducing manner. D. Kil. smeeken 

(blandiri) . 
sned, to prune, cut off, dress by lopping off. D. snyden, 

snoeyen, id. 
speir, spere, to ask, make inquiry. D. sporen, opsporen, to 

trace up. 

spill, spyll, to corrupt. D. verspillen, to squander away. 
stevin (proras obvertere). D. stevenen, to steer. 
syle, (a.) to hoodwink, (b.) to deceive. D. s^ (imbecillis) . 
toot, tout, to sound a horn. D. toeten, id. ; /oe, mouth, 
wp^e, to lift up, exalt. D. opheffen, id. 
upheis, to lift up. D. ophysschen, id. 

B. Substantives, 
afterclap, evil consequence. D. achterklap, id. 


age, edge. D. egge, id. 

anyng, union. Fr. eninghe (contractus) ; enigad (congregatus) . 

barne, bairne, child, young person. Fr. bern, id. 

bak, bank. D. kinnebak, jaw-bone ; bakkebaard, favourites. 

bode, offer from a buyer to a seller. Fr. bod, offering. 

brydal, marriage-feast. Fr. brulloft, id. 

call, kale, cabbage. D. kool, id. ; Fr. koal, id. 

cap, cup. D. kop, id. 

cap, mantle, cloak. D. monnikskap, capouch. 

chaftis, chops. D. schaafsel, chips. 

claver, clover. D. klaver, id. 

cloude, clout, rag. D. kluit, clod of earth. 

dag, thick fog, mist. Fr. dook, id. 

dale, dele, part, division. Fr. deel, id. 

dynn, din, noise. D. deun, tune, song. 

dirk, dagger. D. dirk, id. 

dow, worth, value, avail. Fr. doghet, virtue. 

drotes, nobles, knights. Fr. drochten (dominus). 

eider doun, the smaller feathers of any kind of birds. D. 

eiderdons, id. ; dons, down. 
eild, age. Fr. eld (senis). 
etion, kindred. Fr. etein (procreatus). 
fader, father. Fr. father, id. 

far and, becoming, behaving. Fr. feren (confeetus, formatus). 
farand man, stranger, pilgrim. Fr. farand man^ id. ; fara 


fe, sheep. Fr. fia (pecus) ; D. vee, id. 
frog, upper-coat. D.frack, coat, 
gab, mouth. Fr. gapper, id. 
gaizlings, goslings. Fr. gies (anseres) . 
garth, (a) yard, inclosure, (b) garden. D. gaard, garden. 
glede, gledes, a very small fire, a spark of fire, hot embers, 

Fr. glede, glowing fire. 

grape, a trident fork for cleaning stables. Fr. grype, id. 
hans in kelder, Jack in the cellar. Fr. hansje in de kelder, id. 
hansell, (a) the first money taken, (b) or benefit received 

upon any particular occasion. D. handheld, handsel. 
lallandis, lowlands. D. laage landen, id. 


lauch, law, privilege. Fr. lawa, law. 

lave, remainder. Fr. lefd (relictus). 

leif, leave, permission. D. verlof, id. 

lith, joint. Fr. lithe, id. 

loan, loaning, a vacant piece of ground, close by, or leading 

to, a farm-house. Fr. loane, alley; D. laan, id. 
low, flame, blaze. Fr. loghe, id. 
lute, lent, sluggard. Fr. leuterer, slenterer, loiterer. 
maik, mate, equal. Fr. makker, id. 
maigh, son-in-law. Fr. maag, meg, kin. 
mold, the ground of earth. Fr. molde, moude, mould. 
mone, the moon. Fr. moanne, mona, id. 
morn, to-morrow. Fr. morns (mane). 
mose, moss, (a) a boggy place, a marsh, (b) a heath. D. 

moeras, a marsh. 

muck, mullock, dung. Fr. miuks, id. 
neb, beak, sharp point. D. neb, snavel, bill. 
owke, ouk, week. Fr. wike, id. ; D. week, id. 
, smoke. Fr. reeck, id. 
, rick, stack. Fr. rook, id. 
scharne, dung of cattle. Fr. skern, id. 
schote, the shutter of a window. Fr. schotel, id. 
, lime, mud. D. slyk, mud. 

fefl^, farm-house with dependencies. Fr. state, id., 

with right of voting. 

stew, fumes, cloud of dust. D. stof (pulvis). 
tid, time. Fr. tid, id. 
wald, plain, ground. Fr. wald (nemus). 
wan-luck, misfortune. D. ongeluk, id. 
wan-trow, to distrust. D. wantrouwen, id. 
wan-wyt, want of knowledge. Fr. wanwytschip, id. 
worae, <we, car or carriage. Fr. vaine, wayne, wagon, wain. 

c. Other Words, 
bald, bold. Fr. bold, id. 
blythe, cheerful, merry. Fr. blide, blithe. 
brak, brackish, salt. Fr. brak, id. 
faurd, coloured, complexioned. Fr. verfd, coloured. 


fele,fail, great, very. r.fel,fele (multum) ; D. veel, id. 

forwakyt, exhausted by lying long awake. D. verwaakt, id. 

hais, hoarse. D. heesch, id. 

heal, whole. Fr. hel (illsesus) ; D. heel, id. 

law, humble, low. D. laag, id. 

loune, well sheltered, without wind or wave. D. luuwte, 

place sheltered from, the wind. 
muthe, exhausted with fatigue. Fr. moed, weary; D. ver- 

moeid, id. 
namekouth, famous, well known. D. naamkundig, id. ; Fr. 

burcuth (vicinis notus) . 

ouklie, weekly. Fr. wieks, id. ; D. wekelyks, id. 
raith, quickly, hastily. D. rasch, id. 
sakless, guiltless, free. Fr. secka (accusatus). 
side, hanging, reaching low. Fr. side (profundus) . 
sikken, such kind of. Fr. sokken (talis) . 
smaddit, bedaubed, smutted. Fr. smodsig, id. ; D. besmet, id. 
smittle, infectious. D. besmettelyk, id. 
sprekled, spotted, speckled. D. gesprikkeld, id. 
sute, sweet. D. zoet, id. ; Fr. swiet, id. 
sythyn, ever after that time. D. sedert, since. 
thick, intimate, familiar. Fr. tige, worthy ; D. deeg, id. 
tute-mowit, having prominent lips. D. toet-mond, id. 
op-a-land, at a distance from the sea. D. op in het land (in 

terrain versus). 

wan-schaipen, deformed. D. wanschapen, id. 
wat, weit, wet, to wet. Fr. wet (humidus). 
yeld, eild, barren, that gives no milk. Fr. geld, id. 


A. Verbs. 

bede, pray. Fr. bidda, to pray; D. bidden, id. 

bidden, invited. Fr. bidden, id. ; D. gebeden, id. 

boot, it boots not, is to no use. Fr. batia (prodesse). 

brak, broke. Fr. brak, from breka, to break. 

brosten, burst. Fr. bursten, broken, brutsen, burst, cracked. 

drub, beat. Fr. drope (ictus) . 

feal, hide. Fr. fel (cutis). 


fey, cleanse. Fr. feye, id. ; D. veegen, id. 

gang, go. Fr. ganga, id. 

gee, give. D. geven, id. ; Fr. giva, id. 

glowing, staring. D. gloed, glowing. 

heald, incline. D. hellen, to incline. 

knab, seize hastily. D. snappen, id. 

lake, leak, play. Fr. hlacka, to laugh ; D. lagchen, id. 

/owe, flame. Fr. loghe, id. 

querken'd, suffocated. Fr. querdzed, id. 

nd, remove, prepare. D. redden, opredden, to put in order. 

s##, sow. D. zaaien, to sow. 

sobbed, wet. D. besabd, slabbered. 

sa^, hang down on one side. D. zakken, zygen, to sink down. 

scrab, scratch or claw. D. krabben, klaauwen, to scratch. 

shie, shy, avoid a person. D. schuwen, to avoid. 

s^/^ change one's clothes. D. schiften, to separate. 

skrike, shriek. D. schrikken, to startle. 

slade, carry goods in a sledge. D. sleeden, id. 

smudge, soil, besmear. Fr. smodse, to soil. 

, ask, inquire. D. speuren, sporen, to inquire, 
raise dust. D. stuiven, to raise dust. 

, hang on one side. D. waggelen, to stagger. 
swell, swallow. D. zwelgen, to swallow. 
sye, put milk through a sieve. D. zeeven, id. 
, drop gently, distil. D. zypen, to drop. 

wjo, drink up. Fr. opsupe, to drink up; D. opzuipen, id. 

overturn. Fr. welda (regere, dominare). 

B. Substantives, 
call, obligation : ex. ' as he had no call to do it.' Fr. kalla 


cluve, hoof. Fr. klau, klew, id. 

crib, a rack to hold hay for cows and horses. Fr. krebbe, krib, id. 
dag, dew. Fr. dook, id. ; D. daauw, id. 
dell, low, hollow place. Fr. del (infra) ; D. deel, floor. 
frimfolks, strangers. Fr. fremdfolck, id. 
gob, open, wide mouth. Fr. gapje, to gape ; D. gaapen, id. 

, dung-fork. Fr. grype, gryp, id. 

, nook of land projecting into another field. Fr. herne, 


hauffh, haw'} ^ hillock Fr ^ t higll . D ^ ^ i(L . 


holl y a dry ditch. D. hoi (spelunca) ; holte (cavitas) . 

holt, a wood. Fr. holt, id. 

kern-milk, butter-milk. D. karnemelk, id. ; Fr. suup, soupe, 

tserne molke, id. 

mauf, meaugh, brother-in-law. D. maag ; Fr. meg (cognatus). 
scarn, dung. Fr. skern, id. 
slade, sledge. D. slede, sleed, id. 
stead, place to stand on. Fr. stee, id. 
sted, place or house. D. huisstede (locus in quo domus est 

exstructa) . 

steert, point. Fr. stirt, stut (cauda) ; D. staart (fig. finis). 
stew, cloud of dust or vapour. D. stof, id. 
stub, stump of a tree. Fr. stobbe, thump. 
wark, ache, pain. Fr. } t werkt my in y t lif (dolorem sentio 

in ventre). 
yarth, earth. Fr. irthe, id. ; D. aarde, id. 

c. Other Words. 

bleek, pale, sickly. D. bleek, ziekelyk, id. 

efter, after. Fr. efter, after, id. 

full, drunk. D. vol, dronken, id. 

goel, yellow. Fr. giel, id. ; D. geel, id. 

over, more than. Fr. over, more than enough. 

reet, right. D. regt, id. 

seeVn, seldom. D. zelden, id. 

stolt, stout. Fr. stow/, id. 

stumpy, short and thick. D. stompig, dully. 

swimmy, giddy in the head, having a dimness in the sight. 

Fr. swima, swoon; D. zwym, swoon. 
unrid, disorderly, filthy. Fr. onree, id. 
war, beware ! take care. D. waar \ (cur a). 
worfor, wherefore. Fr. werfor, id. ; D. waarvoor, id. 
yell, barren, or that gives no milk. Fr. geld, id. 



[Read December the 18^.] 

On the reading of Mr. Wedgwood's observations on the 4th 
of November last, on the connexion of the Finnish and Indo- 
Germanic classes of languages, some of the Society now 
present may recollect that I called their attention to several 
remarkable coincidences of words then cited, with their co- 
relatives in the Basque language. Those coincidences, then 
shown to exist so remarkably within so small a compass of 
words, gave good ground for the suggestion that many more 
might be found upon a fuller investigation; and it may, there- 
fore, be well worthy of consideration for those interested in 
such pursuits, to have a detailed account of the words I 
referred to, for the purpose of assisting them in their future 

The Basque has been pronounced by the generality of wri- 
ters to be a language sui generis, though some have hazarded 
other opinions respecting it, into the validity of which it is 
not my intention at present to enter. The language is cer- 
tainly well deserving of the most careful investigation, and 
more especially so as all the theories heretofore published 
respecting it appear to me open to very considerable objec- 
tions. One, however, of those opinions, expressed by Mr. 
Borrow in his work ' The Bible in Spain/ that it is a Tartar 
language, I will venture to cite, as from the analogies now 
shown to exist, there really seems some reason for supposing 
it to have some foundation, though I have not met with any 
higher authority for so curious a fact in philology than the 
dictum of that amusing writer. But the neglect of the Basque 
is the opprobrium of modern philology, in which our English 
philologists must also take their share ; for it seems an incon- 
sistency with right judgment, that while we have been 
exploring the intricacies of the languages of the most remote 
times and countries, we have not, in English researches, any 
account rendered of this extraordinary language of a very ex- 


traordinary people living within a few hours' sail of our shores, 
and forming component parts of the neighbouring kingdoms 
of France and Spain. Even in those kingdoms, though almost 
innumerable works have appeared from time to time relating 
to Basque, I. do not know of one entitled to the least respect 
in a philological point of view. William Humboldt indeed, 
in German, has given the world one of a higher class, yet I 
venture to think that even he, upon this subject, adopted an 
erroneous theory, and was thus led to many controvertible 

Contenting myself for the present with these observations, 
I proceed to point out the coincidences I referred to, and take 
first Mr. Wedgwood's preference, above other explanations of 
the word king from different Teutonic roots, of the radical 
identity, as he calls it, of the word with the Tartar chan. This, 
or khan, is the only word instanced by Mr. Borrow, and in the 
paper before us is associated with the Ostiack chon, l emperor,' 
and other words. The Basques being determined republicans, 
own no king, and the king of Spain is, by their Fueros, only 
lord (jaun orjauna) of Biscay. This word is pronounced with 
a strong aspirate, haun or hauna, by the French writers spelt 
yauna. The only word in Basque for God is Haun-goycoa, 
literally ' the Lord dweller on high.' 

In the same page (ante, p. 174) are the following other words 
instanced, to which I append their corelatives in Basque, 
referring the reader to the analogies given by Mr. Wedgwood, 
connecting them with other languages : 

Eng. pot, Basq. pota. II Eng. oil, Basq. olioa. 

teat, titia. border, borde. 

nasty, nastia. dry, idorra. 

km, u. 

To these I might add several other analogies, from which I 
abstain, as they might not be so readily admitted as the 
preceding. I will however instance E. father, B. ait a, com- 
pared with Lapp attje; and Eng. guise, B. guisa, with the 
Breton giz, kiz (ante, p. 176), adding, that this word seems to 
me to have been adopted into other languages from the 
Basque, in which it is of radical signification. 


The words above enumerated comprise nearly the half of 
those given in the " Observations," to which these may be 
considered a supplement. If they show in connexion with 
them any allowable affinity to words in other languages so 
wide apart as Finnish and others, they show a still more 
marked relationship to the English, and this will prove the 
justice of the remark I made at the outset, of the impropriety 
of our philologists passing over so negligently a language 
spoken opposite our own shores, and to which our lexico- 
graphers have never turned in search of the unde-derivaturs 
they might have sometimes found there. 


There is a new word coined, within few months*, called fanatics, which, 
by the close stickling thereof, seeraeth well cut out and proportioned to 
signify what is meant thereby, even the sectaries of our age. 

Some (most forcedly) will have it Hebrew, derived from the word to see 
or face onef, importing such whose piety consisteth chiefly in visage, looks, 
and outward shows ; others will have it Greek, from <}>, to show and 
appear; their meteor piety consisting only in short blazing, the forerunner 
of their extinction. But most certainly the word is Latin, from fanum, a 
temple ; and fanatici were such who, living in or attending thereabouts, 
were frighted with spectra, or apparitions, which they either saw or fancied 
themselves to have seen. These people, in their fits and wild raptures, 
pretended to strange predictions : 

ut fanaticus cestro 

Percussus, Bellona tuo, divinat, et ingens 

Omen habes, inquit, magni clarique triumph!. Juv. Sat. 4. 

Ut mala quern scabies et morbus regius urget, 

Aut fanaticus error. Hor. in Poet. 

It will be said we have already (more than a good) many nicknames of 
parties, which doth but inflame the difference, and make the breach the 
wider betwixt us. It is confessed ; but withal it is promised, that when 
they withdraw the thing we will substract the name. Let them leave off 
their wild fancies, inconsistent with Scripture, antiquity, and reason itself, 
and then we will endeavour to bury the fanatic, and all other names in 
perpetual oblivion. Fuller's Mixt Contemplations on these Times-, pub- 
lished in May, 1660. 

* Of May 1660. f H3Q vidit. 



[Read February the 22nd.~\ 

As it is the fashion of essays in the present day to begin 
at a point that has apparently no relation with the subject 
matter, I will first offer some remarks on a passage in 
Niebuhr's Roman History. In p. 53 of the translation of his 
second volume, he considers the origin of the word municeps, 
where he tells us that " munus properly signified a duty which 
a citizen was bound to discharge, whether by personal or 
pecuniary services : and municeps was the opposite of immunis, 
which designated a person exempt from such burthens." He 
goes on to say, " that the last syllable (ceps), though it has a 
deceptive look of coming from a verb, is nothing more than 
one of those manifold terminations in which the Latin lan- 
guage luxuriates." In a note, we receive further etymological 
information: "That the additional syllable does not affect 
the meaning is evident in princeps for primus, and in the con- 
secutive ancient ordinals given by Varro, terticeps, quarticeps, 
and the rest. So biceps probably meant nothing more than 
twofold, triceps threefold." Subsequently he refers to anceps 
as another proof of his principle ; and ridicules the idea that 
cap of caput or of caper e enters into any of these words. By 
an awkward omission he leaves out of view auceps and 
praeceps, two words which I suspect he would have found it 
difficult to explain without some reference to the roots just 

Now in the first place, municeps, to speak with accuracy, is 
derived, not from munus, but from the allied neuter sb. muni-, 
more familiar in the plural munia; whence also immunis. 
Then again Niebuhr seems to lose sight of the original 
meaning of these words, viz. "share" or "part"; a meaning 
which again subdivides itself according as the object is de- 
sirable or not, and according as the object is physical or not. 
A share in that which exists and is desirable may be regarded 
as ' a gift/ But duties to be performed may also be divided. 
Here again, if the duty be one of honour, it will be ' an office/ 


and the holder f an officer or charge d'affaires. 3 If it be 
something not desirable, it may be entitled 'a burden or 

Another German writer, whose studies were specially con- 
nected with linguistic principles, Dr. Carl F. Becker, uses 
expressions of a similar character. In his Grammar of the 
German language he thinks it right to include in the list of 
primary substantives, " some which have assumed one of the 
terminations er, el, en ; as messer ' knife/ schenkel ' leg/ bissen 
' bit ; ' " and, in a note upon the passage, he observes : " These 
terminations, er, el, en, differ from affixes of secondary deri- 
vatives, in having no influence on the signification of words 
to which they are added." 

My main object in quoting these passages is to oppose the 
doctrine that any language whatever has dealings with mean- 
ingless terminations ; and the protest is the more called for, 
when the doctrine comes to us with the sanction of such high 
authorities. But the error is a common one, sometimes 
expressed in distinct language, as when we are told that such 
a syllable is ' only a termination/ More frequently the final 
letters of a word are quietly ignored, especially in our dic- 
tionaries, where it is deemed sufficient to explain the forma- 
tion of the first part of a word, or perhaps to give only, what 
is designated by the vague term ' theme/ Yet as regards 
dictionaries, we have the less reason to be censorious, because 
it is the duty rather of grammars to deal with those analogies 
which belong to final syllables. Yet here again there is for 
the most part a sad deficiency, as soon as we leave the par- 
ticular class of suffixes which belong to the conjugation of 
verbs or the declension of nouns. 

But among the terminations which are treated with indif- 
ference, none have met with such neglect as those of dimi- 
nutival power ; and this perhaps chiefly owing to two facts : 
first, that they often lose their distinction as diminutives ; and 
secondly, that they are apt altogether to supplant the primitive 
word. We will not stop now to prove these two assertions ; 
but rather assuming their truth we would point to the causes 
which have produced such results. Diminutives are used 


with various objects, as a, to denote smallness; b, tenderness or 
affection ; c, pity ; d, contempt. But of these four meanings 
the first requires subdivision. An object may be small in 
comparison to others of its own class ; or it may be one of a 
class, all the members of which are regarded as small. In this 
latter case the use of a derived diminutival form beside the 
primitive is in a great measure superfluous; the result of 
which is, that one of the two rivals has soon to give way. In 
a contest of this nature it is commonly the shorter form 
which is abandoned, so that the remaining word, for the 
reason that it stands by itself, seems to claim the honours of 
a primitive. Yet it often happens that some cognate lan- 
guage, or an older stage of the same language, exhibits the 
simpler word ; or, what is equally useful for an analysis, we 
may find the radical part connected with some equivalent 
suffix of different form. But instances may make this clearer. 
When we say ' little robin redbreast/ we use the epithet, not 
to distinguish one redbreast from another, but to compare 
this whole class of birds with classes of larger size. It is in 
this way that the term starling is applicable to any one of 
those birds which were formerly called stares ; but the former 
term alone is now in ordinary use. Again, violet of our own 
tongue and veilchen in German, are partly explained by the 
Latin viola ; but this again needs explanation from the Greek 
lov (ftov), or rather from an obsolete noun via (violet), which 
would correspond to the Greek hov, much as rosa to poSov. 
The English sparrow too and German Sperling alike point to 
a monosyllabic form spar or sper } of which the initial sibilant 
is probably no more an essential portion than it is in the 
Greek adj. a-fuiepos, the Latin sb. spina, or the English vb. 
smelt. Thus we arrive at par, a syllable nearly akin to that 
which is seen in the first part of the Latin pass-er, as we may 
infer from the ready interchange of the sounds s and r. 

So long as the diminutival suffix has maintained itself in 
its full form, or something approaching to fulness, it is a 
tolerably easy matter to detect it ; but from the very circum- 
stance of its being to a great extent an all but superfluous 
addition, it is apt to be compressed and corrupted ; and the 



danger is the greater because the closing syllable of a word 
has rarely the accent. Thus it often appears as a single con- 
sonant or single vowel; nay, at times so completely dis- 
appears, that we have no other evidence of its having belonged 
to a word but in the modification of the root-vowel effected 
by it, as in Jem for Jemmie, from James ; Kit for Kitty, from 
Kate. In such extreme cases it requires not a little nicety 
in the use of the dissecting-knife to demonstrate the dimi- 
nutival element. 

Again, in modern times the power of forming diminutives 
may be in one country a living principle, so that it is per- 
mitted to form such words ad libitum ; while in another, those 
only are admissible which have already received the stamp of 
authority. Thus the suffixes chen and lein are employed with 
almost unlimited freedom in the ordinary language of Ger- 
many ; and in Southern Germany diminutives in el may be at 
pleasure created without fear of the charge of innovation. 

The original purpose of the present paper was solely to 
examine the suffixes of diminutival power in the Latin lan- 
guage, but an inquiry of this nature often derives much 
benefit from the light of comparative grammar. Moreover 
in a dead language, the books of which deal little in the con- 
versational style of private life, we cannot expect this par- 
ticular formation to be exhibited in its fulness. It is not so 
much in the elevated literature of a country, nor indeed in 
public life, but rather by the private fireside, or in the inter- 
course of rustic society, that the free use of diminutives is 
found. Hence, to speak of England in particular, only a 
small proportion of such words is honoured by admission 
into our leading dictionaries. In the provincial dialects they 
still abound ; but we may perhaps affirm, that nowhere more 
than in Scotland is the formation of such words still a living 
principle of the language. The fact is familiar to a native 
Scotchman, but the Southron must accept the assertion on 
the authority of others, as of Jamieson in his Scottish Dic- 

We shall begin then with some inquiry into the several 
classes of diminutives which belong to the Saxon element of 


our own tongue, illustrated occasionally by some of its sister 
dialects on the continent; and here our first duty is to 
consult the elaborate work of Grimm. But the benefit we 
can derive from this source is far from being all we could 
desire. Throughout his book he treats our portion of the 
family with comparative neglect. Thus of the diminutival 
suffix ock } one of the most important we possess, he gives but 
two examples, and several of our other terminations of like 
power he leaves unmentioned. Still, what Grimm places 
before us in some detail of the diminutival forms in the sister 
dialects, ancient and modern, throws much light on the 
inquiry ; and his deficiencies in respect to the English lan- 
guage are to some extent supplied, in this department, by a 
valuable paper published at Cambridge in 1832, in the Philo- 
logical Museum, No. iii. p. 679, from the pen of one of our 
own members, Mr., now Sir G. Cornewall, Lewis. The chief 
sources, then, of which I have availed myself, are those just 
mentioned, Grimm's Grammar, Sir G. C. Lewis's paper, and 
Jamieson's Dictionary, including the Supplement; together 
with some of our provincial glossaries, as Jennings' Somerset- 
shire Glossary, Grose's Glossary, Wilbraham's Cheshire Glos- 
sary, and Moor's Suffolk Words*. But in addition to the 
matter thus obtained, there is also in what follows much for 
which they are not responsible. 

I. Simple diminutival suffixes : 
A. a. ock, as 

haddock, J., fry of the coal fish. 
(See other names below.) 

bannock, J., an oat or barley 
cake. (Cf. our bun and Grael. 

bittock, J., a little bit. 

bladrock,3., a talkative, silly fel- 
low, from blather or blether, 
1 idle talk.' 

brannock, J., the samlet, a small 

fish called in Yorkshire bran- 


bullock, a young bull. 
buttock, the first syllable also in 

bott-om and German bod-en, 
cabock, J., or kebbuck, cheese 

(Gael, cabag). 
cammock, J., a crooked stick 

* These will be abbreviated thus : D.G. Grimm's Deutsche Grammatik ; 
L., Sir G. Lewis's paper ; J. or Scotch, Jamieson's work ; S.W., Moor's 
Suffolk Words, &c. 


(Gael, camag, crooked, from 
cam, crooked). 

castock, J., core of a cabbage- 

charlock, the weed. 
clublock, J., the spotted blenny 


crummock, J., a staff with a 
crooked head, or a cow with 
crooked horns. 
devilock, J., a little devil. 
dunnocJc, hedge-sparrow (Che- 
earock, J., a pullet (Gael, eireag, 

pullet, eun, a bird). 
emmock, J., an emmet or ant. 
fillock, J., a filly. 
finnock, J., or finner, a white 
trout (Gael, fionnag, from 
fionn, 'white')- 
fintock, J., the cloudberry, Eubus 


gavelock, J., an iron crow, an 
earwig (from gavel or galel, 
1 a fork'). 

hassock, J., anything bushy, a 

besom, a turf in form of a seat, 

a kneeling cushion. 

hattock, J., a little hat. 



hirplock, J., one who hirples or 

goes lame. 

hornock, J., old Hornie. 

hummock, J,, or hummie, the 

hand so arranged that the tips 

of all the fingers press the 

point of the thumb. 

Jiumplock, J., a small heap. 
Jamock, J., little James. 
kittock, J., or kittle, a loose 


Jcnublock, J., a little knob. 
laddock, J., a little lad. 
lassock, J., a little lass. 
laverock, J., a lark (Lancashire 

learoc/c) . 
lythocks, J., a poultice, from 

lithe, vb. * to soften.' But in 

Cheshire 'to lithe the pot' 

is to put thickening in it, as 

flour or oatmeal. 
mammock, a piece, a fragment. 
mannock, J., a little man. 

mullock, dirt, rubbish. 
mulock, J., or mulin, a crumb. 
munshock, J., red bilberry. 
paddock, J., a frog (pade A.S. 

paddock, orparrock, a small in- 

closure or park. 
pellock, J., a porpoise, regarded 

as a little whale. 
pellock, J., a ball or bullet. (Cf. 

piltock, J., a coal fish a year old, 

then called lillet at Scarbo- 

playock, orplaik, J., a plaything. 
pollock, J., young of the coal 

fish; and also, I am told, a 

small edible crab. 
queock, J., young cow. 
raplock, J., coarse woollen cloth. 
ruddock, S., red-breast (robin- 

riddick, Jennings). 
rullocks, from an obsol. dim. of 


the vb. row, seen in the Scotch 
role, ' to row.' 

sillock, J., or sillick, fry of the 
coal fish, 


shillock, J., lighter part of oats 
(comp. shell). 

sourock, J., sorrel. 

tammock, J., or tummock, a hill- 
ock (comp. Gael, torn, a knoll, 
and Lat. tumulo-). 

tarrock, the bird Larus tridac- 

tussock, J., a tuft of wheat in 
a cornfield. (Comp. our 
tassel. ) 

wallock, J., lapwing. 

warlock, J., wizard. Note that 
wizards and witches are ge- 
nerally depicted as dwarfs; 
and for the root syllable 
compare the Germ, wahr- 

whilock, J., a little while. 

wifock, J., a little wife. 

winnock, J., window*. 
Hence with the guttural softened to a final w : 
b. ow 3 as : 

minnow (comp. minnikin). 
morrow (comp. mor-n and the 

Grerm. morg-en). 
pillow, comp. A..S. pile, the same. 
sallow, comp. A.S. seal and Gael. 


barrow, 'truck' (bear, vb.). 
barrow, (boar), 
bellows, Grerm. balg. 

farrow, Grerm. ferk-el. 
furrow (comp. rig and fur, Scotch 

and North of England for 

'ridge and furrow,' and the 

~Dsin..fure, 'a furrow'). 
gallows, Grerm. galgen. 
haddow, or haddock, 
harrow (compare harry, harass, 

and A.S. herian). 
killow, ' black earth.' 
mallow, Grerm. malve. 
marrow, A.S. mearh, also smere 

and smeru, 'fat, grease' ; Gael. 

smeor, ' marrow,' and smeur, 

1 smear.' 
marrow, match, fellow, pair. 

sael, 'willow,' Germ, saal- 

scarrow, J., faint light ; comp. 

A.S. scir, sheer, bright. 
shadow, shade, 
shirrow, or skrow, or skrew, 

shrew-mouse ; A.S. screawa. 
sorrow, from sore (A.S. sdr, 'sore, 

sorrow') ; Germ, sorge. 
sparrow, Germ, sper-ling. 
swallow, Germ, schwalbe. 
tallow, Germ. talg. 
willow, comp. Germ, weide, Eng. 

withy ; probably from vb . wind, 
window, Scotch winnocJc, from 

sb. wind. 

callow, Germ. kohl. 

* This list might easily have been doubled with the assistance of 
Jamieson's work. 


fallow, Germ, brack, Dan. brak, 

probably the same word, and burrow, from vb. bore. Comp. 

so related to our adj. bare, 
hollow, from hole; note that in 

Scotch we have verbs holl and 

hoik or houk, ' to dig.' 
mellow, Germ, murbe, which is 

the same word ; Dan. mcer. 
orrow, or orra, J., what is odd 

or over. From the same root 

orr els, what is left over, refuse, 

and a vb. ort, reject, as well 

as the provincial Eng. sb. pi. 

or ts, refuse. Thus orrow seems 

to be for over-ow. Comp. for 

form lark and larrick from 

laverock. See J. 
sallow, Germ. prov. sal, l sallow, 

dirty' ; Eng. soil, sb. and vb. ; 

IV. sale. 

shallow, shoal, adj. in shoal-water. 
yellow, Germ, gel-b, Dan. guul. 

Hence also our gol*d, and se- 
veral Scotch words, asgule or 

^ooZ,the corn-marigold, gulset 

or gulschoch, jaundice, and 

perhaps the heraldic adj. gules, 


3. VERBS. 
bellow, from the old verb to bell. 

I was at first tempted to insert in this list the Cheshire words 
drumbow or drumble 'a dingle/ songow or songle ' gleaned 
corn/ stubbo f stubble/ bricco ' brittle * ; but I was soon satis- 
fied that the ow (o) in these forms was for ol, just as pow in 
the same county stands for poll. Still two provincial verbs in 
"Wilbraham's Glossary seem entitled to a place in the list, 
ballow ' choose' (comp. wale 'choose/ Scotch and Germ. 
wahl-eri), and bradow < brood over.' 

Also by weakening the vowel of ock we have : 

borrow, Germ, borg-en. 

burrow, from vb. bore. 

also bury, 
follow, GreYw..folg-en. 

hallow, Germ, heilig-en, and as 
an adj. heilig and our holy, 
Probably from such a root as 
our whole, Germ, heil, whence 
the Germ. sb. heil, happiness, 
safety, salvation. 

swallow, Germ. schlucJc-en, 
schling-en, and schwel-g-en, 
Dan. sluge or sv&lge. The 
simple vb. survives in vulgar 
life, swill, Somersetsh. swell 
or zwell. 

tarrow, or tarry, J., 'delay,' make 
a difficulty of. The Germ. 
zoger-n seems to imply that 
the first syllable tar has already 
been compressed from a di- 
syllabic form, such as tager, 
and so related to tug or tow, 
draw out; Germ. zieh-en(zug}. 

wallow, 'roll,' Germ. prov. walgen. 

winnow, Germ, wann-en ; So- 
mersetsh. (Jennings) wim ; 
akintoj^m; Lat. vanno-; our 
vb.'to wind"; Lat. vento-; and 


c. ick, as lassick, J., ' a little lass ' ; laddick, J., ' a little lad,' &c. 
whence again by softening the guttural : 

d. ie (or y), as haddie, lassie, laddie, crummie. 

This last form, having the sanction of the Scotch capital, 
is established in Scotch literature, and seems to be extending 
its domain over the provincial dialects of Scotland, so that at 
Glasgow haddie is said to be now superseding the form haddow, 
which till recently prevailed there. But haddock and haddick 
still hold their ground in some quarters. In England a final 
y is preferred to ie, as bury, tarry, worry, penny, lassy, laddy. 

In not a few of the diminutives, a compression has taken 
place either of the root syllable or of the suffix, in which 
cases the form of the latter is often so far modified as in a 
great measure to conceal its connexion with the original form. 
But the changes are scarcely more violent than the varieties 
of sound which attach themselves to the combination ough in 
our anomalous spelling. Again, in different parts of England 
we hear shock, shoof and sheaf of corn; and the loss of a 
guttural in writing a word is the less open to suspicion, as 
sometimes even when written, it is dropped in pronunciation, 
as the fluke of an anchor, commonly called flue. Indeed we 
ought perhaps to have included among the various forms of our 
suffix, as above given, that of och (pronounced as an aspirate), 
for this form is not unfrequently given by Jamieson ; in which 
case the Scotch ock and och would have corresponded with 
some accuracy to the Gaelic diminutival suffixes ag and 
ach. Sometimes indeed, when the compression is limited to 
the first syllable, the siiffix may retain one of the forms 
already enumerated. Thus suspicion, and in some more than 
a suspicion, that a disyllabic form has suffered a compression, 
attaches to the following words, which we would therefore 
propose for examination : 

e. Slock (same as clog and log of wood); clock, 'a beetle' 
(Scotch golacJi see clock, clock-bee, and cZocMeddie, ' a lady-bird,' 
in J.) ; crock, flock (of sheep), flock or lock (of wool), frock, stock. 

f. Blow, vb. (flare) ; bloto, vb. (florere) ; blow, sb. (ictus) ; flow, 
glow, grow, know, throw, low, roiv, sb. ; row, vb. 

g. Brick, click, crick, rick, stick, trick. 


More commonly some other variety presents itself, as 

h. Brook, sb. (br=bur of bur-n, of the same meaning) ; brook, 
vh. (Germ, brauch-en) ; crookl(cr=cur of cur-T) ; nook', rook (a 
variety of the sb. crow). 

i. Pluck, vb. (from ^wZZ) ; pluck, sb. (perhaps of the same 
origin) ; ruck. 

j. Fluke of an anchor, and. fluke a worm, where the root portion 
ft denotes flatness, as in. flat itself, and represents the pi of the 
Latin .piano-, and the pal (pad} of pand-, palam, palma, palud-, 
&c. ; Zw&e-warm (derm. 7#w). 

k. Black, brack, nack, rack, slack, wrack, ' sea-weed' (war, A.S., 
and ware, Scotch, ' sea-weed'). 

1. Brake, ' fern' ; flake, lake, rake (= harrow), to which add 
break and wreak. 

m. Fleck, 'fur of rabbits' (Moor's S.W.) ; freck or freckle 
(G-erm. fleck, flecken), speck. 

n. Bark, vb. (from bell, vb. obsol. ; comp. Germ, bell-en, ' to 
bark') ; bark, sb. ; cark, sb. obsol. from care-, cask ; caw?& (a ship), 
(Scotch vb. calf or coif, the same ; also calflng, for the ' wadding' 
of a gun) ; chink (chine); dark (darn, vb. 'to hide' ; darn, adj. ' se- 
cret' in Scotch, J.) ; hark (hear ; comp. also hearken) ; hoik, Scotch 
'to dig' (holl, the same, J.) ; hulk, sb. (also hull of a ship) ',jerk, 
lark for larrick, Scotch, and that for laverock-, lurk (Scotch vb. 
loure-, Germ. lauer-n,the same) ; milk (A.S.meoloc, as well as meolc ; 
comp. yaXaKT--) ; park (A.S. pearrok or parruc, same as our pad- 
docJc) ; j?m&, vb. as with a dagger (also to pin in the same sense) ; 
sark, prov. ' a shirt,' which is probably the same word ; shank 
(shin) ; shirk or shark (Scotch), vb. ; spark ; stark (as stark-naked, 
stark-mad ; comp. the Germ, starr, ' stiff, rigid,' whence starr- 
blind, 'utterly blind') ; stirk, 'a steer'; talk (tell) ; walk (Old 
Germ, wall-en, ' to go') ; whelk', work (comp. ware, ' the produce 
of labour' ; also ear, vb. obsolete, 'to plough,' or, as the French 
say, labourer la terre) ; wink', yolk (= yellow). 

o. Brog, J., or prog, J. and Moor (our prong), 'a sharp point, 
spike, goad' (from vb. lore ?) ; clog, flog, frog (perhaps the same 
word as the Scotch and provincial paddoclc, first changed to par- 
Took, just as the other word paddock has in fact been) ; grog, log. 

p. Brag, crag (A.S. carr, ' a rock' ; prov. scarre, ' bare rock,' 
N. Grose, GL, and carrock, the same : see Bosworth) ; drag, flag, 
sb. ; flag, vb. ; lag, scrag, slag, snag, stag. 


q. Plug, rug, shrug, ' shrew,' J. ; snug. 

r. Claw, craw, draw, flaw, straw. 

s. Sirrah (contemptuous dim. of Sir). 

t. Brew, crew, new, screw, shreiv, sinew, strew. 

u. Slue, clue, flue (of anchor), ^we (downy matter), glue, rue. 

v. Groo, ' water partially congealed,' J. 

w. Floe, roe (deer), roe (offish), throe. 

x. The syllable ougli, variously sounded, as: plough, rough, 
slough, through (Germ, durch, from dur), trough. 

y. Bring, cling, fling, ring, ivring. 

z. Prong, strong, throng. 

aa. Bluff, fluff, gruff, adj. gruff, ' a mine,' Somersetsh. (akin to 
grave and grub) ; luff, ruff, stuff. 

bb. Calf, half, turf (the simple toor or ture is given in J.), wharf 
(Fr. gare, ' a landing place'), wolf (perhaps from gul, ' yellow'). 

cc. Crave, grave. 

dd. Club, grub, shrub, rub, snub. 

ee. Crib, glib, nib, rib. 

ff. Larrop, vb. of the low language, perhaps contracted from 
an obsolete vb. latherop from leather or lather, vb., the same ; 
(comp. where for whether (Somersetsh.); or for other \ smure, Scotch, 
for smother ; far, Danish, for father \nfar-broder, 'patruus' ; far- 
fader, ' father's father' ; so also in Somersetsh. gramfer, ff rammer 
for ' grandfather, grandmother') ; scallop (shell) ; wallop, vb. (to 
wall or well, Scotch form of weld, beat (metal) into one mass) ; 
wallop, 'boil' (Suffolk), (Germ, wall-en, 'to boil,' our vb. well; 
comp. also pot-walloper) . 

gg. Carp, sb., chirp, help, sharp (comp. shear, vlo.), warp, whelp. 

hh. Bree (=brotv, as in the Scotch ee-bree; A.S. breah) ; knee 
(A.S. cneow) ; tree (A.S. treow), to which addjfe (comp. also 
for the form of these words the G-erm. schnee=snow). 

ii. Cry, dry (Germ. troclcen),fly vb.,j% sb. (A.S.jleoh),fry sb., 
fry vb., ivy, sky, try, wry. 

kk. et* (for ec), as: ladget 'badger,' brisket 'piece of the 

* In our mixed language it is of course important to distinguish between 
the Norman and Saxon element. Thus in reference to the suffix et, we 
must carefully separate from such words as are given above, those which 
represent the French suffixes et, ette, as trumpet, lancet, billet, facet. Yet 
it must be confessed that this separation is at times difficult, seeing that 
the French language possesses not a few Teutonic words, to say nothing 


breast' (for brist-eck or bristicTe, from breast), cricket, emmet 
(comp. Scotch emmock, or immick}, fitchet (fitch, Somersetsh.) 
' polecat,' gimlet (Scotch gemlick}, gobbet ( a piece' (Shakspere), 
comp. gappocks and perhaps gabbocks, Scotch, J., hornet, mam- 
met* (= mammock}, limpet, locket (lock}, mallet (maul or mall, 
prov. the same), pack-et, pock-et, sippet (sop}, smicket (smock}, 
tippet (from top ; comp. hood and cape, both originally signifying 
' head'), wevet 'a spider's web' (Somersetsh.), worret (Suffolk, &c. 
for worry}. 

11. The same contracted to a simple t : graft, haft, left (laevo-), 
lift vb., silt (soil}, tiltv\). (tall}, tuft (tuff, the same, Scotch), wart 
(A.S. wear, ' callosity, knot, wart' ; ware, Scotch, ' knot in a tree'). 

mm. ot (for ock} : ballot (ball), blot, clot, eyot (or ait}, grots (pro- 
nounced grits}, lot, maggot (perhaps for madock, akin to Grerm. 
made, and the precise representative of the A.S. maftu maggot), 
rot, spigot (spike}, spot (= speck}, trot. 

This change of k to t may be illustrated by the double 
forms apricock and apricot, bruckle and brickie (Scotch), now 
corrupted among us, in spite of the word break, to brittle, the 
old sb. make and its modern representative mate, our ordinary 
verbs leak, poke, slack (lime), and the Somersetshire ledt, poodt 
or pote, slait. 

