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

Full text of "American Journal of Philology"

Google 



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

to make the world's books discoverable online. 

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

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

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

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

publisher to a library and finally to you. 

Usage guidelines 

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

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

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

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

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

About Google Book Search 

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

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



\\\ 

^ 



r 



i^ 



[I 



> * • » • 



THE 



AMERICAN 



Journal of Philology 



EDITED BV 

BASIL L. GILDERSLEEVE, 

Profestor of Gretk in tk« yoknt Hopkin* Univtrsity. 



VOL. IV. 



BALTIMORE : THE EDITOR 

New York and London : Macmillan & Co. 

Leipsic : F. A. Brockhaus 

1883 



iv COArTEATTS. 

No. 15. 

I. — The New Revision of King James' Rerision of the New Testa- 
ment. III. By Chablbs Short, 253 

n.— -On the Enfflifth Dative-NominatiTe of the Personal Pronoun. By 

F. B. UUMMBRB, 283 

III. — Participial Periphrases in Attic Prose. By W. J. Alxxaudbil, 291 

IV. — Stichometry. IL By J. Rkndel Harris, 309 

Rrvtbws ant> Book Notices: 332 

Henry Sweet : An Anglo-Saxon Reader in Prose and Verse ; An 
Anglo-Saxon Primer. Richard Morris: Specimens of Early 
English. — Eberhard Schrader: Die Keilinschriften and das 
Aire Testament. — F. Mahlaa and W. Volck : Wilhelm Gesenins' 
Hehrjii<iches und Chaldaisches Handwdrterbuch Qber das Alte 
Testament. — George BUhler : Leitfaden far den Elementarcnrsos 
des Sanskrit mit UebungsstOcken and rwei Glossaren. — Edwin 
Wallace : API2T0TEAH2 HEPI tTXHZ. 

Rkports: 353 

Rheinischcs Museum. — Anglia.-— Hermes. 

corrrspondbnce, 375 

Rbcknt Publications, 380 

No. 16. 

I. — The Nf)Ct€s Atticae of Aulas Gellius. By Henry Nettleship, 391 

n, — On the Final Sentence in Greek. By B. L. Gildersleeve, . 416 

ni. — T. \u Beddoc«,a Survival in Style. I. By Henry Wood, . 445 

IV. — Notes : John Kvclyn*t Plan for the Improvement of the English 
liangiiftge. (H. E. Shepherd.) — Note on Mercator, v. 524. 
(Minton Warren.) 456 

V. — I.ist of Irregular (Strong) Verbs in Beowuli. By J. A. Harrison, 462 

KSVtKWS ANI» MOOK NOTICKS: 478 

()u^^t> History of English Rhythms. — Gartner's Raetoromanische 
Orsmnintik. — Mather's Aeschylus. Prometheus Bound. — Waifflin's 
Archiv fUr Lnteinische Lexikographie u. Grammatik. — Dunbar's 
('(inc(»r(lance to Aristophanes. — Ulrich's Bifrun's Uebersetzang des 
Neuen Testaments. — Kngelbrccht's Studia Terentiana. — Foerster's 
AUfTfln/r»sisrhe Hibliothck. — Ribbeck's Emendationum Mercatoris 
Plautinne Spicilegium. 

Kfvori«» f 503 

Kngllsche StuHien- /clts( hrift dcr deutschen morgenl&ndischen Gc- 
spDschaft — Journn) Asiatinuc — Deutsche Litteraturdenkmale — Ro- 
iiidtiia — v^chuchBidt's Kicoiischc Studien. 

Rm RNT Pl'lUICMIONS 519 



AMERICAN 

JOURNAL OF PHILOLOGY. 



Vol. IV, I. Whole No. 13. 



L— THE COLOR-SYSTEM OF VERGIL." 

Charles Lamb used to say of himself that ^darkness was his helV 
In the genius of this man there is a flavor of bright and sparkling 
childhood ; and in making hell consist of darkness, he uttered a 
feeling that is common both to all children and to all nations in the 
childlike stage of their growth. Thus our hell itself, the Old Norse 
hel^ the Gothic halja^ is probably from the same root as the Greek 
jccXau^ (black), personified as Kijfj, the goddess of death. So, in 
the very foundations of lAdo-germanic speech, darkness is one with 
death ; light is at once the essence and the symbol of physical life. 

This childlike identification of darkness with hell came up, as we 
saw in Charles Lamb's case, from the first impressions of his child- 
hood into the ripeness of the sensitive, thoughtful man. So, with 
the Greeks, the conception of darkness as the awfulness of death, 
the conception of life as the clear effulgence of light, lasted over, as 
an abiding element of their imagination, from primitive days into 
the consummate perfection of their poetry. In Euripides, for 
example, the dying Alkestis, as she feels the approach of death, 
cries out : 

(TKOTia d* fV' Haaois vif^ ((JHfmti (269). 

As she prays for long life for the children that she is leaving, it is : 

Xfupowftf & Tficya, rddt <f)dot 6p^Tov (272). 

As her women pray for her parting soul, her death is for them : 

rhv apdkiov oIkov o^iccrcvciv (43?) > 

'This paper was prepared for the Philological Association of the Johns 
Hopkins University, and read before that body on the 14th of April, 1882. 



2 AMERICAN JOURNAL OF PHILOLOGY, 

and the last ferewell of the dying woman, as she goes down, willing 
but awe-stricken, into the Valley of the Shadow, is 

So too, in Sophokles, the last words of Ajax, before he &lls upon 
his sword, are 

Kol r6¥ dul>p€vnjp *HX«w irpoatrpiww (856). 

Thus, here and in countless other passages of like feeling, there 
comes out a law of the Greek mind, a fundamental fact of the Greek 
iiuagiuation at work: darkness is for the Greek the physical sign 
i4ucl expression of death ; sunlight, in its white glow, in its shining, 
luuuilored radiance, is the physical sign, the essence and expression 
iUlife. 

In passing, however, from the poetic atmosphere of the Greeks 
i(itii the Italians', there is a further progress to be marked in this 
(.(iUt 1 :pti(in of light as the essence, the metonym of life. See, for 
e.wiinjile, how the great Italian poet describes the growing horror 
ol \\\ii hi:enes through which Aeneas went down from the sunlit life 
ui (nan into the regions of the dead — 

rebus nox abitulit atra colorem (Aen. VI 272). 

Here it is no longer mere light, it is color that emerges as the 
toi^^M of lite; and it is the loss of color that is the sign of death. 
I u jhe eyes, to the minds, to the imaginations of the Greeks, the 
I^Mie lit^lu of the sun, falling in undissolved whiteness, piercing, 
|MMtH.aiii«, almost blinding in its sharp etching of shadow and 
jIlMMiUianon, had l>een the glory of the physical universe, the charm 
u\ \\U,, the ayiuhol of all intelligence, the speech, us it were, and the 
jtviJaHon of the fiodhead itself. Form, sharply defined in the 
h|.M k .Old white of the uncolored sunlight, was for Greek imagina- 
\\it\M> the ty|*e of the hij^hest beauty. But the imagination of the 
\u\u\u:i eoiitetj out in their best poetry as something less plastic 
lh,,M the (^reeks', and more sensuous : light pours through thegor- 
lii.uMo Mahna of their fiOedc world, no longer white and undissolved 
|„U lihiken nao prismatic splendors of innumerable tints, reflected 
^Ui\ nJia* ted int<i all ajmbinations of pure and of blended colors, 
\'i,i\^ la no longer so all-sufficing, so correct, so sharply defined as 
I,, iltz \,Ui k and ^ hite, the lights and shado>fc-s, of the Greek ideals, 
t5ut loriM illumined by colored lights, K>rm losing the sharpness of 
\U outimea in tli£ tender v^:ueness of colors that melt and blend 



THE COLOR'SYSTEM OF VERGIL. 3 

with one another, such is the highest beauty of the world as the 
world was seen by Italian eyes and painted by Italian poets. 

Such was the Italian, such above all was VergiFs conception of 
the part that color, as distinguished from pure light, plays in the 
beauty of the universe. For him it was color that made the glory 
of life ; it was the withdrawing of color that made the dreadfulness 
of the dead man's world. For him, however, this separation of 
color from light, this glorification of color at the expense of light, 
was accomplished only by the stress of a grand imagination. But 
for us, strange to say, we can find in the resources of modem 
chemistry the means of changing Vergil's imagination into scientific 
fact ; we can separate light from color, and light up a dismal world 
of horrors from which all color is banished. We can see before us 
in the laboratory the very scene that Vergil makes Aeneas behold 
as he goes down alive in Hell. The process and the result of this 
wonderful experiment are well described in Rood's Modern Chro- 
matics, pp. 102-3. 

From the witnessing of such an experiment we shall come forth 
persuaded for ever of the fundamental truth of Vergil's conception : 
the glory of the physical universe consists, above all, in the diffu- 
sion of color ; and the poetic conception and the presentation of 
life before the sensuous imagination are to be attained in their 
highest effects only by the management and utilization of color. 

Among the great poets of Italy, the poets that saw and lived in 
color rather than in pure light the chief beauty of the universe, 
Vergil is the greatest. He made, as I shall show you, a prodigious 
use of color in his own compositions. He used a rich variety of 
color-terms with a delicate precision of meaning ; and by his ex- 
ample he fixed the use of color-terms and defined the range of color- 
impressions for the poetic literature of his race. I wish, therefore, so 
far as I can, to lay before you the color-system of this great master 
of poetry. I wish to explain the color-terms that he uses, and to 
arrange and analyze the color-impression that he aims to create. 
In conclusion, by comparing the results thus obtained from Vergil 
with the results obtained by other scholars fi*om Homer, I wish to 
show how far Vergil had advanced beyond Homer, both in his 
appreciation of color and in his precision of expression for color. 
By such a comparison, it seems to me that we shall reach a fair 
measure of the progress made by^ civilized mankind, within that 
thousand years, in the adaptation of human language to the color- 
sensations of the eye. 



4 AMERICAN JOURNAL OP PHILOLOGY. 

The first chromatic impression that the mind gets frova readii^ 
Vergil is the impression of his delicate and loving perception of 
color, and of the splendid richness and variety of his color-effects. 
All through his poems, as well the poems of his sensuous youth as 
the poems of his sobered maturity, whether he is dealing with ex- 
ternal nature or with human life, his verses are aglow with an irides- 
cent radiance of color. Sometimes it is still-life, a flower or a fruit 
or a vegetable, pictured with delicate fidelity to nature's coloring. 
Sometimes it is a living animal, bird or snake, gorgeous with bright 
plumage or with lustrous scales. Sometimes it is the human figure 
itself^ maiden or warrior, luminous with richness of flesh-tints or 
splendid in garments of many colors. Sometimes it is a landscape, 
-or a sky-effect, cool with the green light of the forest or glowing 
with all the radiance of the sunset Turn where you will, there is 
always color imprisoned in the sonorous beauty of the verse, until 
the gem -like Vergilian phrases seem to be alive inside with as many 
shifting hues as the beryl-stone of the poet In the Aeneid, for 
example, 8, 22, he tells us how, in a basin fed fi'om the brass lips 
of a fountain, the yellow light, flashing fj-om the brass, is reflected 
fi'om the water and sent flying in tremulous patches amid the carved 
work of the ceiling. In the second Eclogue, v. 50, the white-armed 
Naid lights up the heaps of dark blue-berries with yellow clusters 
of marigolds. In the Copa, v. 20, there is the grouping of wreaths 
of yellow weld with purj^ roses, and brown chesnuts side by side 
with red-cheeked apples, and blood-red mulberries piled up with 
purple grapes and blue egg-plants. In one exquisite poem of the 
Catalecta, VI 10, there stands like a delicate vision of antique beauty 
the statue of Amor carved in white marble with wings of many- 
colored feathers. Rising fi*om still-life to human life, he brings 
before us Aeneas himself (IV 261) : his sword was starred over 
with tawny jasper,and the cloak that hung from his shoulders blazed 
with Tyrian purple, and the threads of the purple were held apart 
by slender threads of gold. So Chloreus (XI 771) shone with red 
and purple, and a gold bow hung from his shoulder, and his cloak 
was saffron-dyed and his leggings were embroidered with many 
colors. And in landscape, as he paints the site of Venice before 
Venice arose, his picture seems to glow with all the deep splendors 
of the Venetian painters: the Po through fat fields pours its 
yellow waters into a purple sea, Georg. 4, 372. And again, with 
what soft tenderness of color he pictures the fountain flowing fi*om 
the dark mouth of the grotto : here there reigned the pinkness of 



THE COLOR'SYSTEM OF VERGIL, 5 

spring ; here the earth brought forth her flowers, and a white pop- 
lar hung over the mouth of the cavern, Eel. 9, 40. And, in another 
place, as the boats full of armed men glide along the forest -river, 
the green woods gaze with amazement upon the shining shields of 
warriors and upon the brightly painted boats gleaming back 
reflected from the quiet waters, VIII 92. And at sunrise, the sea 
grows red beneath the level rays, and the yellow Dawn rides up 
the sky in her rosy chariot, VII 25. Oftenest, however, of all he 
comes back to the richest of all color-effects, to the rainbow, mille 
coloribus arcum^ V 609. Art and science meet in the memorable 
description of Iris, flying dew-bespangled across the sky on saffron- 
colored wings, and drawing after her, as she faces the sun, a trail 
of a thousand colors, IV 700-1. Thus, in small things as in great, 
the visions of physical life and of ideal beauty that floated in the 
imagination of Vergil were visions as splendid in color, as thoroughly 
Italian in sensuous color-effects, as ever came in after-centuries to 
fix themselves upon the canvases of a Giorgione or a Veronese or a 
Titian. 

At this point it will be well for us, before we go further, to settle 
in our minds the conception that lay for the ancient mind in the 
word color. For, in our modem speech, half unconsciously, we 
gather in from the results of science the scientific conception of 
things and the scientific meaning of words. But, if we deceive 
ourselves into applying to ancient words our modem scientific con- 
ceptions, we misunderstand the thinking of antiquity and bring 
confusion and disorder into all their poetry. For us modems, under 
the teaching of Young and of Helmholtz, color b a subjective sen- 
sation produced upon three sets of nerves within the eye by three 
kinds of waves that differ in their length. When the longest wave 
strikes upon the nerves that are fitted to receive its impression, 
we see red ; when the shortest wave strikes upon its nerves, we see 
violet ; when waves of the intermediate length strike upon their 
nerves, we see green. And when waves of different lengths fall 
upon our eyes commingled, we see colors such as yellows, blues, 
purples, etc., made up of violet, green and red. A nobler theory 
was never devised by the wit of man to explain the phenomena of 
nature : the theory, for my part, takes already its place in discov- 
ered truth as a sublime law of nature, most exquisite in its adap- 
tation of wave to nerve. We must, indeed, keep this theory to 
explain the facts of color as they arise in the world ; but in dealing 
with the color-terms of ancient languages and in reconstructing the 



A,V£j?:7A.V rzTxS'AL ZF PHrZIlZCT. 



^ 



cc^OT- system of aroect yj^fs^ -m^ :r--5t 'h.j ris ib^crasdc and 
force ourselves back ino tf^t dl^ilke C':oocc6:cs r">aj arasc in the 
childhood of mankind. Tbs 3:«j=: r^*->r ^ jcs back to the nx:< cal^ 
I. G. J^a/ or skar^ cifar-T^^ to ccz€r. to cc'zzt^ I: is akic to the 
verb celare (hide ■ and occulere ocr.er', . ''/>«^ Sfrm^ker sars Pro£ 
Curtius, p. Ill, ^ycust die Farit aJ: Decke aafr Ccl^r in the 
conception of Indo-gennanic lang'.ia^e is the r^^r^x of tbirigs. So 
in Greek, xpm^ {color) is frooj the saice root as xpim 'jkin\ In the 
I. G. languages, the color of a thing is the cover or skin that over- 
lies or hides the true substance From this conception, howe\-er 
false it may be in science, we have to make our start in explainii^ 
the color-terms of ancient poetry. If we apply it to Vergil, we 
shall find the term color used in se\'eral different senses. In its 
widest sense, Vergil uses color to denote the colored skin or cover 
that overlies all visible objects, e. g, 

rebus nox abstulit atra coloran (VI 272). 

Here color is a kind of skin that may be, as it were, peeled off. 
Hy metaphor drawn from this use, r^A7r denotes the sur&ce as 
(ijiposed to the substance, the appearance as opposed to the reality 
iA things, €, g, 

nimium ne crede colori (Eel. II 17). 

WiK^X, Si^^ a natural transition, color denotes the dye-stuff used to 
^rw: ^ lAiiX to a foreign substance : 

variot dhcet mentiri tana cohres (Eel. IV 42). 

1 1,1 it iu^ii denote the dark Spots lying on the surface of a bright 
\,nt\^ ; t»o of the sun he says: 

tpiiui in vuUu varies errare cohres (Georg. I 452). 