But the sound k is always apt to interchange with the 
sibilant ch or sh. Of this we have the best evidence in the 
double forms which prevail in France, viz. ch in Paris, c in 
Picardy, &c., as chat and cat. Thus it often happens that we 
have by preference adopted the guttural forms in the Norman 
portion of our language, as castle (chateau), captive and caitiff 
(chetif), while not unfrequently we have the two forms exist- 
ing together beside each other, as chevalier and cavalier, cape 
and chief, bank and bench. It is therefore noway surprising 
that among our Saxon stock we have both varieties coexisting, 
as kirk and church, wake and watch, dike and ditch, twig and 
switch. And so too our diminutival suffixes ock and ick have 
given place to sibilants, 

of those primitive words which may be claimed as natives both by the 
Romance and the Teutonic languages. 

* See below for the examination of this word. 


nn. ch, as: scratch (from scar), winch (from wind, vb.), blotch 

oo. sh, as : wish (from will, as the Germ, wilnsch-en from woll- 
en), sulsh vb. (Somersetsh. =soil vb.) and slush sb., blueish, 
blackish, brackish, &c.,frosh (=frog), brush. 

pp. ass, ss, as : harass (=harrow or harry, A.S. herian), morass. 

Before leaving this division of the subject it may be useful 
to point to a few diminutival adjectives, which often fail to be 
recognized as such, viz. any (einig), many* (mannig, now found 
only in compounds), and perhaps the German wenig may be 
the representative of our minny or minnie. 

B. Diminutives formed with el and its varieties. 

a. el (al, il, ul). 

Gothic : magu-s ' boy/ magula ' little boy,' fisk-s l fish,' fiskila 
4 little fish.' 

Old Germ. : lichamo ' body,' lichamilo ' little body,' pure l a fort,' 
pur gild ' little fort.' (See D. G.) 

Modern Germ. : acht-el 'eighth part,' drmel 'sleeve,' bundel 
(bund), dumppel 'puddle' (dump/}, esel (comp. our ass), fessel 
1 fetter,' ferkel ' young pig,' gipfel ' summit,' kummel ' cummin,' 
hiigel 'hiUock,' kettel 'little chain' (kette), lummel 'lubber,' 
merg-el 'marl' (comp. mark 'marrow,' i.e. grease), nabel 'navel,' 
nagel ' nail,' nebel ' mist,' nessel ' nettle,' schenkel (comp. our shank) 
1 leg,' stachel ' stink,' stopfel ' stopper,' mert-el ' fourth part,' wiirfel 

Austrian: mann-el, weib-el, hund-el, aug-el, fuess-el, mannl, 
weibl, &c. 

Tyrolese : waibal, &e. 

English : beetle, darnel, fennel, freckle, funnel, nail, navel, needle, 
nozzle, shovel, snail (snagge in Sussex says Ray), sorrel, speckle, 
spittle, thistle, throstle (thrush), thimble (thumb), wagtail (corrup- 
tion ofwachtel), weasel (vare 'a species of weasel,' Somersetsh.), 
weevil, wennel, prov. 'a weaned calf,' &c. ; adj. little, mickle or 
muckle, evil', vbs. ramble (roam), mingle (Scotch mang), grumble, 
drawl (draw), role, vb. (Scotch for row, J.). 

But the liquid / in all languages is apt to interchange with 
its neighbour liquids, where the word neighbour is used in 

* For the presence of such a suffix in a word denoting excess, see what 
is said below of the adj. mickle. 


reference to the natural order of the liquids, namely, r, I, n, 
ng, m; i.e. the order in which the pronunciation passes from 
the back of the mouth to the front. Hence 

b. er, as : 

G-erm. : bech-er, eit-er, fehl-er, fing-er, geif-er, Tiad-er ' rag, 
wiping-clout' (comp. had-el ' a bunch of ears of corn'), hamm-er, 
jamm-er, mard-er (comp. our marten), kumm-er, maser (comp. our 
measles and the Lat. macula), schlumm-er, splitt-er, wuch-er. 
(These from Grrimm, vol. ii. p. 122, who, however, abstains from 
assigning any special meaning to the suffix.) Many others might 
be added, as : koff-er (=Lat. cqfinus, our coffin and coffer), lager 
or lager ' bed,' &c. (our lair) , mess-er (in old Frisian and, I be- 
lieve, Dutch, mes). 

English: adder, badger , leaker, leaver* (?), bladder (Germ. 
blase), clover, dodder, fetter, Jinger, fresher ' a yo\mgfrog orfrosh 1 
(Suffolk) , fritter 'a small pancake' (a full-sized one called. froize 
in Suffolk), garter {gird', comp. Scotch girten ovgairtairi), heather 
{heath), leather, otter, shoulder, splinter ; and the verbs flatter 
(from a&j.flat ; comp. palpare frompalma ' the flat hand') ; flitter 
(flit), simmer (the primitive sam or zam 'to heat for some time 
over the fire but not to boil,' a Somersetsh. word), slumber 
(Scotch sloom), quiver (quake), shiver (shake), clamber (climb), 
wander, whisper. 

c. en-\\ 

Germ. : besen ' besom or broom,' bissen ' bit,' boden ' bottom,' 

* In dealing with this suffix especial caution is necessary, as it often 
denotes an agent, often a male. Thus the bird diver clearly means * one 
that dives.' Hence the beaver may possibly have received his name from 
his habit of constructing, and so be derived from the German root bau-en, 
in which case it will be only a variety of the German bauer, our boor. Nor 
is this suggestion at variance with the indisputable fact that beaver repre- 
sents the Latin fiber, for this may be but a variety of faber, in which case 
we should be brought to the same result. Instances of er as a suffix de- 
noting the male of course abound. Thus kater, as signifying ' a torn cat,' 
has no title for admission among the German nouns in er which we have 
just quoted from Grimm. Gand-er, and perhaps the Latin anser, may also 
have a suffix possessed of this power. Still, in speaking of the two senses 
agent and male, we would not wish to oppose the doctrine that both these 
senses may result from the idea of ' man' ; kater ' the man-cat,' diver ' the 
dive-man,' formed as our chap-man. 

f The diminutives in et may very possibly, more or less of them, have 


bogen 'bow,' bolzen 'bolt, arrow,' busen 'bosom,' daumen (daum) 
'thumb,' degen ' sword' (but in form our dagger), faden 'fathom,' 
fohlen ' foal,' Jinken (Jink} 'finch,' garten 'garden,' kloben 'clue,' 
kuchen ' tart,' magen ' maw,' stern ' star,' waffen ' weapon,' zeichen 
' token.' 

Eng. : fatten (baf), garden, maiden, token, and esp. speken 'a 
small spike,' besides speke 'a spike' (Suffolk), vb. open, blacken, 
widen, hearken, listen add weapon, reckon, beckon, senon (Scotch) 
=sinetv and compressed; bairn (bear, vb.) ; burn, vb. ; burn, sb. 
(=brook), churn, corn, earn, fawn, fern, learn (Germ, lehr-en, 
our lore}, mourn, run (Somersetsh. hir-n), shun (Germ, scheu-en), 
stern (steer), tarn, turn, warn (ware, vb. obsol.). 

Less frequently. 

d. em (om), as : 

Germ. : athem ' breath,' schirm (our screen), warm 'warm,' adj. ; 
old Germ, varam ' fern.' 

Eng. : besom or broom, blossom and bloom (blow, vb.), bosom, 
bottom, fathom, Jilm (fell ' skin'), and perhaps gleam, seam, team, 
from the several verbs glow, sew, tow. Also warm. 

The convertibility of the suffix el with er and en (occa- 
sionally em or om) appears tolerably evident from the actual 
cases which occur among the examples which have been cited, 
as : fessel, fetter ; kummel, cummin ; lummel, lubber ; stoffel, 
stopper-, hader, hadel', marder, marten; koffer, coffin; degen, 
dagger; besen, besom; boden, bottom; busen, bosom; faden, 
fathom; varam (old Germ.), /mi. 

Still more striking is the evidence when three varieties of 
one word are found to coexist, as in the case of avel f beard of 
barley' (Moor's S.W.), the plural of which appears in Essex 
as ails, in Scotland and the North of England as awns, but 
generally as awms. 

We have ventured to include in the lists which have been 
given the forms little, mickle, evil, in spite of the doubt ex- 
grown out of en, seeing that the letters n and t are very commonly con- 
vertible. Thus brisket has also a provincial form briskin, and other ex- 
amples may be cited. But the same form of suffix, et, is also convertible 
with ec (eck) ; and accordingly the examples of diminutives in et will be 
found above under another head. 


pressed by Grimm (p. 687) as to the fitness of such a suffix to 
enter into a word which denotes greatness. For surely there 
can be no serious objection to softening the idea of greatness, 
as is so clearly the case in our conversational adjective largish 
for ' rather large/ Moreover, if we assign a diminutival 
power to the last syllable of little (Goth, leitil), mickle (Goth. 
mikil), evil (Goth, ubil), which last syllable is by all admitted 
not to be radical, we have at once an explanation of the fact 
that this syllable is dropped, when we form the comparatives 
and superlatives of those adjectives, for such a syllable would 
then be wholly out of place. 

If then the suffixes of these three familiar adjectives have 
been rightly classed with the diminutives, the convertibility of 
the liquids / and n in this class of words receives confirmation 
from the varieties, Swed. mycken, Dan. megen, for mickle ; Sw. 
liten, Dan. liden, for little. Nay, as I am writing, I hear a 
little gentleman (aged two) calling himself licken Liel (little 

This may perhaps be the most convenient place for a remark 
on the diminutival verbs, viz. that the idea of pettiness is in 
them accompanied by that of iteration. In the grammars of 
some languages this is broadly stated. Thus in Finnish 
(VhaePs Gr. p. 60), we are told that derivative verbs with the 
suffix el are habitually formed from simpler verbs with this 
double notion, as from lasken ' dimittere/ laskelen ' paulatim 
dimittere.' So we find in the same language (ibid. p. 66) 
hyppelen ' choreas ducere/ kdwelen ( ambulare/ But the fre- 
quentative character of diminutival verbs is tolerably apparent 
of itself in ransack, mimick, pluck, lurk, harass, worry; 
ramble, gobble ; wander, clamber ; hearken, reckon \ warm, 
gleam. Even among substantives the suffix of diminution 
often implies at the same time something collective : as gravel, 
shingle ; darnel, sorrel, clover, dodder ; fern ; charlock ; sham- 
rock ; farrow ; vraik, silt ; ivy, fry. 

Before proceeding to the suffix ing as employed to denote 
diminutives, we must recall attention to the valuable paper 
with which the fourth volume of our ' Proceedings' opens. 
It will be remembered that Mr. Kemble there explains over 


three hundred geographical names in this island as formed by 
attaching the suffix ing to the name of a former owner of the 
property, and he contends that the principle may in fact be 
extended to the explanation of more than thirteen hundred 
such names. Thus his detailed list contains towns, &c. which 
seem to have for the first element such abbreviated Christian 
names as Ben, Bill, Bob, and such surnames as Agg, Babb, 
Beard, Buck, Budd, Broad, Brett, Bright, Brown, Bunn, 
Bunt, Burt, Butt, Bird, Burr, Case, Cole, Dill, Dodd, Dunn, 
Hall, Home, Horn, Mann, Munn, Part, Peat, Pott, Read, Rust, 
Todd, Wase, Ware, White, Wren, forms still more or less 
familiar in the pages of a modern Directory. In the examples 
to which we are now referring, the names have for their final 
element the Saxon representative of what we commonly write 
as borough, burn, den, fold, ford, ham, hanger, hurst, land, ley, 
mead, meer, moor, stoke, street, ton, wick. Thus the theory 
supported by Mr. Kemble, that the syllable ing is substantially 
a genitival suffix, gives a most intelligible interpretation of a 
vast number of the geographical terms distributed over the 
maps of England. But it may perhaps be objected that the 
theory is an over-bold one which supposes large towns such as 
Warrington, Buckingham, Nottingham, Huntingdon, Chippen- 
ham, Twickenham, Farnham, to have been the property of 
mere individuals. The answer is simple. As Christian names 
are given to individuals when they are infants, so what may 
be now a large city must have begun with being a solitary 
house, and of such solitary house the owner may well have 
been plain Mr. Warre, Mr. Buck, Mr. Nott, Mr. Hunt, Mr. 
Copp, Mr. Tooke, or Mr. Farr. But the argument is con- 
firmed by the fact that our English word town (-ton) and the 
French ville, though now applied to large aggregates of build- 
ings, had for their first meaning ' a farm-house/ Such was 
certainly the meaning of the Latin villa, and is the pre- 
sent use of the word town in Scotland. Again, it not unfre- 
quently happens, as Mr. Kemble has pointed out, that the 
geographical name terminates with the syllable ing, with- 
out the addition of any of the substantives just enume- 
rated ; as, for example, Worthing, Lancing, Reading, Tarring, 


Poling. This also admits of satisfactory explanation on the 
same theory, for nothing is more common than to speak of 
' Mr. Smith's/ meaning his house ; and thus the five names 
just enumerated tell us that early proprietors of the farms or 
houses, out of which they grew, were Mr. Worth, Mr. Lance, 
Mr. Read, Mr. Tarr, Mr. Pole. 

It is felt, however, by some to be a difficulty in Mr. Kemble's 
theory that the Anglo-Saxon genitives of masculine nouns 
commonly end in es, not in en or ing. To me, this, so far 
from being a difficulty, is an advantage on which I greatly 
rely, for I have long contended for the ready convertibility 
of the letters n and s (see our own ' Proceedings/ vol. iii. 
pp. 50 and 51), and especially in the suffix of the genitive 
case, ibid. p. 55, where I have instanced the words mensch-en- 
alter, hasen-lager, monden-licht, as containing a genitival 
suffix in the first portion, as appears also in our own Frier-n 
Barnett compared with Abbot's Langley, Leamington Prior's. 
So also, as I stated in the same page, mine and thine are really 
genitives ; and I might have added that the vulgarisms hisn 
and hern would never have established themselves but from a 
consciousness that the n was well qualified to perform the 
office of a genitival suffix. 

But I find evidence in favour of the claim which the geni- 
tive case has upon the liquid n in the Anglo-Saxon declensions 
themselves. I do not here refer to the declension which Rask 
in his Grammar has honoured with the first place, for the 
syllable an, in one, if not more of the words collected under 
that head, really forms part of the crude form, viz. naman (Lat. 
nomen) ' name/ The error in the view propounded by Rask 
is precisely the same with that of the Latin grammarians, 
who fail to see that the nominatives virgo, homo, ratio have 
lost a final n, and so are unable to explain the forms virgun- 
cula, homun-culus, ratiun-cula. It is on the plural genitives 
of Rask's third declension that I rely : sunen-a ' filiorum/ 
gifen-a ' donorum.' 

Nay, even Sanscrit scholars have been so blind to the con- 
nexion of this liquid with the genitive, that with abundant 
genitives plural in n-am before them, they have yet persisted 


in appealing to the doctrine of epenthesis in order to account 
for the appearance of what they regard as an intruder. 

Much in the same erroneous way it has been proposed at 
times to regard the r of musarum as an intrusive letter, whose 
sole office it is to prevent a disagreeable hiatus between the 
vowels a and u. Such a theory has been perhaps founded on 
the consideration of the Greek fnova-aayv. But here, as in 
many other cases, the Latin has the advantage of the Greek. 
The older form of the genitive plural no doubt was something 
like musas-um, musas being a singular genitive like familias 
(in paterfamilias), and urn the symbol of plurality ; for it is a 
common habit of language to form the plural cases of a noun 
by adding to the corresponding cases of the singular some 
element to denote plurality. But such a form would of course 
be modified according to the genius of each language. Thus 
the Latin, as usual, converts the s to an r, and the Greek, as 
usual, omits it altogether. 

Moreover, in a question of this nature it is scarcely philo- 
sophic in Anglo-Saxon scholars, because the existing books of 
that language limit the n in the genitive to feminine nouns, 
to be stopped by this fact. Declension in its original forms 
must have been totally independent of gender, as completely 
so as prepositions are. 

It is the more necessary to deal with this unphilosophical 
doctrine, as Grimm, in his chapter on diminutives, has re- 
peatedly urged it. Thus, while he admits the suffix chen, or 
rather ichen, and others, to consist of two elements, he speaks 
of the n as something epenthetic or ' shoved in/ contending 
that it was first adopted in the oblique cases as an aid to de- 
clination, and then erroneously extended to the nominative. 
See his remarks on the suffix ilin (pp. 667, 668), elm 
(pp. 670-672), tin or len (p. 674), kin or chen (p. 678), in 
(pp. 683, 684). 

But there has been a difference of opinion among the 
members of our Society whether the suffix ing, as seen in 
such forms as Reading, Buckingham, may not constitute an 
adjective (' Proceedings/ iv. 83) . To any such theory I would 
oppose a counter-theory, that our so-called adjectives in en 

T 2 


are themselves at bottom genitives. Thus, flaxen, golden, 
wooden, may well have been converted from genitives to 
adjectives, having originally signified ' of flax/ ' of gold/ ' of 
wood/ The metamorphosis of a substantive into an adjec- 
tive is well seen in the Latin gen. cujus, subjected to the 
indignity of declension, as cujus, cuja, cujum. I have assumed 
in this argument that the suffixes ing and en may be regarded 
as substantially one. Thus I hold it to be of no material mo- 
ment, that in one county of England we write Buckingham, 
in another Buckenham. Mr. Kemble too has pointed out 
that Surrenden in Kent represents the Anglo-Saxon Swffirse- 
dingden. The fact is, that such a secondary syllable so 
placed is unlikely to obtain a distinct utterance ; and hence 
it is that we find it often still farther reduced, and some- 
times wholly absorbed. Thus Aspenden in Hertfordshire is 
habitually pronounced Aspeden, and what was Oxenford is 
now Oxford. It is precisely in this way that the Anglo- 
Saxon Sunnand&g and old Scotch Sonounday have been com- 
pressed to Sunday. It is thus again that iron and cotton are 
used as adjectives (orig. genitives) in such phrases as iron- 
railing, cotton-gown ; and even leathern will soon be super- 
seded by leather used in the same sense. That Sunday, 
Monday must have once contained a genitive in the first 
element is supported not merely by the corresponding Latin 
forms Solis dies, Lunae dies, but also by the allied forms 
Wednesday, Thursday. Similarly the genitival origin of 
Buckingham, Chippenham, &c. receives confirmation from the 
accompanying forms, such as Broxburn (Brock), Hoddesden 
(Hood), Wadesmill (Wade). 

But the ordinary use of the suffix ing is to form patrony- 
mics. True ; and does not this fact confirm the theory that 
its original power was to form genitives ? In kypoaOevris o 
A?7//,oo-#ej/ou9 the last word performs the office of patronymic. 
Again, in Wales nearly every surname is well known to have 
been at first a patronymic ; yet these are but genitives, as Davis 
(David's), Edwards, Evans, Harris (Harry), Hughes (Hugh), 
Jones (John), Richards, Roberts, Toms, Watts, Wills, Williams. 
Of course when the Christian name already ended in s, it was 


useless to add another, so that Christian and surname are 
blended in Charles, James, Thomas*. It may be useful to 
observe that the Romans also had their genitives used as 
patronymics. Thus, while Marcus, Quintus, Sextus, Decimus, 
Tullus, Atta, correspond to our Christian names, as belonging 
to individual members of a family, so Marcius, Quintius, 
Sextius, Septimius, Octavius, Nonius, Decimius, Tullius, At tins 
came into use as surnames, and originally no doubt as patro- 
nymics ; and indeed they copy cujus in the habit of declen- 
sion, as Tullia Lex, Octavia Porticus. Again, we may as well 
point to the patronymics, which are plainly mere genitives, 
now current as surnames in Germany, Ernesti, Jacobi, Mat- 
thiae, Pauli, &c. 

But if ing, originally a genitival suffix, be well adapted for 
the formation of patronymics, the passage from a patronymic 
or child to a diminutive is easy. Thus when Jamieson argued 
for the identity of the suffix kin in lambkin with the word kin 
or kind f a child/ he put forward a doctrine which had in 
it much that was plausible, although the preponderance of 
evidence must decide us to reject his theory. 

But it is time that we pass from discussion about the form 
and origin of ing to instances of its occurrence in diminutives ; 
and first as a solitary suffix unaided by the suffix el. 

c. -ing ; as Grerm.: ferd-ing, ' farthing ' ; hdring, ' herring'; lem- 
ming (the Mus lemming, Linn.) ; niding (or nidgei), ' a base fellow' ; 
together with some fifty words where this suffix is preceded by el 
('/), so as to constitute the syllable ling. 

Eng.: whiting, bunting, herring, gelding, farthing, lording, riding 
(division of Yorkshire), tithing, shilling (A.S. scill, stilling), morn- 
ing, evening, sweeting. 

The evident derivation of farthing and tithing from fourth 
and tithe or tenth, goes far to prove that riding is but a cor- 
ruption of thrid-ing, from third, the th having been absorbed in 
the prefixed words north, east, and west. Indeed northriding 
should be divided into nor-thriding , for nor is the simpler 
word, as seen or heard in Nor-way, Nor-man, Nor-folk, Nor- 
west, &c., and nor-th is but a derivative from it. 
* See also above the series Dawkins, Edkins. 


Perhaps we ought here to introduce a suffix en, as growing 
out of, or rather only another form, and perhaps an older form 
of ing. Thus farthing with the vulgar is often farden. But 
we have already given the suffix, as being a variety of el. See 
examples under that head. 

We next proceed to the consideration of compound dimi- 
nutival suffixes, in which two or even more elements make a 
contribution to the idea of smallness : 

D. Jcin=&cTc-{-in, Eng. ; chen, German =ick+en. 

This is justly regarded by Grimm as made up of a guttural and 
a nasal element, such as we have above classed under the heads 
ock and en. But the full form is perhaps best seen in the Gaelic. 
The simple diminutival suffixes in this language are acJi, ag, and an, 
which are often united in the uncorrupted form ach-an or ag-an. 
Thus we find from 
fiata, a cudgel or bat .... latacTian, a little staff. 

bata, a boat batachan, a little boat. 

be-ach, a bee beachan, a little bee. 

cuacJi, a cup cuachan, a little cup. 

curacJi, a wicker boat .... curacJian, a little coracle. 

duine, a man duineachan, a mannikin. 

eun, a bird eunachan, a little bird. 

qoblaq. a small fork, or ) 777 

\ goblachan, an earwig. 
goblach, adj. forked ) y 

gunna, a gun gunnachan, a little gun. 

leum, to leap leumachan, a frog. 

meur, a finger meuragan, a thimble. 

pbca, bag, pocket pocaeJian, a little pocket. 

pits, a cat pusachcm, a whining boy. 

ron, a seal, sea-calf ronachan, (a fellow like) a sea-calf. 

sg&il, shade sgaileagan*, a fan or umbrella. 

sgall, baldness sgallaclian*, a bald-headed person. 

Sffuab, a besom sguabachan*, a little besom. 

teine, fire teineachan, a small fire. 

van, a lamb uanachan, a little lamb. 

It will be here seen that the suffixes ach and an are clearly 

* Intermediate forms are sgaleag, a little shade, veil, parasol, umbrella ; 
sgallach, bald; sguabag, a little besom. Comp. with sguab, the English 
swab, and Latin scopae. 


independent of each other, for example, in the instances of 
beach, cuach, and curach, where the simple roots are not to 
be found in the language. So also from hoc, a buck, loch, a 
lake, cnap, a knob, are deduced the diminutives bochan, lochan, 
cnapan ; and again there occur caile, a (coarse) girl, caileag, 
a little girl, lassie ; cam, crooked, camag, a curl, a crook, Sec. ; 
gobhal, a fork, gobhlag, a little fork. Again, it happens at 
times that the suffix an precedes the other. Thus beside 
leum-ach-an, a frog, there 'exists leum-n-ach, of the same 

In Scotch and English too, the forms in ock and ick often 
have an existence independent of the forms in kin. Examples 
of the former we have already seen. 

The following table of words in ikin or kin* is made up 
from Mr. Lewis's paper, and from those which occur in 
Jamieson, with a few additions from other sources. 

1. auchtikin, J. 

2. bodkin, L. 

3. bootikin, L. 

4. brakkins, J. 

5. brotikin, L. 

6. bulkin, M. 

7. bumpkin, L. 

8. buskin, L. 

9. cannikin, J. 

10. catkin, L. 

11. ciderkin. 

12. cutikins, L. 

13. finikin, L. 

14. firkin, L. 

15. flichen, J. 

16. flindrikin, J. 

17. girkin, L. 

18. grimalkin, L. 


19. griskin, L. 

20. hudkin, M. 

21. hulken, M. 

22. jerkin, L. a. 

23. jerkin, L. b. 

24. kilderkin, L. 

25. kinken, J. 

26. lakin, L. 

27. lambkin, L. 

28. malkin or maul- 

kin, L. 

29. mannikin, Tit. 

30. memerkyn, J. 

31. minikin, L. 

32. muskin. 

33. mutchkin, J. 

34. nadkin, J. 

35. napkin, L. 

36. nipprikin, J. 

37. pannikin, M. 

38. pipkin, L. 

39. prettikin, J. 

40. pumpkin, L. 

41. roddikin, J. 

42. siskin, L. 

43. slammikin, L. 

44. slibrikin, J. 

45. smirikin, J. 

46. smootrikin, J. 

47. smulachin, J. 

48. spillikin, L. 

49. thumbikin, L, 

50. toopikin, J. 

51. weerikins, J. 

52. whinkens, J. 

* Such forms as wyfockie, lassickie, are but corruptions of wyfockin, 
lassickin; so that they call for no special consideration. Still the full 
forms in kin seem entitled to a place in our list. 


II. PROPER NAMES : properly Christian names, but sometimes 
Surnames, especially with an added s. 













Malkin i 


Peterkin | 

Ed (ward). 

or Perkin j 











Larry f. 




from Mai or Mol. 



Simeon . 





The following remarks will be easily referred to the items 
of the first Table : 

1. An eighth part of a barrel or half-firkin, from audit, ' eight/ 

2. Probably from our verb bore, the Latin analogue of which 
appears as for are, fodere, fossa. Bodkin, in Shakspere's Hamlet, 
is a dagger, a word which, as well as the German degen, comes 
from the verb dig. 

3. The instrument of torture. 

4. The remains of a feast, the fragments. 

5. = Fr. brodequin. 

6. A young bull, Moor's ' Suffolk Words/ under ' pannikin.' 

7. From boom, a tree ; comp. blockhead. 

8. Perhaps a variety of bootikin. 11. Poor cider, Johnson. 

13. Splatterdashes, from cute, ankle. 

14. A fourth part of a barrel, from/bwr, weakened by the umlaut, 

15. An atom, akin to the German ./fee/i;. 

16. One who flitters about. 

17. From a sb., such as the Germ. gurke or Eng. gourd. 

18. i. e. gris malkin ; see ' malkin,' below. 

19. From grice or gris, a pig. But this is scarcely satisfactory. 
Is it connected with gristle ? 

20. A little hood or hat for the finger, a finger-stall. 

21. A piece of skin, Moor's ' Suffolk Words,' perhaps from G. 
hull, a husk or shell. 

22. A jacket, connected with the Dutch jurk, a frock. 

* Or perhaps John. 
1 Or Lewis. 

f i. e. Lawrence. 

or Simon, or possibly Samuel. 


23. A hawk ; comp. geier, a hawk or vulture, Germ., and our 

24. From kilder=vas. 

25. A small barrel ; perhaps the same as cannikin. 

26. ' By'r lakin,' in Shakspere, i. e. ' by our ladikin,' meaning 
the Virgin Mary. 

28. Of many senses, as : a. little Mai or Mol, i. e. Mary ; b. a 
dirty maid-servant ; c . a mop or clout for cleaning ovens ; d. a 
name for ' pussy,' as Tom with us for a male cat, hence grimalkin, 
first a cat, then a hobgoblin ; e. a hare, in Scotland. The doctrine 
of Hanmer, as quoted by Mr. Lewis, that ' a mop made of clouts 
for sweeping ovens,' was the earlier meaning, and that of ' a dirty 
wench ' derived from it, is upset by the recollection of other 
instances where we apply the name of a servant &c. to what per- 
forms the office of a servant &c., as a dolly, i. e. a washing ma- 
chine, a jack or bottle-jack, a footman, a dumb-waiter, a housewife 
i. e. needle-book. 

30. See Jamieson. 31. Very little, esp. a little pin. 

32. A tit-mouse. 

33. Compared by J. with a Swed. maatt, a pint. 

34. A foul smell, says J., perhaps a nosegay ironically, and so 
from nose. 

35. From Fr. nappe ; but this must have been thoroughly na- 
turalized before taking the Saxon suffix. 

36. A small bit, from nip, the same. 

37. A little pan for warming pap, Moor's * Suffolk "Words/ 

38. TPrompipe. 39. A little trick, fromprattik, a trick. 

40. From a sb. = the Germ, pompe, gourd. 

41. The fourth stomach of a ruminating animal, from reid, the 
same. 42. A bird, so called, perhaps = Germ, susschen. 

43. A drab or slovenly woman, akin to the Germ, schlamm, dirt, 
and our slime ; in Jennings' Glossary spelled slomaking. 

44. 'Slibbrikrn mouse,' a fondling term, probably meaning 
' sleek,' and akin to our slippery and slip, as also to the Latin lu- 
Iricus and labi. 

45. Or smuracliin, a stolen kiss, qu. from the Scotch vb. smure, 
smother or suppress. 

46. ' Smootrikin mouse,' a fondling term. See Jamieson. 

47. Puny, akin to the Gael. smeileacJi, pale, smeilean, a puny or 
pale creature. 

48. From spill, a splinter. 49. Instrument of torture. 


50. A pinnacle, from top. 51. Or whirkins, posteriors ?, J. 
52. Flummery. 

I have not thought it desirable to give any list of German 
nouns in chen, as the language swarms with them. 

E. ling = el + ing. 

In German the number is too great for enumeration here. 

Engl. : bantling, a child in swaddling lands, changeling, chitter- 
lings (Germ. Jcuttel, the same), darling (dear), duckling, dumpling 
(dought}, easterling, 'of the east country/ Jirstling (of a flock), 
fondling, foundling, gosling, grayling, hireling, inkling*, kitling 
(prov. for 'kitten'), lordling, nestling, nurseling, overling (pferlyng 
in the ballad of 'Richard of Almaigne,' Percy, voLu.),popeling, 
'Roman Catholic,' porkling, sanderling, a bird frequenting sea- 
sands, sapling, scantling, seedling, starling (= stare), sterling, a 
little coin marked with a star, stripling, suckling, underling, west- 
ling, 'of the west country,' witling, yearling^. 

r. let = el + et, 

armlet, (bracelet), circlet, (corslet), croslet, eyelet, gimblet, ham- 
let, (islet), martlet, pikelet, ringlet, rivulet, rootlet, runlet, springlet, 
streamlet, tartlet^. 

QC. rel = er -f- el, 

cockerel, mackerel, pickerel. 

Some of the words belonging to the class -let have been 
placed within brackets as of French origin. Yet islet may be 
only a blunder for ey-let } as island is for ey-tand. Mackerel 
(macquereau) is probably a northern rather than a native French 
word ; but how has rivulet found its way into our language ? 

These double diminutives are the more entitled to separate 
consideration, because it is probable that in many cases they 
are not built up by successive additions of single suffixes, but 
formed at once from the root-syllable by the addition of what 
was regarded by the originator of the word as a simple suffix, 
though in truth a double one. Thus it might be permitted 
even now to form new diminutives in ling and let, but scarcely 
so in ing and et. 

* See my Paper on inkling, in Philolog. Soe. Trans. 1857. 
t This collection is almost wholly from Mr. Lewis's paper above re- 
ferred to. 

t Mr. Herbert Coleridge's Paper on -let is in the Society's Trans, for 1857. 


But it is not merely a double expression that satisfies the 
love of diminutives. In Aberdeenshire, where by the way the 
power of inventing new diminutives at pleasure is so tho- 
roughly a living principle, that you may hear talk of a 
caterpillarie, the love of accumulating diminutive upon dimi- 
nutive leads to such combinations as e sic a bonnie little wee 
bit lassickie.' Here, if our fingers are not guilty of mis- 
counting, the idea of smallness is expressed seven times ; and 
it is not for a Scotchman to laugh at sacc-l-in-ch-in* (Grimm, 
D. G. iii. p. 682), ' a wee wee wee wee sack' ; or at 

es-el-in-ch-il-in (ibid. p. 681), a donkey whose smallness needs 
the fifth power to express it. 

A correct estimate of the syllables employed to express 
diminution must be of material value to the etymologist. In 
the first place it will guard him in many cases from being 
misled by erroneous derivations and even erroneous spelling. 
Had Johnson compared shallow with the many adjectives of 
like suffix, callow, hollow, yellow, &c., he would not have 
committed the error of explaining it as a compound of shoal 
and low ; rullocks or rowlocks would not have been divided, 
as it often is, so as to make the second element locks ; and 
a false, though tempting etymology would not have corrupted 
wacht-el into wagtail. 

The word mammet, defined in Todd's Johnson as ' a puppet, 
a figure dressed up,' has been amusingly dealt with in the way 
of etymology. Dr. Johnson's own . derivation, from mamma, 
is sufficiently unsatisfactory, but not so glaringly absurd as 
the doctrine that it is abbreviated from Mahomet, for no 
form of religion was ever more free than Mohammedanism 
from any trace of idol- worship. The right course would 
have been to connect the word with mammock, as gobbet 
with gabbock, emmet with emmock, gimlet with gemlich. In- 
deed Jennings gives us ' mommet or mommick,' a Somer- 
setshire term for ' a scarecrow, something dressed up in 
clothes to personate a human figure.' Again, in one of the 

* It should be observed that Grimm treats the n whenever it enters into 
those words, as something intrusive. But this is a point which has already 
been considered. 


passages quoted in the above-mentioned lexicon, the word 
mommets has the qualifying words, ' consisting of raggs and 
clowts compact together/ This brings us very near to the 
ordinary meaning of the noun mammock, which in the same 
work is denned ' a shapeless piece/ while the vb. mammock is 
1 to tear, to break, to pull to pieces/ Mommacks in Jennings, 
and mammocks in Cocker are explained as c pieces, fragments/ 
while Milton speaks of ' scraps and mammocks/ With these 
facts before us, may we not start from the verb maim*, from 
which the Scotch have deduced a diminutival verb mank ' to 
maim, to wound/ as well as an adjective mank ' deficient' 
(identical no doubt with the Latin manco-) ? Thence by the 
addition of a second diminutival suffix comes mangle f to 
tear/ Again, starting from the primitive maim we have a 
legitimately formed noun in mammock, mommack, &c. ; and 
a scarecrow is little more than a bundle of rags arranged to 
imitate a human being. The main objection to this expla- 
nation is that a scarecrow ought then to be represented by a 
plural. But in fact the bundle of rags so collected forms a 
new unit, and when such is the case, the symbol of plurality 
is soon discarded. Thus, for example, bigae (properly bijugae 
sc. equae] soon gave way to a singular big a ' a chariot' ; and 
while Terence, Varro, Virgil and Columella wrote rastra 
or rastri ' a rake/ in allusion to its many teeth, the later 
writers, as Seneca and Pliny, employed the singular. 

Again, the fact that in different provinces different suffixes 
of diminution are attached to the same root-syllable, may be 
turned to account in etymology. Thus, what we call sorrel 
is to a Scotchman sourock; and a comparison of the two 
forms leads us undoubtingly to regard sor or sour as the 
root-syllable, in which we readily detect the familiar adjective 
1 sour/ so well adapted to characterize the plant ; and if the 
belief needed confirmation, it would be found in the Somer- 
setshire sour-dock (A.S. scearp-docce] , a literal translation of 
the botanical name f E-umex acetosus/ 

It is especially instructive to place beside each other those 

* Even maim is probably a diminutive of mow (A.S. maw-an) ; at any 
rate its fuller form was mayhem. 


diminutives which to a common root attach sometimes the 
suffix ock, sometimes en, as tarrock and tern*, brook and 
burn, brake and fern, morrow and morn, glach (Gael.) and 
glen. As regards the last of these we find the simple root in 
gill, a word to be found not merely, as some have said, in the 
northern parts of England, but equally in our southern 
counties, as, for example, in the ' DeviPs Gill' on the borders 
of Surrey and Sussex. 

And here it may be useful to say a few words on the general 
principle which was assumed in treating those monosyllabic 
words which begin or end with two consonants, I mean the 
doctrine that such words are generally compressed from di- 
syllabic forms. Independently of the evidence found in in- 
dividual cases, it seems a fair inference from the fact that most 
nations find it impossible to pronounce such words otherwise 
than as of two syllables, though written as one ; and indeed, 
even an Englishman, if he carefully examine his own utter- 
ance, will find some slight vowel-sound intervening between 
the two consonants. Thus Villikins (setting aside the initial 
consonant) is a more legitimate form than the favoured Wil- 
kins. Again, it is a well-known fact that missionaries, who 
have to deal with untutored barbarians, often find it essential 
to treat the proper names of the New Testament in this way, 
so, for example, as to write Ecarisito, i.e. five syllables in lieu 
of one. But it may be objected to what is now said, that the 
suggestion of disyllabic forms was carried even beyond these 
limits, for example, in the instances of lock-f, rack, nook. 
The principle which led me to include these words with the 
rest was the belief that in a large number of cases an initial 
liquid has lost a preceding consonant. This is almost demon- 
strably the case as regards an initial r in the Latin language. 
So again the Latin words natus, nosco, nitor are known 
historically to have been originally written in the forms 

* The words may still be identical though applied to birds somewhat 
different. Thus wachtel is e a quail' to a German, a ' water- wagtail' to an 
Englishman. Names in the early stages of society are applied with a 
latitude which is shocking to modern science. 

t Of course ck is to the ear but one consonant. 


gnatus, gnosco, gnitor ; and we may make a similar assump- 
tion for nodus when we see the English knot. For the /, 
although here also the Latin language would lend me support, 
I will confine my remarks to the very word lock as applied to 
wool or hair. This is but the word floccus of Latin, and is 
identical with the English flock (a flock-bed), fluff, flue, and 
flake (of snow). The word flake as used in the phrase ' it 
came off in flakes/ is of course a very different word from the 
flake (of snow) . The idea is now something flat, as it is also 
in fluke or flue (of an anchor), fluke (the worm), flook (any 
flat fish in Scotland), floe (ice-field). 

My next example shall start from a root-syllable, and 
proceed thence to various developments, in which my mat- 
ter is chiefly collected from Jamieson. The Scotch gab, 
Gael, gob, as signifying 'the beak of a bird/ became a 
contemptuous term for the human mouth in talking, as 
in our phrase f the gift of the gab,' 'hold your gab' and 
the derived terms gabble, jabber, gibberish. But the mouth 
has a still more important office. Thus the dim. gabbie or 
gebbie is a Scotch name for the crop of a fowl, where the 
change of meaning is paralleled by that of stomach, which, 
as its Greek form tells us, must originally have meant ' the 
mouth' ; and indeed the Latin use of the word stomachus for 
f the oesophagus/ exhibits the meaning in its transitional 
state. Gabbock or gobbet is ' a mouthful or morsel/ Here 
again we have a change of meaning from that which contains 
to that which is contained ; but this is only what is seen in 
the phrase ' a glass of wine/ meaning a glassful. Further, we 
find gab-stick or gob-stick for ' a wooden spoon/ where the 
notion of wooden is little out of place, seeing that the very 
word spoon in its Icelandic form spann means ' a sliver of 
wood/ The diminutive gebbie might well be reduced to gib, 
as Jemmie to Jem or Jim. Now gib (g hard), says Jamieson, 
is the name for ' the beak' of a male salmon ; and gib (with 
the sibilant g) is by ourselves applied apparently to the mouth 
in the expression ' the cut of his gib ' ; or if this be really a 
simile from the form of a ship's head, we have still the same 
idea, as witness the Latin rostrum, at once ' the beak of a 


bird' and ' the beak of a ship/ But the beak of a bird, con- 
sidered in the habit of pecking, suggests in many languages a 
word to denote any of the pecking or picking instruments. 
Thus the Latin upupa signified first the bird hoopoe and then 
a pickaxe, so that Tindarus, in the ' Captivi ' of Plautus, con- 
soles himself under his troubles with a play on the two senses, 
and for once we are enabled to preserve the joke in the 
English translation by the twofold meaning of our term crow. 
Similarly the Scotch appear to have obtained from this root 
gibble, ' a tool/ giblet, ' any small iron tool/ gemlick or gemblet, 
' a gimblet/ But one prong being for most purposes insuf- 
ficient, two were commonly adopted, and thus we get what 
an Englishman now calls f a fork/ but formerly called 'a 
gable ; (still preserved in the expression gable-end, of a house). 
Here the Gaelic goes with us in the form gobhal, ' a fork/ 
Add yet another diminutival suffix and we have the Scotch 
gavelock, ' an iron crow or lever/ while the Gaelic has, what is 
virtually identical, gobhlag, ' a small fork ' (but also a hay- or 
dung-fork) . This trisyllabic gavelock slips, by an easy pro- 
cess, into gellock or gulock, ' an iron crow-bar/ Such a crow- 
bar, as Jamieson observes, often ends in two teeth, useful 
perhaps for prising the lid of a box where a nail presents 
itself. Be this as it may, the idea of a fork is clearly seen 
when gellock denotes ' an earwig/ a word which must not be 
confounded with golach, ' a beetle/ The Gaelic still adds a 
third suffix of smallness, so that we have goblach-an, ' an ear- 
wig/ or ' a person sitting astride on horseback/ which reminds 
me that the first diminutive gobhal, like our own term ' fork/ 
is used to signify the ' regio perinaei/ The same triple dimi- 
nutive is also used in Gaelic in the phrase goblachan-gaoithe, 
' the swallow ' (with its forked tail) . Jamieson has also noticed 
the connexion between the Scotch gavelock and the A.S. 
gafeloc, ' a spear/ as well as the French javelot and Eng. 
javelin. If the idea of ( a fork ' enters into these words, it 
must be in the inverted form, the spear ending of course in 
one spike, but having two barbs turned the other way, by 
which a wound becomes much more serious. But probably 
the more correct view is to connect these with the original 
meaning, so that one spike will be enough. 