/ |,^H foht ii» narrowed down to denote the warm colors, reds, 
f,M/ *,,;,, I U ., ^'V^^n when darkened almost to black. So in speak- 
fff/ hi * hUtf i$it U^i of the soil's fertility, he says : 

loiJtftf^um eft praediscere^ quis cm col^r (Gj;. II 256) 
. J nith t4outh druexm ah Indis (Gg. 4, 203), (b»^>^^n races, colored people). 

Af.'l h\ M*/ f/^q/J.^^r grape: 

fhHtftt apri/u in c^lZihus irru ccl^rm ^Kcl. IX 4x)V 

hr-'ff * /I//f//*/« \\v^' fcame warm colors illumin.ucd toward white, 

»pnntt «>T«/ zZz: vA«u VU lo:V 



THE COLOR-SYSTEM OF VERGIL, 7 

And, by a special narrowing, color denotes the exquisite tints of 
red and white, rose and lily commingled, that make the complexion 
of a beautiful girl : 

Indam sanguineo veluti violaverit ostro 

Si quis ebur, aut mixta rubent ubi lilia malta 

Alba rosa : tales virgo dabat ore coiores (Aen. XII 67). 

Finally, by a still further mingling of white, color comes to denote 
the palest yellow, almost white itself; so he says of horses 

color deterrimus albis | et gilvo (Gg. Ill 82). 

Vergil, you will remember, in speaking of the rainbow, called it 
milie coloribus arcam. Only by a bold guess, of course, by the 
lucky divination of genius, could he have given as one thousand 
the number of distinct colors in the solar spectrum. Yet, strange 
to say, Aubert has proved by his famous eltperimentS and calcula- 
tions (cf Rood, p. 40), that in the solar spectrum, as spread out by 
artificial means in the laboratory or by natural means in the gor- 
geousness of the perfect rainbow, the unaided human eye can see 
and distinguish 1000 different colors. In addition, however, we 
must bear in mind that one of the colors commonest both in art 
and in nature, the purple, is not seen at all in the solar spectrum. 
Hence, in order to get the number of distinct colors that the eye 
can see in nature, to the 1000 hues of the perfect spectrum we 
must add on about 100 more for the full and the graded hues of 
purple. Let us assume, then, that, in nature seen under ordinary 
daylight, there are for the healthy human eye about 1 100 distin- 
guishable colors. By different degrees of white illumination, this 
number, as Aubert proves, can easily be carried up to 100,000 
tints ; and by illuminating the spectrum by colored light, red, green, 
violet, etc., the number of tints passes up into incalculable millions, 
into a mathematical infinity of possible colors. But let us keep to 
the 1 100 colors as the norm for practical work. 

Given, then, these 1 100 colors that can be seen and known by 
the human eye, it is the task of human speech to furnish expression 
more or less adequate for this number of color-sensations. The 
power of the various languages to do this, a power that varies from 
the rude inadequacy of barbarous dialects up to the subtle discrimi- 
nations and overflowing wealth of the most perfect languages, is no 
bad measure of the perfection of language for the expression of 
human needs. In English, for example, I find that Roget has 
given the number at 102. The power of the modem French, 



8 AMERICAN JOURNAL OF PHILOLOGY, 

however, in the apt and dear expression of color, is recognized by 
philologians as one of the most mavellous £icts of that marvellously 
rich and picturesque language, and is perhaps the highest perfection 
of color-expression ever attained by human speech. The number 
of color-terms in common use in modem French is said to be not 
much short of 500. How, now, does the Latin of Vergil stand in 
this comparison ? Over what number of color-terms does the genius 
of the poet bear sway, in order to find linguistic expression for the 
beauties of color that he discovered in nature ? How many, as it 
were, and what are the pigments that lie upon the palette of this 
exquisite painter of the world for the rendering of all his color 
impressions ? A fnend of Alma Tadema (Collier, Primer of Art, 
p. 55) has lately told us that this exquisite master of color reproduces 
his color- impressions of the antique world by a palette of 12 colors. 
As compared with this dozen real pigments of Alma Tadema, the 
palette of VergiFs vocabulary contains 27 terms of high color — 
rather more than twice as many. But, as I shall prove hereafter, 
the defects of his language compel Vergil to use for the expression 
of definite color many terms for white and black and gray, terms 
which in their scientific sense are not color-terms at all, but which 
in their poetic use are often the expression of real and powerful 
color, as seen under excess and deficiency of illumination. Of 
such terms for black, white and gray, there are 15 in Vergil's 
vocabulary ; and if these 15 be added to the 27 terms of pure color, 
we have 42 pigments in the color-system of the Vergilian poetry. 
Compare now this number of 42 color-terms with the number, fixed 
above, of 1 100 colors to be expressed by them. From the bare 
statement of this rough numerical relation, 42 to iioo, there follows 
one consequence of prime importance for the understanding of 
Vergil's color-system. Each color-term of the 42 must cover, on 
the average, the expression of 26 closely allied tints. For each 
color- term, therefore, we must seek to find one precise color, as 
the norm, and, as it were, the center or axis of its chromatic power. 
This, if fixed by the color of some object in physical nature, 
unchanged and unchangeable, will serve us as the natural or phy- 
sical standard of that particular color-term. For sanguinetts, e, g. 
we shall have sanguis. But on both sides of that physical stan- 
dard we must expect to find a group of allied tints, expressed, 
indeed, by the same color-term, but grading off tint by tint, up 
and down the vertical spectrum, toward the color-terms that lie 
nearest In caeruUus, for example, there must lie not only the 



THE COLOR-SYSTEM OF VERGIL, 9 

meaning of pure blue, as found in caelum^ but also on both sides of 
blue a large number of distinct tints, closely allied to blue, but grading 
off, tint by tint, on the upper side toward green, and on the lower 
side toward violet Here is the natural infirmity of language ; no 
language, however rich, however precise, can ever be so developed 
as to cover the absolute infinitude of man's perceptions and sensations. 
Every human soul in its efforts at utterance is doomed to everlasting 
failure ; the finite expression cannot be so stretched as to cover the 
infinite realities of things. 

In understanding Vergil's poetry, therefore, we are forced to give 
to each color-term that he uses a somewhat widened range of tint- 
variation. Bearing this in mind, let us now arrange the color-terms 
of Vergil in their allied groups, according to proximity of tint in 
the spectrum of nature. 

Following the theory of Helmholtz, a theory that I accept as a 
demonstration, I shall give first the three primary groups of red, 
green and violet ; then the group that lies between red and green ; 
then the group that lies between green and violet ; lastly, the tints 
of so-called purples and blacks and grays and whites that, although 
occurring in nature, do not occur within the spectrum itself But, 
in giving this catalogue of Vergil's color-terms, I wish to do more 
than simply to mark the existence of each particular color in Vergil's 
great picture of the universe. I wish to give a quantitative estimate 
of Vergil's use of the different colors. If we can fix the number of 
times that Vergil makes use of each color, we shall be able to dis- 
cover his color-preferences, and to find out how far the color-sense 
of Vergil corresponds with the diffusion of color in nature ; that is, 
how far the poetic presentation agrees with the realities of things. 
The color- sense of every great artist is an important element of his 
genius, a strongly marked characteristic of his artistic manner and 
range of perception. In painting, for example, it is obvious to all 
that love the art, to all that have striven to understand the work of 
any great man, that to the eye and to the taste of every original 
painter certain groups and ranges of color are distinctly preferable 
to certain others. These preferences and partialities are so strongly 
marked in each man's works as to be a striking characteristic of his 
genius. Amid the vast infinities of nature's coloring, no single eye 
is capable of finding equal joy in all the colors that make up the 
color-effects of the universe ; no genius for coloring is so vast and 
so catholic as to be able to take in and reproduce in an equable 
manner all the coloring of nature. Hence, to use familiar examples, 



lO AMERICAN JOURNAL OF PHtLOLOGY. 

the splendid genius of Turner and the calmer, truer genius of 
Ruisdael are almost complementary to each other in the preference 
shown for the warm colors and for the cool colors of the spectrum, 
for the sunset glories and for the forest-tinted greens of the land- 
scape. The sombre, brownish magnificence of Piloty's Death of 
Wallenstetn differs as much from the glowing splendors of Paul 
Veronese's banqueting-scenes as if the two painters had looked out 
on different worlds of color. 

As it is in painting, so it is in poetry. To each poet's eyes the 
world puts on a different aspect. Each poet, if we watch him at 
his work of representing nature as he sees it, will show a preference 
for certain ranges of color and a distaste for certain others. Thus, 
in studying the color-system of Vergil by its quantitative use of 
color-terms, we shall be able, in a general way, to see with his eyes 
the aspect of the physical universe, to learn his likes and his dislikes 
in color- effects, and to understand how far his genius conforms to 
the realities of nature in the diffusion and prevalence of color as an 
element of beauty. 