I will conclude this part of the subject with the remark, 
that in the examination of English diminutives, as in other 
inquiries connected with this language, there is abundant 
evidence that our Saxon, or to use a more correct, because a 
more general term, our Teutonic ancestors, brought with 
them, not one, but many dialects, so that the English lan- 
guage, whether it be a misfortune, or perhaps the reverse, can 
put forward but a poor claim to homogeneity. 

P.S. Where so many words have been considered or sug- 
gested for consideration, no doubt there are errors ; but these, 
unless they be numerous, will not affect the general con- 
clusions. Moreover, what strikes a reader at first sight as 
erroneous, may perhaps be regarded in a different view, when 
due allowance is made for the following considerations. A 
diminutival word may have been well constructed in reference 
to its original use, and yet subsequently applied where the 
suffix is no longer appropriate. Thus the Greek name for ' a 
sparrow ' was eventually made to include ' the ostrich/ and 
that for ( a lizard ' was afterwards employed for ' the crocodile/ 
Thus too the words circle and orbit possess, each of them, one 
if not two suffixes of diminution, yet we are allowed to talk of 
' great-circle sailing/ and ' the earth's orbit round the sun/ 
Again, the terms great and small are after all but relative, so 
that the one or other term may be appropriate according to 
the point of view, or, to express the idea more suitably, ac- 
cording to the scale by which we measure. Thus Gulliver 
was a giant among Lilliputians, a dwarf at Brobdignag. 
What can be more startling than to find among the alleged 
diminutives in ock, the word pellock, ' a porpoise ' ; for the 
porpoise, both in our own and other languages, is a favourite 
simile for a corpulent person. Yet when we look at the great 
bulk of the words which share the termination, we have irre- 
sistible evidence as to the power of the syllable ock. But the 
difficulty, first raised, disappears, as soon as we are told that 
Scotch fishermen look upon porpoises as young whales. Here 
we happen to have an historical explanation of the difficulty. 
We cannot always hope to be so fortunate. 


ESQ., of Christiania, Norway. 

[Read December the 20^.] 

That the Straits of Behring and the Aleutian Islands con- 
nect the continental masses of the Old and the New World 
by two natural bridges, easily crossed even by the rudest of 
savages, is a geographical probability, the historical importance 
of which does npt depend merely on conclusions derived from 
an inspection of the map of the globe. It is an authenticated 
fact, that the Russians, in the beginning of the eighteenth 
century, were told, by the native Siberians, of the great con- 
tinent lying to the east of Kamschatka, long before any 
European voyage of discovery had demonstrated, that America, 
by its configuration, approached so closely to Asia, that the 
passage from one continent to the other was nothing more than 
a drive in a sledge over the ice, or a coasting voyage from 
island to island, performable by even the rudest canoe. Fur- 
ther to the south, the crossing of the Pacific is so well assisted 
by regular oceanic currents from W. to E., as to have been 
performed in several instances by Japanese fishermen, who 
have been carried safely over in small boats, almost without 
provisions, from their own country to California. 

There is then an extreme probability that, in the course of 
ages, the Pacific ocean has been repeatedly crossed by indi- 
viduals from the opposite shore of Asia. But it is also pro- 
bable, from the habits of the Asiatics, averse to commercial 
navigation, that all these voyages were performed by only a 
few persons in each instance ; more often, perhaps, against 
their own will. It is as well-settled a fact that no Asiatic 
immigration into America has ever been strong enough to 
found, either by settlement or by conquest, an Asiatic com- 
munity, or a state with properly Asiatic manners and polity, 
in the New World. The social state of the American Indians 
at the time of Columbus was so far original, that the more 
recent settlers from Asia, if such there have been, can only 



have mixed as absorbable and subordinate elements with the 
population already established in the New World. If these 
aborigines also be derived from Asia, there is an increased 
probability that their settlement, occurring during a period 
of still more imperfect navigation, and composed of the 
hunting or fishing tribes on the eastern coasts of Siberia, 
must have consisted merely in the drifting over of single 
families, just fit to form the nucleus of roaming bands of 
hunters or fishermen, such as indeed occupied the whole 
northern portion of America at the time of the discovery. 

In the defect of credible traditions of these facts, the only 
available clue to the elucidation of the past, is the comparison 
of those languages of America and Asia that are found in the 
closest geographical contact. From the very beginning of 
ethnological research, this way of investigation has indeed 
been pursued, and it very soon led to the recognition of 
the identity of one American nation with one Asiatic the 
Eskimo with the Tshuktshi. The further development of this 
discovery was however arrested, not merely by the endless 
variety of American tribes and their dialects which rather 
proved too much, viz. that the American Indians are as diver- 
sified mutually as they are when compared with the inhabitants 
of Asia but by the want of complete and scientific data about 
any of the languages, either of America or of Siberia, that 
might be used as starting-points for the comparison. 

This defect has, however, in our days of sudden accumu- 
lation of facts, been in a great measure remedied. The 
labours of Mr. Castren on the languages of the widely-spread 
Samoyed nation, have connected it with all the well-known 
and most important branches of the Finnic and Altaic stock 
in the Old World a conclusion that may be considered 
indisputable, after having been advanced by this profound 
scholar, himself belonging to one of the principal Finnic 
nations. The Samoyed language, as now made known, will 
further be found to offer striking resemblances with all the 
other dialects of Siberia that were considered unconnected 
by the author of the Asia Pofyglotta, viz. the Yeniseyan, the 
Koriak, the Yukagir, and the Kamtshadal. Thus we are able 


to recognize one continuous chain of kindred tribes along the 
whole length of the Arctic coasts in the Old World, from the 
North Cape on the Atlantic to the East Cape on Behring's 

In the mean time the indefatigable exertions of American 
linguists have multiplied the stores of information about the 
aborigines of their own continent. Among these labours 
the work of Mr. Biggs on the Dakota Language stands pre- 
eminent for fulness of information and general literary merit. 
It may be presumed that such excellent productions will 
encourage other individuals in as favourable circumstances 
for observation as Castren and Biggs, to publish similar ex- 
positions of languages likely to serve as closer links between 
Asia and America, although philology can never expect to 
engage in her service more profound thinkers and more ardent 
votaries than those men. When I became acquainted with 
these latest advances of linguistic knowledge, I felt convinced 
that they contained facts sufficient to admit of a close com- 
parison being made between the languages of Northern Asia 
and America; and I thought that such an attempt might 
not be without utility, as a practical proof of the amount of 
evidence that might be gathered from the present stores, for 
forming conclusions on very important points of the history 
of our species. The result of this comparison between the 
rude languages of these tribes of savages shows a series of 
resemblances analogous to, but yet different from, those that 
have been found to obtain among the celebrated languages of 
the great civilized nations. This similitude and discrepancy 
seems to stand in an exact proportion to the wide gulph that 
separates the few families or individuals forming a tribe of 
hunters, from the millions of men that are comprehended 
in any of the civilized communities or populations of Asia or 

When language is confined to the daily use of a family or 
a small knot of acquaintance, it stands in a quite contrary 
relation to the linguistic usages of men, to what it does when 
it is the common medium that combines millions of human 
beings. In the latter case, the individual license in changing 



the adopted sounds and significations of words, by which 
novelties of speech are introduced, is continually checked 
by the impossibility of making all such unnecessary changes 
comprehensible to the mass of those who speak. Thus 
we see that in the present English and French languages, 
this license of adding to what is the common property of 
millions in both hemispheres, is a privilege for only a few 
distinguished inventors of new things, or authors of widely- 
read books. The power of changing language is so much 
repressed, that it can only be observed by comparing two 
remote periods of the history of the language ; just as you ob- 
serve geological changes by considering generations as merely 
a single day. The habit of speaking distinctly is then kept 
up and cultivated as a necessary means of being comprehended 
by the many unknown persons you continually meet with. 

In a small island in the South Sea, or among an insig- 
nificant tribe in the wilds of America or Siberia, the facility 
of changing language may easily be conceived to be next 
to unbounded. Everybody who speaks must be understood, 
because his hearers almost know beforehand what he is to 
say. The most arbitrary changes of language are thus 
introduced continually, as may be proved historically. 

Almost all those languages that are spoken by nations living 
either in a natural (geographical) isolation, or in an arbitrary 
and artificial one, want a great number of letters. For one 
letter in one dialect, is substituted another letter in the 
next ; because every word is as well understood whether you 
pronounce it with the letter r, or /, or v. Accidental and 
individual defects of utterance are thus changed into national 
peculiarities, and a general indistinctness of pronunciation is 
introduced. The sounds that are hardly perceptible to a 
stranger, will, among close relatives, appear sufficiently intel- 
ligible. A few examples will suffice to show the immense 
extent to which this practice is carried, and its vast influence 
upon the languages of all petty tribes. 

In the dialects of Oregon, according to Hale (U. S. Expi. 
Exp.), the way of speaking is so indistinct, that those who 
wrote down the words he was to arrange, could hardly hear 


any difference between the letters v, b, and m ; likewise none 
between n and d. 

Fabricius states that the females in Greenland pronounce k 
at the end of words as ng, and t as n. 

The Polynesian languages in the South Sea (U. S. Expl. 
Exp.) admit only thirteen consonants :f, k, I, m, n, ng,p, s, t 
(h, r, w). Yet, imperfect as is this alphabet, scarcely one 
single dialect admits the first ten of these. 

The Samoan wants k. 

The Tongan changes s to h. 

The New Zealand changes s to h ; I to r ; f to w. 

In Hawaian / and s are changed to h ; ng becomes n\ k is 

" The Mohawk and the Huron (Iroquois) are in a sad state 
of privation, having none of the labials neither b, p, /, v y nor 
m. When conversing, their teeth are always visible. The 
auxiliary office usually performed by the lips, is transferred 
or superadded to that of the tongue and throat. So violent a 
change in the mode of articulation has naturally produced as 
violent a change in their language, and given it at least the 
appearance of a mother tongue." Howse, Cree Gr. p. 317. 

In the dialects spoken on the coast of the Pacific ocean, 
extending from the British possessions, through the Mexican, 
down to the Peruvian, languages, several letters seem to have 
been changed into that uncouth clicking or clashing sound ex- 
pressed by tl or txlj and described by Hale as incredibly 
harsh and indistinct. This sound is moreover added to any 
noun as the most fit termination. 

The strange practices of mutilating the nose and the lips 
must have contributed a great deal to disfigure the enun- 
ciation of language itself. The insertion of one or more 
large pieces of wood into incisions in the lips or the nose, still 
practised on the Pacific coast (Tr. Geogr. Soc. vol. ii. p. 218), 
and from which custom a tribe is called Nez Perce, was no 
doubt more frequent formerly, as we see that all such cruel 
absurdities as tattooing, flattening the heads of children, &c., 
are the first prejudices a Savage abandons when he comes in 
contact with the Whites (Hooper, Tents of the Tuski, p. 270). 


These mutilations would evidently make it next to impossible 
to pronounce any labial consonant, and they would in return 
introduce a nasal articulation. Now, a paucity of labial, and 
a superfluity of nasal, sounds, is just what we observe in many 
American languages. 

Similar permutations of letters of course happen among all 
languages of the world, and in fact form the basis and the 
principal means by which the differences in language are pro- 
duced. But only among those nations who lead an isolated 
life are these changes so violent, as to appear to separate 
tribes, that evidently, from their general habits and manners, 
must be very closely related. Thus the Dakotas, forming 
only a nation of 25,000 individuals, are split into tribes 
divided by such considerable differences of dialect as these : 
one tribe changes d into t, and h into r; another changes 
h into k -, a third changes h into g ; with a fourth d is alto- 
gether rejected, and / substituted in its place ; another band 
only uses g at the end of syllables, and / does not occur ; 
thus the word hda, ' to go home/ becomes kda and gla in 
different dialects. This same tendency will of course intro- 
duce as violent euphonic changes within the same language 
or dialect in the way of declension, conjugation, and the 
formation or composition of words. 

Such regular or irregular transmutations from one system 
of letters to another are common enough in all languages 
and their dialects, forming in fact those connecting and di- 
stinguishing features between allied forms of speech that can 
be made out. But in the dialects of the ruder Mongolians 
these transitions occur in a degree that must be considered 
higher. The European languages present ample instances of 
the permutation of consonants pronounced by the same organs 
of speech, but with a different degree of aspiration, for 
instance, p, b, /, v, w, or t, d, th, and dh. Among the Mon- 
golians, where we observe some of these systems completely 
wanting, we must be prepared for permutations between the 
classes themselves. The most indubitable instances of such 
cases occur even within the same language in its dialectic or 
grammatical forms. The liquid m is changed into the mute b 


in the Lapponic 1st pers. sing, of verbs ; in Samoyed also m be- 
comes by w, u, n,p -, t changes to d, r, I, n ; k to t ; s to h ; and 
/ to n: for instance, num, nom, lorn, nop (dialectical forms for 
God) . In Greenland the women change k into ng, and t into n. 
In the Greenland language no word begins with L or R (nor 
do many, if any, commence with B, D, F, G, H, V). Hence 
it follows that the word lenni, which, among the several mem- 
bers of the Algonkin nation (Delaware, Illinois, Shawnee) 
occurs in the signification man, or to live, and which is 
easily traceable to Asiatic forms (see hereafter the instances 
sub voce MAN), must, according to the rules of the Green- 
land tongue, be mutilated into the forms innuit, innuvok, 
which we observe in its vocabulary. That the I radically 
belonged to the Greenland word also, is seen from two 
isolated words in th*e dictionary : viz. kalalek or karalek*, ' a 
native of Greenland ' ; and kablunak, ' a European ' ; of which 
there can be no doubt that the last syllable must express the 
idea man, although the exact mode of composition or obser- 
vation has not yet been pointed out. 

The Dakota language is one of the richest in independent 
consonants, yet it wants / and r. L is merely found in one 
dialect as a substitute for d or n. When we further consider 
that in a great many American languages all syllables end in 
vowels, or, like the Chinese, in the nasal sound of n, it is 
evident that the number of possible syllables in such Indian 
tongues will, of necessity, be exceedingly circumscribed. As, 
further, the roots of words, or the simple and original ideas 
in most of these languages are monosyllablic or dissyllabic, it 
follows that these original words also must be very few. 

In the syllabic alphabet invented for the Cherokee tongue 
by a native, the whole number of possible syllables is merely 
seventy besides the vowels (Trans. Amer. Ethn. Soc. v. ii. 
119). In the excellent Dakota dictionary of Mr. Riggs, we 
see how a scarcity of radical words and simple ideas is 
made to expand into a language of endless compounds. But 
from the variety of objects to be expressed, these compound 

* That Algonkin tribe, among whom the Swedes settled, pronounced 
Renni instead of Lenni. 


words must, to a great extent, contain the most arbitrary 
descriptions of things : a continual make-shift of substi- 
tutes for the thing that is thus obscurely brought before the 
mind. For instance, the word maza means ' metal of any 
kind, goods, merchandize ' ; hence are derived, by addition 
of other substantives, adjectives, or particles, compounds 
expressing an anchor, iron-pot, bracelet, bell, trap, chair, 
gun and all its parts, pistol, cannon, lock, ramrod, &c., nail, 
steelyard, blacksmith, spade, finger-ring, stove, skates, sword, 
iron, silver, money, dollar, shilling, bank-note, medal, gold, 
lead, bullet, moulds, copper, pewter, button, spoon, pan, brass, 
file, hammer, pincers, tongs. In like manner the syllable ta 
comprehends all ruminating animals and their parts. 

As another instance of arbitrary contrivances, may be quoted 
the Dakota word sungka, that originally comprehended the 
ideas dog, fox, and wolf. But then the dog, being the animal 
first employed for carrying or drawing burdens, its name was, 
after the settlement of the Europeans, also used of the horse 
when it came to be known to the Indians (sungka-wakang=. 
spirit dog, sacred dog = horse] . Thus it became the only radical 
word fit for forming the further compounds denoting horse, 
mare, colt, ass, saddle, whip, lasso, bridle, &c. 

This system is also shown in the Greenland language, the ele- 
mentary sounds whereof differ but little from those of Europe ; 
and so doing, form a tolerably complete alphabet. Never- 
theless the number of radical words is limited by the arbitrary 
rule already noticed, so that only certain consonants can be 
initial. The great prevalence of the sound of k gives to the 
whole language a remarkable palatal character. 

This system of forming language out of a few original ideas 
and sounds, instead of borrowing the name from the neighbour 
who invented or first introduced the object requiring a name, 
easily accounts for the extent of difference observed in all the 
names for more complicated ideas. It is a very improbable 
chance that two persons should hit upon the same combination. 

The similarity between this system of composition and the 
Chinese ideographic writing is very great, as both are based 
upon a natural or arbitrary classification of ideas under certain 


heads. No doubt the peculiarity of the Chinese language 
has given occasion for their way of expressing it to the eye. 
Thus it also points to the identity of the Mongol nations in 
Asia and America, and to the antiquity of their distinguishing 

In the Journal of the Asiatic Society (vol. xvi.) it has been 
remarked by Laidley, that the monosyllabic languages form 
out of a word having a great number of different meanings, 
other more distinct expressions for each of these ideas, by 
adding to the original monosyllabic another synonymous 
word. He calls this system tautologism, and is inclined to 
think that many words in the polysyllabic languages are 
thence derived. The Mongolian languages in the Old and 
the New World offer a great variety of illustrations and 
instances of this practice. Thus for head, is found in the 
Samoyed, as in the Sanskrit, Greek, &c., ngaewa, kapala, 
K(j)a\rj ; but in Finnic and many other languages merely the 
last syllable is employed, pa, &c. In Massachusetts both are 
found, but in inverted order, puhkuk. In Greenland, delighting 
in palatal sounds, only kok. In the Ehnik, Khwakhlamayu, 
Kulanapo, Caddo, and Witchita, merely the palatal letters are 
employed, to the exclusion of the labial. For pigeon Laidley 
gives two Chinese words, ko and pa. He recognizes these 
both united in the Latin columba and palumbus, but their 
combination is still more evidently traceable in the Samoyed 
kafe, ( ptarmigan,' that occurs in the widest extent of the 
American continent in the Chinese signification pigeon. 

Nja in Samoyed has, besides a great many other meanings, 
that of brother ; but to make this last signification unmistake - 
able, you may add teb, 'man/ teb enja. Little is in most 
Mongolian languages expressed by a word like ushi, &c., or 
another like tani ; but as these words also signify the young 
ones of animals, and boy, girl, &c., it is very common to see 
them both combined, as in Chocta iskitini (-^little), and 
Dakota hoksidang ( = boy), askatudang (=young), and cika- 
dang ( = little). (See Vocabulary hereafter.) 

As a source for the formation of new words, with the object 
of greater intelligibility, simple reduplication will still more 


frequently be observed in American languages ; the instances 
of Indian words, consisting of the reduplication of the same 
syllable, being too frequent to require quotation. 

The state of small isolated tribes or clans in which the half- 
savage nations live, will as easily introduce an endless change 
of significations. In a family, or amongst the inmates of the 
same house, it is quite as easy to make arbitrary expressions, 
or slang words, understood and ultimately accepted, as an 
indistinct utterance of the common words. Instead of. father, 
you may say master, governor, husband, the old one, and the 
original word father you may restrict to God only ; instead of 
child you may use any word signifying little or dear, &c. We 
have special accounts of two remarkable instances of the 
action of this principle among the rude tribes. One is, the 
superstitious custom of the South Sea Islanders, on the death 
of a king whose name is composed of a couple of common 
words, to abstain altogether from the use of those words that 
form his name, and to substitute others. The practice is 
either ascribed to reverence for him, or to some religious 
sentiment connected with omens. Such a custom will, of 
course, in many instances, lead to a permanent, instead of a 
temporary, change of language. The other fact upon this 
head is, the sacred languages employed by the conjurors or 
priests. As far as this has been accurately found out, for 
instance, in the Greenlandic, it seems to be chiefly an arbi- 
trary perversion of the significations of old and known words. 
It is then the same principle as in Europe has formed any 
slang, for instance, among vagrants and thieves. Yet these 
words of the conjurors have been so far altered that any 
double meaning is sufficiently avoided*. 

From the effect of these causes it appears probable that, as 
one savage tribe may, from trifling occasions, suddenly split 
into two that separate widely from each other, so also may 
their language, in a comparatively short time, deviate into two 
very different dialects. If there were means of investigating 

* Thus in Greenlandic, tak means darkness, but in the language of con- 
jurors the north ; thence are derived two other words of their secret speech , 
tarsoak (earth), and tarsoarmis (roots). 


the state of a given language of savages at diifereut periods, it 
would perhaps be demonstrable that its formation as a peculiar 
dialect, or a variety of speech, did not require such periods of 
thousands of years which might be supposed necessary by one 
who starts from the fact that a great many Greek and Hebrew 
words have been preserved uncorrupted for thousands of years, 
through the influence of literature and civilization. 

If it be more difficult to point out the radical similarities 
between the words of the semi-savage nations of Asia and 
A.merica, than those between the vocabularies of the more cul- 
tivated tongues, it must be confessed that the next question, 
viz. whether a recognizable word be a recent importation or a 
proof of original affinity of languages in general, is much easier 
to solve in regard to the rude than the more advanced nations, 
because everything that we know of the habits of these petty 
isolated clans in their solitudes and frequent hostility, goes to 
prove how unlikely they are to adopt the words of their 

In the first place, it is clear that limitations are imposed 
upon the introduction of foreign terms by the scantiness of 
the sound-systems just noticed. In the next, we may remark 
that, where similarity exists, it exists between words expressive 
of the primary ideas. It seems, in fact, that natural and fre- 
quent as it is among literary men and cultivated nations to 
adopt foreign words instead of coining new national expres- 
sions, such a proceeding is rare and unnatural to the Savage, 
and even to the unlettered European. 

The discovery of America affords a singular but very com- 
plete proof of this fact. It introduced at once into both 
hemispheres a great many things completely new, and of 
course wanting names both in the languages of Europe and 
America. Did this discovery then overwhelm the Old World 
with Indian names of things, and introduce among the 
savages of the New the refined denominations of the old 
civilization? In the first instance, at least, nothing of the 
kind happened. The common vocabularies of American lan- 
guages do not show the slightest mixture of either pure or 
corrupted English or Spanish ; and yet many of them give the 


appellations of things quite new to the Indian, for instance, the 
horse and the gun. The names are always un-European, and 
of course invented by the Indians according to the rules of 
their speech. The same seems to have occurred with those 
American productions that were first introduced by the 
illiterate discoverers. Take the words turkey, and potato. 
Their names are in most European languages, partly circum- 
scriptions (as Fr. pomme de terre), partly arbitrary com- 
parisons (as Germ. kartoffeln=truffeln, truffles), and partly 
ignorant mistakes (as Eng. turkey), expressions that even 
point in a totally wrong direction for the origin of the dis- 

It is quite another thing that, lately, a considerable quan- 
tity of Indian words has been introduced by men of science 
into the terminology of botanists and druggists. This very 
circumstance shows the tendency of literary men; as does 
their borrowing words from the classical languages. It seems 
that a plain European sailor or soldier very rarely adopts the 
same system, and that his linguistical taste is quite different. 

Another authenticated historical fact is, the Norwegian or 
Icelandic colony having existed for half a thousand years in 
Greenland, and having left numerous remains in ruins of 
houses and other material objects. When these Scandinavians 
settled there, they found the Eskimos on the coast. The two 
nationalities must then have coexisted in a sort of contact. 
When Greenland was rediscovered and again settled by the 
Danes in the 18th century, the first missionaries thought that 
they observed sufficient traits in the physiognomy of some 
individuals among the Eskimos to warrant the conclusion 
that a mixture of races, had taken place, and that there still 
remained traces of the Teutonic blood among the natives. Yet 
the most complete dictionaries that were collected of the Green- 
land tongue hardly show more than a single indubitable Norwe- 
gian word having crept into the language, viz. the word kona, 
forming a synonym for woman. This word, curiously enough, 
points to the circumstance that was most likely to attend the 
extermination of the colonists, viz. the fact of merely a few 
women being spared, perhaps as slaves. Of the existence of 



the former colonists, the Greenlanders had plenty of traditions, 
but under a name of their own invention, Kablunet. 

An important fact in the opposite direction has been often 
quoted from the narrative of Hale, viz. that in Oregon there 
has been invented a ' trade language,' or ' lingua franca/ con- 
sisting of a rude mixture of different Indian and European 
words. But before we calculate the importance of this cir- 
cumstance in our own days, it would be safe to ask for the 
influence and the real signification of the original lingua franca, 
that has existed in the Mohammedan towns near the Mediter- 
ranean since the time of the Crusades. Has this lingua franca 
ever been adopted by either Arabs or Turks, or Italians and En- 
glish among themselves ? Has it exerted any perceptible influ- 
ence upon the real and vernacular language of those individuals 
that transact their shopping in a gibberish that is barely 
intelligible, but at the same time despised and ridiculed by all 
of them? The trade-language on the Columbia river will 
likewise, in all probability, never become the sole and only 
means of expressing the thought of one single individual. It 
is absurd to imagine two Englishmen conversing together in 
this childish jargon, nor is it credible that two Chinooks 
would prefer it to their own. It must then be looked upon 
as a mere make-shift, a substitute for a language employed 
only occasionally, by two men, of whom each has, besides, a lan- 
guage of his own, which he makes use of in the common 
intercourse of life, and with those whom he considers as his 
countrymen. Whether words from the trade language may 
ultimately creep into the common every-day Indian, is a cir- 
cumstance of which proof and experience are yet wanting. But 
from the difficult and strange system of sounds, and the pecu- 
liar grammatical forms that regulate every one of the Indian 
tongues, it is evident that it is as difficult for one nation to 
pronounce the words of another, as for an Englishman to 
pronounce Russian. The transferring of words from one 
language to the other finds then as powerful an obstacle in the 
form of the language as in the manner and prejudices of 
those who speak it. We also see, in the reports of travellers, 
that even a nation weakened next to extermination, does not 


corrupt its own language, but that it, in some instances, will 
adopt altogether that of a tribe with which this nation 
becomes incorporated, or by whom it is enslaved. (Reports 
of Expl. and Surv. 1853-4, vol. i. pp. 411, 425.) 

In the Chocta Definer there certainly occur, under the 
head of Religion (p. 176), many English words; but these 
also are owing to the instrumentality of the literate teachers 
and missionaries of the Indians. 

The observation, that men in the ruder states of social 
manners will seldom adopt foreign words, gives a high value 
to the results of comparative philology. Whenever a simi- 
larity of language is proved, it cannot be derived from later 
importations and changes; the similarity must be original. 
For those who deny the common origin of the nations in 
question, nothing remains but to declare all these similarities 

With these preliminaries we may now consider the import 
of the following tables. That the comparison has also been 
extended to the Japetian languages will not startle any reader 
aware of the affinities pointed out by Prof. H. Key and Mr. 
H. Wedgwood, between the Finnic and the Indo-European 
tongues. For some additions to the information about the 
Lapp, I am indebted to Mr. Friis, Professor of this language 
at the University of Christiania. 

Words denoting the ideas of Life, Man, fyc. 

1 *. English I live, alive Man, Indian, native. 

Samoyed . . iljiro, jirido, jileadm, ") nienec, ennete (compare ma, 
jelinje. ) friend) . 

Jukagir . . link, endsit 

Koriak .... jolgat 

Lapland .. alam,jalab, alleme . . olmaj (mV), olmus (homo). 

Finnic, &c. elam'd sulahane, ihminen (vir). 

Esthonian,. . olio innimene (homo) . 

Mordwin . . walmes loman (homo). 

Tsheremiss . ilimas ulmo (vir) . 

Permian . . olom 

* In all these lists the figure 1 denotes the Asiatic ; 2, the American 


Wogul .... elmalna ellem cholles. 

Hungarian elet 

Ostjakian . . wulta 

Tungusian inen, insem tungus, donki. 

Kamcadal . kakolin, sont-lonem. . el ku. 

2. Shawno . lennamawe illeni. 

Chippeway innini, man ; neeje, friend. 

Knisteneaux ethin. 

Potawotomi neeah. 

Satsikaa ninaw (homo) ; napi, husband. 

Menomeni enaineew. 

Dakota .... ni t ti hihna. 

Nottoway eniha. 

Tuscarora aineehau. 

Iroquois nene/cin, ecinak. 

Athabasca dinni. 

Beaver tine. 

Kutchin tenghi. 

Kinai teena. 

Sikanni sikkane. 

Greenland, innuvok innuk, uvek, angut. 

Kolosh tlinket. 

Tsihaili kolmukh. 

Slavs, oloa, father. 

Tahkali, anna, ninastsa onla, mother. 

Kalapuya ialei. 

Wailatpu inaiu. 

Aztek, nemi 

Huasteca, elel, birth inic. 

Maya uinic. 

Poconchi uinoc. 

Quiche uinac. 

Pima intui .... mother. 

Shoshoni inea .... friend. 

In the Aztek language no word begins with /. 

Compare with these forms the English live, the Sanscrit 
giv, the Latin vivo (vigeo), the Slavonic ziv (living), man (?) 
homo (?). 

For the allied ideas of great, and old, and for many others 
growing out of them, we have 


1 . Samoyed, agga, great ; wesa- 
ko, buiza, husband ; baikua, old. 

Lapland, akka, wife. 

akko, old woman. 

Ostjak, anaya, stepmother. 

ika, hushand. 

Finnic, ukko, 1, old man ; 2, God. 
Lesghian ( Akush), ukna, okna, old. 
Mizdz'eghi, kani, tkene, old. 
Turkish (Uigur), acha, elder 


(Kirgiz) aga, elder brother. 
(Osmanli) agha, master. 
Tshiiktshi, aganagak, girl. 

apakaka, grandfather. 

aganak, woman. 

aghat, God. 

2. Dakota, tangka, great. 

wica, male. 

Assiniboin, wincha, man. 
Greenland, aka, uncle ; angut, 


okok, mother ; angajuk, el- 
der brother. 

Dakota, tawicu, wingy, wakct- 
angka, woman. 

Assiniboin, weak. 

Shawno, (n)ewa, ~\ 

Chippeway, (w)ewan, J 

gah, mother. 

Otawa, ague, woman. 

gachi, mother. 


^ woman. 

> mother. 

Satsikaa, akuea -} 


Algonkin, squaw 

gah, mother 

Iroquois, yongwe 


Adahi, quaechuke 
Umqua, ekhe 
Tahkali, sak ~\ 

Tsehaili, naxonaxo >wife. 
Tshinuk, kakilak. J 
Tahkali, skaka 
Tsehaili, soqo, skui 
Wailaptu, quks 
Tshinuk, akxo 
Kizh, aok 
Kadjak, aghajun, God. 
aganak, woman, girl. 

abaga, grandfather. 

angaga, elder brother. 

agugux (Aleutian), God. 

Huasteca, uxum* I 
Maya, ixal 
Poconchi, ixoc i 
Chorti, ishoc 
Otomi, oqha, God. 
Maya, ku, God. 
Seneca, ungouh 
Wyandot, aingahon 
Mohawk, oonguich. 
Tunghaasa, uncan, chief. 
Okanagan, uncus, warrior f. 



Add to these, as more distantly allied 

In Arrapaho, enanitah, man ; enenitah, Indian ; Costano, imhen ; 
Navaho, tennay ; Jecorilla, tinlay ; Tesuque, say en ; Jemez, tah- 

* x in Spanish = k h. 

1* Compare Sanskr. vaks, vali, to heap, to grow; Lat. augeo; I si. auka, 
grow; Germ, auch, also; wucher, usury ; English to wax. 


Man, native, hunter, warrior. 

1 . Samoyed, hasawa, kasa, nganang, nganasang ; Aino of Kamc'atka 
andTarakai, ainu, okajuh, ozukai. 2. Seneca, haujenauh; hawneuauh, 
white man ; Cayuga, hajina ; Mohawk, wakeniakon, I have a hus- 
band; Muskoghe, honunwau ; Chocta, hatak ; Adahi, haasing; hase- 
kino, husband; Wailaptu, inaiu; Talatui, sawie; California, S. L. 
Obispo, sapi f father ; Ehnik, ahwunsh ; Lutuami, hishu atsus, man ; 
Arrapaho, nash, husband ; Tahlewah, astowah, Indian ; Chinuk, 
uchu shaash, warrior ; Kawitchen, nooz sho wawa, hunter ; Tlao- 
quatch, haioha, 1 ? Haeeltzuk, hailthloscun, 10? Klikitat, aswan, 
boy ; Sahaptin, hawahush, warrior. In the Jurakian dialect of the 
Samoyed husuwaei is a pronoun (everybody)*. 

Compare the Polynesian tangata, kanata-, vulgarly called 

English, old man 1. Samoyed, dra, ira ; jieru, jierwu, chief; 
juru, njirung, friend ; Hungarian, ferj, man ; Turkish, er, ir, man 
(compare Latin vir). 2. Californian, ehnek, woman ; ijei, ngoroite, 
efieije, man ; Greenland, arnak, woman. 

1. Samoyed, teb, teppa, old man. (In the Ostjak Sam. dialect 
tep, tap, is the pronoun of the third person, he or this.} Yeniseyan, 

btet t hadkip, man . 2. Blackfoot, muttuppe, man; Chim- 

mesyan, tzib ; Billechula, tlimdash, man. 

English, woman. 1. Samoyed, nelgum, naigum; nieleu, I take a 
wife ; Greenland, nulliak, wife ; nulliarau, copulate ; nallegak, 
master. 2, Algonkin, nihillal<juenk, our Lord, God ; (nihillapewi, 
translated by Zeisbeger, I am free : it probably means I am (tny 
own) master ; Chocta, nahullo, master. 

Another fertile root comprehends the ideas little, young, 
child, boy, son, girl, daughter, in almost all the Mongol 

Samoyed, ngaceky \ 

u6il> utcei j young ngaceke, boy ; m, child. 

Lapland, ucca, little 

Tsheremis, isi, little. . 

Permian, icet, uoet, izot, little. . . . 

Hungarian, kistin, little 

Turkish, jas, young ; kicik, kizik, 1 . 

kici, little . . . . } *" 

Tungusian, asatkan, asatkan,acatkan,%\v\\ 



Lesghian, mici, little yaso, yase, yosi, girl. 

uasa, bici, moci, wazu, usi, boy. 
Kamtshadal, uicinan, ucinolo, little uci pec, daughter. 

Yeniseyan, kisigena, little 

Chinese, see, little dzd, child. 

Tn America this root exhibits a numerous offspring in almost 
every language with which we are sufficiently acquainted. 

Greenland, mike, little 

Tshuktshi, acik, young 

Athabascan, tsoota, little azay, son. 

Tlatskani, astekwo, little yaase, son ; tsukais la, daughter. 

Tahkali, ensoole, little astoque, ete, sie, child ; eyoze, 

son, cekus, tsikesle, girl, &c. 

Delaware, wuski ~ oowasis (Knist.) 

Ojibbewa, oskenege > young. anese, muckiese (Narrag.] 
Massachusets,fc0w,s#eo J my washish (Micmac) 

Abenaki, nemetessan, he is the 7 . / i > >\ 

[ awansis (Abenaki) 
youngest. 3 

Blackfeet, pistakwiu equssis (Knist.), son, etc. 

enaksti-pokas, child (Blackfeet) ; tsia, girl. 
Tahkali, cekui, tsikesle, girl. 

Nottoway, osae, young yaweetseutho (Wyandot), girl. 

qu'dtsiageyung, my daughter. 


aqu'dtsiaskaya, my son. 
atsatsa, boy. 

ayayutsa, girl. 
joost ekuh, child. 

Chocta, iskitini, little ushe, ussi, child. 

Muskhoghe, -. oshetik, his daughter ; schah- 

chostie, my daughter. 

Uche, tesunung, son. 

Blackfeet, pistakwin enakssti-pokas, child. 

Natchez, tsikistiktenu > 

Chetimaca, aksekamche / ? Q % ' ' **&*<****> son. 

Dakota, askatudang, young ; cis- "| 

ting, cikadang, little / hMdan 9> h 7- 

Upsaroka sJcakkatte. 

Aleutian, kutakh (the diminutivel Kulish , daughter*. 

termination) >Chimmesyan . tzoushk, small. 

J Kinai sija, son. 

* Achsi is the form found in the vocabulary ; but with the same ach, ik, 


Chemraesyan tzoushk, small. 

Kinai sija, son. 

Lutuami, kitskan kitskenishnawats, girl. 

Pailaik, tsoktsa kistka, child. 

Tsihaili, pustsitl, cells isa, daughter ; waxtetl, child. 

I etsokha, son. 

Chinuk, . . . J asa > daughter. 

| tlkaskus, boy. 

\j)sko kus, girl. 

Sehsh . . tasika, skokosea, kokwasso, boy. 

Wailatpu skutxla, child. 

Arrapaho issaha, girl. 

Shyenne XS a, girl. 

Sasti . . atokwiakh, boy. 

Jakon . . tlomkhato, boy. 

Talatui, weso, new 

San Rafael, yokeko, child. 

San Antonio skitano. 

Otomi hy t iso, son. 

Palaik, yauitsa, son. 

Another series of synonyms for the same ideas is 

1. Samoyed, tanio, tanu, tenne, little pirib-tjea, girl. 

, or kanak, kanang, takanang, } 

j- tt | e > tati, dati, a younger wife. 

, or ngoliu, ngoloko, njoloko . . 

Lapponic, unna njejda, girl. 

Jukagir, andelgoin, lukun, young. . baitaya, girl. 

Korjak, liuchin 

Jeniseyan, khennenam 

Kamcadal, linetlen, young 

ucinolo, little 

Finlandic, tytto, tytar tyttu, tytcir, girl, daughter*. 

2. Delaware, tangtitti (Knist.) tanis, girl ; (Chippe- 

w&j)janis, son. 

Chippeway danis, daughter. 

Blackfeet ntani. 

Chickasah take, girl. 

also commence all (nearly twenty) words having respect to man or parts 
of the human body. It is most probably then = the Dakota wica, ' human,' 
added, for instance, to ista, ' eye,' wicista, ' human eye.' 

* The Japetian Gr. Ovydrrjp, Sanskr. duhitr, Russ. docj, Eng. daughter. 


Arrapaho . . nahtahnah, my daughter. 

Chocta, iskitine 

Natchez, tsiki stiktenu 

Dakota, tonana 

Athabasca tenaiu, boy. 

Tlatskanai, teneuai, young 

Tahkali dinias, boy. 

Tsihaili, kiki ana, tauma 

Nsetshaw tunuwon, son. 

Talatui (California) tune, child. 

Palaik tatii, mother. 

Shoshoni natsi, boy ; nanai, daughter. 

Attacapa tigu, daughter. 

Skittagets tinekti, child. 

Tlaoquatch tannais, child. 

Sahaptin tata, son. 

Wakash (Nootka) tanassis, child. 

San Antonio (California) sketana, child. 

Another word of the same signification is English boy, 
Swedish pojke, Danish pige, girl, Gr. irals, Lat. puer (por, a 
slave), pupillus, Germ. bube. It is found in Finnic as son : 
for instance 

Finland, pojka ^ ~ 

Zyrian and Votjak, pi i papa, younger brother or son. 

Wogul, pum, pu, py s ^^ bi, brother-in-law. 

Yeniseian, pijwo | ^ | bing ing, son-in-law. 

Hungarian, fiu J ^ L 

Ostjak, pox, pax ; pada, paga, brother-in-law ; 
Kamcadal,^ec,jjaca ; uci-pec, daughter; Tshuktshi,panni&a, daughter. 

In America there is likewise in Eskimo Greenl. pannik, 
daughter ; in Algonkin (Penobscott, Powhattan, Mohican, and 
Abenaki), pchanum, phainem, panum y po. In other Algonkin 
dialects (Mohican, Nanticoke) the word for girl is formed 
from pekh or pinsh, and the word squaw, woman; Massach. 
penumpun, girl Narragansett,^j9oo*, child; Chippeway, bo- 
beloshin, children; Blackfoot,j90&aA, child; Upsaroka, bakatta, 
child. In the Eslen (California), Tr. Am. Ethn. Soc. vol. ii. 
p. 127, panna is son, and tapanna, daughter. 

In San Miguel (California), son and daughter are paser, pasel. 


In Wailatpu daughter puena. 

In Kalapuya (Willamet) . . daughter tshitapinna. 

In Muskoghe son chahpozhe,my . 

For daughter-in-law, occurs in Samoyede, meje'd ; in Ostjak, menj, 
minjing ; Finnic, mini'd ; Zyrjan, monj ; Hungarian, meny. 

By the current change from m to n, this is probably Lapon. 
njeida, girl; Samoyed, nddek, nitteng, girl; mijda, younger 
brother ; muanga, brother-in-law. The Sioux family in Ame- 
rica offer for woman 

Mandan, meha, submihe, girl ; Minetari, meeyai ; Iowa, mega ; 
Upsaroka (Blackfoot), meyakatte, meekatay, and moah, wife. 

The Japetian languages offer for girl 

Sanskrit, mahild, woman ; Germ, magd, mddchen ; Eng. maid, 
maiden ; O.N. (Icel.) megfta, mey. 

For brother-in-law or son-in-law, O.N. (Icel.) mdgr ; Goth, megs ; 
Anglo- Sax. mag, relation ; Gaelic, mac, son. 

FATHER. Sam. as, ese, es, atja ; Lap. acce ; Turk, ata ; Juk. 
efcea; Jen. es, God; Ostj. essig, old man; Hungar. osz ; Greek 
arra ; Russ. otjets ; Dakota, ate ; Ojibbeway, os ; Delaware, och ; 
Pottawatomi, (n)osah ; Cherokee, tawta ; Greenland, attatak ; 
Tshuktshi, atta ; istla, God ; Aleutian, adax. 

In ten Athabascan dialects tah, &c. 

Loucheux. tsay ; Tsihaili, to*, katsha; Talatui, fata; Nus- 
dalum, outset. 

Another Samoyede word for father is 6a = Tahkali, apa. 

But this last root also appears in Samoyed as elder sister. 
Sam. ngaba, oba, apa, ada } and in the Jurakian dialect njabako 
(being a composition of nje, woman, aba, and agga, great, 

1. Lap. obba, oabba', Ostj. opa, aba-, Juk. awuca. 2. Dakota, 
tawinohting ; Abenaki, nabaenemun-, nadangus, cousin; Knist. nimis ; 
Delaware, tawima ; Shawno, newa, my wife ; Ojibbeway, wewan, his 
wife ; Chocta, ipo, sister-in-law ; Greenl. nejak, nukak ; alleka, elder 
sister ; Lutuami, tobaksip ; Shoshoni, patsi ; Yamkalle, yet apai, 
brother ; Kalapuya, opomeik, daughter ; Pima, uba, woman. 