I. Red group of color-terms. — Ruber, used 36 times ; rudlus, 5 ; 
sanguineus, 14 ; cruentus, 2 ; sandix, 1 ; minium, 2 ; ferrugo, 5 ; 
roseus, 12, Total of red group 77. 

II. Greengroupofcolor-terms. — Viridis, used 63 times; vitreua, 
I ; hyalus, i. Total of green group 65. 

III. Violet group of color-terms, — None used with exclusive 
and absolute precision of tint 

IV. Group of color-terms between red and green, — Igneus used 
e; spadix, i; fulvus, 20; flavus, 19; croceus, 16; luteus, 5; 
iim,33; gilvus,i; cereus, 1. Total of red-green group 97. 

'. Group of color-terms between green and violet. — PaJlidus 
d 24 times; lividus, 2; caeruleus, 31. Total group of green - 

et color-terms 57. 

n. Outside group of color-terms not belonging in spectrum, 

formed by composition, Purpureus used 33 times, puniceus?, 
■um ir, murex 3; red and blue 54, Albus 38, candidus 37, 
:us 18, decolor i, argenteus 2, lacteus 5, marmoreus 6; formed 
blending, more or less perfect, of complementary colors 107. 
ms 13, glaucus 9, ater 72, fuscus 4, fumeus i, mger 41, pullus i, 
^s 2 ; formed by the darkening, more or less complete, of each 
)ri43. 

'bus, in Vergil, if we include the so-called blacks, whites and- 
ys, there are 600 uses of color-efTect ; but, if we leave out the 



THE COLOR'SYSTEM OF VERGIL, II 

blacks, whites and grays, there are 350 uses of bright color in the 
poems. 

Later on we shall compare this diffusion of the different colors 
in Vergil's poetry with the actual diffusion of the same colors in 
the physical universe, and work out some interesting results. But, 
before this comparison can be made, we must find out some philo- 
logical system, some method of hermeneutics, by which we can fix 
more precisely the color-significance of each color- term itself. 

The color of anything in nature, whose color is visible to the 
human eye in any given position, depends on at least three elements 
that are always present, the three so-called constants of color, cf. 
Rood, ch. Ill, and pp. 209-10. Think, for example, of a mass of 
green leaves hanging as foliage upon a tree. The color of that 
foliage will depend 

ist, on the wave-length of the waves of light reflected from those 
leaves into your eyes. That is hiUy absolute color, as element in 
color-perception. 

2d, on the amount pf green light that is reflected firom the leaves 
into your eyes. That is the brightness or luminosity of the color 
as an element in the color-impression. 

3d, on the amount of -white light that is mingled with the green 
light in the final color-impression. This is the purity of the color. 
If all white light could be removed, a thing impossible to do, the 
color would be absolutely pure ; as the amount of white light is 
increased, the purity becomes less and less. At last, by excess of 
white light, all color may be made to vanish into whiteness. 

From the clear understanding of these three elements of color, 
you will see that there can be no objective fixedness in our color- 
impression of any visible object. The hue may abide the same, 
but accidental variations in purity or in luminosity may make the 
color-impression vary in ever-shifting tints from what approaches 
blackness up to what approaches whiteness. As the distance varies, 
as the laws of aerial perspective work their will upon objects more 
near or more remote, dark objects become lighter when afar off. 
light objects become more vivid when near by. Now the color- 
terms of the poet, like the pigments of the painter, are not meant 
to define the absolute color of things ; the mere wave-lengths of the 
waves of light that each thing reflects are not to be defined in the 
language of art The poet like the painter records not what the thing 
is in itself, but what the thing seems to be in his eyes under all the 
circumstances that surround it From this it comes to pass that 



12 AMERICAN JOURNAL OP PHILOLOGY. 

objects which seem to us altogether different in color are often 
presented by the poet under the same color-term. Thus, under 
process of darkening, from defect of luminosity, the green of the 
fresh -cut foliage and the red of flowing blood are both called by 
Vergil ater, aier sanguis, Gg. Ill 507, zndfrondibus atris, Aen. VI 
215. And again, under variations of purity, by contrast with 
different surrounding colors, the color of the same object may be 
expressed by two different color-terms. For example, in the same 
book of the Aeneid, V 309 and 494, the foliage of the olive is at 
one time viridis, at another /tt/z/a. Unless you can make plain to 
your minds the variations in our color-perceptions that are produced 
by variations in luminosity and purity, above all the variations 
produced by contrast, you can never hope to understand the use of 
coloring in painting nor of color-terms in poetry. Who, for example, 
that has seen the blue-green waves of the ocean blanch into white- 
ness under the red light of the setting or the rising sun can fail to 
understand why Vergil calls the sea martnoreum aequor f The 
poet*s eyes saw in nature and painted in language the exact effect 
which is produced in the laboratory by throwing red light on a 
blue-green surfece, Rood, p. 153. 

From the point that we have reached we can come back to the 
practical problem that color-terms present to the mind of the philo- 
logian. If color be a thing so variable and so unfixed, by what 
means known to philology arc we to fix the meaning, and to 
understand the use» in any ancient writer, say in Vergil, of the 
color-terms that he employs ? Upon this problem as it arises in 
Greek philology-, the minds of many Greek philologians have been 
of late keenly dinvted- But in their investigations, full of learning 
and fill! of charm as they are» it seems to me that there have been 
such gra\^ mist.ikes ot^ method as K> rob their conclusions of per- 
manent and solid value, 1 am not without hoj>e, therefore, that by 
examining the color-sx-stx^ni <>t'a SiUj^lc >;Teat ^xx^t, esjx^ciaily of one 
90 fond of cvCv^ and so ver>:^l in c\>K>r as VoreiK we mav be able 
to learn some t»ct:s. pe:hA:\:i even u> ^"^taKuvh s^^nx* Uw^ of inter- 
pretation, that ni,%y be of u<>e ;n dcti:*»inji the Oi>W>tvmts of other 
ancient wnTer> and of oihcr aixSent lanj^xiAjix^ alx^\^ all, where w^e 
need it mrts:. in :be Gr*r>e^. My o*n vNM>v\i;s;vXt ;s ih,u> in order 
tc undensiind :be niear.:;-;^ vV arv ^^wxr^n ^^\v^^-tx^:r,'i.\^t^ :r.u<: Wv^rk 
rrr to i.oe £njil rxssul: bv n\t^ ^K\>fSN,\e $^a<x:^ of n:c:h.xiical 

isL ilie en^TDvuC^' vV :bi^ cv-CvV ^^i^-n n";;^^; be >:,vI-od, If we 



THE COLOR'SYSTEM OF VERGIL, 13 

can trace the word backward to its root and discover and compare 
its cognates, we shall find the fundamental conception, the concrete 
significance, that underlies its use, e, g, the derivation oi gilvus 
from the I. G. root gar (shine, bright) and its connection with bills 
in Latin, with ycXciy (=:Xa/uw€ty, avQtiv Hesych.) and figuratively 
y€\av in Greek, with yellow in English, are fiill of help. 

2d. The physical standard of the color must be fixed. We 
must try to find in nature some fixed and permanent standard of 
the color-term in question, some visible object that may fix the 
color for our eyes, and make us able to visualize, as it were, the 
color-impression of the ancient poet, e. g, the splendid pure red of 
sandix is fixed for ever in the mind of one that has ever looked 
upon the mineral (realgar^ red sulphuret of arsenic) in which it 
occurs. 

3d. The extension of the given color up and down the gamut 
of the spectrum must be determined, e, g. caeruUus running up 
almost to green and running down almost to violet 

4th. The variation of the color under different degrees of purity 
and luminosity must be determined, e. g, purpureus may lighten 
up into the softest shades of pink, or darken until it loses itself in 
a kind of violet blackness. 

5lh. The variation of each color by contrast, in combination 
with other colors, must be studied ; and it will explain the most 
violent use of the color-terms, e, g. a wreath of olive leaves around 
a head of black or brown hair will seem yellowish and be called 
fulims ; but around a head of yellow hair will seem intensely green 
and be called viridis. 

In separate essays, I hope to publish, so far as Vergil is concerned, 
my determination of his various color-terms according to this five- 
fold method of investigation. For the present I cannot do more 
than give in the fewest words the etymology and physical standard 
of each term. 

Ruber, I. G. rudh, name of blood and of planet Mars, cognate 
with ipvSpSsf redy ruddy ^ etc. Physical standard, the color of the 
crab when cooked, rubenies urefoco cancros, Gg. 4, 47. 

Ruiilus. I. G. rudh^ akin to ruber^ with red verging toward 
yellow. Physical standard, the splendid color of the Italian bee, 
rutilis clarus squamis, Gg. 4, 93. 

Sanguineus, from sanguis. I. G. sag (drop, flow), cf. Latin 
sucus. Physical standard, color of fresh blood dropping from 
wound, blood itself, mulberry. 



14 AMERICAN JOURNAL OF PHILOLOGY. 

Cruentus^ from cruor. I. G. kru (hard), cf. crudelis, Kpvos, xpO- 
oToKkoi, etc., Eng. raw. Physical standard, cruoTy blood hardening 
and darkening after it has left the body. Myrtle-berries show very 
dark red, cruenia myrta, Gg. i, 306. 

SandiXy Gr. trav^v^. Etymology unknown to me. Phjrsical 
standard, realgar y red sulphuret of arsenic, used as dye-stuff. Eel. 

4. 45- 

Minium, Etymology unknown to me. It is said to be an Ibe- 
rian word. Physical standard, cinnabar, vermilion, used as rouge. 
Ed. 10, 27. 

FerrugOy iron rust, from ferrum. Physical standard fixed in 
nature. The color of the larkspur, dark red, rather dull, ferni- 
gineus hyacinthus, Gg. 4, 183. 

RoseuSy from rosay probably of Semitic origin. Physical standard, 
rose-red with blue tinge, light or dark, red lips of beautiful women. 
Aen. 9, 5, etc. 

I 'iridis. I. G. gkvary ghar (sprout, grow). Physical standard, 
young leaves of trees, young grasses. 

ViiretiSy from vitrum (glass). I. G. vid (to see). Physical 
standard, greenish antique glass, of transparent green, vitrea Fuci- 
nus nuda, Aen. 7, 759. 

HyahiSy Gr. CoXw (raindrop, glass). I. G. su (drop). Physical 
standard, glass of deep green color, hyali saturo fucaia colore. 

Gg. 4. 355- 

Igneus^ from i^is. I. G. ag (move, flickerV Physical standard, 
fiery red of sun presaging wind-storm, i^Sol' igneus Euros. Gg. 

i> 453- 
S/aJSx, Gr. an^, date-p.\lm, name from Semitic Physical 

standard, date, rich red brown, used of bay horse, honesii spadices. 

Gg. 3, 82, 

FlavMs, Tooi (:kar ^^grow, sprout'^, c£ z^iriJis. Ph\-sical standard, 
ripening grain, Tilx^r at RonK\ accv^rvling to Fronto's definition 
n\ade up of gnxni and nxi nitjiovi by white, /.j;»//j//ii« flavescet 
camf^ms an\^tj. EcK 4, iS. 

/Wcur.^, rxxn ^kar i^virvnv, spanu\ cf nn\i:s and ,-:,k^^ Physi- 
cal standatxi. skin of lion, j^umc5 ol' cav;U\ acwrxiing to Fronto's 
definition, the s,imc iis ^^rj^of, with whitx^ Ktt ^hu, fs,.':-m^ /^k-is aUs. 
Aen. 12, i47, 

t > avjikc, trvMn , » a i*jc v^s,aVi\>n^, Scnwtiv^ vixmxK rh\^\\l standard 
SAt^rxMi it^)t', \»>c\i AS dyt>^tutl, xvllv^w ln\<x\l uith red, crc^umque 
r%hcHTtm^ Gji» 4» iJnj* 



THE COLOR-SYSTEM OF VERGIL. 1 5 

Luiens, from Itiiutn (weld). Root lu for hiu for ghlu for ghar 
(sprout). Physical standard, weld itself, used as dye-stuff, yellow 
with less red than croceus^ e, g. Aurora luiea in resets bigis. Aen. 
7, 26. Yellow seems more yellow seen against red. 

Aurum (gold). Root aur = aus = vas (bum, blaze), cf. Aurora, 
etc. Physical standard, gold itself, yellow with lustre, becoming 
reddish by reflection, sol aureus, Gg. i, 232. 

Cereus, from cera, I. G. kar (to separate), ici/p/r. Physical 
standard, wax, yellow plum. 

Pallidus, Root pal (greenish-blue, or bluish-green darkened). 
Germ, fahl. Physical standard, brunette complexion after death 
or in violent fright, Dido pallida morte futura. Aen. 4, 644. 

Lividus, Root liv (gray-blue). Physical standard, lead, glan- 
des liventis plumbi, Aen. 7, 687. 

Caeruleus, from caelum. I. G. kav (hollow, vault). Physical 
standard, blue sky, normal tint of Mediterranean, caeruleo sunt 
Tiatnina ponio. Aen. 12, 182. 

Purpureusy purpura, Gr. irop<f>vpa, I. G. dkar (wave, agitate). 
Physical standard, murex, or sea-snail, magnificent color formed by 
union of red and blue, ripe grzpe, purpureae vites, Gg. 2, 95. 

Puniceus, from Punicus, from Poenus, origin of the color from 
(Tarthage or Phoenicia. Physical standard, murex itself It seems 
to have more red thaca purpureus, puniceis rosetis. Eel. 5, 17. 

Murex, I. G. mar, smar (rub, stain, smear), name given to the 
dye made from sea-snail. Physical standard, itself, as dye -stuff, 
Tyrio ardebat murice laena, Aen. 4, 262. 

O strum. Gr. Sarptiov, oyster, shell-fish. l,G, as (cast off, throw 
away). Physical standard, same as above, inclining strongly to red, 
ebur violare sanguineo osiro, Aen. 12, 67. 

Albus, I. G. albh (white, dull white, approached either through 
yellow or through blue), cf. SK^i, dX<t)6s, etc. Physical standard, 
white, or albumen, of egg, blue-white. Sulphur, yellow-white, 
sulfurea Nar albus aqua, Aen. 7, 517. 

Candidus, I. G. kand, skand (shine, glow with heat), cf incen- 
dere, Kovdapos (live coal). Physical standard, lily glowing white 
against dark green leaves, white horse in sunshine, feeding in field 
of grass, etc., Candida viridi in litore conspicitor sus. Aen. 8, 83 
{/etu albo). 

Nvueus, firom nix, I. G. ^«/^(wash, cleanse). Physical standard, 
nix, generally a cold, bluish white^but changing with atmospheric 
conditions. 



l6 AMERICAN JOURNAL OF PHILOLOGY. 

« 

Argenieus, from argentum, I. G. arg (shine, glitter), c£ a^r^U. 
Physical standard, silver, white with metallic lustre, goose, dolphin. 

LacUus, from iac. I. G. gliiki (milk). Physical standard, milk 
itself, soft, creamy white. 

Marmorais^ from marmor, I. G. tnar (shine, shimmer), c£ 
iMopfuufm (shine), fuiX<k (white), etc Physical standard, white marble, 
white with lustrous surface. 

Dfcoior^ from color. Physical standard, the dull, Weached 
whiteness of old age, in complexion and hair, decolor artus, Aen. 
8,326. 

Camus for casmus, I. G. kas (shine), c£ cascus (old). Ph)rsical 
standard, sil\*er-gray hair and bud, shinii^ but yellowish. 

Ghucus. Gr. yXci^x^k, root gal (c£ gilvus), bright Physical 
standard « sea agitated with foam -crested waves, bluish gray. 

Atc9\ Root acd L G. idk (bum^, blackness as the result of 
burning, rh)*sic;U stand,ird, ashes, cinis aUr^ what is left after all 
cvCv^r is bunit out := */cv*wr, notion not positiv'e but negative. 

•\Xv<''« 1* G. n:k. HJik vslay, kiir, cf. tux^ nox, nocere^ m'ciy, etc, 
Kackiurss as colv^r ot' nii^ht and sien of death and e\-il, blackness 
appi\vAoh<xl thivHi^h vk^Ict ^vkxa\ through green ^^^ex), through 
l\ue >^atciv thi\>u>;h rxxi ^ncgTv^", through yellow .^wet sand*. 

/V<,;^.x :■ •>' ^v,jk,s rvv4 •*•-. 1, G, b\^r ^shake, agitate) — cfl 
/A*;^^-MaokiH^:^dpj>r\\Avh^\i thrvv.ioh red and brown. Physical 
:?4a;>vU:\l. iH>;tv'5 v\^r,\jvc\x^;i, Mv^r. ^^v lu^iht-blrd s wings. 

c \ «f , ^.v\ u\MU * X n Avv 1 , G. -c' t* ^K^ ;1. sos c-ke"* Physical stan- 

,*V Vv\ L Vk /^a.\ oC i\:\:x.N etc Kuish c^ay approaching 
K^^c^;^>3i. rh\^v\i* star,vU:vl. s;\>C:? vX". 5;heep. Gg. ^ 5SC;. 