The same root further occurs in Samoyed, as 
aunt, mother's sister, njaba ; and aunt, father's sister, abijo, awijo. 


From the first word for father, is derived for uncle 

Sara, isi, ise, aca ; Lap. akke, cukke ; Dakota, ate ; Abenaki, nesis, 
my uncle ; Knist. nesim, my elder sister or brother ; Potawatomi, 
sesah ; Satsikaa, nisa, brother ; Chocta, imoshi ; iki, father. 

A singularly fertile root in Samoyede is nje, ne, which is 
given for 

Woman Sam. ne, nie, njaru, inia. 

Wife njejeru, ne'd, nei-kum ; nieljeu, to marry. 

Mother njebea. 

Aunt njaba, njejea. 

Girl nitting. 

Daughter nje, njenju. 

Sister, elder .... njabaka, njanja. 

Sister, younger . . njenja. 

Friend, comrade. . nja, nje. 

Brother, elder . . nja, nienne. 

Brother-in-law . . njinjiadea. 

This word, then, cannot possibly be more than a word of 
endearment, whose general use in so many cases shows the 
amount of indistinctness in expressing ideas, or the ease with 
which such a small knot of acquaintance as forms the given 
tribe, can make itself understood. There is a possibility, by 
compositions, of making some of these words more distinct or 
expressive. Thus, the addition of this root to teb, man, tabenja, 
makes it indubitable that you mean brother. In other cases 
there are synonyms for the same purposes, as for ( friend* 
kai, iljUjjuru, njirung. For brother, the American languages 

Delaware, nimat, minut ; 

Satsikna, nausah; 

Chocta, imunni, elder brother. 

In Greenland ningauk is brother-in-law, ningiok, old woman. 
For mother 

Dakota, ina. 
Shawno, niwa, nig a. 
Menomini, nihia. 
Delaware, anna. 

Pottawatomi, nanna ; niowah, 


Tuscarora, ianu. 
Adahi, amanie. 



Greenland, ananak. 
Athabasca, enne. 
Tlatskanai, nana (and similar in 
five other Athabascan dialects). 
Koltshani, niji. 

Willamet, sinni. 
Kalapuya, innim. 
Kawitchen, Noosdalum, intan. 
Californian, nene ; niook, father. 
Seneca, hanec, father. 

Friend is in Delaware nitis, in Ojibbeway neeje. Corre- 
sponding to the other Samoyed root kai, there is for friend 

Dakota, koda. 
Shawno, necana. 
Seneca, gache. 
Chocta, ingkana. 
Chetimaca, Natches, keta. 
Kolosh, ekawu. 
The following words relate to the chief primary ideas 

Tahkali, kanane. 
Tsehaili, noquai. 
Atnah, tasskanaan. 
Ugalenz, sekoanak. 
Takahli, chutaissi. 

English, beard. 

1. Samoy., munac, mudute. 
Jukagir, manallae, hair. 

2. Delaware, wuttoney. 
Knist., michetoune. 
Satsikaa, mongatsi. 
Chocta, nutakhish. 
Greenland, umik, 
Bodega, ummu. 
Kalapuya, mundi. 
Shoshoni, muntsu. 
California, numus. 
Kulanapo (Cal.), musuh. 
Talatui (Cal.), mono, hair. 
Shy enne, meatsa. 
Arrapaho, wasesanon. 
Shasti, makh, hair ; mak, head. 
Kulanapo, musuh, hair. 

Pimo, mouk, head ; ptmuk, hair. 
Caddo, beunno, hair. 

English, hair, wool. 
1. Sam., tar, opte. 
Finnic, tukka. 
Lapp, vuofta. 
Jenis, tenge. 

Korjak, kitigir, kacugui. 

2. Satsikaa, otokan. 

Greenl., tinge. 

Shoshoni, tupia. 

Apach, seesga. 

Athabasca, thiegah. 

Tahkali, thiga ; tamagaie, beard. 

Dogrib, theoga ; tarra, beard. 

Umqua, Kinai, Atnah, Koltshani, 

zuga,zygo, zega, ciga; ktatahi, 


Kolosh, sachagu. 
Umqua, etaga, beard. 
Kwakhlamayu, shuka. 
Weitspek, tegueh, head. 
Calif., tiih, tomi, tomoi. 

English, belly. 

1. Sa,m.,munedi, wand, my, nand. 
Lapp, coavgje. 

Ost., xon. 
Korj., nam, kam. 
Jen., wui. 

2. Dakota, cowohe. 
Etchimin, nut. 
Narraganset, iviinnaks. 


Iroquois, atquonta. 
Greenl., akoak, nak. 
Athabasca, beeth. 

Cree, mithkoo. 
Tuscarora, cotnuh. 
Greenl., auk. 
Tahkali, sko. 
Sahaptin, kiket. 
Calif., kiio. 
Chocta, homma. 
Weitspek, happl. 
Kulanapo, bahlaik. 
Copeh, sank. 

English, blood. 

1. Sam., ki, kam,kap, hem. 
Jukagir, liopkul. 

Lapp, vara. 

2. Dakota, we. 
Delaware, mokum. 

This same word blood is also largely employed to denote 
several colours. Thus in Chocta, homma is further given for 
bay and red, and enters into the compound words for copper, and 
purple, and swarthy. In some Algonkin dialects mokkum is 
black (Delaware, mokum, blood). In others it forms a part 
of the words, as 

English, black. 

1. Sara., saga, seak, hag. 
, smankua. 

, newai. 

Lapp, cap. 
Korj., nowukain. 

2. Dakota, sapa. 
loway, sewi. 
Delaware, suck gek. 
Narrag., suckesu. 
Long Isl., shikayo. 

, squayo, red. 

Natchez, tsokokop. 
Muskoghe, echatau. 
Tsihaili, cmaka. 
Sahaptin, cmuk. 
Wailatpu, skupskupu. 

English, breast, bosom, teat. 

1. Sam., sudo, suso. 
Ost., tju, tji. 
Lapp, cidze. 

2. Ojibbewa, totosh. 
Chocta, huship, ipishik, teat. 

Greenl., sikkik, sokkiek. 
Tahkali, tsoo. 

English, body. 

1. Sam., ngaja, aja. 

2. Ottowa, eeio. 
Ojibbewa, yoa, yohi, yas. 
Abenaki, haghe. 
Satsikaa, iniwia. 
Onondago, ojatah. 

English, ear, hear. 

1. Sam., ku, ho, ha. 
3ak.,golendzi; Lap., gullat, hear. 
Ostjak, xudem, hear. 
Georgian, quri; Lazian, gur, hear. 
Zend, gaosa ; Sanskr. sru, hear. 
Osset., gos. 

Persian, gus. 

Lithuanian, ausis ; girdeti, hear. 

Latin, auris. 

2. Dakota, noae, nakpa. 
Mandan, nakoha. 
Yankton, nougkopa. 



Osage, naughta. 
Long I si., catawoe. 
Shawno, towakah. 
Iroquois, ohuchta. 
Pawnee, at karoo. 
Cherokee, gule. 
Adahi, calat. 
Muskoghe, kutseo. 
Chocta, haklo. 
Tahkali, ocho. 
Kolosh, kuk. 
Tlaskanai, xonade. 
Tsihaili, qoalan. 

English, eye, see. 

1 . Sam., hai, saeu, sai, saime ; 
thence sea, sja, face. 

Lapp, calbme. 
Finn, silma. 
Ostj., sem. 
Hung., szem. 

2. Dakota, ista ; ite, face. 
Mandan, istume, estah. 
Upsaroka, meishta, esa. 

Algonkin, (w)iskinki-, (ne)sisse- 

guk ; Abenaki, (my) face. 
Menomeni, maishkaishaik. 
Miami, nesique. 
Arrapaho, mishishi. 
Seneca, kaka. 
Chocta, nishkin ; nashuka, face ; 

issokuh (Chickasaw), face. 
Greenl., irse. 
Eskimo, eieega. 
Athabasca (Dogrib), nhae. 
Tahkali, naxai. 
Umqua, Kinai, Atnah, Inkilik, 

Koltshan, nage, nag a, nega, 

noga, ntagi. 
Tshinuk, iaxot. 
Shasti, oi. 
Tsihaili, qalom. 
Sahaptin, cilu ; atcas, face ; sua, 


Wailatpu, takai, forehead. 
Copeh, sah. 
Tesuque, chay. 
Jemez, saech. 

The words for finger, fist, hand, foot, nail, are explained in 
conjunction with the numerals; the subject of a paper else- 

English, head (1.). 
1 . Sam , ngaewa, eba. 
Lapp, oajve. 
Finn, pa, poja. 
Chinese, he, hep. 
Lat. caput ; Gr. *ce0d\^ ; Sanskr. 
kapdla, skull; Germ, kopf, 

2. Dakota, pa. 
Yankton, Omaha, pah. 
Mandan, pan. 
Minetare, apeeh, neck. 

Arrapaho, pahhih. 
Massachusets, puhkuk. 
Algonkin, uppa. 
Chocta, nishkubo. 
Muskog., ecau. 
Tahkali, pitsa. 
Athabasca (Dogrib), betthie. 
Uchee, pseotan. 
Pawnee, pakshu. 
Riccaree, pahgh. 
Chickasah, skoboch. 
Natchez, tomme apoo. 


Caddo, cundo. 

Witchita, etsJcase. 

Shoshoni, pampi. 

California, pakon, buhk, pok. 

Shasti, uiak. 

Tesuque, pto ; po, hair. 

Among the Japetians 
blended thus 
English, . . neck . . throat. 
Latin, .... collum. . gula. 
French, . . cou, col. 

Sanskr gala. 

German . . hals. 

English, head. 

1 . Icel., Mir. 

French, col (the Alps), top of a 


English, skull, scalp. 
Hindostan, kallah. 
Russian, golowa. 

2. Delaware, wil, head. 
Shawno, wilan, head. 
Palaik, ul, forehead. 
Sekumme, tsol, forehead. 
Cushna, chole. 
Costano, ulc. 
Tshokoyem, moloh. 
Pujuni, cucul, head. 
Greenl., auvak, neck. 
, niakok, head. 

English, heart. 

1. Sam., sa, seal. 
Jenis., sitabu. 
Ostjak, sem. 
Finnic, syd'dn. 

2. Dakota, cangte. 
Massachus., tah. 
Ojibbeway, da. 

English, neck (2.). 
1. Sam., ol; aolj, awoi, awai, 


Jukagir, monoli, head. 
Jeniseian, kolka, head. 
Lapp, oalge, shoulder. 
Finn, kaula ; kallo, scalp, 
these two significations are also 

Ottawa, (nin}de, my head. 
Tlatskanai, stsaie. 
Dogrib, e-dzai. 
Umqua, sci. 
Kenai, see xtee. 

English, mouth, tongue. 
1. Sam., nia, ngang-, njami, 

Lapp, njalbme; njuovc, njuok- 

cam, tongue. 
Juk., anga. 
Jen., khan, hohuj. 
Ostiak, nadam, tongue 
Hungar., nych, tongue. 
Chinese, kheo. 
2. Massach., minan, tongue. 
Ottowa, tenanian, tongue. 
Mohican, ninanuh, my tongue. 
Echemin, nyllal, tongue. 
Attacapa, nedle, tongue. 
Dogrib, eththadu, tongue. 
Ugalenz, ka-nat, tongue. 
Adahi, tenanat. 
Willamet, mandi. 
Greenl., kanek. 
Tsihaili, kanuk. 
Jakon, qai. 
Shasti, au. 

English, tongue, mouth. 
1. Sam., se, sie, sioro, siolo. 


Chinese, ski. 

2. Dakota, cezi. 

Arrapalio, dehzeh. 

Algonkin, uton, don, also mouth. 

Chocta, iti, mouth. 

Pawnee, hatoo. 

Tahkali, tsoola. 

Kenai, zylio. 

Inkulit, tljulja. 

Arrapaho, nathun ; netlee, mouth. 

Shyenne, vetunno ; martke,mouih. 

Navaho, hu-zzay, mouth. 

Jecorilla, hu-zzy. 

Tesuque, sho. 

Caddo, (ocko)tunna. 

English, tooth. 

1 . Sam., tibea, tim, tin, tiw. 
Lapp, badne. 

Ostjak, penk. 

2. Delaware, wipit, his tooth. 
Algonkin, tibit ; nepit, my tooth. 
Tuscarora, otoatseh. 
Sahaptin, tit. 

Wailatpu, tenif. 
Lutuami, tut. 
Caddo, (ocko)deta. 
Nootka, cicice. 
Kalapuya, puti, tenti. 
Cora, tenita. 
Mexican, tentli, lip. 

English, day, light. 

1. Sam.,jale, eel; jalina, white, 
Lapp, culgas, jalakas, light j 

vielgad, white. 
Korjak, hallo. 
Jukagir, jelonsa, sun. 

2. Yankton, ohjajo, light. 
Dakota, eang. 
Onondago, jolacharota, light. 

Attacapa, iff I, day. 

Dogrib, zeunai. 

Athabascan, dzine. 

Tahkali, janes. 

Kutchin, tzin. 

Kenai, can. 

Atnah, cajane. 

Koltshani, tiljkan. 

Tsihaili, sxaltxalt ; xal, light. 

Wailatpu, tlaxa. 

Kulanapo, la, sun ; luelah, moon ; 

dahmul, day. 
Tshoyem, hiahnah, day. 
Caconoons, hial, day. 

English, sun, moon, star. 

1. Sam., eel, hajer, hajar, kou, 
kuja ; khi, moon ; keska, star. 

Zyrj. and Perm., tolys, moon. 
Ostj., xat ; xus, star. 
Finn and Mordwin, kou, ku, 

2. Winnebago, weehah, sun. 
Micmac, koushet, moon. 

Cree, kesekow, it is day; kijik, 

day and light. 
Ojibbeway, Illinois, kisis, sun and 

moon ; gezhig, sky. 
Echimin, Abenaki, kisos, moon ; 

watawesu, star. 
Kickapoo, kishek, heaven. 
Shawno, gilswa, sun. 
Arrapaho, ishi, day; nishi-ish, 

Mohawk, kelauquaw, sun and 


Seneka, kachqua, moon. 
Chocta, hushi, sun and moon. 
Natchez, kevasip, moon. 
Adahi, nachaoat, moon. 
Chicasa, husha, sun. 


Muskoghe, hahsce, sun. 
Greenl., kaua, south ; kail, day ; 

kaumaty moon ; kauma, light ; 

kaulor, white. 

Athabasca, saw, sun and moon. 
Navajo, chay, haei, sun and moon ; 

e, star. 
Ugalenz, kacha, moon ; kaketlj, 


Sahaptin, alkhaikh, moon. 
S. Barb. Californ., aguai, moon. 

Kulanapo, ucyahho, star. 
Costano, kolina, moon ; agweh, 


Tshokoyem, hitlish, star. 
Haidah, kosugh, moon. 
Kliketat, uchych. 
Cathlascon, kaium. 
Kutshin, shethie, sun ; shetsill, 

moon ; keemshaet t stars. 
Wailatpu, kaki, star. 
Noosdalum, kokweh, sun. 

Weitspek, kamoi, star. 

English spring summer, . . . autumn. 

1. Samoy tagai * ..... tang a, tag a gnutu. 

Lapp gidhag, adv. . . gasig cafcag, cakcag. 

Juk nada. 

Ostj taven kahatsaan. 

Hung tavasz. 

2. Algonkin .... thequan, sequan tagwage, tahgagi. 

(in eight dialects.) 

Satsikaa atahi. 

Arrapaho tahuni. 

Dakota mdoketu. 

Osage tondah. 

Seneca unguitikne . . kahayneh . . gankneh. 

Onondaga teoganhouiti. 

Chocta tofahpi tofah .... onafahpi. 

Muskh tasachuy. 

Chicasaw tomepulleh. 

Attacapa tsampska. 

Eskimo kuiya, kegmi, 

Greenland tseykerek okiak. 

Tshuktshi anchtoka. 

Kolosh takity \ takoonehate. 

Athabascan .... tacata. 

Tlatskanai citaxat. 

Umqua ghainsghaltsi. 

Tahkali cago tsinta. 

Sahaptin taiom. 

* Castren's Ostjak Grammar; omitted in the Samoy ed Dictionary. 


Chinook tsagwaix. 

Shasti atahi. 

Shoshoni taza, tatsu. 

Wailatpu tong. 

Palaik. kaitui. 

Tsihaili tlakam. 

Chimmesyan sughone. 

Comanche taneharro. . . . taareh. 

English, hot, warm. II Micmac, epekit. 

1. S&m.,jipi, efi,jefi. Etchemin, kesipetac (kisi, see 

2. Dakota, petiskang. || sun). 

In most Algonkin dialects nipin is summer ; in Micmac 
nepinowe, spring; Greenland, aupak. In the Pueblo lan- 
guages, pah, Sec. is sun and moon. 

English., hot, warm. 

1. Sam., lahum*. 
Korjak, nomling. 

2. Sahaptin, laxoex. 
Wailatpu, lokaia. 
Tsihaili, xwala. 
Tahkali, wola. 
Lutuami, soalkas. 
Chocta, lushpa. 

English, cold. 

1. Sam., tin, tjasiti, tjasaga. 

kail, kai, kanie, hanie. 

kandak, I freeze. 
Lapp, coaskes, galmas. 
Korj., khuelgin-, tintan, ice. 
Jen., tajim, kucidin. 

2. Dakota, tasaka, sni, caga. 
Mandan, copcaze, snow. 
Knist., hisina, kiJcatsh. 
Delaware, ten. 

Abenaki, teki. 
Satsikaa, cane, snow. 
Chocta, kupussa. 
Muskhog, kussupe. 

Chetimacha, kasteke. 
Caddo, hehno. 
Natchez, kowa, snow. 
Greenland, keja, kajorpok, 
Tshuktshi, anu ; anighu, snow. 
Korjak, cigu, ice. 
Tshugatsh, caguk, ice. 
Ugaleuts, tets, ice. 
Atna, Kenaij, ten, ice. 
Kolosh, tyk; Jcakak, ice, kusjat. 
Loucheux, kabeitlec. 
Tahkali, hungkox ; ton, ice. 
Tsihaili, tatsuwaii. 
Sahaptin, tsuaia. 
Chinuk, c'ds. 
Lutuami, kataks. 
Billechoola, kai, snow. 
Haeeltzuk, naie, snow, 
Yamkallie, kano, khan. 
San Diego, Calif., xetchur. 

English, snow. 

1. Sam., hawa, hada, juomze, 
kodung (comna, it snows). 

2. Dakota, icamna. 

* Samoyed Grammar, p. 87, omitted in the Dictionary. 


Algonkin, gun ; kwam, ice. 
Knisten., koona. 
Satsikaa, konis. 
Greenland, kannik. 
Severnow, komua, winter. 
Tshuktshi, ukiumi, winter. 
California, yamim. 
Athabasca, thun, ice. 

English, ice. 

1 . Sam., ser, song, sok ; sira, 

2. Dakota, caff a. 
Satsikaa, sakoo cootah. 
Greenland, sermek, sikko. 
Tshuktshi, cikuta. 
Saliaptin, tok, 

English, God, thunder, heaven. 

1. Sam., num, thunder, heaven; 
num, nom, nop, lorn, God. 

Lapponic, jubma, thunder, hea- 
ven ; jubmel, God. 

Finnic, jomala, God. 

Jukagir, jendu. 

Lat. nu bes, nebula, cloud; numen, 

Russian, niebo, heaven. 

2. Algonkin, manitu; but accord- 
ing to Schoolcraft, man, only. 

Onondago, moh. 
Oneida, neeyoh. 
Chocta, shilombish, spirit. 
Umkwa, yaamee, heaven. 
Athabasca, yaha, heaven. 
Kenai, jugan, heaven. 

English, God, idol. 
1 . Sam., nga, kudai, hahe, koika, 

idol ; kolmu, spirit. 
Lapp, gavva, image; vuojgnga, 


Korjak, anggan. 
Juk., koil. 

German, god. Persian, koda. 
2. Dakota, wakaghapi. 
Powhattan, kiwassa, idol. 
Utchee, kauhwu hoo. 
Eskimo, aghat. 

Chocta, (chito)kaka, God (chito 

1. Sam., itarma, spirit. 
Lapp, ittet, to appear. 

2. Greenland, tornak, ghost ; 
tarnek, soul. 

English, fire, sun. 

1 . Sam., tu, su ; tamta 'am, I 
make fire. 

Jem's., ku. 
Ostj., tugit, tut. 
Lapp, dolla. 

2. Dakota, peta. 
Delaware, tendey. 
Nanticoke, tent. 
Etchimin, skut. 
Cree, scout ay. 
Long Island, sut. 
Iroquois, iotecka. 
Muskoghe, totkah. 
Kulanapo, khoh. 
Hichitee, edih. 
Weitspek, okho. 
Attacapa, cam. 

Greenland, ikkuma ; tarkikpok, 

to make fire. 
Athabasca, kon, ku (in eleven 

Tesuque, tah. 
Jemez, twaah. 
Navaho, konh. 
Apach, kou. 
Tsihaili, teekwu. 


Wailatpu, tec. 
California, toina. 
Kulanapo, khoh. 
Tuolumne, wuhah. 
Talatui, wike. 
Pimo, tald. 

English, water. 

1. Sam., bi, wit; bigai, river; bit- 
lorn, birebo, bedeam, drink. 

Lapp, Jielbmaa, river. 
Finn, wesi. 
Juk., use. 
Jenis., wesk, river. 
Korj., mimel ; wejim, river. 
Sanskr. pi ; Gr. vivw ; Lat. bibo ; 
Germ, wasser; Sl&v.pitj, woda. 

2. Dakota, mini. 
Minetari, beedeehee, drink. 
Quappa, nih. 

Osage, nebnatah, drink. 
Delaware, bi ; minatey, island. 
Miami, nape. 
Ojibbeway, sipi, abo* ; minnis, 

island ; sipi, river ; minikway, 

Mississippi and many other local 


Muskhoghe, weway. 
Ahnenin, nitsa. 
Chemmesyan, use. 
Okanagan, utz la hap, river. 
Shoshoni, pah ; itnipi, drink. 
Arrapaho, banna, drink. 
Greenland, sarbak, river ; imiek- 

mok, drink. 
Lutuami, ampo. 

Costano, sii. 
Ruslen, ziy. 
Pimo, suutik. 

1. Sam., ut, water; jaha, djaga, 

Lapp, jokka, river; jugham, I 


Ostj., jeaga, river. 
Jen.,jat, chuge, river. 
Nottoway, joke, river. 
Seneca, uttanote, drink. 
Bodega, duka, water. 
Severnow, aka, water. 
Nay ah o, tonh. 
Apach, toah. 

English, river. 
1 Sam., ky, kuelj, kold^. 
Jenis., dugalno, ul. 
Germ. Qvclle; O.N.(Icd.),*W; 
Eng. well. 

2. Satsikaa, ohkeah, water. 
Pima, vo, lake. 
Natchez, wol. 

Chocta, hucha, kuli, bok. 
Ugalents, kaja, water. 
Greenland, kok. 
Tshuktshi, kiuk. 
Tahkali, a/cox. 
Tsihaili, cuax. 
Lutuami, kohai. 
Tuolumne, kikah, water. 
Talatui, wakaci ; kik, water. 
Kawitchen, Noosdalum, Squalya- 

mish, kah, water. 
Tlaoquatch, aook. 

Dieguno, kha. 

* According to Gallatin (Synopsis, p. 228), is found in the Ojibbeway 
compositions, shominabo, wine (shomin, grape), and totoshabo, milk : 
totosh, female breast. The similarity of this word, which Gallatin declares 
himself unable to explain, with bi, pi, water, is sufficiently near. 

f Samoyed Grammar, p. 64, omitted in the Dictionary. 


In all the seven Californian dialects, of which vocabularies 
are given in the Journ. Geograph. Soc., vol. ii., the syllable xa 
occurs in the words for water, river, sea, lake. The Pujuni 
and Copeh, however, use meny, moni, belonging to the root bi ; 
Dakota, mini. 

English, lake. 

1. Sam., tu, turku, tudjo, tuse. 
Jenis., dee, sea ; kurtju. 
Ostjak, teu. 

Hungar., to. 

Germ, see ; Sanskr. sava, water 
Pehlvi zera, sea. 

2. Dakota, mde. 
Winnebago, tehha. 
Huron, utaw rawya. 
Mohawk, kanyatarle. 
Greenland, tarajok, sea ; tessek, 

In nine Athabascan dialects, to, 


Chinuk, tzalil. 
Tsihaili, taugit. 
Billechula, tzalh. 
Chemmesyan, tzumdah. 
Haidah, shoo ; tungha, sea. 

1 . Sam., jam. 
Korjak, jamam, salt. 
Chinese, yang. 

2. Winnebago, tehchuna. 
Knist., gaming. 

Ojibbeway (Schoolcraft), guma, 


Cherokee, ahmaquaohe. 
Greenland, imak. 
Tshuktshi, imah. 
Wailatpu, jamuc. 
Ugalents, jaa. 

English, tree, wood. 
1 . Sam., paeidu, mati, man. 

Lapp, miestag, tree ; vuovde, fo- 
Korjak, utut, nguft. 

English, wood; Germ, wald, 
baum ; Fr. bois ; Gr. QVTOV. 

2. Dakota (sacred language), 

Knist., mistick. 

Satsikaa, mistis. 

Abenaki, abassi. 

Etchemin, apas. 

Chocta, upi. 

Ahnenin, biss. 

Greenland, orpik, nappo ; masik, 
the cross-beam in a canoe. 

Wailatpu, mos. 

Kalapuya, awatiki. 

Sahaptin, paps. 

English, pine-tree. 

1. Sam., kue, kut, tede, tju. 
Lapp, guossa. 

Ostj., xut. 

Jen., tin; chon, cedar. 

O.N. (Icel.), kvofta, pitch. Com- 
pare Germ, fahte and pech, 
TTLTVS and iriaaa. 

2. Dakota, wazi. 
Delaware, cuwe. 
Narraganset, cowan. 
Satsikaa, toitsha. 
Nottoway, ohotee. 
Chocta, tiak. 
Tahkali, tosse. 
Tsihaili, qama. 


English, stone, mountain. 

1. Sam., pi, fala, fudar, bagir 
mor, bor. 

Lapp, varre. 

Jukagir, pea ; pudan, high. 

Korjak, bukkon, pinugi. 

2. Dakota, paha. 
Chippeway, wudju. 
Wailatpu, apit. 
Shoshoni, tipi. 

1 . Sam., hoi, kawa, ki, sa t ta. 
Jen., kar, kai. 

Juk., kail. 

Korj. guwon. .' 

Chinese, shi. 

O.N. (Icel.), haugr, hill ; Germ. 

hoch ; Eng. high ; Pers. koh y 


2. Dakota, he. 
Otto, ohai. 
Minetare, avocavee. 
Algonkin, hockunk, height. 
Cayuga, kaura. 
Attacapa, kat. 

Chocta, chaha, high. 

Caddo, hio, high. 

Greenland, kakak, karsok. 

Haeeltzuk, koquish. 

Tsihaili, ckom. 

Jakon, kwots. 

Shoshoni, kaiba. 

Several Califor. dialects, haix, &c. 

Straits of Fuca, govachas. 

English, earth. 
1 . Sam., mou ; njanga, clay. 
Lapp, nane, mainland. 
Korj., nutenut. 
Teuton, land. Compare the word 

God for the transition from n 

to I. 

2. Dakota, maka. 

Yankton, mongca. 

Seneca, uenjah. 

Chocta, nunne, hill. 

Greenland, Kodjak, Tshugats, 

Tshuktshi, nuna ; marak, (Gr.) 


Athabascan, ninne. 
Umqua, nanee, noc. 
Navajo, ne. 
Jecorilla, nay. 
Tesuque, nah. 

Kenai, altnen, alslin, alshnan. 
Atnah, nann. 
Koltshani, nynkaket. 
Kolosh, llen-ketaanny . 

English, bird. 

1. Sam., kus. 
Lapp, cicas. 

2. Dakota, zetkadang. 
Massachus., psukses. 
Iroquois, tshigasko. 
Chocta, hushi. 
Adahi, washang. 
Tahkali, ogaze, eggs. 

Tlatsk., tshiasi; wo-skaiake, egg. 

Kenai, kakashi ; kgasja, egg. 

Atnah, tshetsha. 

Ugalents, kan-ny ; kota-ut, egg. 

Koltshani, tshoje. 

Haeeltzuk, Billechola, tseco. 

Chemmesyan, tzotz. 

Aleutian, cissu. 

Nootka, akutap. 

Kolosh, kot, egg. 

Shoshoni, kasa, wing. 

Jakon, kokoaia. 

Straits of Fuca, ucutap. 

Salish, tlasqoqa. 

Sahaptin, kakia ; kotkot, feathers. 



English, egg. 

1. Sam., eng, eang. 
Jen., eegh, eng. 

2. Dakota, witka. 
Delaware, waTih. 
Oneida, onhoncons. 
Umqua, exa. 

English, dog. 

1. Sam., kanak, wueniuk, baggeo, 

Lapp, bana, b'ddnag. 
Chinese, kan. 

Sanskr. basa ; Eng. bitch ; Russ. 

2. Delaware, mekanne. 
Huron, gaguenon. 
Greenland, kemmek, kemmo. 
Kawitshin, Squalyamysh, skomai. 
Chinook, kamokus, 

Atnah, Noosdalum, scacah. 
Haidah, watts. 
Lutuami, watsak. 
Palaik, watsaga. 
California, wasi. 

English, duck. 

1 . Sam., njaby, sipa (related to 
bi, water, andjabidm, drink). 

2. Dakota, skiska. 
Knist., sisip. 
Ojibbeway, shisip. 
Tuski, cikuta. 
Tsihaili, sistxlom. 
Chocta, shilaklak, goose. 

English, fish. 
1. Sam., kole, hale, kuel. 
Lapp, guolle. 
Juk., olloga. 
Korj., kokajalgating. 
Jen., ilti. 

Ostj., xutj. 

2. Dakota, hoghang. 

Knist., kenose. 

Ojibbeway, kikon. 

Mohawk, keyunk. 

Oneidah, kunjoon. 

Chocta, kullo, garfish. 

Muskoghe, tlaklo. 

Cherokee, agaula, perch. 

Greenland, aulisegak. 

Eskimo, khallu, 

Athabasca, Dogrib, cloua. 

Tahkali, cloolay -, tallo, salmon. 

Kutchin, tleukhko. 

Tlatskanai, selokwa, salmon. 

Urnqua, txlee, salmon. 

Kinai, tluka-, Atnah, Ugalenz, 

Inkilik, Inkalit and Koltshan, 


Tsihaili, kaixalis. 
Palaik, alls. 
"Wailatpu, waibalf. 
Noosdalum, chaaloh. 
Cathlascon, call a. 

English, flesh, meat. 

1. Sam., wati. 
Lapp, oadze. 

2. Ojibbeway, wiyas. 
Arrapaho, wonunyah. 
Miami, weensama. 
Iroquois, owachra. 
Natchez, wintse. 
Greenland, uinek, nikke. 
Loucheux, beh. 
Athabasca, bid, bet. 

English, ptarmigan. 

1. Sam., kafe, hondie, aba. 
Chinese, ko, pa, pigeon ; Latin, 


2. Dakota, wakiyedang, pigeon. 


Narragansett, wskowan, pigeon. 
Nanticoke, pakquun, turkey. 
Satsikaa, katokin, partridge. 
Oneida, oquas, partridge. 
Chocta, kafi, hen, quail ; fakit, 


Muskoghe, kowyguy, partridge. 
Adaize, owachuk, turkey. 
Greenland, kauio, ptarmigan. 
Sahaptin, kuinu, pigeon. 
Chinook, kaxamau, pigeon. 
Shoshoni, ihooc. 

English, reindeer. 

1. Sam., tho, ta, ty. 
Jen., dsol. 

Greek, 6>)f> ; Eng. deer. 

2. Dakota, ta (the moose and any 
ruminating animal). 

Winnebago, ca, deer. 

Cree, attik, deer. 

Massachus., attuk, deer. 

Illinois, moussoah, deer. 

Menomini, upahissaoh, deer. 

Satsikaa, hipasto. 

Seneca, chinnoundoh, elk j nin- 

dunhe, moose. 
Muskoghe, itzo, deer ; ponatta, 

wild beast. 
Chocta, issi, deer. 
Natchez, tza, deer. 
Caddo, dah, deer. 
Hichithe, echu, deer. 
Riccaree, w at ash, buifalo. 
Greenland, tukto, reindeer. 
Athabasca, Dogrib, edthun, et- 

thin t reindeer. 

Kutchin, bitzey, reindeer. 
Tahkali, yestshi, reindeer. 
Tlatskanai, tshesle, deer. 
Umqua, intshi, deer. 
Kenai, motchish, deer. 
Koltshani, batshich, deer. 
Kolosh, tave, watzich, deer. 
Kawitchin, tla, deer. 
Tlaoquatch, tloq, deer. 
Salish, atsulia, deer. 
Tsihaili, toixa, buffalo. 
Kitunaha, tsopokai, deer, 
Sahaptin, tatapai, deer. 
Kalapuya, atalim, deer. 
Palaik, tasi, &c. 
Haidah, cisk. 
Klikitat, cato. 

English, night, evening, dark. 

- paebi, faemei. 

- ud, nodi. 
Ostj., idai. 
Korjak, tyngfouty. 

2. Dakota, tpaza, dark ; htayetu, 


Ojibbeway, tepikat. 
Ottawa, tepik. 
Abenaki, pesede, evening. 
Delaware, nepawi, in the night. 
Menomini, pekotek, in the night. 
Utche, pato. 
Chocta, opia. 
Lutuami, psin. 
Athabascan dialects, tata, tac, 

&c. . 

If, by a more intimate and accurate knowledge of the lan- 
guages and social manners prevailing among the aborigines of 
Siberia and North America, it shall be ultimately proved, as 



has been here merely shadowed forth from a comparison of 
those few languages that have as yet been scientifically 
described, that the American Indians are certainly settlers 
from Asia, it is evident that this emigration, rather than colo- 
nization, must have happened in an age of the remotest 

The manners of the Americans at the time of the discovery 
of the continent by Columbus prove this beyond contradiction. 
They had no cultivated vegetable, they had no domesticated 
animal (except perhaps the dog), which belonged to the Old 
World, or which was in use among the nations of Asia and 
Europe. ' The Americans were either hunters without agri- 
culture and without cattle, or if they had either, it was 
evidently (as the llama, the maize, the potato, the tobacco, 
&c.) of American origin. The adaptation of these natural 
productions of the western world to the uses of a more im- 
proved civilization, must then have been a native invention. 

But it is altogether improbable that the emigrants from 
Asia should have neglected or forgotten to carry along with 
them their most valuable property. And even if it had been 
impossible for them to transport across Behring's Straits, 
corn, the ox, the sheep, and the hog, they would have preserved 
the idea of the importance of these primary and principal 
inventions of humanity. In the new country they would have 
tried to domesticate the buffalo and the Californian sheep, or 
the peccari ; a thing that appears to be by no means difficult, 
according to the experiments that have been made with some 
of the animals proper to America. 

But nothing of this kind happened. The inhabitants of 
the whole northern continent down to Panama, at the dis- 
covery in 1492, were without any domestic animals at all, 
except the dog. Agriculture only existed in the most 
southern and fertile regions, consisting evidently in produc- 
tions originally tropical, which had spread from thence to some 
of the more northern tribes. 

The American civilization, then, as far as it went, evidently 
was a production of the native mind, developed in America by 
the innate resources of the human intellect, the discoveries to 


which, they led being a proof that the mental powers of the 
Americans are similar to those of the nations in the Old 
World. Both showed a capability to adapt the resources of 
their peculiar climate and soil to the particular uses of man. 

The immigration from Asia, then, happened before the 
domestication of cattle and the cultivation of corn. It was 
anterior to both the nomadic and the agricultural state of 

Yet even then arts existed; and mankind was widely 
advanced from mere animal existence. Even in these arts 
there is a marked similarity. The rude implements and arms, 
wrought principally of stone, that are dug out of the tumuli 
in the north of Europe, closely resemble the instruments 
made use of by the Eskimos and other American tribes when 
first they became acquainted with Europeans. This likeness, 
however, may be considered accidental, or as a proof of the 
identity of the human intellect; the same inventions being 
made by different men under the same circumstances. , 

But here comparative philology assists in solving the pro- 
blem. If it is found that there also exists a similarity in 
the name, then this combination of accidents or coincidences 
becomes an improbability amounting to an absurdity. 

The existing vocabularies offer but a few words belonging 
to the arts, but the arts of savages also were few ; yet some 
of the most important are noted down. 

The boat is called by the Samoyeds ngano, ngandui, angi, 
anzej Lap. vanas, vantsa; Jukagirs, acel- } Korjaks, agwat. 

In America there is Dakota, wata; Algonkin, amochol, 
aguiden-j Iroquois, kauuwau; Eskimo (Greenland), kajak, 
umiak ; Chinook, ikanewe ; Shasti, ikhui ; Chimmesyan, 
nohwio-, Klikitat, wassas ; Cathlascou, cunaim; California, 
waxat. The word canoe, adopted by the early navigators 
from an American tongue, corresponds closely with the 
Samoyed ngano. 

The house is called by the Samoyeds ed, ede r iede, but the 
fuller and more original form is probably the Lapponic viesso, 
goatte, that also occurs in the Samoyedic koac, kuac (village) ; 
the Ostiak xat ; the Jeniseyan khus, hukut, corresponding to 


the English cot, cottage, hut ; Sanskrit kuta, kota. The same 
word is found in the Dakota wizi. The common American 
expression wigwam is from the Algonkin, but includes probably 
the possessive his. The proper form is igwam, Cree igi, Mas- 
sachusets ik, Chippeway ainda (home) ; in Eskimo (Greenl.) 
iglo. Another Greenland form for the same idea, inne, 
resembles also the German in ; Icel. inni ; Kolush, it ; Lou- 
cheux, jetz ; Oregon (Paliaik), tsitzu ; Nootka, mukati ; Tsi- 
haili, xax j Haeeltzuk, gook-qua (house), gookquilla (village) ; 
Klikitat, coosie ; Kalapuya, keowtan ; California, ketcha, kivit ; 
Athabasca (Tahkali), kux, yah, yok, cooin-, Shoshoni, kuo. 

Another Samoyed word for the tent or house is mat, mea, 
ma, meaja ; Jukagir, memd, Tskuktshi, mautaak. This appears 
in the Algonkin (Blackfoot or Satsikaa), muyai-, Athabasca 
(Tahkali), m; Oregon, Shasti, oma; Nootka, mukati ; Haidah, 
natee; Tlaoquatch, maas; Umqua, ma. 

A Samoyede word for village, town, is kera, kereme, and 
talo, related to tura, chamber; Lapp, dallo, garden; Finn, 
kartano', Jeniseyan, kelet; and khus, hukut, house; Turkish 
and Jen. tura ; Hungarian, kert, garden. In the European 
languages, Scand. gard-, Fr. court-, Ital. corte; Lat. hortus ; 
Rus. gorod*. 

In the Sioux family there is Quappa ton, Osage, towah ; in 
Algonkin, Narragansetts, otan; Muskhoghe, talofah; Sahaptin, 
tlaknit ; Billechola, Haidah, shoolh ; Chimmesyan, wul ; Yam- 
kallie, kulha, house; Kawitchen, kueh, tola, Urns; Athabasca 
(Tahkali), tlane. 

The word for door is connected with this Sam. ngoa, mada, 
muada; Iroquois (Seneca), kawhoah-, Caddo, duswatcha ; Es- 
kimo, matto ; Oregon, Billechoola, mum ood ota ; Tlaoquatch, 

From the earliest discoverers and navigators, it is sufficient- 
ly known that the more southern, and also more improved, 
tribes of North America, worked metals for implements and 
ornaments, the quantity of gold in use among the natives 
being in fact the principal source of admiration to the Spa- 
niards among the wonders of the New World. But it is 
* See Wedgwood on the Finn and Lapp, Philol. Soc. Trans. 1855, p. 6. 


equally certain that this knowledge was extended up to the 
very northernmost tribes, whose manufacturing industry was, 
however, from natural causes, chiefly confined to the copper 
found among them in an almost pure state. It thus becomes 
quite as reasonable a course to derive the employment of 
metals among the Athabascans, Eskimos, and other northern 
tribes, from Asia as from Mexico. The word tagai was em- 
ployed by the Samoyedes to denote a metal. But which ? 
Iron was not the oldest. In several languages a word of the 
same root denotes both iron and copper : 

Kolc'an. Atna. 

Iron. . . . cacej .... ketic. 
Copper. . clean .... cety. 

In Dakota all metals are called maza, and merely distin- 
guished with adjectives: maza-sapa (black m = iron) ; m-ska 
(white m= silver) ; m-sa (red m= copper), &c. 

In Collinson's Account of the Proceedings of H.M.S. En- 
terprise (Journ. Roy. Geog. Soc. 1855, p. 201), it is stated of 
an Eskimo tribe, " this being their first communication with 
white men," " very few iron implements were found among 
them, the most warlike being a spear-shaped knife made of 
native copper." Holmberg states (p. 27) that the Kolosh 
(Thlinkithi), before they were acquainted with Europeans or 
the use of iron, knew how to manufacture the native copper 
which is thrown out by the Athna or Copper River, and even 
now is highly valued by the Indians. 

Iron is called by the Samoyeds, basa. In Finn, s'dppi 
= smith; in Dacota, maza= metal; in Upsaroka, mitsce; in 
Minetari, matsi = knife ; Narraganset, mowashuk (knife) ; Sus- 
see, marsh (knife) ; Athabaskan, bes (knife) ; Lutuami, wate 
(iron and knife) ; Fall Indians, warth, knife ; Ahnenin, wa- 
hata, knife. 

In Eskimo, savek = iron and knife, sabbiortok = smith ; 
Tshuktshi, cepiak= knife; Choctah, bushpo. 

But the knife was originally made of stone. Hence in 
Samoyed it is tagai, har, falli, laku, kales ; Yenisean, ton 
Korjak, waiia ; Ostjak, kedje ; Madjlar, kes. 


Among the Dakotas the sword is called maza-sagye (maza 
= metal); the knife by the Algonkins (Micmac), wag an; 
Athabascans (Tahkali), texe ; Kinay, kissaki ; Loucheux, tlay ; 
Oregon (Sahaptin), wals, tekek, and tsaise (arrows) ; Haeltzuk, 
taio ; Okanagan, tzuk, (arrows), tuchte; Muskhoghe, islelafka ; 
Klikitat, techy e (spear), tooks (iron). 

The very interesting question, what nations in America 
were acquainted with the working of metals when they left 
the shores of Asia, might thus be brought nearer to a con- 
clusion, if we knew in their languages all the synonyms for 
the words smith, metal, iron, knife, sword. 

The axe, hatchet, is an implement formed of stone by the 
rudest savages and the oldest nations, as is sufficiently proved 
by comparing the arms from the South Sea with those that 
are dug out of tumuli. Its name among a great many nations 
is evidently related to this word for knife. 

Among the Samoyedes the axe is called sumba, tubka; 
Jukagirs, numundzhi ; Yeniseyans, tok, cok ; Ostjakians, tajem; 
Algonkins, togkimk, tamahican (thence the word tomahawk) ; 
Iroquois, atauhoiu ; Woccons, tauunta ; the Eskimos (Cukc'i), 
kalkapak-, the Tahkali, seradY/; Neatku, tawish ; Tsihaili, tlomen. 

To make a sort of thread out of the sinews of animals is an 
art as old as the bowstring and the attempts to form a dress 
out of the skin of an animal. 

To spin, is in Samoyede panau; Lapponic, banam; Ostjak, 
puntem; Finnic, punon; Hungarian, fon. In Dakota it is 
pahmung ; Eskimo (Greenl.), perdluk, nugit. The word shows 
great similarity with the English spin, and seems to be derived 
from pa, finger (the thread being made by fingering) *. 

The awl, also used as a needle or a gimlet, must be as old. In 
Samoyede, parte ; Ostjak, por,par; Finnic, pura; Hung.fura; 
Lappon, bogham-\ (to make holes) ; in Eskimo (Greenl.), putlout. 

* In * Det Norske Sprogs vsesentligste Ordforraad ' of C. A. Holmboe, 
Wien, 1852, sub voce spana, it is shown that this word is connected with a 
great many having the signification to stretch, to dilate. The common 
origin may be as before stated,* the finger being the plainest of all these 

t Compare the German bohre, bohr ; Lat. forare ; Gr. irfipa 
Tropos), hindust. phorna. 


If the Dakota word ta-hing-spa had been found only in a 
common trades vocabulary (and there, perhaps, mis-spelt), 
it would have been considered as a most convincing proof of 
the radical dissimilarity of the Siberian and the American lan- 
guages. Now, as we fortunately have an excellent dictionary 
of the Dakota, we see that the word ta-hing-spa is a compo- 
sition of ta-hiny, buffalo or deer's hair. Ta-hing-spa then is an 
awl for making holes for the thread of deers' hair ; spa alone 
is then the proper word for the awl, and sufficiently like the 
Asiatic appellations. The first part of this composition ta, 
deer, is also pure Siberian, being the Samoyed name for the 
rein-deer. By a further composition, the Dakotas, of this 
word, form ta-hing-spa-cikaday (little awl needle). 

A bag is, in Samoyed, ngaese, koza, pad, foadac ; Dakota, 
unksu, wozuha ; Cree, wut', Greenl.^o^;; compare Old Norse 
(\.CQ\.)poki,posi,pungr' } Swed.ficka; Chocta, bahta; English, 
bag, pouch (poke) ; French, poche (pouque) . 

That the religious ideas of the North American Indians show 
a very close similarity to the system of conjuring or witch- 
craft (Shamanism) known to prevail among the Siberians 
and all the other Finnic tribes, will be conceded on a com- 
parison of the best accounts we have got of them. The con- 
juror (Shaman) is called, by the Samoyedes, abes, tadibea ; 
in Dakota, wapiye sa-, in Koltshani, tizenne. Among the 
Greenlanders, the most common name was angekok, derived 
evidently from the root ang, signifying old, in a great many 
of their words. But another name for their priests was 
tarajok (derived from tak= darkness, and tar ak= shade). If 
the Samoyed tadibea is derived from Sam. tasi, below, the 
ideas are as similar as the words. 

Although the life of savages in general, and most particu- 
larly that of the Indians of America, is commonly considered 
as the very ideal of liberty, yet it is sufficiently certain that 
domestic servitude, in its worst form, exists in full develop- 
ment among many of even the rudest tribes, and seems to be 
known to all; although the number of bondsmen may be 
comparatively small, both because there is little use for them, 
and because most of the prisoners of war are killed from brutal 
revenge. In the Finnic Kalevala, slaves are mentioned as 


anterior to the conversion of the Finns to Christianity. Slaves 
are found among the Siberians. In North America this 
system has been developed to a considerable extent among the 
Kolosh and Konjagis on the Pacific coast. That it is common 
to all the nations in question is sufficiently apparent from their 
languages. The slave is called, by the Samoyeds, kadsh, kotie, 
tidio, tandiaa, habi; the Lapps, goccostak (a servant); Da- 
kotas, (wica) toka, htani (work) ; Eskimo, kevgal; Green- 
land, kiggak , Konjav, kajur ; Kolush, kux ; Haeeltzuk, kagh- 
kah; Chimmesyan, uchack; Tunghaase, kooch; Iroquois, 
wawun teotaut (labour). 

The words that have thus been found to present similarities 
between Asiatic and North American languages (fire, metal, 
copper, knife, axe, boat, house, tent, village, door, spin, awl, 
bag, dog, slave, God, priest), relate to arts, institutions, and 
superstitions, of which no Asiatic or North American tribe is 
ignorant. To this series of similar manners may be added 
some others, which, if not so generally prevalent, yet offer a still 
more singular coincidence, as being more arbitrary and being 
far from necessary to the well-being of man, rather prejudices 
hurtful to national improvement and individual happiness. 
To this class of similitudes belong the clanish institution 
and its concomitant badges and signs ; the tattooing, the era- 
dication of the beard, the shaving of the head leaving a single 
lock of hair, and the prejudice against using milk for food*. 
From the contemporaneous prevalence of such absurd customs 
among hundreds of nations and millions of men on both sides 
of the Pacific, it may be still safer to argue to a common 
descent, than from natural workings of the human mind. If 
direct evidence fails for a historical fact, and recourse is to be 
had to circumstances that can only be explained by assuming 
the fact, then the mass of circumstances is no more important 
than those vestiges that are so arbitrary that they exclude the 
hypothesis of simultaneous hitting upon the same idea. To 
this class belong both similar sounds and arbitrary fashions. 

The circumstances we know of the civilization or the bar- 
barism of the American aborigines point then irrefragably to 

* ' On the Liberal Government of the Ruder Nations,' by Lewis K Daa. 
New Edinb. Philos. Journ. v., 1857. 


its author and to the place whence it is derived. As for the 
time when the immigration took place, the low grade of all 
the Indian tribes putting aside the Peruvians and Mexicans 
proves that the settlement must have happened in the very 
remotest ages. Yet from this very reason it cannot be viewed 
as a premeditated and combined national effort, but as a suc- 
cession of wanderings across the ocean bridges the Aleutian 
Islands andBehring's Straits by single individuals or families, 
partly the effect of accident on hunting or fishing expeditions, 
and partly of expulsion in wars. The infinite multiplicity of 
nations in America and their mutual hostility, as well as the 
surprising paucity of their numbers, equally correspond with 
this view of the subject. 

It is then sufficiently probable that the grouping of the 
Indians in well-defined ethnological families, an examination 
of their manners at the commencement of their contact with 
Europeans, and of their language, will lead to important histo- 
rical conclusions. It may be allowed me here merely to hint at 
two obvious facts ; that the lowest savages unacquainted 
with houses and garments are found in South America only, 
in Brazil and Guyana, farthest off from Asia ; and that the 
fishing tribes that border the Arctic and Pacific oceans from 
Labrador to Oregon the Eskimos, the Athabascans, and 
their kindred being in the closest contact with Asia, are 
also the most improved, if we take into account their hard 
climate. The Kolush, for instance, do not appear to be 
inferior in any way to the Asiatics of Kamschatka and its 
adjacent islands. Does not this observation point out the 
beginning and the end of the immigration ? 

Yet even the latest of these settlers has not arrived at the 
pastoral development of the Samoyedes, but stops short on a 
still more primaeval grade. Why then did not the immi- 
gration of Asiatics into America continue after the acquisition 
of a higher culture ? 

The answer to this question is twofold. First, there never 
was founded on the eastern coast of Asia any trading or con- 
quering state, that was inclined to make the discovery of, and 
to form regular settlements in, a foreign land of which a vague 
report only might be heard. As for the accidental drift- 


ings over of fishermen and hunters, as well as the pushing 
forward of the nearest Aleutian islanders, these additions to 
the population of America were met with a great obstacle 
when the coast and even the interior wildernesses were in a 
manner occupied by a set of cruel possessors or claimants of 
the soil. The more the older nations of America multiplied 
by their own increase, the greater became the chance that 
any new-comer would be exterminated on landing, or perhaps 
adopted into an existing tribe, and thus leave no trace of a 
peculiar nationality after him. 

That the tribes of New England ultimately repulsed the 
attempts of the Norwegian discoverers of Finland to settle 
on their territory, is a well-known proof of their ability to 
resist a small band of colonists, or in fact any that did not 
either adopt the roving habits of the natives, that offer 
some chances of escaping, but also of becoming a savage 
tribe ; or on the other hand, who did not establish themselves 
in fortifications impossible to take by assault or surprise. 

The chief languages collated are the Samoyed and Dakota. 
To the comparison of these two, the other languages, less fully 
illustrated, and less sufficiently known, are subsidiary. In a 
paper published elsewhere, the Asiatic affinities of the Atha- 
baskan tongues, interjacent to the Dakota area and Behring's 
Straits, are indicated. They are as decided as the preceding. 
If the data were equal, they would probably be more so. The 
evidence, too, of the numerals is omitted, forming a separate 
notice, involving certain points of criticism, the exposition of 
which would be extraneous here. 

CHERTE : have in cAerte=hold dear : 

Thou comyst to late, for gadryd up be 
The most fresh flourys by personys thre 
Of which tweyne ban fynysshed here fate, 
But be )>rydde hath datropos yet in cherte 
As Gower, Chauncer, and Joon Lytgate. 

Bokenam, Lyvys of Seyntys (A.D. 1447), p. 117. 

Fr. avoir quelqu'un en cherte, cierte, avoir cher (Burguy Gloss, and 
Gram, de la Langue d'Oil, i. 278) : 

Je ne t'ain [aime] tant ne tant n' ai en cierte 
Que je te die mon cuer ne mon pense. 

La Chanson Ogier de Danemarche, par Raimbaut de Paris (Paris, 1842), 
vv. 8786, 8787 


[Read February the 22nd*.] 

In my paper on English Diminutives, which in its first 
sketch was intended solely as an introduction to the Dimi- 
nutives of the Latin language, some advantage was found in 
starting from the Gaelic suffix ach or ag, or, to combine both 
in one earlier form, agh. The very nature of this medial 
aspirate, scarcely belonging to articulate sound, accounted in 
some measure for the great variety of forms by which it was 
found to be represented among ourselves. Passing through 
och, ock, ow, ick and ie or y (lass-ock, lass-ow, lass-ick, 
lass-ie, lass-y), it subsequently appeared in almost every 
variety that our alphabet can denote. This of course was 
startling, but the strangeness of the fact might perhaps have 
been in part accounted for by the subordinate character 
of the syllable both in position and in power. The latter part 
of a word is naturally liable to a less careful pronunciation ; 
and even if the suffix had always preserved its definite meaning, 
that meaning would have been of less moment than the 
leading idea to which it was attached. But in truth the 
diminutival character of the suffix was often lost sight of, 
especially in the numerous cases where the primitive had 

Now the law of language, for it seems to be a law by 
which the simple substantives are supplanted by derivatives 
originally coined for the purpose of denoting diminutives, 
obtains on Italian ground to an extent not surpassed by any 
other nation. The ideas, to take a familiar example, of bro- 
ther and sister, can be expressed solely by what are clearly 
diminutives, fratello and sorella. 

But can we reasonably adopt for Latin the same course of 
argument which served our purpose in the treatment of lowland 

* After this date the paper was recast, and read at three of the Society's 
Meetings in 1857. Several alterations and additions have since been made. 


Scotch and English ? The Gael is the immediate neighbour 
of the Scot, but has always been too far distant to have had 
any direct influence upon the language of the Italian penin- 
sula. This is true : but on the other hand, in the early ages 
of Roman history, a Gallic nation held undisputed possession 
of a large portion of that country, and has left to this day 
a record of the fact in the name of Senegaglia. Moreover it 
is now an admitted truth, that there existed a strong affinity 
between the Keltic and the classical languages ; and this affi- 
nity, it is believed, must not be limited to the vocabulary or 
roots of the language, but will be found in the details of 
structure, by which, from common roots, the longer words 
were built up. 

No doubt there are broad distinctions between the Keltic 
languages and the soft dialects of Southern Italy. This very 
syllable agh for instance, is one which a Roman of old would 
have found it as impossible to pronounce, as a Southron in 
England of the present day. The Roman indeed dealt little 
in aspirates. His h was but a symbol, devoid of all living 
power, and his/, something different from a Greek <, was 
all but limited in use to the initial place in a syllable* ; while 
the sounds for which the Sanskrit alphabet possessed simple 
definite characters, but which we can only denote most awk- 
wardly by gh and bh, were alike strangers to the eye and to 
the ear of both Greeks and Romans. Had fate handed down 
to us specimens of Latin, as spoken in ancient Etruria, we 
should probably have had a much rougher specimen of the lan- 
guage, than we find in Cicero and Virgil ; and the contents of 
this paper might have been matter so patent to every scholar, 
as not to need discussion. As it is, 1 must request a patient 
hearing of the whole paper before an opinion on its truth is 
finally adopted. 

But if the Keltic languages are so loaded with gutturals 
and aspirates t as to have in sound little that is common to the 
soft and harmonious Italian, is it not perverse to commence 

* Rtf-us is an exception, but here also we have rub-er. 
t It were much to be wished that the orthography asperate could be 
reestablished in its rights, as against the usurper aspirate. 


the inquiry on that side ; and above all to take as the starting- 
point a syllable which was confessedly unpronounceable by a 
Roman ? The answer is, that this is the very reason why I 
select a Keltic form of the suffix. It is in the rougher va- 
rieties that the earlier forms of language are found, but though 
the asperities are smoothed down in the later and more cor- 
rupt dialects, traces are often left, which are only intelligible 
when compared with the more rugged specimens. The soft 
sounds heard in Clovis, and still more in the modern Louis, 
give us a very different and a less exact idea of the genuine 
w r ord than Chlodovicus, Ludovicus, and Ludwiff. Similarly 
Merove'e, as the French write it, was the founder of the Mero- 
ving-ian dynasty, and called (says Sismondi) by his Teutonic 
countrymen Meer-wig, ' the warrior of the sea/ An English- 
man, like a Roman, is apt to dispense with many guttural 
sounds. He deals in such words as the substantives way, day, 
honey ; the adjectives any, manly ; and the verbs may, slay, 
lie, see. He writes, but only writes, the aspirated guttural 
of Armagh, Youghal, Brougham, Strachan, might, slaughter, 
though. At other times, while writing gh, he substitutes the 
sound of a labial aspirate, in lieu of what is too rough for his 
throat, as in laugh, cough, rough. Meanwhile there are kin- 
dred languages which retain in the representatives of our soft 
words all the original asperity, as iveg, tag, honig-, einig, 
mannlich; mogen, schlagen, legen, sehen', schlacht, doch, la- 
cheln, &c. Nay, at times our own language in a derived form, 
restores in some degree the consonant which the simpler word 
has discarded, as Norway, but Norweg-ian. 

Precisely in this way I hope to show, by the evidence of 
kindred languages, and by the fuller forms of Latin derivatives, 
that the suffix agh, or something near it, must have belonged 
at one time to a very large number of Latin substantives, 
adjectives, and verbs. The exact form indeed, the very 
letters agh, one can have no hope of presenting ; and we must 
also recollect that the Latin, like our own language, has a 
strong tendency to suppress a guttural at the end of a syllable. 
As we say may, for a verb whose stem is really mag-, so the 
Romans, who possessed the very same root in the adj. magnus 



and maximus, had no trace of the g in maior (pronounced 
mayor), or in mavolo and malo. Still we shall not unfre- 
quently find some guttural in the words which contain our 
suffix ; and even when it is no longer visible, I hope in many 
cases to trace it by the evidence of derivatives. 

II. AGH, as seen in Latin Substantives. 
A first search for agh* in the Latin vocabulary has no great 
appearance of success, for we can lay our hands on but two 
examples, limac-^ a slug, and fornac- a furnace, or perhaps 
at first rather a melting-pot. But this deficiency will be fully 
compensated before long ; and in the interval we may point to 
the abundant supply of examples which are found in the sister 
tongue ; examples too in which the diminutival power of the 
suffix is self-evident. (See Pott, Etym. Forsch. ii. 506, 507.) 

(TTpuTv\\ctK-,m. a general en petit. 
Ko\dK- t m. flatterer. 
7rXdfc-,y. a flat; (for TraX-o*-?) 
a splinter. vXdic-, m. a barker = whelp ? 
nv\a.K-, or aXoK'-, f. a furrow. 
6v\dtc-,f. bag, pouch (Ilesych.). 
cfKv\a.K-, m.f. young dog. 
pvXdx:-, m. mill-stone : cf. yuvX/. 
tyvXuK-, m. guard. 
/3wXdc-,/. clod : dim.of/3wXo-,m. 
-, m.f. a pole. 
:-, m. a mat. 

,/. meadow. 
K\ifjiaK-,f. ladder. 

a/3d/c-J, m. a slab. 
Trila.K-,f. a spring. 
Oiate-,, a lettuce. 

o-, n 

7ruj'dfc-, m. bottom. 
po$a.K-,f. dwarf-rose : c 

. a small stone : cf. \i6o-. 

K-, m. child of a Helot. 

- 1 f. pomegranate : =(oot . 
<77raXdi>, m. mole. 
<7a\(k-, m. miner's sieve. 
<TjutXa/c-,/. yew-tree. 
TT aXXd/c-, tn.f. youth, maiden,Eust. 
/ueXXaic-, m. a youth: cf.fjietpa.K-. 

* Ach, ag, an, and ag-an, are diminutival suffixes of Gaelic substantives ; 
ach of Gaelic adjectives ; ach and ig of Welsh substantives ; agh of Manx 
consuetudinal verbs ; ek of Breton adjectives, ik, ig, of substantives. 

t This word in its first syllable no doubt containing the same element 
as lim-O' mud, leim, German, glue, lime Eng., employed in making the 
sticking material mortar, and slime. 

J These words are arranged alphabetically, according to the final let- 
ters, an order which is always to be preferred where suffixes are under 
examination. The list is derived chiefly from Hoogeveen and Liddell and 
Scott. But see also Lobeck's ' Paralipomena,' p. 275. 

See below, XI. 



ep/iaK-,/. pi. heap of stones. avQpaK-, m. charcoal (piece of?). 

a small raised place : axvQpaK- or 1 
v f > m. a youngster. 

dim. Of /3fa>JUO-*. (TKVpOctK- J 

fjieipaK-, m.1 f. boy?, girl. 
KopaK-, m. a raven, crow, 
rerpdio, m. a kind of grouse. 
:-, m. a shrew-mouse. 

, m. lower spike of a spear. 
tTTvpaK-,f. the storax-shrub. 

, f. a seat : dim of dujtpo-, 

a chick. 

TTTa.K- t f. a hare, &c. : =7rrwK-. 
iropra.K-,f. a calf: cf. Troprt-,f. a 


paaTOK-tf. mouth, &c. Compare 
JUI/OTOK-, m. upper lip, &c. 

, m. a stand. 
-, m. a sea-muscle. 
pvdx-, m. a stream. 
(TVO.K-, m. a kind of pulse. 
deXtyaK-, m. f. a young pig. 
o/Li0ao, /. an unripe grape, olive, 

The list just given is confined to words in which the suffix 
arc is known or believed to have a short vowel. It might be 
largely increased by words equally available for our purpose, 
which have a suffix CLK as oid/c- m. a tiller ; or dy as \ardy- f. 
the splash made by drops of wine ; or dy, as pdy-^ a berry. 
And over and above these, there are derivatives from nouns 
in ate, &c. which have been superseded by tertiary forms de- 
rived from them, as pavvaK-to- n. a little necklace, /ca^a/cio-, 
n. and tcatyaica- (nom. Ka^a/c'r]^} m., beside a simpler tca^a- 
a box or chest ; a-copafco- m. a basket ; /SaTiatcy- a sort of cup ; 
TriOa/cva- or <f)t,Barcva- f. a wine-jar ; beside TnOo- m. the same : 
in fact, all these words seem to guarantee the previous 

* See below, XI, 

t A word beginning with p has always lost one or two preceding let- 
ters, so that even in this apparent monosyllable dy is but a suffix. 

- 1 , , 

V wz. a rocky place. 

'- J 

Tru'ttic-, m. a plank, board. 
f)piva.K-,f. or TptvaK-,f. a three- 
pronged fork. 
0w^oik"-, m. small thunny-fish : 

dim. of Qwvo-*. 
Soi'aK-, m. a reed. 
\apvaK-, f. m. a coiFer, a box. 
K\uvaK-, a young shoot (Hesych.) : 

dim. of K\d)v-. 
(3oaK-, m. a kind of fish. 

. a woodcock. 
. a shrub : PWTT-, a shrub. 
-, m. a pitch-plaster. 
-, vn.f. a pointed stake. 
fipaK-, m. a young animal : 
dim. of vefipo-, a fawn*, 
m. a clenched hand. 
. a blister, a pimple. 
. a horn (Hesych.~) . 


existence of the simpler /u-avva/e-, /ccnjra/e-, o-wpaic-, fiaria/c-, 

But are there any Latin substantives which exhibit the 
suffix agh, shorn of all trace of both aspirate and guttural ? 
After much reflection on the subject, T venture to affirm that 
nearly all the words which constitute the first declension 
come under this head; but the assertion is one which can 
only hope to obtain the conviction of readers, when it has 
been well supported by argument*. 

I have already pointed to our own words way and day, as 
having lost a final guttural. But these very words are repre- 
sented in Latin by via- and die- ; or we may even say dia-, 
considering the form of Dia-na- ' the goddess of light/ and 
the adjective quotidia-no. 

These form, no doubt, but a slender basis for my argument. 
Let it next be asked whether a consideration of the adjectival 
forms lig-neo-, made of wood, pic-eo-, made of pitch, does 
not justify the belief that a final guttural once attached itself 
to the four-and-twenty nouns of the a or first declension, 
whence are formed the following adjectives : 

fab-ac-eo-. viol-ac-eo-. lapp-ac~eo-. 

herb-ac-eo-. ferul-ac-eo-. heder-ac-eo-. 

canic-ac-eo-. form-ac-eo. ros-ac-eo. 

* The origin of the final a in this declension, as well as of the final o in 
the second declension, two classes which include a decided majority of all 
the Latin substantives, has long been a desideratum ; nor does there seem 
good ground for assenting to the doctrine, which I have heard propounded 
in conversation, that the little suffixes in question may be the feminine and 
masculine definite articles attached to the end of the noun, as is the habit 
of the Scandinavian languages. At any rate, the idea expressed in the 
definite article forms no essential part of the words so ending. If it be 
thought that the vowels o and a, though not connected with the article, 
were yet added for the sake of distinguishing genders, the answer is, that 
the old Latin had no aversion to masculines in a, witness Cinna, Sulla, P. 
Cornelius Scipio Asina, Nasica, scriba, advena, &c., nor to feminines in o, 
as seen in humo-, piro- a pear-tree. &c., to say nothing of the Greek 660-. 
vqcro-, Kepacro-, and the numerous Greek adjectives, such as 6 /cat 17 mretpos, 
Indeed, that it was not the office of the vowel o to denote masculines, nor 
of the vowel a to denote feminines, is shown by the fact, that in Gothic, a 
by preference is attached to masculines, and o to feminines. 


ole-ac-eo-. membran-ac-eo-. bet-ac-eo. 

tili-ac-eo-. aren-ac-eo-. cret-ac-eo-. 

argill-ac-eo- . aven-ac-eo-. chart-ac-eo-. 

favill-ac-eo-. gallin-ac-eo. test-ac-eo-. 

ampull-ac-eo-. resin-ac-eo-. malv-ac-eo-*. 

Of these, be it observed, a large half belong to the botanical 

Two of the nouns which appear in the list just given, have 
other derivatives which confirm the suspicion of a lost final 
guttural, viz. ferul-ag-on- f. 'a sort of fennel/ and lapp-ag-on- 
1 a plant of the bur kind; 5 alongside of which we may place 
cunila- and cunilag-on- f., simila- and similag-on- f. } the 
plants so called, serra- and serrag-on-, ' saw-dust/ That the 
Latin language possessed a simple suffix on is clearly seen 
in turb-on-, scaturig-on-, (beside scatureg- nom. scaturex), 
asperg-on-, and harpag-on-. Furthermore, before the paper is 
closed, other reasons will be given for disbelieving the received 
doctrine that gon is a simple suffix of the Latin language. 
Moreover, a question virtually the same, was considered in the 
paper on English and Scotch Diminutives, where such a form 
as lass-ick-in was analysed and divided as here marked, in 
accordance with Grimm's views and with the formation which 
prevails in Gaelic, as cor-ag-an. I therefore claim to write 
ferul-ag-on-, &c., the first two syllables of which correspond to 
our own fennel, rather than ferula-gon-. But if the division 
suggested for ferul-ag-on- be adopted, I must put in a similar 
claim in behalf of citre-ag-on-, ostri-ag-on-, sel-ag-on-, tus-sil- 
ag-on-, lactil-ag-on-, ustil-ag-on-, mutil-ag-on-, must ell- ag -on-, 
sol-ag-on-, capr-ag-on-, lustr-ag-on- } laur-ag-on-, trix-ag-on- ; 
and this the more, because, like ferul-ag-on-, they are all 
names of plants. Ole-ag-on- too and tili-ag-on- are implied 
in the adjectives oleagin-eo-, tiliagin-eo-. 

Nor let it be objected that nearly all of these are little 
known to ordinary scholars. In an inquiry of the present 
kind, words that belong to the lower currents of society and 
to the rustic, deserve even more attention than the words of 
polite society, for the latter are often of foreign origin, and 
* So verna- has vern-ac-ulo- beside it. 


even when really home-grown, are still subject to those cor- 
rupting abbreviations which mark the fast life of a city; 
whereas the countryman, setting a less value on time, is con- 
tented with those drawls which belong to the fuller forms of 

I have already pointed out that the double suffix ag-on- 
correspoiids with great precision to the double suffix ag-an- 
of the Gaelic, which was discussed in the preceding paper (p. 
240) ; let me here add that the simpler forms ferulag-, simi- 
lag-, cunilag-, Sec., of which I have quoted eighteen denoting 
plants, stand well beside our own plant-names which end in 
ock, as charlock, shamrock, sourock (sorrel), the more so as g 
Latin should correspond to k English. 

But I must leave the argument at present incomplete, 
because I do not wish to anticipate what will have to be said 
of forms which in adopting the suffix agh have modified the 
vowel. Indeed the truth of the doctrine will not appear in 
its full force, until the paper be before the mind as a whole; 
for if my views are right, every separate portion of the argu- 
ment throws light on every other portion. 

III. AGH, as seen in Latin Verbs. 

I next proceed to consider the formation of secondary verbs 
by the addition of the same suffix. And two points here 
require previous notice. In the paper on English Dimi- 
nutives, it was contended that the addition of a diminutival 
suffix to verbs often introduced the idea expressed in the 
Latin paulatim, and so produced verbs of a repetitive, in- 
ceptive, imperfect, or continued character. In confirmation 
of this view, it may be as well to quote the authority of Dr. 
Johnson, who had certainly no preconceived theory to bias 
him. In the grammar prefixed to his dictionary, speaking of 
the formation of verbs by the addition of an element con- 
taining the liquid /, that is, the very suffix which plays so 
leading a part in our own diminutival substantives, he says : 
" If there be an /, as in jingle, tingle, mingle, sprinkle, twinkle, 

* For example, in cities we pronounce meat just as we do the adj. meet', 
but a clown says me-at. 


there is implied a frequency or iteration of small acts, &c." 
In confirmation of this view, reference was made to the 
regular habit which prevails in the Finn language, of em- 
ploying the diminutival suffix el with verbs, to express this 
very idea, as lask-en ' dimittere,' lask-el-en ' paulatim di- 
mittere/ On a former occasion (Proceedings, iv. 93), a similar 
argument was drawn from the Manx variety of the Keltic 
family, where the suffix agh added to the stem of any verb 
whatever, produces what, by one of the grammarians of that 
language (Leo), is called the ( modus consuetudinalis.' 

That verbs expressive of certain ideas should be prone to 
assume a suffix of a power equivalent to the word paulatim, 
will, on a little reflexion, appear to be very intelligible. Take 
for example, the ideas expressed by our own verbs walking, fol- 
lowing, writing, drawing, digging, rubbing, growing, breath- 
ing, all of which express aggregates of many petty acts. 

This premised, I proceed to deal with the Latin verbs which 
appear to have taken the suffix agh, or rather its represent- 
ative ag and ah. 

Flag- of plango, seen also in the sb.plaga. A monosyllabic 
verb is not always a root; and scarcely ever so when it 
begins or ends with two consonants. In the word under 
discussion, I contend that a fuller form of the verb is pal-ag-, 
and that the syllable pal- alone is radical, with the notion of 
flatness, as in palma- the flat-hand, opposed to the clenched 
fist, pal-ud- a marsh, pal-am openly, as in a plain ; and as d 
and / are readily interchangeable in Latin, we have pad- 
(pando) with the same idea of flatness, or an expanse. Plaudo, 
beginning with the same consonants as plango, also denotes 
the striking with a flat surface. Thus the idea expressed by 
our own verb clapping, belongs to both. The clapping of a 
bird's wings is expressed by plango, of a man's hands bjplaudo. 

Straff-. The anomaly of a perfect and supine stravi. stra- 
tum, beside an imperfect sternere, is also seen in the derivatives 
sir amen and str amentum, but the sb. strag-e- and adj. strag- 
ulo- present us with the desired consonant. That e- and ulo- 
constitute the suffixes of these two words, is clearly seen from 
the parallel cases of fid-e-, faci-e-, speci-e- on the one hand, 


and cred-ulo-, bib-ulo-, on the other. Another argument in 
favour of the form strag is to be found in our own verb strew, 
or, as older writers had it, straw ; for the final w of English 
words is generally accompanied by forms with a corresponding 
g in kindred languages. Thus stern- of sternere must be 
regarded as a compression of something like ster-en- and 
strag- as one of star-ag-. Similarly I should deal with sparg- 
scatter, as reduced from spar-ag-, the first syllable of which 
is identical with the Greek cnrep of ajreipw and perhaps we 
have the same root in wap-ay- of ariraparray . Nay, if we 
unite in one family sternere to strew, spargere to scatter, and 
<T7raparrew to tear to pieces, we shall be doir*g no great 
violence either to meanings or to forms'*. 

Frag- of frango, also tempts one to ask whether a vowel 
did not once divide the / and r, so as to leave ag for a possible 
suffix. Now prjjvvfjii, in Homer and the earlier Greek writers 
is rather ' break or burst through/ than ' break in pieces ' ; 
and the same idea prevails in our own day-break, as well as 
the German der Tag bricht an, phrases which exactly cor- 
respond to the French ' point de jour ' and ' poindre ' from the 
Latin ' pungere.' Hence it is suggested that frag- may well 
be a compression offor-aff-, and so a secondary verb from the 
base for- orfod- pierce (see xxix. xxxi.). An English irre- 
gular verb usually has its original form best preserved in the 
perfect participle, and thus brok-en invites our attention to 
bor-ock, i. e. a derivative from bore. 

Trah-, with traxi and tractum, seem by form to claim 
kindred with our own drag and draw, and the German trag-en. 
But a difficulty occurs in the different senses of the words, for 
the German tragen has, for its own sense, ' to carry or bear/ 
A closer examination however of the Latin trahere will 
supply instances where the notion of bearing is indisputable, 
as the indocili jugum collo trahentes of Horace, compared with 
the ferre jugum dolosi of the same author. The substantives 
trah-a and trah-ea, ' a sledge/ unite the two senses. I will 
not rely on our own word dray, as now used, seeing that the 

* Compare for the initial consonants sp and st, the parallel case of two 
words all but identical in power, sprain and strain. 


brewers still use at times a sort of carriage without wheels. 
But in tractare we find evidence which seems to show that 
' to bear ' was the earlier meaning of the root. This at least 
is the power of the word in such phrases as tractare arma, 
Hor., tractare personam and tractare paries secundas, Cic. 
But if this view be correct, then trah- (trag] may well be a 
compression of tol-ag-, where tol is the radical part of tollo, 
tuli (tetuli), tolera-. 

But here again the fragmentary evidence I have put for- 
ward would be wholly insufficient, if unsupported, to sustain 
my argument. I therefore proceed to call evidence of a very 
different nature, which will certainly not be liable to the 
charge of narrowness. As I contended that many, if not all, 
the nouns of the first declension had lost a final guttural, so I 
now make a similar assertion about the first conjugation. 

One thing at any rate will be readily admitted, namely, 
that not a few verbs of this conjugation co-exist, or at least 
co-existed, with shorter forms of the third conjugation. For 
example, this may be affirmed of all those verbs which are 
said to be irregular in having perfects in ui, supines in itum. 
The term irregular is indeed misapplied, for the perfects and 
supines belong to that third-conjugation verb, and not unfre- 
quently this simpler form appears in older writers, even in 
the imperfect tenses, as sonit, sonunt in Ennius; sonere in 
Lucretius ; tonimus in Varro, &c. ; and lavere is of frequent 
use in the poets. But if these roots were originally triliteral, 
for what purpose was the a added ? Such additions are never 
made without a purpose, yet no writer has ever suggested an 
answer to this question. I am the more entitled then to 
request the attention of scholars to the doctrine here pro- 
pounded, so far as the thirteen* disyllabic verbs in question 
are concerned. In some of them the repetitive idea is well 
marked, as in micare, fricare, crepare, lavare. Again, cubare, 
when contrasted with cumbere, as seen in the compounds pro-, 
ad-, re-, in-cumbere, Sec., tells its own tale, and that tale is in 
my favour, for procumbere denotes the single act of falling 
down, whereas cubare is always 'to keep your bed/ So 
* See anv Latin Grammar. 


vetare also denotes a persistent idea, the prohibition con- 
tinuing long after the order is given. The substantives 
spir-itus and hal-itus, by their short penults, bear witness 
that there were once shorter verbs in existence, from which, 
by the addition of our suffix, were deduced spir-a-re and 
hal-a-re ; and certainly the idea of breathing involves the 
idea of iteration. 

This argument, however, is rather of a negative character. 
If it be not enough to establish my doctrine, and I readily 
admit that it is not, still I may put forward the assertion, 
that the doctrine, if true, would account for the appearance of 
the a in the fifteen verbs before us, whereas it is at present 
wholly without explanation. 

Let us next ask whether the process employed with the 
substantives in a- is applicable to the verbs in a-. The de- 
rived forms ferulac-eo- and ferulag-on- were brought forward 
to prove that ferulag must have been an older form than 
ferula. Do the derivatives from the verbs of the first con- 
jugation in a similar manner exhibit traces of a lost guttural ? 
I answer confidently, they do. 

In the first place, calling to mind the frequent formation 
from verbs of substantives in ulus, ula, and ulum, as cap^ulus, 
teg-ula, spec-ulum, I claim the right of making a similar di- 
vision in the nouns : 

subligac-ulo-, spirac-ulo-, hospitac-ulo-, 

piac-ulo-, orac-ulo-, sustentac-ulo-, 

cenac-ulo-, augurac-ulo-, receptac-ulo-, 

propugnac-ulo-, objectac-ulo-, ambulac-ro-, 

gubernac-ulo-, spectac-ulo-, simulac-ro-, 

mirac-ulo-, crepitac-ulo-, lavac-ro- ; 

in the last three of which an r* has supplanted the /, simply 
because the words already possess an /, pi ecisely as puellaris 
and familiaris stand beside juvenilis and rivalis, and laquear 
beside puteal. 

The adjective grac-ilis seems by termination to classify 
itself with such forms as ut-ilis, fac-ilis, &c. But if this be 

* Hence the Spanish milagro for miraculo-, the change of r to I in the 
first syllable leading to the converse change in the last. 


the case, grac- should be a verb. If so, it seems to be 
identical with our own verb grow, and the meaning suits, as 
growing-fast is generally the cause of a person being slim and 
slender. Moreover, the same root is apparently found in 
gramen, if we may look to the form of the word ; nor is the 
meaning repugnant to the idea. Grass, being a collection of 
multifarious plants, may well have received a name common 
to them all ; ' growth ' being, in this respect, not unlike 
our own word ' vegetables/ and the Latin olera*. But our 
own grow would seem to be a secondary formation like 
know from ken, already noticed. This also has its analogy in 
Latin, where we find ger-men, which implies a simpler verb 
ger- (whether identical with gen of gigno I will not say). 
Thus the supposed Latin verb grac- would be a compression 
of ger-ac-, and here again the idea oipaulatim is self-evident, 
as Horace says, Crescit occulto velut arbor aevo. 

Beside gracilis I place alacer, which stands for alac-ilis, 
the r having supplanted the /, as in ambulacra-, &c. Thus 
al-ac- will be a secondary verb from al- raise, and the idea 
expressed in alacer, excited, roused, in opposition to downcast, 
depressed, is satisfactorily explained. So we have verbs in a, 
and adjectives in ac, often running beside each other, as 
pr oca-re andprocac-, vigila-re and vigilac-,fuga-re andfugac-, 
nuga-ri and nugac-, (retinac-ulo- and tenac-), sona-re and so- 
nac-, consterna-re and sternac-, crepa-re and crepac-, fura-ri 
-, incur sa-re and incursac--\. 

IV. AGH supplanted by ABH or AB in Verbs. 

So far we have searched for a guttural as the non-aspirated 
complement of the a suffix. Let me next draw attention to a 
change which the peculiar sound gh not unfrequently under- 
goes. In our own words laugh, cough, and rough, we see a 
guttural aspirate, but hear a labial aspirate. In the same 
way, as has been often noticed, a Greek initial % sometimes 

* The root is seen in al-ere to raise, ol-esc-, co-al-esc-, sub-ol-e-, &c. 

f See below ( vi.) the adjectives in ac and the statement there made, 
that Manx adjectives and the consuetudinal mood of Manx verbs, alike end 
in agh. 

2 A 


gives place to an/ in Latin, as v^cwo- funi- ; ^akivo- freno- 
XV- ofxa) } fud- oifundo-, ^0X77 and/e/. But as the Romans 
limited the use of /to the first place in a syllable, they seem 
to have been tempted to take a b* as the substitute for this 
labialized gh, and indeed bh would have been a more reason- 
able substitute than ph or /. In this way I would account 
for such forms as medicab-ili-, revocab-ili-, laudab-ili-, &c., a 
class of words which in the pages of Forcellini exceeds four 

Secondly, the neuter nouns, such as vocab-ulo-, conciliab- 
ulo-, venab-ulo-, tintinnab-ulo-, &c., admit of explanation on 
the same principle. Thus it is a mere accident that the 
Romans said mirac-ulum rather than mirab-ulum ; and indeed 
convertibility of the two sounds accounts for the form of the 
Italian maraviglia, French merveille, and our own marvel. 

In favour of calab- (cala-re) and dolab- (dola-re), I may 
point to Curia Calab-ra, and the sb. dolab-ra, So also 
cadav-er seems to imply a secondary verb, like cad-agh, from 
cad- fall. 

The same argument may be applied to such forms as con- 
tionab-undo-j volutab-undo-, plorab-undo-, where the consue- 
tudinal character is not to be disputed ; and the number of 
instances exceeds sixty. The existence of the suffix undo, as 
well as endo, for participles, is seen in faciundo-, regundo-, &c., 
as also in the so-called adjective see-undo- from sequi. 

But if in contionab-undus Sec. the suffix ab be adapted to 
express continuity of action, it must be equally well fitted to 
enter into the formation of imperfect tenses. May not then 
the middle syllable of am-ab-a-m be the same element ? This 
at least is certain, that the following a, which immediately 
precedes the personal endings, is employed both in Greek and 
Latin as the symbol of past time, as is seen in e-ru-de-a, 
-T-Tv<f>6-a } (ra-, whence by contraction v)v of the singular, 
and without contraction ecrav of the plural, while er-a-m 
throughout exhibits the , and by the length of this a so 

* B in Latin is the ordinary equivalent for a Greek < at the end of a 
syllable, as the datival suffix bi for <i, nebula for vffaXrj, umbilicus for 
9, i. e. ovv<pa\os navel, sorbeo beside po<ea>. 


far justifies the theory that a following consonant has been 

But the so-called future am-ab-o has an equal claim to 
consideration, and I admit the claim, even though there be 
here no additional suffix to denote futurity. On theory alone 
it may be maintained that an action declared to be imperfect 
at the present moment, can only be completed, if completed 
it is to be, in the future. Moreover in practice we often 
find presents used with future power. To say nothing of the 
Greek et/u, ' I shall go/ and such cases as duco uxorem, ' I am 
going to be married/ and scribendum* est mihi, which, without 
any element to denote futurity, is still practically used of the 
future alone, I may point both to er-o, ( I shall be/ in the 
one language, and e<r-o/ww. in the other, as forms essentially 
present, just as much so as scrib-o and kir-o^ai. 