,"\,;Asx :;\vn .Vix ^-^ . KG. .\- ^^O^^^^rnr, c£ r-CJU^ rx#w», etc 
TVann al >t.i\xU'Aiv ^- ; ^ il><%V v; ;t\ . vtw^ b^ck. swtat mingled with 

\nvi ;^,VN >>\ "^ :Sv^ v^\ ;^vvv\;x\0 \vwa\xitbc rhx-^rjJ standard 
yV vNtcN v\\\'^ tv\-^^ ;\x: \ c'V'.* ^>»<^ ;Sr *"o:h^>d vt: ocr ir:\iestigation 
>fc\N. \\ \\^v\ ;** tv^ n\ Kv \\iob \\\\v ^"^ I. '- :$ v"C vjLritDoa under 
^\vw As- s\\< \^x> vv js. ;\ xv \'. ^' ,N>.:\ jt'd vX^ vvrn-jsst. In my 
>^%'^ v;,x v> w ;^.x v^^\\ \ S^xv >fc\\xv\: vX ; xr :'"^ wiv. ill the 
\ s vv^ ns\x A s* <,\^ V ^^ \^;vv vM xWx '^ vvw^ ;,^v xWxX ^"^i — ^^ Tr.c cetails 
x\ ^'^ .X N..v\ N\\\\ N >\v, A -'\v'^^ v-^-^",^ vc :>^ <CJCcr;t that 
VNvvK x^ ,> ,x^ ^ >N nN^^'N ^ ,i ^'v i^ N .-WN^ >v jV\V.\\;v ^.ih the 



• THE COLOR-SYSTEM OF VERGIL. 17 

monographs than to a general discussion of VergiPs color-system. 
As general result of my separate studies, I may boldly say that 
in no single case does Vergil ever extend the use of any color- 
term beyond what science recognizes as the possible limit of its 
extension. His application of his color-terms is often imagina- 
tive and sometimes startling ; but there is no Vergilian application 
of any color-term that is not correct and even accurate according 
to the principles of chromatics. Leaving this subject, then, as unfit 
for discussion except in special treatises, I pass on to the last great 
question that is involved in the color-system of the poems. Is any 
comparison possible between color as imagined in the poems of 
Vergil and color as diffused in the visible objects of the universe ? 
How far does the color-system of Vergil agree with the color-system 
of nature? In the distribution and quantities of each color respec- 
tively, how far does the world as represented in Vergil's poems 
agree with the world that actually exists ? 

Color, as we know, arises in nature from the breaking up of white 
sunlight, by absorption and reflection, into colored lights of various 
hues. The spectrum gives us power to break up the sunlight at 
pleasure ; it even gives us power to measure with absolute accuracy 
the fractional amoupt of each color that is present in every unit of 
white light. In every thousand parts of sunlight, each color, red, 
green, violet, etc., is present in a fixed proportion ; consequently, 
amid all the infinite play of colors that makes the ever- varying 
charm of the visible universe, however various the combinations, 
there is a fixed and unvarying quantity of each color always present. 
In every 1000 parts of white sunlight that beats upon the earth, 
there is of red light 330 parts, red-yellow 155, yellow no, green 
87, green-blue 67, blue 74, violet 177. 

As s^id above, if we leave out the terms for white, gray and black, 
there are in Vergil's poems 350 uses of high color. That is, in every 
thousand parts of light, as the world showed itself to Vergil, of red 
l^ht there were 220 parts as opposed to 330 in the spectrum ; red- 
yellow, 117 as opposed to 155 ; yellow, 160 — no; green, 186 — 87 ; 
green-blue, 74—67; blue, 88 — 74; violet, 0—177. That is, the 
world of Vergil's imagination, as compared with the real world, is 
defective in red, red-green and violet, and is excessive in yellow, 
green, blue and blue-green ; or, to put the same facts into a more 
convenient form, Vergil's sense of color is fullest at the middle of 
the spectrum in yellow, green, green-blue and blue, and is defective 
at both ends, in red and red-yellow and especially in violet. 



l8 AMERICAN JOURXAL OF PHILOLOGY. 

The excess is to be explained by the poet's preference for warm 
colors over cold and for the more Imninoas o\'er the less luminous 
colors. The warm colors are, in painter's lai^^uage, the reds, red- 
ydlows and ydlows that make the upper half of tBe spectrum. Of 
these there are in Ver^ 497 parts in the thousand ; but Vergil's 
purple also is a warm color because of the red that is in it, and of 
purple he has 154 parts. Thus, in every thousand parts di light, 
Vergil saw 651 parts of warm color as opposed to 349 parts of 
cold color. But in the ^>ectrum there are 595 parts of warm color 
as opposed to 405 parts of cold color. Thus the genius of Vergil, 
in his picture of the universe, errs from the exact trudi of nature 
by an excess of warm coloring. The poet's imagination, to that 
extent, idealizes the &cts of nature by a warmer and brighter 
presentatioa of the visible world. Hb world stands to the real 
world as a pwtrait by Titian to the ^ure of the human originaL 
Again, in respect of luminosity, the co1(H3 of the ^)ectrum grade 
downward from yellow, the most luminous, through green and red 
and blue to vic^et, the least luminous of alL Here, too, as idealizing 
artist, Vergil prefers the luminous colors. In yellow, for example, 
he has 160 parts against 1 10 of the spectrum ; in green he has 186 
parts against 87 in the spectrum ; in reds including purples he has 
374 parts against 330 in the spectrum. But in blues he barely 
reaches the prc^x>rticm of the spectrum, and in violet he is totally 
deficient (o against 177). 

These fects are all significant of Vergil's genius. His perceptions 
of color are dearest and strongest at the middle of the spectrum ; 
even in his sensuous imagination he is temperate and reserved, 
avoiding the extremes of sensation, and dwelling by preference 
upon the mean terms, the wudia via of visual perception. But, 
in lighting up his imaginary world, he is, in his perfect art, not 
realistic but boldly an idealist. By unconscious selection he floods 
his canvas with the warm glow of reds and purples and jrellows, 
and brings down the use of cold colors for below the measure of 
their actual diffusion. And again, with the eye of the poet, antici- 
pating the analysis of science, he discerned the colors that had the 
highest degree of luminosity, and lifted his glowing picture of the 
world far above the actual light of nature, by giving preference to 
the colors that are the mi^t luminous* the most effective and fer- 
reaching as well upon the eyes as u[)on the imaginations of mankind. 
Thus, even in this point oi' color, the works of Vergil's genius stand 
out as creations of a nobly ideal art, temperate in all things, won- 



THE COLOR-SYSTEM OF VERGIL, 19 

derfully true to nature, but rising boldly above nature both in the 
luminosity and in the warmth of their coloring. 

But the color-system of Vergil differs most from the color-system 
of nature by the total absence from it of violet. In the solar 
spectrum, as we saw, in 1000 parts of white light there are 177 
parts of violet ; but in Vergil*s spectrum there is no violet at all, 
and the spectrum ends after caeruleus in niger. This absence of 
violet from the Vergilian system is strong confirmation of the theory 
that the conscious perception and the naming of the colors have 
followed a law of natural sequence. Those colors that produce the 
strongest effect upon the eye were the first to be noticed and the 
first to be named ; and the strength of the color-effect and, there- 
fore, the priority of the color-term are dependent upon the wave- 
length of the color itself. Thus red was the first color to be noticed 
and to be named ; then orange, then yellow, then green, then blue, 
then at last violet. For example, when the Homeric poems were 
composed, red, orange and yellow were the only colors much 
noticed or distinctly named ; green was the frontier-color ; blue and 
violet were unnoticed and unnamed. But mankind grew on both 
in its observation of color and in its power of giving names to the 
colors observed. 

A thousand years passed by. Vergil came forward, as the poet 
of the Roman empire and of the Italian people, to give us his poetic 
representation of the world. See now the advance in color from 
Homer's time to Vergil's. Green is added on to red and yellow 
as part of the color-system of nature, and blue is added on to green. 
But at blue the progress is arrested. Of violet, the last color of 
the spectrum, the color of the shortest wave-length, there is in 
Vergil no conscious vision and no distinct name. But the genius 
of Vergil stands on the very verge of the final discovery. The 
lovely color that lies below blue, closing the glories of the spectrum, 
as we can see by many allusions in Vergil's poems, was dimly 
present before his eyes and in his imagination. His genius was 
already laboring at the task of giving to it expression in language. 
Sometimes it is by niger that he seeks to express the mysterious 
color through which blue light passes into blackness. Sometimes 
by a confusion of sight which still prevails among more than one- 
half of civilized men, he confounds violet with purple, and calls it 
purpuretLS. In one line of exquisite beauty he combines the two 
methods, in order to find expression for the violet color that floated 
as a distinct impression before his mind : 



20 AMERICAN JOURNAL OF PHILOLOGY. 

violoi tublueet purpura nigral (Gg. IV 275), 
" The pnrple of the black riolet tones down its color.** 

Here, then, in this attempt of Vergil to complete his color-system 
by the discovery and naming of violet, we must take our leave of 
the great poet that stands half-way, as it were, between Homer and 
Gothe. 

He had advanced beyond the point reached by Homer in sub- 
stituting varieties of color for degrees of light, as the prime beauty 
of the physical universe. He had enlarged the spectrum of poetry 
by carrying his perception of color down from green through blue 
to the very verge of violet. Above all, he had given precision to 
the color-terms of his own language, and painted for us a world in 
which simple colors and combination of colors are worked into an 
almost perfect glory of color-effects. And even in respect of violet, 
although he did not succeed in expressing what he saw, he felt the 
color-impression that he could not name ; the poet's eyes had been 
blessed with the dimly felt sensation of that color which is the last 

exquisite perfection of the spectrum. 

Thomas R. Price. 



II.— HISTORICAL AND CRITICAL REMARKS INTRO- 
DUCTORY TO A COMPARATIVE STUDY OF 

GREEK ACCENT. 

I. 

Accent is a universal phenomenon in language, and one which 
is in dose union with what is treated by grammar under the head 
of sound or phonology. 

The sounds of a word without accent are merely separate stones 
which accent cements into a linguistic entity, either a word or a 
sentence. W. v. Humboldt says : * The unity of the word is pro- 
duced by the accent. This, by itself, is of a more spiritual nature 
than the sounds, and it is therefore called the soul of speech, not 
only because it is really the element which carries intelligibility 
into speech, but because it is, more than other Victors in speech, the 
immediate expression of feeling ' (cited by Gdttling, Accent der 
griechischen Sprache, p. 8). 

The word accent in modem terminology is unfortunately com- 
pelled to do duty for more than one linguistic fact First, in the 
case of the word, it signifies the relaiive stress and pitch character- 
istics of its various syllables, with no restriction to that syllable 
which has the strongest stress or the highest pitch. This is the 
most scientific function of the word. A doser study of the life of 
the word cannot be satisfied with a theoretical analysis of its sounds 
and syllables and a superfidal recognition as to which of the syl- 
lables has the highest pitch or strongest stress, but it must be known 
also in what way or to whai extent thb syllable is elevated above 
those surrounding it Furthermore, the rdations of the remaining 
syllables to one another will always show that the same character- 
istics which distingubh the tone-syllable kox i^xh^ attach themselves 
in a lesser degree to some one or more of the remaining syllables ; 
in short, I would define word-accent in this wider sense as the history 
of stress and pitch in the immediate practical subdivisions of the 
word, its syllables. This definition of accent has necessarily to be 
kept apart from that other more &miliar one by which, in the cur- 
rent parlance of grammar, the pitch or stress of the most accented 
% syllable is designated. This, of course, is not all. For just as the 



22 AUERICAy JOURNAL OF PHILOLOGY. 

word has its history of pitch and stress, so has the sentence. The 
members of the sentence stand in a relation to the sentence as a 
whole which b not unlike that m which the syllables stand to the 
word. Here, <rf course, the word ' accent ' has again to do double 
duty: first, it indicates the relative diaracteristics of the words 
which make up the sentence, and, secondly, the word is also em- 
|Jo3red to mark that ^v'cxed monber of the sentence which holds 
the most prominent position, L e. the one which corresponds to the 

* tone-syllable ' in the word- 

In the sentence ' he did it, not she,' we may speak erf accent in 
its most pregnant sense and refer merely to the two summits 
' he ' and * she,' or on the other hand we may call before our minds 
a pictiire of the exact relation of each of the words in pitch and 
stress, not gi^*ing our attention merdy to the summits, but watchii^ 
the undulation of the tone-line in which the sentence mo\~es aU 
along, firom the beginning to the gkL This fe the study of accent 
in its scientific sense. 

That the accent (rf a sentence is as much under the influence of 
an organic law of some kind as the accent of the word is seen as 
sooc as <»ie attempts to disturb the natural cadence of a sentence 
such as the one dted above. By transferring the summit pitdi 
and ictus to the second word of the sentence we destroy the organic 
lijfe of the sentence ftiHy as mu<^ as though we change the summit 
pitch and stress in a single word. * He d::d it, not she ' is as much 
not an English sentence as * JlA^opment ' is not an Elnglish word. 
Frequency the change of reiaiion in pitch and stress d.>es not go so 
fur as to destrov the sentence, it simrCv makes anxber sentence 
Oct c< h, as for instance when the sumrr,:t tone is shifted successively 
frcca one word to anoth^a* in the grvxip erf wc«"ds • gi\ie me that 
book.* We obtain four difierent seniences c»rre?ronding to the 
iccr dinerest |>as:t:ottS of the summit tone. 

Wirh this last case rnay S? con^panxi :he w,\y in which, c, g. in 
Greek, the change of accent chan*:^s enrlrx:Iv the chanicter of certain 
wrrds ocher^K^se the same, and in Uct enters as a cv^nsSderably 
£r:rrul iactoc inw wocd-Jocrr-JiSon. Fv>r i:'£?:^»c>e, r.vv-« i^ ^^- agent- 
arcn cc pirticipial Nxnaarioc meanin;;: " running,' * a rjinner'; rjc^^r 
san act::n«^c<:2s or absract, "a nn.n.:':^:.' * a cvxirse ; cs.%v« n^eans 

* bearir-c : ovcwc * a bearinc/ * * tr.b<:^f ' ; Vvxh cocv-'ets are JL> 
6rcs iienncjL in e\^enr rejpect b;:t :he> acve.n ; :he a.x^rn: 
tbe same phcoeoc grocp? inrc* tw wvco> a^ vi^ :v:Iy vi.rVireniiixed 
in niDctfcc as two pr.rrvirT rkXin-icmvArrcc:^ n-vX^: the stn^ rcv:< can 



STUDY OF GREEK ACCENT, 23 

be. And, lest it be suspected that it was merely the superfine 
linguistic genius of the Greeks which brought in so delicate a factor 
as a power in word-formation, it may be stated at once that this 
difference is prehistoric, and Indo-European ; the couplet ^o^^ and 
0<$poff makes a perfect proportion with Sanskrit bhards * bearing ' 
and bhdras * a bearing,' * a burden.* In the same manner cf. in 
Greek fufrpoKrSvot * killing his mother * as epithet of Orestes, and 
fjufrpoKTovoi * slain by a mother ' as epithet of the children of Medea, 
the accent alone is the factor which has produced two distinct 
categories in noun-composition, also prehistoric and Indo-Euro- 
pean, and up to date not understood by the familiar guides for the 
study of Greek.* 

The chapter on sentence-accent is one of the most difficult and 
obscure in the study of grammar, and has been brought within the 
range of scientific discussion only very lately. Of course certain 
obtrusive phenomena which belong under this head had been 
noticed and discussed long ago ; as for instance the fact that certain 
words lose their independent accent in the sentence, namely, the 

' yUirpo-KT&vog means literally * mother-slaying *; it is the kind of compound 
which is called taipurum by the Hindu grammarians, that is, a simple compound 
in which the first member stands to the second in the relation of a case depen- 
dent upon it. MtfTpd^icTovoi is a secondary adjective compound, what is called 
in Hindu grammar a bahuznihi compound, one upon which the idea of possession 
and the like is secondarily engrafted ; the meaning is strictly speaking * pos- 
sessing/ t. e. being affected by a mother-slaying. The stem ktwo' in the two 
compounds is not the same ; in the first instance it is the nomen agentis ktov6^ 
• slaying,' in the second it is the nomen actionis Kvdvoq • a slaying.* The differ- 
ence of tone in the two compounds represents one of the most noteworthy 
archaisms in Greek nominal accentuation. Simple dependent compounds like 
luirpo-KTdvo^ were originally accented on the second member of the entire 
compound ; this law is so strongly alive in the Greek compounds of this class, 
whose second member is a noun of agency in -d-, that the law for recessive 
accentuation is observed only so far as it does not annul the older law according 
to which the tone must be on the second member, therefore fit/rpo-Krdvoc is 
against the recessive tendency. On the other hand, possessive compounds 
were originally accented on the first member, and in accordance with that, such 
compounds follow freely the laws of recessive accentuation, as fAtfTpd-KTovoi . 
The same law reveals itself in such accentual difference as is contained in Sk. 
yajnakdmds * desire of sacrifice,' zxidi yajUd-kdnuu * having desire of sacrifice'; 
the former is a simple dependent, the latter a secondary possessive compound. 
The Sanskrit regularly differentiates such compounds by varying accentuation, 
while in Greek the archaic differentiation of accent is preserved only sporadi- 
cally. See L. V. Schroeder in Kuhn*s Zeitschrift, loi fg., esp. pp. 106, no and 
116; Whitney, Sanskrit Grammar, §§1247, 1264 fg. and 1293 fg. 