But the habit of using simple forms as futures is more 
marked in the Slavic family. Thus Dobrowsky, in his 'Insti- 
tutiones/ p. 374, says : Futurum simplex a forma prsesentis 
non differt, and he soon after gives as examples, dam, ' dabo/ 
in opposition to daio, ' do ; ' bud-u, ' ero ' to yesmi, ' sum/ In 
fact, the Old Slavic, from which these examples are taken, has 
frequently two forms of the present, one for ideas of mo- 
mentary action, called by Dobrowsky ' verba singularia/ the 
other for a continued state of things, ' prsesentia imperfecta.' 
Yet even with this advantage, the habit of the language does 
not always confine the use as futures to the one or the other 
form. Thus the same writer, p. 376, says : Utuntur vero 
Slavi subinde prsesenti verbi singularis pro futuro, eadem ni- 

mirum forma exprimendo praesens et futurum as gryadu, 

' venio et veniam/ 

If my theory, which explains the middle vowel of am-a-re as 
a corruption of abh or agh, be correct, we must not be sur- 
prised to find all trace of the consonant lost in the future reg- 
a-m. Two difficulties however present themselves in this part 
of my argument ; in the other conjugations the past imperfect 

* I have elsewhere shown that this form in itself denotes, not futurity, 
but the imperfect state of an action, like our verbal substantives in ing, as 



presents the termination ebam*, not abam; and again in the 
future of the second or e conjugation, ebo not abo ; while in 
the other conjugation the a of the first person gives place in 
the other persons to an e. Still this substitution of an e for a 
is no violent change, seeing that d and TJ in the Greek tongue 
are often but dialectic varieties, while the Latin also constantly 
mixes together the first and fifth declensions, as materies or 
materia; and again, in the subjunctive mood, writes both 
fuam and siem. 

But the case of regam, followed by reges, reget, &c., may be 
explained on another, and it is thought more satisfactory 
principle. As the second and third persons have for their 
simplest suffixes is and it, in opposition to the o of the first 
person, the a of the fuller suffix abh would be subject to the 
so-called umlaut, and thus give place to that sound which a 
German represents by d. This influence of the umlaut is well 
known to have caused in the very same persons of many 
German verbs the same result : as, ich fange, but du fangst, 
er fdngt. The appearance of the changed vowel in the 
Latin plural indeed, is not to be defended in this way, but a 
modification once established is apt to overreach the proper 

V. AGH supplanted by AB in Substantives. 

Lastly, among substantives we have arrhab-on- beside arrha 
earnest-money, and Varro's apex-ab-on- a sausage, whexe the 
double suffix ab-on seems to correspond to the double suffix 
ag-on, which was considered above. To which must be added 
cacabus, cannabis or cannabum, carabus a sort of crab, and 
carabus a coracle, in the last words it seems highly pro- 
bable that ab of the one represents ac of the other. It 
should be noticed too, that in all these examples a preceding 
guttural x or c furnishes a fair excuse for the substitution of 
ab. Yet in trab- tree (see xii.), we have ab without this 

* The change of vowel in the various forms of our suffix, will be dis- 
cussed more fully below ( xxix.). 


VI. AGH as seen in Adjectives. 

But the Manx language, to which I but now referred, not 
only forms the modus consuetudinalis of the verb by the 
addition of the suffix agh, but employs the same suffix for a 
large majority* of its adjectives ; and with reason, as the office 
of the adjective is to denote a permanent state of things. So 
the Gaelic swarms with adjectives in ach\ and according to 
Leo, the German adjectives in ig represent the class. This 
suffix is our y, steinig stony ; and if the German isch is but a 
variety of ig, then our own ish must also be one of the family. 
The Welsh too has adjectives in off, iff, and ac, but forming a 
minority of their class ; while the Breton has a respectable 
number in ek. 

Does this formation exist in the classical languages? I 
answer, yes. The Greek has not a few words which have 
much of the adjectival character, but are commonly limited to 
human beings, yet so as to denote an habitual condition : 

\aXay-, prattler. 0iAa:-, watcher. ffTvirdx-, rope-seller. 

apirdy-, robbing, /ucu/xait-, boisterous. QopraK-, porter. 

veaK-, youngster. 0a\a/zd*:- =0aAapra-. TrXovrdfc-, rich churl. 

Xi0ac-, stony. cw//<k-, debauchee. XWTO.K-, flute-player. 

^aaKaK-, gaper. <f>eva,K-, cheat. \\evaK-, mocker. 

/3\aic-, dullf. <mva<c-, nimble. 0\vdfc-, jester. 

KoXaK', flatterer. j(avva.K- t gaper. oroyu^afc-, big-talker. 

In the Latin language they form a familiar class, but one 
which has been subject to some misunderstanding, in that a 
faulty or vicious character is often attributed to the suffix. We 
shall perhaps be more correct, if, following the suggestion of 
the Manx, we call them adjectives of habit. At any rate,/er0# 
ager is in no sense ' bad land', although edax, ' habitually 
eating ', does not imply a praiseworthy habit. 

For the sake of brevity, the part to which the suffix is 
added, is alone given in the following list, while to a few of 
the examples a word or two of comment is attached : 

* Bei weitem die mehrzal aller adjectiva 1st so gebildet. Leo, Ferien- 
schriften, erstes heft, S. 181. 
t For /AaXcK- = Latin molli-. 

















pell- 2 . 

proc- 1 . 









stern- 4 . 



incurs- 6 . 


cat- 7. 




lingu- 8 . 















1 With an old verb proca-re to demand. 2 Not from pellicere, which 
could only have led to a formpellex, as from illicere, illeac. Perhaps a va- 
riety otfollax. 3 Olax, Mart. Cap. 4 No way connected with sterner e ; 
rather from sternare, the simple form of consternare to start or startle, as 
a horse. 5 Vatrax ' pedibus vitiosis.' Lucil. ap. Non. 6 Incursax, Sidon. 
7 Catax (=cadax?) limping. Lucil. 9 Linguax, Gell. 

VII. AGH (AC) supplanted by EC (EG, 1C) in Substantives. 

A comparison of the Doric fivp^a^ and the Ionic or Attic 
fivpfjirj^ affords something like a suitable stepping-stone to 
the next matter for consideration. Our examples of substan- 
tives with the diminutival ac- for the Latin language, made 
but a poor show beside the sixty and more Greek substantives 
in a/c-. The reason was simply that the Latin language pre- 
ferred for the most part a variety of vowel. Thus the Greek 
TraXXaf, vpaj;, and TrvvBal; have, for their Latin equivalents, 
pelleXy soreXj and podex, while to auXaf corresponds the rare 
yet truly Latin word aulix. With this clue we are led to 
a list of words sufficiently numerous : 

ibex, m. the wild goat. pulex, m. a flea. 

vibexy f. a weal. 
codex, or 

or 7 

> m. 
y ) 

a trunk. 

podeXy m. = 7rvva*. 
forfeXy f. scissors, pinchers. 
aleXy m. f. herring. 
ilex, f. evergreen oak. 
si lex, m. f. a flint. 
ulex, m. some shrub. 
y m. f. a gnat. 

fameXy m. blood of a bruise. 

i. abronchial vessel, &c. 
Vy m. f. a bug. 

pumexy m. f. a pumice-stone. 
rumeXy m. f. sorrel. 
senex, m. an old man. 
apex, m. a summit. 

hirpeXy or \ , 

f m. a narrow. 
urpexy J 

forpeXy f. curling-tongs. 
* That is, they denote the same general idea, though differently applied. 



vitex, f. the agnus castus. 
pant ex, m. a sausage. 
dent ex, m. some sea-fish. 

rupex, m. a boor. 

carex, f. a rush. 

imbrex, f. m. a gutter-tile. 

grex, m. f. crew. 

sorex, m. a shrew-mouse. 

murexy m. the purple-fish. 

scaturex, m. a gushing spring. 

latex, m. f. a shooting streamlet. 

And with the vowel again slightly modified* : 


cortex, m. f. bark. 
frutex, m. f. a shrub. 
vervex, m. a wether-sheep. 

culix, m. some plant. 
fulix, f. a coot. 
tomix, f. a cord. 
pternix, f. a sort of thistle ? 
comix, f. a crow. 
for nix, m. a vault. 
spinturnix, f. some bird. 
coturnix, f. a quail, 
tariff, m. f. a larch. 
tamarix, f. tamarisk. 
van#, m. f. a twisted vein. 
sorix, some bird. 
struix, a pile")*. 

In these two lists the notion of smallness is often exhibited 
in the most decided manner ; for instance, in the names for a 
bug, a flea, a gnat. A comparison too with the series of 
English diminutives may be useful. In speaking of the 
latter, attention was drawn to the frequent occurrence of 
names of plants, of birds, of fishes. Now our two above- given 
series of names in ex or ix contain, of plants twelve, of birds 
six, of fish three. Again, five of the class are the analogues 
of English nouns already claimed as diminutives. Thus sorex 

* In many instances the nominative is made to end in ex or ix, by the 
sole authority of dictionary makers, no Latin author supplying any nomi- 
native; in other cases there is authority for both. 

t There are some other words which at first sight seem to claim ad- 
mission to our lists, but are really derived from verbs, as obices and subices 
which contain jac-ere, elices and pollex, from compounds of the obsolete 
lac-ere, index and judex from the root dec- show, whence the Greek SetK-i/u- 
and Latin die- say. 

'tf, f. a root. 
scandix, f. chervil. 
pendix, f. a tumour ? 
appendix, f. a make-weight. 
coxendix, f. hip-bone. 
foe&X f. a blanket ? 
perdix, m. f. a partridge. 
cZi#, m. a cup. 
salix, f. willow or sallow. 
turdelix, ap. Varr. 
helix, f. ivy. 

wfelix, f. fern. 
Z>, m. a furrow. 


is 'shrew'; grex } 'crew'; filix, 'brake'; pulex, 'flea', 
aulix=av\aj;, i. e. a-Fa\a%, 'furrow'. Even culex seems to be 
identical with 'gnat/ if the latter, as is probable, be a 
corruption of gnack, and this again identical with KCOV-COTT-. 
Apex may perhaps be regarded as a corruption, and a 
very reasonable corruption of ac-ex (pronounced ak-ex], and 
if so, it is connected with the numerous words which begin 
with ac-, and denote sharpness. The change of the &-sound 
and p is well seen in another word or words of the list. 
Forfex, forpex, and forceps are treated in some dictionaries 
and etymological works as of independent origin. Thus we 
are told that forfex is from foris and facio, forceps from 
foris and capio, forpex from foris and pecto. A more rea- 
sonable derivation makes them all varieties of one and the 
same word, and that a diminutive of the noun that we 
write fork (furca), formed by our suffix ec. But as fore-ex, 
like the theoretic ac-ex, is intolerable to the ear, we get a lip- 
letter substituted, at one time for the first of the two gutturals, 
at another for the second. Then, as to meaning, the idea of 
a fork is well marked in all the various uses of these words, 
whether tongs, scissors, curling-irons, claws of a beetle or a 
crab ; nor was it without reason that forficula was adopted by 
modern writers as the name of the ear- wig, seeing that our 
own gabel, a fork, supplied a derivative for the same creature. 
Of the claim of senex and ilex to be regarded as diminutives, 
something will be said below. 

VIII. AGH reduced to UG (UC) in Substantives. 

But a suffix agh should, in the nature of things, be quite 
as liable to interchange with ug or uc, as with ec } eg or ic ; and 
our English derivatives pluck and brook (both as verbs and 
substantives) support the claim. Accordingly we find crux 
and fruges demanding our attention. The first is only a 
compression of some such form as colux, corresponding to the 
Greek cncok-o-^r* stake, a word so often used of punishment. 
The root-syllable of this Greek word, KO\, probably denoted 
wood. We say this, looking to the Greek icakov wood, to the 
: crux :: (nceAos : crus :: crKuXev-: scru- of scrutari. 


Latin calo (see Festus), to the secondary forms clavus a tree- 
nail or wooden nail, for the addition of ferreus was required 
to denote a nail such as we use. Moreover, the theoretic KO\- 
wood, stands in the required relation to the German hol-z. 
And now returning to crux, we see how justly it may be con- 
sidered as equivalent to the infelix arbor. Fruges will be 
more conveniently discussed under the verb fruor. 

IX. AGH reduced to C in Substantives. 

We found our English diminutive ock or ick frequently 
cut down to a simple guttural, as in park, abbreviated from 
parrock or paddock. Here too the Latin seems to agree with 
our own tongue, as in arx, calx the heel, calx a stone, falx } 
lanx, merx-, and these words stand in our dictionaries, for the 
most part without etymological explanation. Perhaps the 
consideration that the guttural is no true part of the words 
may render the problem easier. Now the essence of arx is 
height ; and the root ap- of aip- raise, supplies what we need, 
but with the disadvantage that it is a Greek verb. This 
difficulty however disappears if al-ere to raise, be the same 
word, and ard-uus a derivative from it, for as r and d are both 
interchangeable with l } rd may well be also. We have what 
is precisely parallel in sord-es dirt, beside the verb ob-sol-escere 
to become dirty, and solum soil. (See our Trans. 1854.) Of 
calx heel, the essential part cal is the fitting correlative of 
our own word heel. Freund indeed would regard calx as 
little more than a metathesis of the Greek adverb Xaf . The 
more correct view would be to treat Xaf as a corruption of 
/c\af, or rather /caXaf . Calx (calculus) a stone, receives 
satisfactory explanation from what is found in the Keltic lan- 
guages. In Gaelic we have clach a stone, also a testicle; 
in Welsh calch lime, and careg a stone or testicle. In these 
three nouns our diminutival suffix forms an element. But 
the Welsh has also caill, and the Breton kail or kell, a testicle*, 
originally no doubt a stone of any kind. The Greek too 
has aXfc? small stone, gravel. Fal-x, if regarded as a 

* Perhaps it was owing to this peculiar use of the word, that for de- 
cency's sake it ceased to be employed for the more general idea. 



contraction of fal-ax, has probably in fal the same root as 
fol of folium, with the notion of a flat surface or blade. Lanx, 
while it has taken to itself a foreign guttural at one end, has 
probably lost a labial aspirate at the other, in this latter re- 
spect corresponding to lana, lorum, lamina, &c. So too the 
Spanish llano and the Welsh Llan have something more than 
a liquid / for the commencing consonant, if we trust the ear. 
But the Spanish word is known to be the representative of the 
Latin adjective piano-. Thus to Ian of lanx we may venture 
to affix either an / or a p, so as to establish an affinity between 
it and our own flat or plate, or, if a diminutive be preferred, 
platter. Mer-x in the last place, is brought to a form identical 
with that of mer-eo, the original meaning of which is, to earn, 
that is, to labour. We have the same root in our own lan- 
guage in the words work and ware, the first of which like 
mere- possesses our diminutival suffix in a compressed form, 
while the Greeks, first pronouncing Fepyov, eventually cast off 
the initial w ; and the Romans copying them in this respect, 
formed a compound verb ex-erc-e- work out. 

X. EC as a suffix of Substantives, followed by other 

We will now go back to the suffix in the form of ec, in 
order to trace it in cases where it has been for the most part 
disguised by the loss of the consonant, viz. 
aesculetum. citretum. funetum*. opobalsame- (saxetum.) 
arboretum. cornetum. ilicetum. turn. senticetum. 

arundinetum. coruletum. juncetum. palmetum. (sepulcretum.) 
(aspretum.) cupressetum. lauretum. 
(bucetum.) dumetum. lilietum. 
buxetum. ficetum. moretum. 

cannetum. (fimetum.) murtetum. 
carduetum. fructetum. nucetum. 
castanetum. fruticetum. olivetum. 

It will be seen that all but five of these, included in 
brackets, connect themselves with the names of plants or trees, 

* Funis properly denoted a reed, being the same word as o-xoivos. 
t Equisetum is purposely omitted from the list, inasmuch as it is formed 
from equi seta. 














so that they may be considered to represent the type upon 
which the others have been modelled. That something like 
ec-tum was the original termination of these words is esta- 
blished by the existing forms car-ec-tum, dum-ec-tum^l-ic-tum, 
frut-ec-tum, lum-ec-tum, sal-ic-tum, vir-ec-tum. Of these, four 
stand in immediate relation to the nouns carex rush, filix 
fern, frutex shrub, salix willow. For dumectum we have the 
express authority of Festus: "dumecta antiqui appellabant 
quae nos dumeta" j lumectum, used by Varro, is in fact only 
a dialectic variety of the same word; and lastly virecta is 
the form supported by the best MSS. in Virgil and other 
writers, as shown by Wagner. 

But it is utterly erroneous to regard ec-tum as a contraction 
from icetum } though the error has the sanction of Festus 
among the ancients (v. dumetum), and Wagner, &c. among 
moderns (v. virecta) ; for it is a somewhat ludicrous ana- 
chronism to explain the old formations from those of later 
date. Indeed we fall into one of those never-ending etymologies 
which are self-convicted. If viretum stand for virectum, and 
virectum for viricetum ; this again will be for viricectum, and 
this for viricicetum, and so on ad infinitum. The fact is, that 
as carectum is formed immediately from carec- (nom. carex) 
by the addition of a neuter suffix to- (nom. turn), so salic-, 
arbos- lead at once to salic-tum, arbus-tum. And yet it is 
quite true that eventually etwn came to be regarded as a whole ; 
and so arose some few instances of superfluous growth, as 
fruticetum, ilicetum, senticetum, which really contain a du- 
plication of the suffix. 

So much for external form. The fitness of a diminutival 
suffix for these forms seems explained by the several con- 
siderations, 1, that many deal with vegetation in its smaller 
varieties ; 2, that the larger trees are often first trained as a mass 
of nurslings to be subsequently transplanted ; and 3, that trees 
growing thickly together rarely attain a full development. 

XI. OCR and OW of English, how represented in Latin 

So far we have dealt with ac, ec, ic and uc, as diminutival 


suffixes of substantives, but have passed over oc. Of this we 
have but one example in esox a fish, and that a foreign word. 
What then has become of the class of words which should 
correspond to our English diminutives in ock, as hillock, had- 
dock ? In the first place, it is scarcely to be expected that any 
one allied language should possess analogues of all the four 
varieties of a suffix which appear in the different dialects of 
Scotch- Saxon ; and we have already seen a rich supply of 
substantives in ec (ic) corresponding to haddick, lassick. 
Still, if we fail to find representatives of the suffix ock, we may 
be more successful with the corrupted form of ock, viz. ow, 
in which only o is heard. Add to this that an examination 
of the first declension has led to the belief that a final gut- 
tural has been there lost. Is it not then a priori probable 
that the suffix o, which constitutes the second declension, has 
its origin in ogh or ow ? and this the more, as no explanation 
of the o has yet been proposed, save indeed the suggestion 
that it is the masculine article ; but this suggestion loses all 
its probability, if the a of the first declension be not the 
feminine article. In truth, the o and the a final which di- 
stinguish the first two declensions are but dialectic varieties 
of the same word, just as we say one, two, stone ; a Scotch- 
man, ane, twa, stane. But languages often avail themselves 
of such dialectic varieties, so as to attach to each some slight 
peculiarity of meaning. For example, bag, bay and bow -, 
canal, kennel, channel ; 6pa<ro<; and Oap&os, are severally one 
in origin, yet practically distinguished in use when allowed to 

I venture then to claim the great bulk of the second de- 
clension as representing the Scotch and English substantives 
in ow, and corrupted from och or ock. That this declension, 
like the first, has suffered the loss of a guttural seems shown 
by the derived adjective aprug-no-*, beside the substantive 
apero- a boar. 

* It may be useful to compare callo-, sb. n. hardened skin, with the Greek 
-, the thick skin on the neck of an ox or hog ; and the more so as 
-oK-, having too strong a supply of gutturals, would naturally pass into 
-. And here by the way I would suggest, that the notion of callous 


Nor is this a solitary instance. A more numerous series 
is seen in the adjectives in ac-eo-, formed from masculine and 
neuter nouns of this declension, as from cacabus, sebum, bulbus, 
intubus, hordeum, to/us, lilium, milium, folium, lolium, minium, 
amygdalum, capillus, surculus, pampinus, farnus, porrum, pa- 
pyrus, argentum, frumentum, murtus. To which add a sprink- 
ling of adjectives in ic-io- from substantives in o, as rapic-io-, 

It is no contradiction to what is here said, that poSatc-, 
/3o)Xa/c-, Ovvvaic-, veftpaic-, S^pa/c-, are regarded as dimi- 
nutives of poSo-, pco\o-, 6vvvo- s vefSpo-y B(,(f>po-. The words 
are in strictness duplicates of each other, mere dialectic 
varieties ; yet as the fuller form is preserved in ySwXa/c-, &c., so 
the sense of the diminutival power is more deeply impressed. 

XII. 1C, EC, UC of substantives reduced to T, E, U. 

But if agh reduced to a, and ock reduced to o, supply the 
nouns which constitute the first and second declensions, why 
should the suffixes ic (struic-), ec (vertec-) and uc (cruc-) be 
exempt from a similar loss of the guttural ? We shall on 
inquiry find good evidence that they also are liable to the 
same curtailment. As the Scotch lassick is reduced to lassie 
or lassy, so there is ground for suspecting that ensi-* t cani-, 
reti-, to take these as single examples of large classes, must 
have been at one time pronounced ensic-, canic-, retic-, when 
we find derivatives from them in the form ensic-ulus, canic- 
ula, and retic -ulum. 

Similarly, diec-ula, rec-ula, anic-ula and cornic-ulum seem 
to imply that the simple words die-, re-, anu- and cornu- have 
lost a final guttural. And especially note trabec-ula, corre- 
sponding to the noun trabe-, nom. trabes, with regard to which 
the assumption of a form trabec- is confirmed by the Greek 
Tpair^K- or rpa^Tj/c- a small beam, a spearshaft, stake, post, 
which can scarcely have any relation to the verb rpeTrco. 

skin may be the cause why cal-c- came to signify the heel ; and if so, the 
notion of hardness may unite what are commonly considered independent 
words, calc- heel and calc- stone, the latter being compared to other earths. 
* Compare TroXt-, i. e. TTO\IX-, beside TroXt^-vj? and iro\ix-viov. 


But to return to corniculum : this noun ought more regu- 
larly to have appeared as cornuc-ulum, but we must remember 
that in the same declension cornibus has superseded cornubus, 
to say nothing of the habitual degradation of u into i in so 
many Latin words. Thus the old language gave umus as the 
first person plural of verbs, as seen in sumus, volumus, quaes- 
umus, but imus in scribimus, &c. So decumus, optumus and 
recuperare gave place to decimus, optimus, reciperare. In these 
cases, the u occupies commonly the penult place, where the 
want of an accent leaves the syllable specially liable to a 
careless pronunciation. But a u in all parts of a word seems 
apt to slide into the weak pronunciation of a short i. Thus 
the modern Greeks always pronounce u as i (continental 
sound). The French u is not far removed from it, and our 
nouns tree and knee (in Anglo-Saxon treow and cneow) have 
attained to the same vowel sound, although they represent 
Snpv and <yovv. Moreover, we may safely assert that as genu 
is the Latin representative of knee, so gen-uc-ulum must once 
have corresponded to our kn-uck-le. But over and above this, 
there is in the forms with which we are dealing a special 
reason why the u should lose its power. The diminutival 
suffix seen in ulus, ula, ulum, has for its truer form el, as is 
seen in ocellus beside oculus, in the German tafel beside ta- 
bula, in 2t/ceXo? beside Siculus. Now the weak vowel of el 
would tend to effect an umlaut in the preceding syllable, pre- 
cisely as in the German knochel. But there yet remains one 
instance of the u being preserved in a diminutive from the 
fourth declension, and that because the u was long. While 
our dictionaries ascribe to Plautus the use of an adjective 
which they are pleased to write meticulosus, Bitschl justly 
contends that the second syllable of this word is long, but he 
is wrong in writing it meticulosus. He would have done 
better to follow his excellent MSS. C and D (Most. V. 1. 52), 
and write, 

" Nescis quam metuculosa res sit ire ad iudicem." 
So again when he publishes the Amphitruo, I would sug- 
gest that he should substitute for nullus est, in the line I. 1. 
137, what is more in accordance with the Plautian habit, 


; and then we should have : 

" M. Nullust hoe metuculosus aeque. S. Quern? in men- 
tern venit." 

That in some nouns of this declension the u should be long, 
in others short, is parallel to what happens with the nouns of 
the i or third declension, where we find ndvicula and canlcula. 

But it may be well to take a cursory glance at those nouns 
in the fourth declension which stand apart from the so-called 
supines or nouns in tu, as auditu- : 

Ac-u- (cf. ac-esc-, ac-ido-}, an-u-* (cf. sen-ec- and ypa-v-),, 
arc-u-, cib-u- (cf. cib-o-), col-u-, corn-u- (cf. horn), dom-u- (cf. 
dom-o- and the Greek vb. &//--), fic-u- (cf. fic-o- and our fig), 
gel-u- (^jgen-u-^r-u-, lac-u- (cf. lav-ere), laur-u-(cf. 
laur-o-}, man-u- (cf. A.S. mand ( hand'), met-u- } nur-u-, pec-u- 
(cf. TTto-v^tpen-u-foo^portic-u-jquerc-u-, spec-u-(cf. cr7re-e<7-), 
trib-u-, ver-u- (cf. ver of ver-t-, our own veer, wear ship, Fr. 
vir-er, Germ, wirr-en). 

In some of these an undoubted etymology tells us that the 
u is an element foreign to the root-syllable; in many, the 
fitness of a diminutival suffix is evident, as ac-u- a needle; 
trib-u- a third, or rather ' thriding' ; and for affection's sake, 
nur-u- . Quercu- is the one word which opposes this view, and 
in this respect agrees with ilex and the Greek Spvs, Welsh 
derw. In these words our doctrine finds its chief obstacle ; 
but so far as the last word is concerned, the solution is not 
difficult. A/ou- properly means a treef, and is at bottom the 
same word with Sopv- spear, and with the Norse dor a spear, 
which by its umlaut tells us that some final vowel, such as u, 
has fallen from it. The Sanskrit again has taru, with tarav- 
as as the nom. pi., a form the more interesting, as it assures 
us that the Latin trab- is the same word. The habit of 
translating this Latin word by ' beam/ has tended to conceal 
from us its true meaning. But when Ennius, Virgil and 

* The appearance of a diminutival suffix in these three words is justified 
by the fact that the long-liver is commonly spare, and the more spare 
the older he is. 

t Compare the Greek derivative Ap.a8pva8-. Indeed Eustathius ex- 
pressly asserts that the first meaning of 8pvs was ' tree.' 


Ovid all use the word of ' trees/ there ought to be no hesi- 
tation in restoring this as the original meaning of the word, 
as indeed it is of our own word beam (cf. hornbeam and Germ. 
baum). As regards quercu-, it is very possible on the one 
hand, that the final u is not diminutival, and on the other 
hand, that the word originally denoted a dwarf oak. Indeed 
the word querc-uk may have in its first part a representative 
of the Welsh cor-ach or cor-ig, both existing words for a 
' dwarf/ while uk may be an analogue of our own term oak. 
In our own acorn, that is oak-corn, Germ, eich-horn, the first 
element has been sadly reduced. All this* is put forward 
solely as a possibility, nor indeed would it be reasonable to 
reject a theory in consequence of a difficulty growing out of 
a single word. 

As the nouns in ex and ix in the oblique cases give to our 
suffix an identity of form, it is not strange that 011 the loss 
of the guttural we should have a class of words whose no- 
minative ends indifferently in is or es, as plebi- or plebe-, nubi- 
or nube-, aedi- or aede-, cani- or cane- ; whence the derived 
forms plebec-ula, nubec-ula, aedic-ula, canic-ula. That the-i 
or e in this class of words is the remnant of a distinct suffix, 
seems to follow from the fact, that we also find such nomi- 
natives as plebs, nubs, trabs, and the Greek KVWV, KVV-OS, 
corresponding to our own hound; and we can now account 
for such forms as the gen. pi. can-um, ap-um, juven-um, which, 
as deduced from the primitive nouns, were not entitled to an i. 
Again the Greek vav-5 speaks in favour of a simple nau- in 
naufragus, of which I hold nav-i- to be in origin a diminutive, 
just as lvss-ie is of lass. So again au- in au-ceps, au-spex, 
au-gur, may be regarded as the word from which av-i- was 

In order to show the general fitness of the nouns in i, so 
far as regards meaning, to possess a diminutival suffix, and 
also to show by etymology that the i in many cases is no 
genuine portion of the root, I give the following list : 

Amn-i- (cf. our Avon), angu-i-, ass-i- a unit (Fr. as, our 
ace), ass-i- or ax-i- axle (cf. A.S. eax], caul-i- or col-i- a stalk 
(only a variety of cod-ec-), clav-i- (cf. K\tjf-t8-), clun-i- buttock; 


coll-i-, hill ; corb-i-, basket (cf. German korb) ; crin-i-, band of 
hair; cut-i- (cf. our hide, and Germ, haut) ; ens-i- } fasc-i- 
bundle; fauc-i- gullet; febr-i- (comip&refebric-it-a-re),fin-i-, 
foll-i- pi. bellows (the Greek has ^oTUu/e-, nom. <oX)uf in the 
sense of a follicle), for-i-,fun-i-, ign-i- (cf. our oven), imber-i-, 
juven-i-, lact-i- pi., lintr-i-, mun-i- part ; nar-i-, nav-i-, orb-i-, 
oss-i- (gen. pi. ossium), ov-i-, pan-i-, pelv-i-, pisc-i- (cf. A.S. 
fisk, our fish), rat-i-, rav-i-, rud-i- staff, and what seems only 
a variety of the same word, sud-i- stake ; scob-i-, scrob-i-, 
secur-i- hatchet ; sem-i-, sent-i-, sit-i- 9 torr-i-, trud-i- a pike 
(cf. vb. trud-) ; turr-i- (cf. Fr. tour), tuss-i-, venter-i-, vepr-i-, 
verm-i-, vit-i-, ungu-i- nail; uter-i- skin; to which add the 
neuters il-i-a, mar-i- originally water rather than sea, ret-i- 
sal-i-, nom. sale salt (Ennius). 

Nouns which interchange i and e are : 

Aed-i-, ap-i- (cf. gen. pi. ap-um), caed-i-, call-i- a little path ; 
can-i- (cf. can-um) crat-i- hurdle, clad-i-, fam-i-, fel-i-, fid-i-, 
lab-i-, lu-i-, mel-i-, nub-i-, pleb-i-, pub-i-, sed-i- (cf. sed-um), 
sord-i-, stru-i- (cf. stru-ic-), torqu-i-, vall-i-, verr-i-, volp-i-. 

This list would probably have been more extensive if the 
writings of the Romans had come down to us in greater 
abundance. Thus we might probably have found naves for a 
nominative as well as navis, trabis as well as trabes. Indeed, 
in the case of many nouns of this class, the particular form 
assigned by our dictionaries to the nominative is simply an 
unauthorised assumption, no instances of any nominative oc- 
curring. With regard to moles and saepes we are the more 
entitled to assume nominatives molis and saepis, when we look 
to the verbs moli-ri and saepi-re. 

XIII. The suffix AGHin Substantives virtually repeated. 

We must pass hastily over a class of words which to the 
suffix in the form ac, Ic or ic, oc, uc or c alone, add a second 
suffix a or o. Nor let it be objected that upon this theory a 
word will be taking the very same suffix twice over ; for after 
all, this is exceeded in the case of ocellulus, which repeats the 
other diminutival suffix el three times, oc-el-el-el-us. 



portul-ac-a* , 








Jistuca ; 

lingul-ac-a *, 




to which add 




ridiea(c. rudi- a stake), 




porca, a furrow. 



porca, a farrow. 


formica (cf./iup/i^K-). 












lorlca (cf. OwpaK-), 



urica = eruca. 


tritico-.* ; 




verruca (cf. ware \ 

a Add again to these 


callosity, Scotch) 

, malva*,^fjiaXa'xr)* 


p,o\o^ = mallow. 

XIV. AGH or AC in Latin Adjectives, how corrupted. 

Having thus considered at some length the form which our 
diminutival suffix has taken over and above the original form, 
where a is followed by a guttural, in substantives, the question 
arises whether, besides the more regular formation seen in 
ed-ac- or ed-aci-, the Latin adjectives present any corrupted 
forms of the suffix corresponding to the varieties which have 
shown themselves among the substantives f ; or, another shape 
may be given to the same inquiry : we may begin with asking 
whether the Latin has any adjectives to represent our English 
adjectives in ow, as shallow, yellow. 

A claim to this position is put forward in favour of the 
following among others : aceri-, brevi-, comi-, dulci-, forti-, 
grandi-, gravi-,jugi-, laevi-, leni-, levi-, limi-, mani- (DiManes), 
molli-j oci- (of odor, ociter), pingui-, rudi-, suavi-, tenui-, 
tristi-, turpi-, vili-. In the first place, the final i of these 
words can scarcely be a radical letter. If then it be a suffix, 

* All plants, and fourteen of them. 

t A slight change of the vowel is seen in/eroc-, nom.ferox. 


we may reasonably ask once more, why it was added, if devoid 
of significance ; and if significant, will not the power of a di- 
minutive give at least an intelligible and not inappropriate 
sense ; one also that might easily be lost sight of, just as has 
been the case with the ow of our own adjectives ? Secondly, 
for many of these words we find a trace of a final guttural in 
the derivatives acric-ulo-, brevic-ulo- } dulcic-ulo-, fortic-ulo-, 
grandic-ulo- } levic-ulo-, mollic-ulo-, tenuic-ulo-, tristic-ulo-, 
turpic-ulo-. Thirdly, we know that some Latin adjectives in 
i have lost a final guttural, as quali- and tali-, corrupted 
from such forms as qua-lik-* like what; ta-lik- like this, 
corresponding to our own old form whilk and thilk ; as also 
to the German welcher (i. e. we-lich-er) which; and soldier 
(i. e. so-lich-er) such. So our silly is the German selig. 

But we have light thrown on these words by their Greek 
analogues, such as to bring their suffix into all but identity 
with what we see in our words, shallow, &c., for brevi- = 
vi-^ekaxy-, dulci-=j\vKv- y forti- = 0pa(7U- } gravi- 
pingui- = TTCT^V-, oci- = COKV-, suavi- = Fa$v- } and 
densi-, implied in the verb dense- make thick, = Sacrv-. 

Of course what has been said of substantives of the second 
declension is applicable to adjectives of the same form ; longo- 
for example, has in its last letter the very sound of the final 
syllable of our shallow, and with longo- must be included 
the large stock of disyllabic adjectives in o. This word 
' disyllabic ' brings before the mind the strange fact that Latin 
is utterly devoid of monosyllabic adjectives; in this respect 
differing so widely from English and even French. Yet this 
distinction has rarely, if ever, been the subject of comment, 
though it might well have been so. The theory here pro- 
pounded accounts for it. It also gives a satisfactory solution 
of the fact, that in the formation of comparatives and super- 
latives, the final vowei of the positive longo-, tristi-, is disre- 
garded. If longo- strictly means ' long-ish/ it is clear that 

* These are examples of a large class, including all those which have the 
suffix li, signifying 'like,' zspuerili- boy-like ; puellari- girl-like; aequa-li- 
(from aevo-, sb. n.) of like age. 

2 B 2 


such a suffix would be superfluous, if not entirely out of place, 
in longior- and longissimo-. 

In asserting that the Latin language had not a single 
instance of a monosyllabic adjective, I did not forget the 
adjective true-, for this word has suffered compression from 
a fuller form tor-uc-, which may be usefully placed beside 
tor-v-o-. The two words, it is well known, have in practice a 
special relation to the eye, expressing that rolling of the 
organ which marks a cruel purpose, and so at last they came 
to signify savage-looking. Thus Desdemona says : " For you 
are fatal then when your eyes roll so." If this definition of 
true- and torvo- be correct, the root-syllable is the verb ter-, in 
the sense of turning (Freund, B. 2), whence tor-no- the lathe, 
rep-par- the turning-point or limit, &c., and eventually our 
own ordinary verb turn. Or again, if the idea of piercing be 
expressed by the trux oculus, we have still the same root ; for 
one of the forms of piercing is by boring, as is seen indeed in 
the related word rop-ev-ew. On the other hand, the final 
syllable of tor-uc- represents our suffix in a less corrupted form 
than was seen in the recently cited adjectives in i. 

XV. The Suffix AGH virtually repeated in Adjectives. 

But if tor-uc- possesses our diminutival suffix, tor-v-o- has it 
in duplicate ; first in the v t secondly in the o. The same may 
be said of other numerous adjectives in uo or vo, as ard-u-o-, 
curvo-, and especially of the five allied words, gil-v-o- (also 
gil-b-o-) pale yellow, fl-av-o- golden yellow, ful-v-o- reddish 
yellow, tawny, fur-v-o- swarthy, hel-v-o- } defined by Festus as 
inter rufum et album. These words are probably but dialectic 
varieties of each other ; and also represent (setting aside the 
final o) our own ' yellow/ Probably gul, or something like 
it, is to be regarded as the root-syllable, and this identical word 
in Swedish (Danish guul) signifies ' yellow.' Even the Greek 
possesses it in <yv\-iir7ro<s } as was long ago pointed out (Philol. 
Museum, iii. 687) by one to whom linguistic studies owe much, 
and would owe more but for his present regretted silence. So 
again the Scotch have gool the corn-marigold, gule-fittit yel- 


low-footed*. We are here dealing with the root-syllable, and 
so I do not refer to such derived forms as the German gelb, 
our own gold and yolk (of an egg). 

But before we leave the Latin adjectives which possess the 
suffix in duplicate, I may point to other instances of this 
repetition, as in the Greek /u,aX-a/c-o-, and what is probably 
the very same word, the Latin fl-acc-o-. Further, we must 
include a number of adjectives which, in the first syllable 
representing our suffix, drop the vowel, but retain the guttural, 
as planco- flat, manco-, pauco-, fusco-. When writing planco- 
in this list, I do not so much refer to the cognomen of that 
form, as to what Festus says (p. 231, ed. Miiller), "plancae, 
tabulae planae-f" from which we clearly learn that our own sb. 
plank, Fr. planche, is but a secondary form of the adjective 
plane. In cor-usc-o- } and perhaps l-usc-o-, the three letters 
use probably correspond to the uc of tor-uc- just considered J. 

XVI. Some Adjectives in not deduced from AGH. 

But I do not claim all adjectives or substantives in o. In 
former papers I have called attention to the formation of 
adjectives from the genitive case, or, to express the matter 
more correctly, the habit of treating a genitive as an adjective 
so as to force it into the process called declension. Cujus 
cuja cujum is a familiar instance of this. It was contended 
too (Proceedings, vol. iii. p. 56) that such surnames as Tullius, 
Sextius, were in origin only genitives like nullms, used first 
as patronymics, son of Tullus, Sextus, &c., and finally as 
permanent surnames, precisely like our Welsh surnames, 
Roberts, Williams, &c. In the same way I would deal with 
lign-eus, ign-eus, violac-eus,rosac-eus,and perhaps with civic-us, 
hostic-us, bellic-us, apiac-us of parsley ; and still more cer- 
tainly patr-iuSj of a father, and words similarly formed. So 
again, our own mine, thine, wooden, flaxen, are in origin but 
genitives, yet now classed with adjectives. But the form 

* It is amusing to see modern philologists rushing off to Sanskrit where 
they would find in Europe more distinctly what they want. Thus, is not 
the Scandinavian gul and Scotch gool of more avail here than the Sanskrit 

f See Andrews' translation of Freund's lexicon, sub voce. 

I For sc=c or g, see below, xxviii. note. 


uullius beside nulllus suggests a question which bears upon 
our problem. It is the ordinary habit of language with the 
progress of time to pass from long to shorter sounds; and 
again, where poets differ in the forms of words from ordinary 
life, the difference usually consists in the adoption of old or 
obsolete varieties. Now unms is found in the poets alone, 
thus traversing what we have just said. Is the explanation 
this, that unms is the genitive of the more primitive un- 
one, while unms, i. e. unoius, is the genitive from the secondary 
form uno- ? That ius rather than is or us (os) is the more 
correct form of the genitival suffix, seems to follow from the 
old relatival form quo-ius. 

But over and above this, it seems highly probable that 
when the diminutival power of the suffix, lost, as the Germans 
say, its consciousness, it was still retained, or even assumed 
in new cases, solely with a view to the advantages it offered 
both for distinction of gender and convenience of declension. 
What has been said on the o growing out of ogh or ock and a 
from agh, &c. with the power of a diminutive, applies in all 
its strictness only to the earlier stages of the language. We 
have a parallel case in another allied family of languages. 
While the Old Slavic has, what I venture to write with 
ordinary Greek type, opi% nut, Trpa^ dust, crrpa^ fright, 
french-bean ; yw-te^ leathern-bag, ypie^ sin -, KO%V% skin, 
a plant, Su% spirit, the modern Servian has in their 
several places opa, irpa } arpa, ypa ; /-i^te, 7/077*6 ; KO^V, peTrv, 
Bv. (See Wuk's Servian Grammar, translated by Grimm, 
pp. 24, 25.) Nay even the Gaelic has begun to drop the 
final guttural. Thus the words bara barrow, cadha narrow- 
pass, dula noose, betray the lost consonant in the plurals 
barach-an, cadhach-an, dulach-an. 

XVII. AGH in Verbs corrupted to UG, UC or U. 

So far as we have hitherto considered the verbs and traces 
of verbs in which our original suffix agh appeared, the con- 
sonant indeed has been variously modified or even absorbed, 
but the vowel has been retained without disguise. After what 
we have seen of the change of vowel in both substantives 
and adjectives, we must not expect the verbs to be free 


from variety in the same respect. Let us proceed then to 
examine the verbs which exhibit the suffix as ug, uc or u. 

Flu-ere, fluxi, with a participle fluxus used as an adjective, 
and a substantive fluctus. Here a crude form flue- or fluff- is 
not to be disputed ; and our own language has kindred words 
in flow and billow, for the b in the latter is a more regular 
representative of a Latin/ than / itself. Outflow possesses 
our suffix in its usual English form, so that we may assume,/?, 
in both Latin and English, to have lost a central vowel. 
Thus the crude form of flu- must have beeu/o/-wc- orfal-uc-, 
or else fud-uc- (according to the propensity of the Latin lan- 
guage to interchange I and d) . But fud- is the essential 
syllable of fandere, fudi, fasum, to pour ; and what meaning 
could be more suitable to our purpose, seeing that circumfluere 
and circumfundi, profluere and profundi, are words of identical 
import? We have the same root in %u\-e<r- juice, %vro- 
(also XVO--TO-), %ur-Xo- n., and in a more corrupted form 
in %eo). 

Frui, with the substantives frug-es and fruc-tu-s. Frug- or 
fruc- being assumed as the base, we may set down for the En- 
glish and German analogues, brauchen and brook. f To eat ' 
was probably the original meaning of the Latin verb ; and we 
see this sense metaphorically retained in our own phrase ' to 
brook an insult/ i. e. to swallow and digest it. The ordinary 
sense of the German, 'to use/ is not far remote from ' enjoying/ 
which is commonly denoted by the Latin verb. But frug-, 
to follow the analogy of the preceding verb, must have been 
compressed, and we have to search for the radical portion. I 
would propose as the original form vor-ug. This might be only 
a variety of vorag-, which appears with more or less distinct- 
ness in vora-re and vorag-on-, sb. f. But as the idea of bolting 
or gulping scarcely fits itself to the uses of frui, it may be 
that the word is of different origin. As es- of esse, the first 
meaning of which is rather ' to eat' than ' to be/ had originally 
a digamma (which is seen in the German wes-en, our past 
tense was, the Norse ver-a to be, &c., and also in the Latin 
vesc-or, beside esca food), our frug-, = vor-ug-, may have for 
its first syllable what is a reasonable variety of ves- eat. 


Fug-, to fly, seems to claim connexion with the German 
flieh-en, fluch-t, and our own flee or fly, fligh-t. But if this be 
true, fug- has lost an /, and must be a corruption offlug-, a 
change of no great violence for Italy, where an I after / ha- 
bitually vanishes, as in Florentia, Fiorenze, Firenze. But the 
supposed fluff- might well arise out of vol-ug-, in other words, 
out of vol-ag-, whence vol-a-re to fly, the first syllable alone 
of which can be radical. All this is strongly confirmed by 
voluc-ri- a bird, literally a verbal adjective, l flying,' ri repre- 
senting the familiar suffix Hi. 

Loqu-i, locu-to-, is no doubt immediately akin to \ey-etv, 
but must not be considered as deduced from the Greek. 
E-log-ium is thoroughly a Latin word, and sufficiently esta- 
blishes the native rights of a lost verb leg-ere to speak, if 
indeed it be a lost verb, for it is very possibly identical with 
that well-known verb which we translate ' to read/ considering 
that this very verb 'to read' originally meant 'to speak' 
(Germ, reden). As the reflective form of loqui is well adapted 
to denote the mutuality of the act, c talk to each other/ so its 
diminutival suffix marks at once its unpretending and its 
iterative character, and so agrees with our own word tal-k. 

Lu-ere (Xu-eti/) to loosen. The Germ. adj. l&cJc-er, loose, 
seems to contain in the syllable ock a fuller expression of our 
suffix. But if ock be a suffix, we do not leave enough to con- 
stitute the base of the word, unless we assume that some 
letters have been lost before the /. This is often the case 
with an initial /, and in the present case we see a way to a 
recovery of the lost letters. As rep- of repere, repsi, is iden- 
tical with e/37T- and serp-, i. e. ep-67r- and ser-ep- ; so lu- may 
be abridged from solu- (solver e) loosen. The Latin adjective 
liber has an older orthography loeber, and as this diphthong 
oe always in Latin coexists with u } we may safely assume a 
variety luber, which stands to eXevflepo?, much as ruber to 
epvOpos. The initial vowel of e\-ev0-epos is a remnant there- 
fore of the root. The connexion of meaning between liber 
and solutus needs no discussion. 

The verb minuere seems by its power well entitled to our 
suffix, but in spite of tl is it must be rejected from our list, 


as having in all probability been formed immediately from the 
comparative minus, with the loss of the sibilant. Compare 
our verbs to ' lessen' and to ' better/ 

Nu-ere (veu-etv) to nod. 1 have elsewhere given reasons 
for believing that the initial letter of evepcn evepOe is radical, 
as well as that of the Greek preposition evt, and that the 
original sense is ' lowering down/ In the same paper it was 
urged that inferi and its related words were compressed from 
en-ef-eri, &c. To this same stock nuere, i. e. en-u-ere belongs. 

Plu-ere, to rain, is only a variety vifluere, and meant simply 
to pour ; just as we say ' it is pouring.' 

Ru-ere, to rush, or cause to rush, like every Latin and Greek 
verb that presents an initial r, is subject to a strong suspicion 
that the liquid was originally the final letter of the root- 
syllable, which had a w or * or c or h for the first consonant. 
The Greek epv-w already contains a prefixed vowel ; and we 
may not only readily accept the assertion of a lexicon that 
epv-(o is in general synonymous with eX-o), but even extend the 
assertion so far as to say that they are but varieties of the 
same secondary verb FO\-VK- or feX-e/c-, the primitive of which 
is best seen in vel of the Latin vellere; of the occasional 
violence of the act expressed in this verb, evidence will be 
found in the next section, where ulciscor is treated. The 
connexion of the Latin ru-ere with the Greek epv-ew (to draw) 
is confirmed by the familiar phrase trahere ruinam*. 

Scrutari to poke and poke again, has its original power 
best exhibited in the material phrase scrutari ignem, to poke 
or stir the fire. The simpler verb is not to be found in Latin, 
but appears in the Greek a-Kokev-ew avOpaicas. 2/eaXXetv, 
to dig, gives us a yet simpler form, the stem of which is again 
seen in <rKa\-[j,ij dagger, a-/ca\-i8- a hoe, as also in the Latin 

Sol-u-ere has already been noticed under lu-ere. It remains 
to ask what is the primitive. The following suggestion as to 
this point is offered for consideration. To let loose implies a 
previous restraint, and is commonly followed by rapid action. 
Now the root sal- (salio) is commonly translated ' to leap/ 
* See also xviii. on the verb rup-. 


but probably meant, as did the English verb itself*, to ' run/ 
quite as much as to leap (compare, for example, prosilire). 
The Greek equivalent is 6op, seen in eOopov. The noun sors, 
that which ' leaps ' out of the urn, and the current use of the 
French vb. sortir to go out, include the same idea. It is 
asked then, whether the original sense of solvere may not have 
been ' to let run' ? The English representative of solv-ere is 
to ' slack' : Note also sol-ub-ilis. 

Spu-ere as well as Trrv-ew, to spit, and the perhaps kindred 
English vb. spew, as well as spit itself, seem both by meaning 
and form to belong to the list. 

Stru-ere, struxi, struc-tum, to pile up, is the aggregate of 
many small raisings. As the s may be thrown out of view, 
it seems that tru- may well be a contraction of tol-u-, where 
tol is the essential element of tollere to raise. That tol-u-ere 
must once have existed is proved by the adverb tolu-tim, as 
used in the phrase ire tolutim, of a rough-trotting horse. 

Viv-ere, vixit, and sb. vic-tus. This verb was considered in 
a former paper (Proceed. Phil. Soc. vol. iv. p. 93), and compared 
with the Manx verb be-agh ' to eat habitually/ If what I have 
there said be true, viv- is but a variety of vesc-. The loss of 
an s, which constitutes the sole important difference between 
them, is seen in the comparison ofpascor and pastor with pavi 
and pabulum. Our own be too, which belongs to the same 
family with viv-, is proved by the Old German to have been 
once bis. Observe also that the French vecu has recovered, 
or rather preserved, the guttural. 

Volv-ere, volu-tus, Greek etXu-etv, is the exact representative 
of our wallow, which, though now used only in connexion 
with the idea of dirt, denotes in itself merely repeated rollings. 
The root vel- (feX-?) or ver- turn, and its numerous progeny, 
are too well known to need discussion here. But I may point 
attention to the forms volub-ilis and in-voluc-rum, as also to 
the French en-velop-er, the Italian volg-ere, and our wrap. 

Ferv-ere, de-ferb-ui, fyc. This verb corresponds in form 
with much accuracy to our own br-ew, the first meaning 

* Compare the German laufen and our own compound elope, Germ. 
ent-laufen, Dutch ontlopen. 


of which was probably to ' boil or heat/ without any special 
reference to the making of beer. Indeed the Welsh berw-i, 
to boil, is never used of brewing. 

XVIII. AGHin Verbs has its vowel corrupted to E or I. 

As ec and ic were the commonest forms of the suffix with 
Latin substantives, we might expect them with some fre- 
quency in the list of verbs. But the instances are few : as 

Spec-, whence species, specta-re, spectrum, and so many com- 
pounds, inspicere, &c. But both this verb and its Greek analogue 
crtceTT- of o-KC'jrrofjLaL appear to have undergone a certain change. 
If, as seems probable, sec- be the ultimate stem corresponding 
to the German seh-en, and our own see and sigh-t, the disyl- 
labic verb should have been sec-ec-; but as two successive 
gutturals were intolerable (see forceps, forfex, &c., and apex 
above), the Greeks soften the one guttural, the Romans the 
other, so that instead of scec- we have (nceir- and spec-. 

Plec- of plectere, plexus, plait, braid, and of plecti, get 
flogged, for the two words are of one origin, may be 
regarded as a compression of pel-ec-, and so substantially 
identical with pal-ag- ofplanao. The meaning in both is, to 
place one flat surface on another, with quietness in plaiting, 
with violence in plecti and planaere. 

Flee- of flectere, flexus, is probably also compressed from a 
disyllabic form. I cannot assent to those who would make 
it a variety ofplecto. 

Nee- of necto claims kindred with our own knit and net ; 
and so lays claim to our suffix. 

Nic- of nitor, nixus, or rather gnitor, gnixus. There can be 
no doubt that this verb is a derivative, as Festus assures us, 
from genu (yenuc-) or genie- (genic-ulum), and really means 
first to kneel, and then by kneeling to obtain a purchase. In 
this case however the ic is to be regarded as a substantival 

XIX. AGHin Verbs reduced to a guttural G, C, or Q. 
We next take the cases where the vowel wholly disappears, 
but a guttural, g, c, or q, is retained. 


Merg-, sparg-, ter-g-, verg- incline, verg- pour. For these 
see our 'Proceedings/ vol. iii. p. 131. 

Fulg-ere, fulsi. Of course I connect with this <f>\6y-, 
flag-rare, which, severally expanded, give us such forms as 
fol-ug-, fel-eg-, fal-ag- ; and in the ful-si we seem to have the 
simplest form of the verb. It is not unlikely that fel- may be 
the root of this verb and identical with fer- offervere. 

Pare- save, is so evidently akin to our spare, that the c 
must be a foreign addition ; and the meaning fits most aptly, 
as the verb distinctly denotes a. series of petty acts. 

Pose- might be a derivative from pet-ere. But does the 
meaning suit ? If ' to demand' were the first sense of this verb 
and its derivative^o^w/are, we shouldbe compelledto admit that 
the idea is sadly at variance with all notion of pettiness. But 
I would start from what is seen in such a phrase as posce deos 
veniam, where all is humble, though iterative. And the violence 
of the idea so often expressed by these two words, may have ac- 
crued to them in this way. An address to a court of justice is of 
course worded with all humility, and indeed our own phrase 
is, to ' petition the Court/ But what is humility in reference 
to the judge, is often an act of extreme violence towards the 
other party in the suit ; and hence as regards him, postulare 
is translated to ' arraign, impeach, prosecute, demand one's 
rights.' How words of one origin may attain a great differ- 
ence of meaning is seen in our own verbs require and request. 

Ves-c-i, to feed oneself, is a repetition of petty acts, and 
for the form of the word see ' Proceedings/ vol. iv. p. 92. 

Ulc-iscor, ul-tus. That ul- rather than ulc- is the ultimate 
root-syllable, seems implied in the participle. But the first 
meaning of the word is doubtful. If we take this to be some 
severe punishment, we may connect the verb with the neut. sb. 
ulcus or hulcus, Greek e\o? ' a wound or sore/ which, though 
sometimes treated as a root, is evidently deduced from the 
verb e\K- in the sense of ' tear.' But this is a secondary form 
of vel- (vellere) pull or tear, a root as well known in Greek as 
in Latin, and indeed the parent of vol-n-us a wound. Thus 
vel- tear, may be regarded as the parent of ulciscor. 

Tor-qu-e-re has in the first five letters a compression of 


tor-ogh, and is represented in English by th(o)row, a word of 
identical import, first in the sense of twisting, as we say 
'throw silk/ secondly, f to whirl, by means of an attached 
thong, and so throw ' a spear, torquere hast am. The ultimate 
root is seen in the simple ter- turn, whence tor-no- a lathe ; 
and also in tor-si, tor-to-, tor-tor-, tor-men-to-, &c. On the 
other hand, the secondary, or rather tertiary verb torqu-e-re 
is formed immediately from the sb. torqu-e-. 

XX. AGH in Verbs changes its guttural for a tennis lip- 
letter, besides changing or dropping its vowel. 

On several occasions in this paper I have drawn attention 
to the substitution of a labial tenuis for a guttural tenuis, to 
soften off the roughness of articulation. Thus fore-ex gave 
place to forceps or forpex (also forfex), ac-ex to apex. So 
again in Greek, from the root a/caX- dig, might have been 
deduced a fitting name for the mole in cr/coX-at;, but the ear 
insisted on o-TraX-af ; and a labial once established as a 
variety of our suffix, at times extended its dominion beyond 
the limits so justified by euphony. I proceed to consider 
such cases. 

Car-p-ere. The power of this verb, so far as it is at once 
diminutival and frequentative, is clear beyond all controversy. 
Yet the original meaning and the source of the word are not so 
evident. The simple verb is no longer to be found in Latin, 
but we probably have it in the Greek Kep- (/ceipa)), the mean- 
ing of which, before sharp tools were invented, must have 
been rather ' tear ' than ' cut ' ; and, of course, in the com- 
parison of Kep- and car-p-, we must not expect to find in the 
diminutival verb the violent action of the simple verb. In 
Liddell and Scott's Lexicon the eye catches the successive 
meanings, for Kep-, of ' cut, devour, eat up, detract from/ Now 
car-p- also is used of shearing, or rather plucking, sheep; 
of eating; of detraction. But there are uses of the Latin 
family which seem not to have had all the notice they deserve. 
With the Homeric jvire fjirap erceipov before us, we cannot 
be surprised to find the noun ear-on- (caro, carnis) signifying 
flesh. On the other hand, carpere, as applied to eating, pre- 


pares us for the noun corp-os- flesh, for such, rather than 
' body/ is its correct translation in many phrases. If our own 
verb carve, as seems probable, be connected with carpere, we 
come again to the idea of cutting, and that not uiifrequently 
in immediate relation to eating. 

Rup-, of rumpere, is but a fuller form of ru-, or in other 
words an equivalent for rue-. The connexion of rumpere and 
ruere is well seen in the compounds. Thus prorumpit se and 
proruit are not to be distinguished in power ; and ifprorumpere 
be a causative verb, so also at times is proruere as well as 
other compounds of ru-. We derived ru- ( xvii.) ultimately 
from the verb vel- pull ; but to pull a flexible surface is to 
force it into rucks. Thus we recover the guttural in the 
English words ruck and wrinkle, as well as in the Latin rug-a \ 
while the guttural, instead of being destroyed, is replaced by 
another tenuis in our rumple and rut, and in the Greek 
pvr-iB-. Note also the phrase trahere rugam. 

Scalp-ere, already noticed, and its compounds exsculpo, &c., 
have the simple verb represented in the Greek crica\- (of 
ovca,AA,&>) dig. 

Serp-ere, epirew, and rep- of repere, repsi, &c., compared 
together, inform us, as I have already said, that ser-ep- is the 
non-compressed verb. Though the simple verb ser- refuses 
itself to our inquiries, we distinctly feel in the idea of creeping 
the repetition of petty doings. 

Trep- (rpeTr-) must at one time have been a Latin as well 
as Greek verb, as proved by the existence of trepido- and 
tre-pidare 'to be confused/ more literally 'to turn first 
one way and then another/ The full form ter-ep- at once 
claims connexion with the root ter-ere to turn. See tor- 
quere above. 

XXI. AGH in Verbs exchanges its guttural for a medial 
lip-letter, besides modifying the vowel. 

I have dwelt at some length on the substitution of b* (bh) 

* The appearance of a b in substantives ; but these probably derived 
from verbs, is seen in tur-b-a (rapa;^), tur-b-on-, both ultimately from the 
verb ter- or tor- turn; ver-b-o- (n.) (comp. cpe-a>) ; ver-b-er (n.)> compare 


for the aspirated guttural of agh, as in am-ab-ilis, am-ab-a-m, 
am-ab-o, and might likewise quote ten-eb-rae and lug-ub-ris, 
vol-ub-ilis, sol-ub-ilis, fl-eb-ilis, &c., with the same object. It 
may be useful to point to a few verbs, where the b is admitted 
by all to be secondary, and, as I hold, is still referable to the 
same origin. Scribo, essentially the same with ypa(f>a>, scalpo, 
7\u</>&>, has nothing radical in what follows the r. Whether 
we should regard seal- (<TKCL\- dig) or scar- as the root-syllable, 
or perhaps rather without an s, cal or car, is for our present 
purpose of no moment. In any case ib is but a suffix, and 
my interpretation of it agrees well with the notion of the verb. 
The word yp-a<f)-co indeed lends strong support to my theory, 
for as the syllable a(f> must by all be admitted to be repre- 
sented by ib of scrib-, whence scribere (cf. for quantity con- 
scribillo), so on the other hand this same a</> is a most fitting 
substitute for agh, considering our own laugh. 

Trib- rub, so familiar to the Greek, virtually exists in the 
Latin, where we find tri-vi, trl-tus, and the sb. trib-ulum. 
It is also clearly a compression of ter-ib- from ter-o. Nay, 
the original g appears in two derivatives from a compound of 
our verb, in-tr-ig-on- y inter -tr-ig -on-. The sb. ter-eb-ra is of 
the same origin. 

Gl-ub-ere to skin, as has been elsewhere noted in the So- 
ciety's Transactions, is probably a compression of col-ub-, and 
so identical in root with col-or skin, cul-eus a skin, and the 
Greek GKV\- (<TKV\\(O) to skin, ovcuX-ecr- a skin, as well as our 
own hull ' to shell/ &c. Liber, the thin bark of a tree, is 
probably but a corruption of an obsolete glub-er, from this 
very verb. 

Illeceb-ra tends to establish illicib-, as an old form of illici-, 
the compound of the obsolete lad-. A similar argument may 
be founded on the forms elec-eb-ra, perlec-eb-ra. 

Sal-eb-ra may be quoted in favour of sal-ib- = sail- ; and 
indeed may we not here have the origin of our own leap and 
the German laufen ? Compare what is said of luere above. 

Glob-o-, sb. m. (comp. lud-o-), gleb-a clod (comp. fug-a), 

Greek apao-o-- ; bar-b-a, the last three of which with us take the forms 
rvord, rod, beard. 


and the adj. celeb-eri- (its suffix eri=ili), all point to a 
verb as the source from which they are derived, though that 
verb may be now lost past recovery. Moreover, glob-o- is 
commonly admitted to be of the same kin with glom-es-, sb. n., 
'a clue/ which by its suffix again points to a verb. This 
secondary verb I hope to have an early opportunity of 

Morb-o-y sb. m. may well come from that secondary verb 
morub- or morib-, which we have assumed as the parent of 
morib-undo-. As to the difference of meaning, I will merely 
note, that in Appleyard's ' Kafir Language/ the fourth ex- 
ample (p. 70) of their ' free use of tropes and figures/ is : fa, 
literal meaning ' to be dying/ figurative meaning ' to be sick., 

XXII. AGH in Verbs changes its guttural aspirate for an 
M, besides modifying the vowel. 

The pair of words glob-o- and glom-es- may serve as an 
introduction to our suffix when it has for its consonant m. 

But we might have arrived at this liquid by another route. 
Strange as it may appear, the Greek ^ is often represented 
by an m as well as an / in Latin, nay, often gives way to a fju 
in Greek. The case of mili-a, &c. by the side of ^ikioi has 
been often noticed ; and other cases have been the subject of 
comment in our own Proceedings (vol. iii. p. 116). See also 
Buttmann's Lexilogus, ii. 265. 7. But we have also within 
the limits of the Greek vocabulary 

Ep-6fji-y /3p-a%-, and ftp-v^-, all signifying 'roar/ corre- 
sponding on the one hand to the Latin frem-, and on the other 
to our bell-ow, and probably bar-k. 

A more interesting example is seen in rp%- beside e-Spa/jL-ov, 
where the true root of the verb is concealed in the consonants 
rp and Bp. Perhaps rpe^- is only a euphonic substitute for 
Kp-e^-j the dental tenuis being adopted, to avoid the repetition 
of two gutturals. If so, Kp may be identical with the Latin 
cur- run, and /cp-e^-, itself identical with what I deem a 
secondary verb in the first two syllables of curric-ulum. 

Pr-em- of premere seems to have nothing very substantial 
in its m, seeing that it utterly disappears in the perfect 

liY T. HEWITT KEY, ESQ. 339 

and participle. Add to this, that frequens, thronging, is a 
solitary participle without a verb to which we may refer 
it, unless it belong to premere. A search for the analogues 
ofpremere in kindred languages is attended with much diffi- 
culty ; and indeed such difficulty occurs in nearly all the cases 
where a mute and liquid commence a Latin verb. We have 
just had the case of rpe^w. And although there can be no 
doubt that trah- is represented by our drag and draw, and by 
the German trag-en, yet the assumption involves a viola- 
tion of Grimm's law which says that t Latin = th English. 
Indeed it seems not unlikely that a Latin tenuis before an r 
may have had something of an aspirate or rough breathing. 
Of this we have perhaps an example in the Greek Opacra-a for 
rapaaa-a) ; for if the had been merely a transference of the 
aspirate seen in rap-a^-rj, there seems to be no satisfactory 
reason why dapaaa-a) too was not preferred to rapaacrw. If 
so, frequens represents the sound more correctly ; and then we 
have a guide in the fact that a Latin / often supplied the 
place of a 9, and to a 6 should correspond a German d. I 
would suggest therefore, as the correlative of freq- } frequens, 
the German druck-en, druck-en, and dring-en, all denoting 
( to press/ and represented in English by throng. 

Tr-em- is a verb belonging alike to Greek and Latin. If 
tr contain the root, it may be a substitute for cr-em- } and 
deduced from the root quer-, which meant f to shake/ as is 
shown in querquera febris, the ague. (See BelPs ' Journal of 
Education/ vol. xiii. p. 315, &c.) Cf. Kap/catpa) and cor-uscus. 

Cr-em-or thick juice, as a sb. in or, implies a verb crem- 
(comp. amor, timor, pallor, &c.) ; and creb-ero- thick, sup- 
ports the claim. But creb-er is but a variety of celeberi-, 
that is cel-eb-ili-. 

Crem-a-re-, to burn, in its first four letters may well be 
akin to TrprjO-, Trip-Trpvj-fju,, the c corresponding, as it should 
do, to a Greek TT, while 6 may well be represented by the lip- 
letter m, much as ir\^0- of 7r\r)0- ecr- by pleb- of plebs or 
plebes. But of Trp^O- &c., vrp alone in my opinion can belong 
to the root-syllable ; and though our lexicographers call 777377- 
the root, they identify it with our own burn, in which as- 

2 c 


suredly the non-radical character of the n is established by 
the Latin com-bur-o and bustum. 

Let us next bestow a few thoughts on the neuter nouns of 
the Latin in men and en, or in Greek yttar and ar. Of the 
shorter forms we have examples in unguen-, sanguen-, inguen-, 
and in Greek ^Tr-ar-, <f>pe-aT-, crre-ar-. The instances in 
men and par are too familiar to need quotation. Now the 
coexistence of two suffixes, in which the sole difference is the 
possession of an initial consonant, should perhaps always lead 
to an inquiry whether this consonant does not in truth belong 
to the preceding syllable. In the present instance I have some- 
thing like a conviction that gestam-en-, crim-en-*, TroLrjjjL-ar-, 
ov-ofji-aT-j &c., are more correctly divided than gesta-men-, 
cri-men-, Trotrj-fjuaT-, ov-o-/uaT-t j and of course in saying this, 
I say also, that the more correct division of ornam-entum, 
monum-entum, is such as implies the existence of obsolete verbs 
ornam-j monum-, equivalent to orn-ab- and mon-ub-, or orn- 
agh-, mon-ugh- ; precisely as ungu-entum implies a verb ungu-. 

The same question should also be considered in relation to 
nouns in p-i) and 77, afc/JL-rj and <t>v<y-?] } to nouns in /-to? and 
05, GaXap-o- and Xoy-o-, to adjectives in t//.-o?, &c. and 05, 
fjua^ifi-o- and /eev-o- ; as also to the Latin nouns and adjectives 
of like form. 

But perhaps the Greek infinitive is most deserving of con- 
sideration under this head. Starting from the Doric infinitive 
in V, as Xa//,/3av-ev, Xey-ev (Buttmann, 81. Anm. 10. p. 358), 
and from the infinitives of the verbs in pi, Ti6ev-ai, StSov-at, 
&c., and then passing to the longer infinitives Tvirrefju-ev and 
Tvirrefju-ev-aij I am strongly tempted to regard the syllable e//, 
as of independent origin and power, so that from TfTrr-e/it-ev 
through the loss of the //, we pass to rvTrreev, rvTrreuv. In 
other words, is not rvTrr-e^- a secondary verb, from which 
is deduced an infinitive 

* Compare crib-rum. 

t The old word must have been yoi/-o/x-ar-, which would agree with the 
Latin gnosco (gnom-eri), so that yov would represent our con learn, 
=ken. The Aeolic ov-v^-ar- by its vowel v obeys the prevailing vowel- 
law seen in op-vo-o--, o8-vp-, ov-v^-, mon-um-entum, 


Lastly, in the theory now before us, applied to neuter 
nouns of the second declension, we may find an explanation 
of the anomaly which presents us with the nominatives aevom, 
helium, for the suffix agh may pass not merely into oak, but 
also into om (urn) (see glom-es, xix. ad fin.) ; and op in 
Greek, where a final //, in inadmissible, would of course be- 
come ov, precisely as the theoretic ervTrro/j, (cf. ervKTop-ev) 
takes the form erwirrov. 

This form of the suffix for our own language was not duly 
noticed in the paper on English Diminutives. It is seen in 
our words bottom, fathom, bosom, besom (or broom), blossom, 
bloom, gloom, room, and abridged in arm, harm, swarm, worm, 
helm, halm, qualm, film (suggested, except the last, by Grimm's 
D. G. ii. p. 145). Of this list the first example has its Latin 
counterpart, not merely in fundo-, but also in the fourth 
example of words in ec ( vi.). I had previously thought 
that our suffix om was but a corruption of the German en, now 
seen in the representatives of these words ; but it is clear that 
the stream runs the other way. Thus the Old German var-am 
corresponds with all fitness to our brake (bar-agh) ; and it is 
only in the later German, that we find farren-hraut, or in 
shorter form farn ; while with us the influence of the umlaut 
has produced fer (e)n. So again pod-um or pot-am and vad-um 
in Old German preceded the existing forms boden and faden 
(p. 150). 

XXIII. AGH in Verbs passes through EC to E. 

But we have passed over those cases where ec or ic would 
degenerate into e or i. We will commence with the former, 
and I first present 

Ver-e-or I fear me, to use a somewhat archaic phrase, 
but one which most literally represents the Latin exhibits 
the guttural in ver-ec-undus. Moreover, the simple verb is 
seen in the very word by which we translate it, and our 
language possesses also the secondary verb in the perf. 
part, a-fr-ai-d, and virtually in the derivative fr-igh-t and 
German fur-ch-t. As to afraid, we need not hesitate to 
regard the a as the prefix of perfect participles corresponding 

2c 2 


to the German ge, and our own obsolete y (y-clept], when we 
find it in agone, now reduced to ago, and in the vulgar, but not 
less to be respected, afeard. There seems no necessity to go 
to the French effraye to explain our afraid. 

Suad-e-re, being derived from suavic- (suavi-), must be re- 
garded as having lost a guttural. 

Dens-e-re, as I have said, implies an adjective densi-, cor- 
responding to the Greek Sa<rv-, and so also is claimed as 
containing our suffix. 

Re- in reor, ratus, calculate, is proved by its English repre- 
sentative reck, reckon, reckless, to have lost a final guttural ; 
and every word beginning with an r may be assumed to have 
also lost some previous letters, so that ec would be a suffix*. 

Rub-e-, compared with robig-on-. But if we analysed this 
word with strictness, we should write er-ub-e- as an older form, 
in which er alone is radical : cf. ep-vd-po-. 

Ten-e- has in its first syllable the idea of tendo, rewa), i. e. 
straining, and the static character of the verb, ' to hold with 
a strain, to grasp tight/ is due solely to the suffix. Moreover 
ten-eb-rae exhibits something like that which I have already 
claimed in am-ab-ilis. 

Lat-e-. Another verb of static power, with lat-eb-ra to 
support its claim. 

Scat-e-, stands beside a simpler scat- (scatit, Lucr., scatere, 
Enn.(?) ), and again is supported by scat-eb-ra. 

But the two grounds for claiming tene-, late-, &c., apply 
one or other to a large number of the verbs of the second 
conjugation, for the e in this conjugation is very generally 
considered to be something added, and that something the 
representative of a permanent state, as in hab-e-, jac-e-, 
plac-e-, tac-e-, luc-e-, pend-e-, lug-e-,fl-e-. Secondly, we have 
the support of the forms fl-eb-ilis, lug-ub-ris, expl-eb-ilis, as 
also the derived sb. pl-eb-e-, which merely means ' the many/ 
like the kindred noun 7r\-r)0-e<;-. But we shall have to return 
to the verbs in e. 

I will here only notice, that as stravi beside sterno owed its 

* I am half inclined to connect it with the very family of ca/c-ulation, 
when I find the word calc- taking the form car eg in Welsh. 


peculiar form to a secondary verb ster-ag-, so crevi, cretus 
(secret us], beside cerno, imply some forms such as cer-ec- or 
cer-eb-, which, losing the final consonant, left the vowel long. 
From this secondary verb cer-eb- or cer-ib- is deduced 
cr-ib-rum. On the other hand, cer-tus used as an adj. is in 
reality the participle of the verb in its simplest form cer-. 
Similarly sprevi and spretus imply a secondary verb sper-ec- 
or sper-eb-. 

XXIV. AGHin Verbs passes through IG or 1C to I. 

In the fourth conjugation I may first urge, that if the 
adjectives leni-, molli-, and the substantives tussi-, fini-, parti- 
part, sorti- lot, auri-, rudi-, moli- (or mole-) heap, muni-, saepi- 
(or saepe) fence, reti-, siti-, senti-, vesti-, are justly claimed as 
having lost a final guttural, a similar loss would not be 
strange in the derived verbs lenire, mollire, tussire, finire, 
partiri, sortiri, audire, erudire, moliri, munire, saepire, irretire, 
sitire, sentire, vestire. At the same time I admit that this 
argument is itself but weak ; but not so, when confirmed by 
the derived nouns prur-ig-on-, or-ig-on-, esur-ig-on-, which go 
far to establish the verb prurig- rather than pruri-, orig- 
rather than ori- (orior), esurig- rather than esuri-. Those 
who think they have got only root-syllables in our rise and 
the Latin reg-o, cannot have observed sufficiently the tendency 
of consonants and vowels to fall off from before the letter r. 
Otherwise they would feel that o in orior and opeyw is no 
euphonic addition, but an important element of the root- 
syllable. But I shall return to these verbs in i, to produce 
still stronger evidence. 

XXV. AGH in Verbs exchanges its guttural for a sibilant. 

But we have yet another kind of consonant presenting itself, 
the examination of which will be more fruitful on Greek soil. 
The verb e\avv- push, as pronounced by a modern Greek, 
presents the sound a<$> (e\a$vw), while some of its tenses 
\-a(T-<ra, r)\-a(r-0r)v, and many derived words exhibit the syl- 
lable aa. These, though apparently so discordant, appear to 


be both deducible from a common form agh, while eX of the 
Homeric etX- press, supplies a root fitting alike in form and 
sense. More than once in this paper attention has been 
directed to the truncated character of words which appear 
with an initial /. I do not hesitate therefore to connect the 
German l-ach-el-n, our own l-augh, with its labialized guttural, 
and the several Greek forms ye\-a-a), eye\-acr-(ra, ^ek-aa-pa, 
ye\-a<r~Ti>]s, and the Doric <ye\-aa). All these varieties re- 
ceive their explanation in an assumed gel-agh. I have said 
assumed, but have we not the very word in the Scotch yelloch, 
as used for example in Jamieson's quotation from Blackwood : 
" Who was merrier . . . . ? They laughed, they leaped, and 
shouted, and yelloched." But this brings us to the primitive 
yell, so familiar with ourselves. The connected family of 
words fipao-- (fipaacrco), our own brew, Lat. ferv-, and French. 
brass- er, furnish evidence of like import. Thus although the 
verbs <ye\a-, e\a-, Safjua- } epa-, have a apparently for their 
characteristic, yet by the varieties yeXacr-, eXacr-, Safiacr-, epaa-, 
exhibited in tenses and other derivatives, they all tend to con- 
firm the suspicion that something has been lost after the a of 
the Latin first conjugation, especially when the languages 
have a common verb, as Sapa- and dorna-, tame. 

As to the interchange of 7 with cr or crcr, the examples are 
past enumeration, as //,aXacro-, fipacrcra) (fipacr-), ^apaa-aw, 
Trpaa-o-a), epecrcro) (eper-), eXt<ro-a>, TrXtcra-ft), aidvcrcrw, opva-aw, 
KVWG<TW. The connexion of some of these sibilants with the 
guttural, so that fia\ay- for example may be regarded as the 
crude form of jjuakaaaw, is of course generally admitted, and 
perhaps we should not be very wrong in assigning to era the 
sound of our own sh, just as is the case with the name Kossuth. 
This form paX-ay- brings us almost to the very form of the 
suffix from which we started. And now I may point to the 
Latin verbs capesso, incipisso, petesso, &c. 

What is here seen in the Latin, is also traceable in one of 
the daughters of the Latin language, namely in that class of 
French verbs which is conjugated like finir, producing such 
forms as finissons, fimssez, finissais, finissant; the suffix of 
which has passed into the Norman part of our own tongue in 


finish*, &c. It was to these forms that I was looking when I 
said that more would be said of such verbs as finire. 

But the Italian in this very verb writes finiscolr, finiscono, so 
as to bring us back to what are called inceptives in Latin, 
lucisco, repuerasco, refrigesco\. 

Even this inceptive idea is not ill- suited to our suffix. But 
the usage of the forms in ecric-ov among the old Ionic writers 
is still more favourable to my argument, as they always 
express a repeated action : " indem sie immer eine wiederholte 
handlung bezeichnen," to use the words of Buttmann ( 94. 4, 
p. 381). Thus he employs the very term Iterativa as de- 
scriptive of the power expressed by the syllable e&K in these 

XXVI. EC or 1C, $c. exchange the guttural for a T. 

Let us now turn to another class of consonants which 
supersede the gutturals of agh, ac, ec, ic, &c. In the paper 
on English Diminutives (Transactions of the Society, 1856, 
p. 229), under the kk, instances were quoted : 1. where both 
forms coexisted beside each other, as emmet (Scotch emmock 
or immick), gimlet (Sc. gemlick), gobbet (Sc. gabbock), mammet 
(or mammock), apricock or apricot; 2. where without such 
historical evidence, a preceding guttural served as a cause and 
excuse for making the second guttural give place to a dental, 
as cricket, locket, packet, pocket, smicket, clot, grot, spigot 
and 3. cases without such excuse, which tend to show that 
suffixes once admitted under the pressure of peculiar circum- 
stances, are then carried beyond the original limits, as mallet, 
tippet, silt, wart, blot, eyot. 

* This strengthens the claim already put in for our English adjectives 
in ish in the paper on English Diminutives. 

f Thus we come to a form isc, already familiar in Greek as a suffix of 
diminutives, as veav-KTK-o-, 2vp-iovc-o-. 

J The interchange of ks, sJc, and g (as in goose) is common, as in 
misc-e-re, mix-tus, our own mix, and fjny-vvp-t -, aug-e-re and avt-~av-a> ; 
our own frog compared with the A.S. /rose and frox. So ag might well 
change with asc-. 

Our ant is but an abbreviation of emmet, precisely as our other word 
aunt is of the Latin amita. 


That the Latin abiet- is a corruption of abiec- seems esta- 
blished by the form abieg-no-. Again, vellica-, fodica-, 
morsica-, by their meaning show their title to be considered 
frequentative verbs, and in their form exhibit our suffix, so 
well suited to the purpose, and indeed exhibit it in duplicate, 
in the ic and in the final a. On the other hand, the presence 
of a preceding guttural justifies the substitution of a dental in 
place of the legitimate c in such verbs as locita-, clamita-, 
dicta-, quaerita-, agita-, rogita-. Of frequentatives ending in 
ita, which may so defend themselves on the ground of having 
a guttural in the preceding syllables, there exist over three 
hundred examples. I say this after a careful enumeration ; 
and thus the Latin language might well be tempted to carry 
the formation beyond the limits originally justified. 

Again, I have supported the doctrine, that aedili- and brevi- 
had at one time a final c, by the derived forms aedilic-io-, 
brevic-ulo-. But aedilicio- was eventually changed to aedilitio-. 
Arguing from these facts, I venture to suggest that puerilit-er, 
brevit-er, and such adverbs, have in the t a substitute for the 
c, so that er alone is the adverbial suffix. 

XXVII. Adjectives in I and : further proof of their 
having lost a final guttural. 

When claiming for adjectives, whether of the form bono- 
or tristi-, an original guttural (bonogh, tristigh), I refrained 
from using an argument of which I avail myself now. Such 
comparatives as laetic-ior, tristic-ior, if established, would as- 
suredly give a strong support to my theory, but the evidence 
of such forms is somewhat remarkable. A reading amicitior, 
in place of amicior, is supported by the three Palatine MSS. 
and seven others in Liv. ii. 15. 6, amicitior and inimicitior by 
the same Palatine and three others in xxvii. 4. 6 ; amicitior 
has also some MS. authority in Cic. ad Fam. iii. 2 and 3, to 
say nothing of later writers. Again, the MS. Harl. i. has 
laeticior in Liv. ii. 1. 2; tristicior in iv. 52. 5 and ix. 6. 3 ; 
justicior, in iv. 53. 4. Nay, in Sallust, Jug. x. 5, amicitior 
has .the support of not less than twelve MSS., and is rejected 
by Cortius as being ' contra analogiam et meliores codices.' 


Whether the latter argument be true for Sallust, I doubt, and 
certainly deny that the word is against analogy *. Moreover, 
it should be observed, that beside the four adjectives just enu- 
merated, we have the abstract substantives amicit-ia, laetit-ia, 
tristit-ia, justit-i a, in which the appearance of the it, hitherto 
unexplained, is now justified, supposing my theory to be 

But an unexplained it occurs in not a few Latin forms. 
Thus we have verbs in t-ula and ula as us-t-u-la-, pos-t -\-ula-, 
and amb-ula- ; masculine nouns in t-a and a, as nav-it-a and 
scrib-a ; neuter nouns in it-io and io, as serv-it-io- and im- 
per-io- -, feminines in t-i and i, as mor-t-i- and for-i- ; in 
it-ion and ion, as mon-it-ion- and opin-ion- % ; in it-ut and 
ut, as serv-it-ut and sal-ut- ; in it-ud-on-, as well as in idon, 
edon, ugon, &c., as mult-it udon-, cup-idon-, grav-edon-, alb- 
ugon- ; in t-ela and ela, as tut-ela- and quer-ela- ; adjectives 
in t-ili and Hi, asfer-t-ili-, aquat-ili-, ut-ili- ; adjectives in it-io 
and io, &spatr-it-io- and patr-io- \\ . In the last case, although 
patritio- is said to have the authority of the Monumentum 
Ancyranum, it is admitted that patricio- is the more genuine 

* It is not meant to exclude amicior, laetior, &c. Nay, for Livy, they 
have the best authority in the support of the Putean and Medicean MSS. 
Yet the appearance of the other forms in so many MSS. of Livy, receives 
perhaps its best explanation in the supposition that some of the tran- 
scribers came to their task with a preference, it may be a provincial pre- 
ference, for the longer forms. 

f Here pos-t- is probably the same as pos-c- of posc-ere. 

J Coercio, in Liv. iv. 53. 7, though rejected by editors, has the support 
of the two best MSS., the Putean and Medicean. 

Probably we should also divide civit-at-, densit-at-, &c. so as to make 
at alone the suffix. 

|| The opprobrium of Grammars is the assumption of disyllabic suffixes. 
These, as knowledge improves, will always fall apart into two. Sculp-tura 
should be divided as sculp-tur-a (from sculptor), vic-tric as mc-t(o}r-ic- 
(comp. doct(o)r-ina), regina as reg-in-a (Mi=the German suffix inn), while 
fac-in-os-, it-in-er-, imply obsolete secondary verbs fac-in-, it-in-, just as 
Tep.-ev-f(T- is from the secondary verb rep.()v-a>, root rep- cut. Again, 
diurno- and kiberno- are no doubt deduced from obsolete nouns (probably 
neuter), di-ur and hib-er, still surviving in the French jour and hiver. 
Should we divide the Greek comparative o-o<coT-epo so as to make the 
suffix fp as in our own wis-er, &c. ? * 


form, as also that condic-ion- is older than condit-ion- ; and 
indeed we have a similar degradation in such words as sub- 
diticio-j altered in later writers to subdititio-. We have here 
then distinct evidence of ic passing into it. 

Let us give a little special thought to the forms of which 
subditicio- is an example. There is this difference between 
a participle and an adjective, that the former refers to some 
individual act, while an adjective denotes a general con- 
dition. But this latter idea is very fitly expressed by our 
suffix, which is so often used to denote a habit or state. Thus 
edens means ' while eating,' but ed-ax ' habitually eating/ If 
then we accept ic of subdit-ic-io- as representing agh, there 
remains the suffix io, which we may well interpret as ' of or 
belonging to/ just as in patr-io-, and what has virtually the 
same suffix, ign-eo-. Thus wine imported on a given day at a 
given place by a given person is vinum ab eo importatum ; but 
if we wish to express generally the idea of foreign wine, without 
reference to the particular circumstances of its importation, 
we must say vinum import aticium, belonging to the class of 
imported wines. Thus novic-io- (where novic- represents novo-) 
denotes merely ' belonging to the class novi, a novice. But 
there still remains a little problem. Ifposit-o-, alit-o-, taking 
these as examples of a perfect participle, owe their final o to the 
same source as our theory assigns to the other nouns of this 
declension, it follows that posit-, alit-, alone represent the 
true participle. Shall we hesitate to accept this result, when 
we have the identical suffix it, so familiar in Old Scotch, as 
in abasit, heapit, straikit, gule-fittit (yellow-footed), howe- 
backit (hollow-backed) ; also in Norse, as hold-it* (held), 
brunn-it (burnt) ; in Manx, as moyll-it (praised) ; while in 
Breton the sole difference is, that et is preferred to it, as kar-et 
loved. I have passed over the Welsh, not however because it 
fails to support my view, but because it gives so decided a 
support to it, that I wished to look at it separately. The Welsh 

* Let it not be objected that haldit is the neuter ; for a neuter, though it 
may have less, has never more than the crude form, unless indeed it be a 
nominatival s, as \\\praesens,felix, and potis used as a neuter in the comic 


participle gared-ig-, loved, represents the form posito-, so 
soon as the o is regarded as having lost a guttural, for posit- 
o(gh) might well pass on the one side to posito-, on the other 
to positic-. 

XXVIII. AGH, %c. change the guttural to a D. 

But if the tenuis guttural of ac, ec } ic, pass into a t, the 
medial gutturals agh and ogh, &c. might well give place to a 
d. The Greek language abounds in examples, both among 
verbs and substantives. Thus we have 7r<uo>, Tratfo/xat, 
eTratfa, eTrai^Orjv, and iraiyviov, but vrat?, TRU^O?. 

But let us rather look to the Latin. While Forcellini has 
forty-one nouns in agon, thirty-one in igon, and nine in ugon, 
and some four in gon without any prefixed vowel, there are 
twenty-eight in edon, five in idon, and nine in don without a 
preceding vowel. That on is a suffix of the Latin language is 
proved by offend-on-, a word as old as Afranius, and con- 
firmed, it would seem, by the asperg-on- of Virgil. Now, 
setting aside for a moment the 112 nouns in tudon, and 
looking to the other nouns which present a d rather than g, 
a large half of them have the better excuse for preferring a d, 
in that the words in the first part already possess a guttural, 
g, c, or h. And if I am right in claiming multic-ud-on as the 
older form of multit-ud-on, in this class of words the same 
excuse is found. Moreover, we can at times assign a pre- 
cedence in time to those with a g over those with a d. For 
example, robigon- or rubigon-, has the authority of nearly all 
the best writers from Plautus downwards, and albugon- that 
of Pliny, while rubedon- and albedon- can appeal only to such 
writers as Firmicius and Cassiodorus. 

XXIX. The many variations of AGH, $c. in respect of 
Vowels considered. 

Here I pause to consider the vast amount of ground I have 
traversed in the treatment of our suffix, especially in the 
department of verbs. The very fact that I have assigned to 
it all the vowels in succession, and most of the consonants, 
is, to say the least, startling, and if left without further dis- 
cussion, may be a bar to the assent of my hearers. As 


regards the vowels, there is however a consideration which 
removes much of the difficulty; I mean the fact that the 
change of vowel in the suffix is an adaptation to the vowel of 
the root, in obedience to that general law which more or less 
affects all languages. Thus padaa-ao) (from pamw), /^aXac-cra), 
aXaa'O'a), \aifjiaa ~a a>, Kavacraa) (ica- 
apaaorw, rapacro-ci), ^apaa-cra), 

a(f>ao-(Tci) 3 are sufficient to establish the case of aero- ; 
stands almost alone ; Se^to-cro), eXtcrcra) or hXicrcra), 
may suffice for iacr; and oSvcro'OfjLaij /j,op/jLvo'cra) and 
\VTTOfjLai, opvaaa), Kopvaao), fiopvcrcra), (fropva'O'a) for 

It was a feeling of this truth which has guided me for some 
time past in my attempts to expand those words, where, by 
the loss of the root-vowel, the initial and final consonants have 
been brought into juxtaposition. Thus the neuter noun 
0pao-os might at once be suspected to be a compression of 
Oapaa-os, and this is confirmed both by the other form 0ap<ro9, 
and by the root which has its simplest form in our own dare. 
Again, in Opaa-aco we might have presumed the loss of an a 
between the and p } even without the knowledge of the 
fuller form rapacrcrw. 

But evidence to the same effect is visible in Latin, as in 
1. amabam, alacer, arare, amare, aratrum: 2. gemebundus, 
fremebundus, tremebundus, vereor, verecundus, teneo, tenebam, 
tenebra, terebra, celeber, expetesso, necesse, c(e)revi, sp(e)revi, 
veretrum, feretrum, f(e)retus ; 3. nitibundus, ridibundus, ridi- 
culus, incipisso, vicissim ; 4. opo<o? (beside epetyco), 6(o)pwcnc(0 
(cf. eOopov), Kcova)-^; 5. lugubris, lucubrare ; and with kindred, 
though not identical, vowels ; 6. algeo, ardeo, pateo, maneo, 
lateo, latebra, scatebra, salebra, facesso, capesso, lacesso; 
7. queribundus, c(e]ribrum, t(e)rivi, t(e]ribulum, vertigo; 8. 
volv-ere, volum-en, volub-ilis, involuc-rum, volup-e, voluc-ris, 
solu-tus, solub-ilis, tolu-tim, docum-entum, monum-entum, in- 
colum-is, mollugo, molluscus, coruscus, columba, columna. 

A comparison of allied words in Greek and Latin will con- 
firm what is here said : pocfrea) and sorbeo point to a fuller 
form <ropo(f>- ; epjr-j serp- } and rep- } to a disyllabic serep- ; 
dp7ra%- and rap- to a trisyllabic dpaTray- ; paTrrco (pa(f>-), and 
sarcio, sartum, to aapa<f>- or sarac-. 


No doubt the Latin language and even the Greek contain 
many violations of the principle, but of these some might 
easily be explained. Thus veta-re ceases to be anomalous in 
the older form votare ; moribundus may have been preceded 
by morub-undus, ludibundus by ludub-undus, as genibus was 
by genubus. It seems indeed to have been an idiosyncrasy 
of the Latin language to change broad to narrow vowels, as 
seen in sine, lingua, ignis, in (the preposition), and in (not), 
mihi, tibi, sibi, vester, verto, veto. 

But for evidence to show how readily adjoining vowels 
assimilate with each other, nothing perhaps can be more con- 
vincing within the same compass, than the following little 
series of words, in themselves substantially identical, though 
they run through the whole gamut of vowels : a. TraXXaf, 
e. Lat. pdlex, i. Eng. filly or fillie*, o. Scotch pollock, Gr. 
TTftjXo?, M. Lat. pullus. Here etymology justifies no more than 
the translation ' little young one/ It belongs to external 
causes alone that the uses of the several words have become 
variously limited as to sex and species; so that 7raXXa is 
perhaps always masculine, pellex and filly always feminine, 
and so on. Again, a Roman is thinking of a ' concubine' 
when he says pellex, and means by pullus a ' colt/ or may be 
a ' chicken/ while to a Scotchman the word pollock at once 
raises the idea of ' a young fish/ &c. 

What I have said, if true, may lead to the correction of 
what I regard as serious errors. The first of these is a matter 
on which I wrote but recently, the undue extension of the 
doctrine of metathesis, as, for example, in the case of dap&os 
and Opacros. The presence of the vowel a in both forms was 
some excuse for the doctrine that p and a had changed places ; 
but such cases are more truly explained by the theory that 
the vowel originally belonged to both syllables. An instance 
there quoted in proof of this position was the word through 
beside the German durch, while thorough contained two 
vowels. It is only by taking these fuller, though sometimes 
theoretic forms, that we have any hope of analysing such 
words, for on the other theory we should have elements devoid 
of vowels, that is, impossibility. 

* For p Lat. =/ English. 


Secondly, the knowledge that aa-cr, veer, &c. in these verbs 
are suffixes, would prevent such errors as deriving oSucro-oyiiat, 
I am grieved, I hate, from the adverb Sv<$ and the Sanskrit 
verb duish, when the radical part, denoting pain, is probably 
only oS, as in the Latin odi. 

Thirdly, I wish that some philologer would reconsider the 
doctrine of prefixed euphonic vowels, for the purpose of con- 
fining it within narrower limits. Thus the o of opvcraw is often 
treated as euphonic, but we have here in all probability the 
stem Pop-, seen in the Latin forare, foris, and our own bore. 
Nay, the Latin fod- of fodere is but a dialectic variety of it, 
just as the verb audi- stands in immediate relation to the sb. 
auri-. And to confirm this view, we have bod- as well as bore 
in our own language. Thus the piercing instrument of a 
lady's workbox, and, in Shakspeare, a dagger, bear the name 
bod-kin*. Even fora(gh]re has one of its many representatives 
among us in the secondary form to broach (a cask), i. e. bor-och. 
So again, without any initial digamma or/, the Latin has in os, 
oris, a compression of a fuller form, a neuter or-os- or 6s-6s-, 
i. e. the base of the verb with such a neuter suffix as enters 
intofriy-os-. But op-uacrco is only one example. It is a some- 
what suspicious fact that the words for which the euphonic 
initial vowels are claimed, begin for the most part with one 
of the liquids r, I, n, that is, the very letters before which a 
loss of prefixed letters is so common. Such cases as a<T7raXaf, 
(KTTrjp, stand on different ground. 

XXX. The many variations of AGH in respect of 
Consonants considered. 

But the consonants assigned to the several varieties of our 
suffix are also most numerous ; and indeed, if the matter were 
more thoroughly investigated, perhaps there is no form of 
consonant except the liquids r, /, and perhaps n, which does 
not enter into it. This also is startling ; but it should be re- 

* A young gentleman, L. E. A., aged three, was recently heard, after ask- 
ing what this instrument was, and being told a bodkin, to make the truly 
philological inquiry : ' Do you bod with it ? * Then again, to quote similar 
authority, a young lady, C. W., of the same age, was heard a few days ago 
to say : ' Mama will teach us to dite, y i. e. (w)rite. The talk of young 
children is philological evidence of the most genuine nature. 


membered, that among the consonants none are more liable 
to extensive change than the aspirates, and we may perhaps 
specially affirm this of the medial aspirates. This has been 
long noticed by Sanskrit grammarians. Thus bhri and dhri 
let us rather write them bhir and dhir are but varieties of 
one stem signifying ' to bear/ the former of which is repre- 
sented by <e/o-, fer-, and our own bear ; while the latter, fol- 
lowing that very common law which substitutes I for r, has 
for its Greek, Latin, and English analogues, roX or raX, seen 
in ToX-/z,a- and ra\-av-, tol of tollo tetuli (tuli), and our obso- 
lete verb thole ' suffer/ German duld-en; while the addition 
of our verbal suffix agh, or its equivalents, leads on the 
one hand to the Homeric fap-ea/c-ov, to the Latin fr-e-tus 
(fer-egh-tus) , to our own br-ing, br-ough-t, to the German 
br-ing-en, br-ach-te, and on the other, to the Greek e-rX-^-i/, 
Te-T\-r)-/ca, rX-Tj-TO? (tal-agh-), and to the Latin (t)l-a-tus, i. e. 
tal-agh-tus. In this it is seen that the common doctrine which 
finds two or more independent roots in the combination fero, 
tuli, latum, is without foundation. Nay, even in yvey/cov, 
(TvveveiKo/uLai,, it is probable that a form fev-e^K- may be a 
substitute for fap-egh or fer-agh = owc br-ing. The substi- 
tution of v for p in Fev-ey/c- has its exact parallel in our own 
fennel, beside the Latin ferul-a, ferul-ag-on. If the Latin ger- 
and our wear, be, as is probable, only another variety of the 
root, we have gutturals, labials, and dentals, all intermingled. 
A few familiar examples of the interchange of aspirates 
within the limits of Greek and Latin may still be useful ; as 
of </> with 6, (j>\a-, 6\a- ; <Xt/3-, 0Xt/3- ; fap, Oijp ; <j)owr}, Ooivrj 
$i suffix and Ot, ; 0aX- of 0aXX&>, / of fl-os ; epvO-po-, rufo- ; 
Ovpa-, fori-, and fora- offoras ; Opaav- , forti- ; OV/JLO-, fumo- ; 
6 with ^, opvix-os, opvi6-o<$', e/o(u)%-ojLuu, t]\v0-ov; KaX^Sov-, 
Carthagon-; % with </>, e^t?, ocfris ; ^akivo-, freno--, %?- 
(X e(0 ) fud- (fundo) ; aypivo-, funi- ; %oXo-, feJ- ; xpa- (r.), 
fru- (r.); a-ftiS- (o^tg) , fid- (findo) ; hordeo-, fordeo-, &c. 

XXXI. Conclusion. 

The present paper might receive much light from a fuller 
examination of the Greek vocabulary ; but this task must be 
left to others, as also the consideration of the evidence which 


the Sanskrit would furnish. Unhappily, Sanskrit scholars 
are for the most part so wedded to the theories of the native 
grammarians, that they hesitate too often to form an inde- 
pendent opinion. With my own knowledge of that language, 
I should be utterly without justification, were I to undertake 
the desired inquiry, yet I have seen enough to satisfy me 
that my theory would obtain valuable support from this 
quarter. Even the nouns in u, taken by themselves, afford 
evidence which to me seems irresistible. 

If my views be right, the various representatives of the 
earlier suffix agh, which is not unfrequently repeated in the 
same word, supply no small per-centage to the elements of 
the Latin and other vocabularies. Often indeed the spirit of 
the diminutival power has evaporated; but the result has 
been to supply the language with a variety of words, in them- 
selves possessing little or no distinction of sense, yet available 
for very different purposes, as the occasions of life may sug- 
gest. Thus in our own language the verbs, burrow, bury, 
broach, break, as well as the substantives furrow, breach, 
broach, prong, fork, row, rank*, &c. are probably formed by 
the juxtaposition of a common root, of which an English re- 
presentative is bore, with a common suffix denoting ' little ' ; 
yet how divergent are the applications of these words. 

Lastly, are we not entitled to set a higher philological 
value on our own language, when we find that it possesses 
simpler forms than most of its sisters ? While we have the 
primitive verbs fear, tame, dare, veer, bell, con, ken, hear, see, 
wake, the primitive substantives ware, heel, mill, hill, hide, 
hand, nose, and monosyllabic adjectives without number, the 
classical languages in all these cases exhibit but secondary 
formations ; and the superiority would be still greater, if we 
treated with less neglect our Scotch and provincial dialects. 

* The last two words in this series may prove their connexion with fur- 
row through the evidence of the Greek opxos, i. e. opoxos or opv\os. The 
grammarians seem to connect this word with opvo-o-co (see Liddell and 
Scott) ; and if we translate it ' a trench or furrow,' especially for planting 
vines, we have a meaning which well accords with its use. The Latin 
or-d-on- too is of the same stock, and so the word is used with its proper 
power in such sentences as ' quae arbores in ordinem satae sunt,' Var. R R. 
i. 714, and Horace's 'Est ut viro vir latius ordinet arbusta sulcis.' 



A final, in Latin nouns, did not ori- 

Derivations, English, continued. 

ginally mark gender, 300 n. 
a, as a suffix of Latin verbs, 302 &c. 

father, 216. 
fillie, 351. 

nasty, 174, 217. 
nave, 182. 

ab, as a suffix of Latin substantives, 

flock, 248. 

nickname, 183. 

ay, as a suffix of Latin verbs, 302 &c. 
agh or ach (Keltic) ' little,' its repre- 
sentatives in Latin, 295-354. 
Algonquin group of languages, ad- 
ditions to; theBethuck 58, Shyenne 
61, Blackfoot 61, Arrapaho 62, 

gable-end, 249. 
gib, 248. 
glen, 247. 
gore, 182. 
gules, 226. 

oil, 173. 
ordure, 189. 
orts, 188. 

rest, 143. 

Fitzhugh-Sound 64. 
America; Dr. Latham on the lan- 

holy, 226. 

row, 354 n. 

guages of Northern, Western, and 
Central, 57-115. 

javelin, 249. 

-some, 178, 179. 
sorcerer, 186. 

Assimilation of vowels in Latin, 349 

sorrel, 246. 


Athabaskan group of languages, 65-70. 
Atna group of languages, 71. 

kill, quell, 175. 
king, 174, 217. 
know-ledge, 177. 

s-par-r-ow, 221. 
such = so-like, 

ATJFRECHT, THEODOR; on the deri- 

vation of sons, 115-118 ; of otium, 
143, 144 ; on the terminations tia, 

lack, 185. 
laugh, 344. 

tarry, 226. 
turnip, 182. 

tio-, 144-146. 

leisure, 184. 

lie (falsehood), 


California, languages of, 74-87. 


woe, 221. 

Old lai 

iguages of, 87-91. 
orm explained, 322. 
A alamenca language 

like, 177. 
-lock, -lok, -ly, 
-ly (adverbial), 

winnow, 226. 
w<?d-lock, 177. 
wh-ich, 175. 

canwm (Lat.), the 
Costa- Rica, the 1 
of, 112. 

176, (adjecti- 

wise, .grmse, 176, 

Derivations of words : 

val), 177. 

178, 217. 
work, 316. 


mammet, 245. 

worthe (wel,wo), 

afraid, 341. 

bur (flock of 

mangle, 246. 


ant, 345 n. 

wool), 181. 

meed, 174. 

auger, 181. 

burin, 180. 

must, 184. 

year, 175. 

and see Key on English Diminutives, 

bite, bit, 180. 

cherte, 294. 

219-250 ; Index, p. 356, col. 2. 

bodkin, 352. 

cran-, crow-ber- 

booty, 150. 

ry, 173. 


bore, 174, 180. 
break, 304. 

en (adject.), 237. 

bourre, 181. 
burin 181. 

javelot, 249. 

brew, 344. 

ev-il, 234. 

ZomV, 184. 

bring, 353. 

broach, 352. 

fanatics, 218. 

finissant, 344. 

aw, 183. 




Derivations continued. 

Derivations, Latin, continued. 


merx, 316. 

serpo, 330, 336, 

apa, 174. 

rjveyicov, 353. 

metuculus, 320. 
morbus, 338. 

similis, 177. 

affTV, i&o. 

idetv, 137. 

municeps, 219. 

socer, 140 n. 

IffT-nui 138 n. 

mu-^o, 180. 

soZvo, 330, 331. 

/3opa, (3ptoffK(t), 

50^5, 115-118. 


KTav-, 118. 

no/pus, 182. 

sonticus, 117. 

^SpcKTffw, 344. 

necto, 333. 

sorbeo, 350. 

Xtav, 183. 

wZw, 333. 

wrea^, 312. 

yeXaw, 344. 

1 Sft 

wwo, 331. 

soro, 332. 

jUJJKfaJV, lot). 

nurus, 186. 

spargo, 304. 

Spvs, 321. 

fivla, fivs, 115 n. 

specio, 333. 

OtKOS 135. 

opera, 185. 

sternax, 312. 

i >/* 
eap, loO. 

opwTffw, 352. 

07*0*0, 354 n. 

sterno, 303. 

5> QOQ 

ecpauov, ooo. 

opxos, 354 n. 

otium, 143, 144. 

sfr-Mo, 332. 

eiKOfft, 134. 

6s, 142. 

ta-li-s, 175. 

6KWV, 138. 
X > t ) 

o\|/, ocrffa, 130. 

parco, 334. 

tenebrce, 342. 

eXauvw, 34o. 
evvvpi, 138. 

7raXXa, 351. 

pellex, 351. 
plango, 303. 

torqueo, 334. 
torvus, 326. 

67TOS, 617TOV, 129. 

TroXis, 319 n. 

^Zefte.9, 339, 342. 

trctbes 319. 

epyov, 131. 

plecto, 333. 

troiho, 304. 

effojuai, 309. 

trivis, 118. 

podex, 312. 

evw, iffx^f 139 n. 

(TTTOtXa^, 335. 

porcus, 173. 

tremo, oo9. 
trepidus, 336. 

lind-wurm, 186. 

posco, 334. 
premo, 338. 

^5o, 337. 
tristitior, 346. 


quercus, 322. 
quisquilice, 187. 


a&ies, 346. 
aZacer, 307. 
amicitia, 347. 
amicitior, 346. 
aper, 318. 
arbiter, 186. 
ar#, 316. 

caZ-c-, 319 n. 
callum, 318. 
ea&r, 315. 

flaccus, 327. 
/atjws, 326. 
/wo, 329. 
/or/ear &c., 314. 
/oro, 352. 
frango, 304. 
/remo, 338. 
frequens, 339. 
/rwor, 329. 
/w^t'o, 330. 

m>r, 342. 
rumpo, 336. 
runcare, 189. 
rwo, 331. 

salictum, 317. 
saZio, 337. 
sarcio, 350. 
scalpo, 336. 
scrutor, 331. 

ulciscor, 334. 

verier, 336 n. 
verbum, 336 n. 
vereor, 341. 
verecundus, 341. 
volucris, 330. 
coZvo, 332. 
vorare, forare, 

capesso, 344. 
carpo, 335. 

gaudeo, 174. 
#;Zs, 326. 

Digamma, in Homer's verse, 125 ; 
words requiring it : aarv, 127 ; PTTOS, 

can>, 335. 

glubo, 337. 

eiTrov, o\//, offffa, 129 ; epyov, 131 ; 

convitium, 145. 

gracilis, 306. 

erijs, 132 ; eros, 133 ; eiKofft, 133 ; 

corpus, 336. 
coruscus, 339. 

gramen, 307. 

HXis, 134 ; OIKOS, 134 ; apves, 135 ; 
ayvv/ii, 135; apaios, 136; fap, 136; 

eras, 314. 

illecelra, 337. 

e^vov, 137 ; e^eiv, 137 ; evvvp,i, 137. 

cremo, 339. 

iw-muni-s, 219. 

Diminutives, English ; Key on, 219- 

cremor, 339. 

in-col-umis, 180. 


cribrum, 343. 

-oc/fc, 223-225. 

,y^ w _ oc ^- 1 ^ 

curriculum, 338. 

latus, 353. 

-ow, 225-227. 


Ztfier, 337. 

-ic&, 227. 

-chen, &er. = icA 

domo, 344. 

ZiJer, 330. 

-te, 227. 

-few, 240-244. 

m>, 309. 

ZiceZ, 184. 

-eZ (aZ, iZ, Z), 

ach, -ag, -an, 

lilium, 173. 


Gael. 240. 

facesso, 344. 

-Zi-s, 176. 

-er, 232. 

-Z# = eZ -f- iw^, 

/aZtf, 315. 

loquor, 330. 

-e, 232. 


/ero, 353. 

lugubris, 342. 

-ew (om), 232. 

. Ze = eZ-(- e, 244. 

fervo, 332. 

Zo, 330. 

-%, 234, 239. -reZ = er+*Z,244. 



Diminutives connected with iteration, | 

English Idioms, continued. 


a, proposed for ex- 

prise or price, 
hold at, be in, 

ta"kekepe = heed, 
159 ; upon, 

amination, 227-239. 

&c., 157. 

tent, 159. 

Key on, 295-354. 

teach of, 164. 


prove of by, 

tives, 295. 
Distributed ; Dr. Latham on the word rebelnesse to = 

tell, by, 159 ; 
store of, deynte 
of, 160. 

as used hi logic, 190-195. against, 172. 

thank of = for, 

remember of, up- 
English Idioms; Perowne on some, on, on, 163. 

to, after verbs, 

146-172. || repreve of, 162. 

&c., 169-172. 

able to, 171. 

give o/=/or,163; rich of=in, 167. 

touch to, 170. 

accord to = with. 

of, 164. 

trespass to, unto, 


glade of her, 168. serve fo, 169. 


annoy to, 170. 

served of '= with, 

be to, 171. 
beseech of, 162. 

hand, have upon, 166. 
151 ; take up- set, of, in, be- 
on, bere on, hinde,&c.!58. 

upon, 172. 

bid to, 169. 

bring to, 152. snow o/, 164. 

well, the be, 

bilovid with = by, 

have to son, snybbe (snub) of 

worthe, 160. 


blame, to fall in, 

daughter, &c., =for t 162. 
171. spare to, 171. 

wite of, 162. 
wo, be, worthe, 

set in, 149. 
boot, to do, make, 

helpen of = cure spede to, 170. 
of, 162. surmountede of, 


worthy of honde, 


herken of = to A 163. 


cast, chere, 150 ; 
loke, wittes, 


hope of = for, 

take of, 158; 

write of with, 

loue, lesinge, 
151 ; cast off 


English substantives in om, 341. 

the herte, 151. 
cease of = from, 

inure with = to, 

jft, 344. 
of, 354. 


, an iqui y 


change with = 

make, amendis 

fanatics, introduction and derivation 

for, 148. 

of^for, 153; 

of the word, 218. 

chosen of = by, 
comaunde to, 
couns e 

it straunge, 
wys, 153; him 
153; wo, 153; 
of=for, 164. 
most of synne, 

Finnish language 
with the Indo- 
, miscellan 
lustrated from, 

Affinities ^ 

s, their connexion 
3-ermanic, 172-179. 
eous etymologies il- 
ithBasque, 216-218. 
Friesic the fons et 
iglish, 196-215. 
verbs, how formed, 

Friesic, the Old- 
origo of Old-Ei 
Future of Latin 

do, boot, 150; 
-wo, 154; do 

nede to = for, 

309, &c. 

to go, maken, 


Gallatin's Essays on American Philo- 

154 ; do = 
make, 155, = 
put, 155; used 
with nouns,ab- 
solutely, pas- 
sive, as an aux- 

obeye, unto, to, 
of, after verbs, 
161-165; after 
adjectives, 166 


logy, a supplement to, 57-115. 
Greek adjectives in ajc- and ay, 311. 
diminutives in ate, 298. 
infinitives, 340. 
iterative verbs in eaicov, 345. 
substantives in /war-, 340. 

iliary, 156. 


Gutturals, final, disappear, 297, 328. 

enjoin to, 169. 
espy of, 162. 

pass qf=in, 163. 
payd of, on,= 
with 165. 

HETTEMA, M. DE HAAN; Hints on 
the Thesis "the Old-Friesic above 

flatter with, 148. 
forgive to, 170. 

pray of = for, 
161 ; to, 162. 

all others the font et origo of the 
Old-English," 196-215. 



Homer ; Yates on the irregularities of 
his versification, 119. 

Honduras and San Salvador, lan- 
guages of, 109. 

KENNEDY, JAMES ; on some affinities 
in the Basque language with words 
referred to the Finnish and Indo- 
G-ermanic languages, 216-218. 

KEY, T. HEWITT ; on English Dimi- 
nutives, 219-250. 

, on the representatives of the 

Keltic suffix agh or ach ' little,' in 
the Latin vocabulary, 295-354. 

knuckle, its representative in Latin, 

Jet, its transmutations in Greek and 
Latin, 116. 

Kutani language, 70. 

LATHAM, Dr. E. G-. ; on the languages 
of Northern, Western, and Central 
America, 57-115. 

; on the word distributed as 

used in logic, 190-195. 

Latin representatives of the Keltic 
suffix agh or ach ' little,' 295-354. 

adjectives in -ac-eo-, 300-319; 

in aci- or ac-, 307, 311, 312; in 
ab-ili-, 308 ; in ab-undo-, 308 ; in 
icio, 346 ; in till-, 347 ; in ib-undo-, 

adjectives in i-, origin of, 324, &c. 

substantives in a have lost a final 

guttural, 300, &c. ; as also those in 
e, 319 ; in i, 319 ; and in u, 319 ; 
and those in i and u are diminutival, 

substantives in ag-on-, 301 ; in 
ac-ulo and ac-ro, 306; in ec- and 
ic-, 312, 313 ; in ec-to- or eto-, 316 ; 
in men-, 340 ; in ig-on, 343 ; in tura, 
347n. ; in tion-, 347 ; in gon- and 
don-, 349. 

substantives in o, origin of, 318. 

i- and e-, as nubis 

or nubes, 323. 

verbs in a have lost a final gut- 
tural, 306 ; as also those in i, 343. 

frequentative verbs, 346. 

metuculus, a Latin word, 320. 
Mexico, languages of Northern pro- 
vinces of, 91. 

-, New, languages of, 96. 
-, G-uatimala, &c., 107. 

Moskito country, languages of, 111. 

Norse neuter participles in it, 348. 
nulllus, origin of this genitive, 327. 

otium, Aufrecht on the derivation of, 

Past -imperfect of Latin verbs, how 

formed, 309. 
Perfect-passive participles of Latin, 

PEEOWNE, J. J. S. ; on some English 

idioms, 146-172. 
<t> = x = 0, 353. 

quercus, its possible origin, 322. 

a or erf, words requiring this prefix, 

inHomer, eKvpos, 139 ; riSvs, 140n.; 

eOos, 140 n. ; eicas, eicaaros, 140 ; 

ow, 01, e, os, 141, 140 n. ; 'Hpa, 142. 
Sanskrit ksh, equivalents in Greek and 

Latin, 115. 
sons; Aufrecht on the derivation of, 


terminations, E. -ch, -Ik, -ly, -lok, L. 

-U-, G. -XiK- ; Wedgwood on, 175- 


, English ; Key on, 229-250. 

Latin, represented by the 

Keltic agh or ach ; Key on, 295-354. 
, -tia, -tio- ; Aufrecht on, 


Texas, languages of, 100. 
tree, its representative in Greek and 

Latin, 320, 321. 

Utah, languages of, 96. 
Yeragua, languages of, 115. 

WEDGWOOD, HENSLEIGH ; further ob- 
servations on the connexion of the 
Finnish and In do- Germanic classes 
of languages, 172-179. 

; miscellaneous etymologies 

illustrated from the Finnish lan- 
guages, 179-189 ; (bore, lurin, bur, 
180 ; auger, 181 ; turnip, 182 ; 
Xiav, 183 ; nickname, 183 ; leisure, 
184 ; lie, 185 ; lack, 185 ; G. lind- 
wurm, 186 ; /LIT/KWV, 186 ; L. nurus, 
186 ; arbiter, 186 ; quisquilise, 187 ; 
runcare,189 ; orts,188; ordure,189.) 

YATES, JAMES ; on the irregularities 
in the versification of Homer, 119- 


Friday, January 11, 1856 (at the London Library, St. James's Square). 

Professor KEY in the Chair. 

Theodore Aufrecht, Esq. and M. Lothar Bucher were duly elected 
Members of the Society. 

The Paper read was " On the Connexion of the Lapp and Finn with 
the other Indo-European Languages;" by Hensleigh Wedgwood, Esq. 

January 25, 1856. 

JAMES KENNEDY, Esq. in the Chair. 

The Paper read was " On the Liquids, especially in relation to certain 
Mutes ; " by R. F. Weymouth, Esq. 

February 8, 1856. 

HENSLEIGH WEDGWOOD, Esq. in the Chair. 

The Paper read was " On English Idioms their change since Chaucer 
and Gower," Part I. ; by the Rev. J. J. S. Perowne. 

February 22, 1856. 

JAMES KENNEDY, Esq. in the Chair. 

The Rev. J. R. Peake was duly elected a Member of the Society. 
Mr. Burckhardt Barker's Practical Grammar of the Turkish Language, 
1854, was presented by Mr. Quaritch the publisher ; and the thanks of the 
Members were returned for the same. 

The Transactions of the Kilkenny Archaeological Society for the years 
1852 and 1853 (exchanged for the Philological Society's Transactions) 
were laid on the table. 

The Paper read was " On Latin Diminutives," Part I. ; by Professor 

March 14, 1856. 
Professor KEY in the Chair. 

Signer Bernardino Biondelli's "Poesie Lombarde Inedite" was presented 
by him, and the thanks of the Members returned for the same. 

The Paper read was Extracts from " Remarks upon certain Words fallen 
out of good usage and preserved in the speech of the common people living 
on the southern border of the county of York ; " by Joseph Hunter, Esq.* 

March 28, 1856. 

THOMAS WATTS, Esq. in the Chair. 

George Metivier, Esq. of Guernsey, was duly elected a Member of the 

The Papers read were I. " On the Etymology of ST^OS-; " by Professor 
Key. II. Further Extracts from " Mr. Hunter's Provincialisms of the 
southern border of the county of York*." 

April 11, 1856. 
Professor KEY in the Chair. 

W. J. Brodribb, Esq., M.A., Fellow of St. John's College, Cambridge, 
was duly elected a Member of the Society. 

A Pamphlet by Professor Lassen, " On Lycian Inscriptions," was pre- 
sented, and the thanks of the Meeting returned to the donor. 

The Paper read was "Miscellaneous English Etymologies;" by Hens- 
leigh Wedgwood, Esq. 

* Not intended for printing in the Society's Transactions. 


April 25, 1856. 

Sir J. F. DAVIS, Bart, in the Chair. 

The Rev. J. Dundas Watherston was duly elected a Member of the 

The following Works were presented, and the thanks of the Society 
returned for the same : 

Sanskrit Derivations of English Words, 1856 ; by Thomas Bellott, Esq. 
The Journal of the Ethnological Society, vol. iv. ; by the Ethnological Society. 
Report of the Proceedings of the Annual General Meeting of University Col- 
lege, London, Feb. 27, 1856 ; by the College. 

The Papers read were I. " On the cause of some of the Irregularities 
in the Versification of Homer ; " by James Yates, Esq. II. " On the Pro- 
vincialisms of Hallamshire ; " by Joseph Hunter, Esq.* 

May 9, 1856. 

Professor GOLDSTUCKER in the Chair. 

The Papers read were I. " On the Languages of Northern, Western, 
and Central America;" by Dr. Latham. II. "On the Derivation and 
Meaning of 7777109," and " On the Welsh Affix of Equality ; " by Dr. 

May 23, 1856. (Anniversary Meeting.} 
JAMES KENNEDY, Esq. in the Chair. 

The Treasurer's Cash Account, as approved by the Auditors, was read 
and adopted. [See last page.] 

The following Members of the Society were elected its Officers for the 
ensuing year : 

President. The Right Rev. The Lord Bishop of St. David's. 

Vice-Presidents . 

The Right Rev. The Lord Bishop of London. (Bloomfield.) 
The Right Hon. Lord Lyttelton. 

Edwin Guest, Esq., LL.D., Master of Caius College, Cambridge. 
H. H. Wilson, Esq., Professor of Sanskrit, Oxford. 

Ordinary Members of Council. 

Theodore Aufrecht, Esq. | J. M. Kemble, Esq. 

Rev. J. W. Blakesley, B.D. James Kennedy^ Esq. 

E. H. Bunbury, Esq. 
Campbell Clarke, Esq. 
P. J. Chabot, Esq. 
Rev. T. O. Cockayne. 
Sir John F. Davis, Bart. 
E. B. Eastwick, Esq. 
Theodore Goldstiicker, Esq. 
Joseph Hunter, Esq. 

R. G. Latham, Esq., M.D. 

Rt. Hon. Sir G. C. Lewis, Bart. 

E. L. Lushington, Esq. 

Henry Maiden, Esq. 

Rev. J. J. S. Perowne. 

Rev. R. Scott, D.D. 

Rev. A. P. Stanley. 

Thomas Watts, Esq. 

Treasurer. Hensleigh Wedgwood, Esq. 
Hon. Secretaries. T. Hewitt Key, Esq. ; Frederick J. Furnivall, Esq. 

The Chairman read the following letter from the Secretary of the Royal 
Astronomical Society : 

To the President and Council of the Philological Society. 

Eoyal Astronomical Society, May 1856. 

GENTLEMEN, I am instructed by the Council of the Royal Astronomical So- 
ciety to offer you the use of their apartments for the Meetings of the Philological 
Society, so long as such arrangements shall be found to suit the convenience of 
both Societies, and subject to your defraying the expenses of your Society for 

* Not intended for printing in the Society's Transactions. 


gas, coals, &c. Of course this offer implies that your days of meeting are not to 
be those of the Koyal Astronomical Society, which are on the second Friday of 
each month from November to July, both inclusive, and on the Wednesday pre- 
vious when the second Friday of April happens to be Good Friday. 

The possibility of urgent necessity requiring a Special Meeting of the Koyal 
Astronomical Society on one of your days of meeting, of course exists, but the 
contingency, judging from experience, is extremely remote. 

I am, Sir, your obedient Servant, 

WABREN DE LA KITE, Secretary. 

The Society resolved I. That the Philological Society gratefully accept 
the offer of the Council of the Royal Astronomical Society ; and that 
the Honorary Secretaries signify such acceptance to them, and express 
the great gratification with which the Philological Society has received 
this offer of help from so long established and so numerous a Society as 
the Royal Astronomical. 

II. That the Council be empowered to make such arrangements for quit- 
ting the London Library, holding the Society's Meetings at the Rooms 
of the Royal Astronomical Society in Somerset House, altering the 
Society's days of meeting, and otherwise in the matter, as they shall 
think fit. 
The Papers read were I. " On a Zaza Vocabulary," by Dr. Sandwith ; 

communicated by Dr. Latham. II. " On the Nasalization of Initial Mutes 

in Welsh ; " by Theodore Aufrecht, Esq. III. " On Latin Diminutives," 

Part II. ; by Professor Key. 

June 13, 1856. 
Professor KEY in the Chair. 

The following book was presented : " Studii Linguistici," Milano, 1856, 
by Signor Bernardino Biondelli, and the thanks of the Society returned to 
the donor. 

The Papers read were I. " On the Etymology of the Adverb actutum ; " 
II. " On the Etymology of the Noun otiurn ; " by Theodore Aufrecht, 
Esq. III. " On Latin Diminutives," Part III., by Professor Key. 

June 27, 1856. 
E. B. EASTWICK, Esq. in the Chair. 

Mr. Furnivall stated that the Council, in pursuance of the Society's 
Resolution in that behalf on the 23rd of May last, had arranged (1), with 
the Committee of the London Library that the Society should quit the 
Library on paying rent up to Michaelmas next ; (2), that the Society's days 
of meeting should be the first and third Thursdays (instead of the second 
and fourth Fridays) in every month from November to June, both inclusive ; 
(3), that on and after Thursday, Nov. 6, 1856, the Society's Meetings would 
be held at the rooms of the Royal Astronomical Society in Somerset House. 

The Papers read were I. " On the Etymology of the Latin sons ; " by 
Theodore Aufrecht, Esq. II. "On the Interchange of n and d;" by Professor 
Key. III. A Theory on the origin and principles of the Sanskrit declension, 
&c. (being an enlargement and revision of the Paper read before the So- 
ciety on the 24th of November, 1854, and not yet printed*), by Professor 

Thursday, Nov. 6, 1856. 

First Meeting of the Society in the Rooms of the Royal Astronomical 
Society, Somerset House. 

* This Paper, recast, will probably be printed in Two Parts in the Society's 
Transactions for 1858 and 1859. 


Professor KEY in the Chair. 

The following presents were announced, and the thanks of the Society 
voted to the donors for the same : 

Oude Friesche Wetten, 3 vols. 8vo ; 2 vols. 4to. 

Jurisprudentia Frisica, 3 vols. in one. 

Het fivelingoer en oldampster landregt. Het Emsiger Landregt van 1312. 

Oude en nieuw Friesland of aardrrjkskundige Beschrijving van die provincie. 

Dait oajlaon wangeroog. Het Eiland Ameland. 

Thiu Foeroiske spreke. Het Meer Flevum en het Eiland Flevo. 

Friesche spraakleer van Rask. All from and by M. de Haan Hettema. 

Prieuwecke fen Friesche Eijmmelerije. 

Vergleijking van oud-noordsche met oud-Freische Eigennamen. 

As. Siemme it Lyemen in Blyspul mit it ingelo fen Wm. Shakspeare. 

De Keapman fen Venetien in Julius Ca3sar, twa toneelstikken fen Willem 

Shakspeare. All from the Rev. It. Postumus. 
Vergleichende Grrammatik, Part I. From Prof. Franz Bopp. 
The one Language before the Flood, Five Nos. Rev. J. Smisby. 
Eeport of the Literary Institution of the Friends of Poland. The Institution. 
Chinese Numismatics. John Williams, Esq. 

Several parts of the Transactions of the Kilkenny Archaeological Society 
and of the Notices of the Meetings of the Royal Institution, with their 
Annual Report, given in exchange for the Society's publications, were laid 
on the table. 

The Papers read were I. " Further Observations on the Connexion of 
the Finnish and Indo-Germanic Classes of Languages ; " by Hensleigh 
Wedgwood, Esq. II. " On the Latin abstract Nouns in tia, tio- ; " by 
Theodore Aufrecht, Esq. 

Nov. 20, 1856. 

THOMAS WATTS, Esq. in the Chair. 

J. P. Bidlake, Esq. was duly elected a Member of the Society. 
The Assistant-Secretary exhibited a Cingalese Book written on the 
leaves of the Palmetto. 

The Paper read was " On the Affinities between the Northern Lan- 
guages of the Old and New Continents ; " by Lewis Kr. Daa, Esq. of 

Dec. 4, 1856. 
JAMES KENNEDY, Esq. in the Chair. 

The following Present was announced, and the thanks of the Society 
voted to the donor : Redhouse's English and Turkish Dictionary (large 
paper copy). From Mr. B. Quaritch. 

The Papers read were I. " Hints on the Thesis ' The Old-Friesic above 
all others ihefons et origo of the Old-English ' ; " by M. de Haan Hettema. 

-II. A Memorandum by Capt. Chapman, R.E., accompanying his present 
of Ten Copies of a plate of the Modifications of the Sanskrit Alphabet from 
B.C. 543 to A.D. 1200, by the late James Prinsep, Esq. III. " Miscel- 
laneous Etymologies illustrated from the Finnish Languages ; " by 
Hensleigh Wedgwood, Esq. 

Dec. 18, 1856. 

The Rev. T. OSWALD COCKAYNE in the Chair. 

The Papers read were I. " On some Words common to the Basque and 
Finnish ; " by James Kennedy, Esq. II. " On English Idioms," Part II. ; 
by the Rev. J. J. S. Perowne. III. "The Etymological and Logical 
Meanings of the term ' Distributed ' reconciled;' 1 by Dr. R. G. Latham. 






Philological Society, London