74 AMERICAN JOURNAL OF PHILOLOGY. 

enclitics and prodidcs ;' certain words change their accent accord- 
ing to their position in the sentence : the so-called anastrophe* of 

1 That the proclitics do not lack an etymological accent (cf. below, p. 56), bat 
that they lose their accent from 83mtactical causes, f. /. from their relation to 
other words in the sentence, can often be shown easily, either by parsning their 
hiktory within the language itself, or by comparison with corresponding words 
in other languages. For instance, ov proclitic appears at the end of a sentence 
and in some other cases as o( ; <^ and i^ when they follow the governed word 
appear as t^ and l^ (Oeb^ 6(, Koxiw If^. That the proclisis of d, 1} is not due to 
some etymological peculiarity of these words is shown by the Sanskrit corres- 
pondents j</, sd; ol, al the special Greek new formations for older toI, rai (Sic 
masc. //= rot) are made analogically after d, 1), and borrow from them their 
proclisis. In the same manner no doubt all proclitics lose their accent owing 
to syntactical relations, 1. e, their lack of accent is due to Greek laws of sentence 
accentuation. About enclisis we will have much more to say below. 

[It is almost needless to add that the word * proclitic' is a modem invention 
brought into currency by G. Hermann (GAttling, p. 387). That does not militate 
against the existence of the thing ; only there seems to have been no recogni- 
tion of it in antiquity, and the omission of the accent in the cursive MSS was 
due to differentiation, to the desire of distinguishing not only between 6 and 6, 
il and ^, ol and oZ, al and a2, but also between ov and ov, tic and ei(, kv and ev, 
if and If, the spiriius asper not being heard at that time. See G. Uhlig, Zur 
Wiederherstellung des iltesten Compendiums der Grammatik, Festschrift zur 
BegrOssung der XXXVI Philologenversammlung, p. 80. — 6. L. G.] 

' The true explanation of anastrophe is as follows : Originally * prepositions ' 
were oftener or as often * postpositions,' f . e, the position of these small words 
in the sentence was a free one. This is clear, especially from the Vedic San- 
skrit, where some of the most common ones occur oftener after their nouns than 
before them (e,g, J * to' occurs in the Rig- Veda 1S6 times after its case and 
only 13 times before it). The mere fact that in later periods of language {e, g, 
Greek and classical Sanskrit) the tendency is to place them before their cases 
in itself proves nothing against this natural assumption. The case of a mono- 
syllabic preposition like ^f , which receives its natural accent after the word it 
governs, but is proclitic when it precedes it, points to the probability that the 
true accent of these Greek particles must be looked for in their postpositive 
position. Indeed, just as ef (orthotone), so do all bisyllabic prepositions appear 
with their true accent when they follow their cases, and just like if (proclitic) 
do all bisyllabic prepositions exhibit a snbsHhtte for proclisis when they accent 
their ultimate. The grammars which regard the oxytonesis as the original 
accentuation, of course explain it as due to a desire on the part of the language 
(o point to the word governed by means of the accent, but such an explanation 
fiee<l8 hardly to be refuted. 

The originality of the tone of bisyllabic prepositions in anastrophe is proved 
m addition by the fact that this accent is demanded by the corresponding 
bauakrit words whenever the etymology is clear. So Sanskrit dpa is not to be 
c(yU)jiared with Greek an6 but with Ato; Sk. dpi not with M but with Int ; in 
I he k4me manner the archaic character of the accentuation in iri^ irdpa and 



STUDY OF GREEK ACCENT. 25 

ox3rtone bisyllabic prepositions, which, as is now generally believed, 
preserves the original accentuation of these prepositions. The 
change of an acute to a grave on an ox3rtone before another word, 
though a phenomenon totally unexplained,' contains no doubt a 

vsro is warranted by Vt^ic pdri^ pdrd and 4pa ; the etymology of i^ha and K&ra 
is obscnre, bat they probably, like those preceding, have preserved their original 
form in paroxytonesis ; imr.p is not to be directly compared with Sk. t^dri^ 
which is reflected exactly in the oxytone imtip ; imtp may have preserved an 
originally different accentuation, or it may have followed secondarily the accent 
of the other prepositions which suffer anastrophe, aided perhaps by the accent 
of vn-epoc = Sk. lipara. On the other hand d/^/, which does not suffer anastrophe, 
is borne out in its oxytonesis by Sk. abhi ; avri to be sure is oxytone after the 
case which it governs, against the accent of Sanskrit dnii; but it may have left 
the company of the prepositions with anastrophe, because it differs from all of 
them in having its first syllable long (by position). In fact it appears to be a 
law, unnoticed even by Benfey, the author of this explanation of anastrophe, 
that only prepositions of two short syllables are affected by it {}mtip always 
oxytone, but vrrep—vfrkp with anastrophe). The etymol<^ of avd and t^id is 
obscure, but there is again no reason to doubt that their oxytonesis is based 
on good etymological grounds. The fact that these prepositions were originaUy 
paroxytone is proved also by the fact that they are so accented in adverbial 
function. Prepositions were originsdly adverbs, which have become attached 
to certain cases secondarily and in relatively later periods of language. Many 
common prepositions in Greek are still adverbs in Vedic Sanskrit: dpa^prd^ 
pdrd (diro, ?r/)^, vhpc^^ while pdri {^ipC) does function for both ; conversely the 
Vedic dti {frC) is both adverb and preposition, while in Greek it has remained 
adverb only. 

The assumption that such accentuation as dird, irapd, etc., contains a substi- 
tute for proclisis is easily vindicated. As a matter of fact only monosyllables 
are toneless in proclisis ; the treatment of bisyllabic words in the same position 
is in perfect accord with the treatment of enclitics when these contain a too 
great number of morae. Just as enclisis is restricted to three morae and two 
syllables (therefore y^dyoq ti^, but ^yoi rivic* cf. below, p. 42), so proclisis is 
restricted to one syllable and two morae (therefore i« irdwuv, but nepl irdvruv). 
The author of this ingenious explanation of anastrophe is Benfey (* Die eigent- 
liche Accentuation des Indicativ Praesentis von kg sein und ^ sprechen 
sowie einiger griechischen Praepositionen,' GOttinger Gelehrte Nacbrichten, 
Febr. 27, 1878, p. 165 fg., reprinted in Vedica und Linguistica, p. 90 fg.); he 
closes his article with the following remark: *' . . . es ist nicht besonders 
rflhmlich fQr die griechische Philologie, dass, nachdem sie mehr als zwei Jahr- 
tausende mit verh&ltnissm&ssig geringer Unterbrechung geQbt ist, noch in ihren 
jQngsten Lexicis und Grammatiken, die Formen d?rd, M, ?rapd, irep/, im6^ nard, 
furd anfgestellt werden, welche in der Sprache weder je vorkommen noch vor- 
kommen konnten.' 

'An elaborate discussion of this difficult question, which space forbids us to 
reproduce even in a condensed form, is contained in the essay of Leonhard 
Masing: Die Hauptformen des Serbisch-Chorwatischen Accents, nebst einleit- 



26 AMERICAN JOURNAL OF PHILOLOGY. 

difficulty whose solution will depend upon further investigation in 
sentence-accent The difference between interrogative and indefi- 
nite pronouns (interrogatives, orthotone ; indefinites, enclitics) is a 
case where sentence-acceni^ apparently, has given the language a 
method for differentiating an originally single category into two ; 
this also is not understood, but the archaic character of this phe- 
nomenon is warranted by similar methods in other languages.* 
And it has been urged lately that two different word-forms which 
perform the same function, may owe their difference in form to 
different intonation in sentence nexus.' 

enden Bemerkungen zur Accentlehre des Griechischen und des Sanskrit, St. 
Petersburg, 1876, p. 19 fg. 

* The relation of Wf, orthotone and interrogative, to r<c, enclitic and indefinite, 
is evidently the same as that of the German interrogative * wer ' to the indefinite 
• wer * in such sentences as the following : * W& ist gekommen ? ' and * Es ist 
wer gekommen.' We recognize at once that the enclisis of the indefinite is 
due to its peculiarly subordinate position in the sentence and not to any etymo- 
logical deficiency, it is therefore a feature of sentence-accent. Cf. the still less 
clear method of the Sanskrit for differentiating interrogatives from indefinites. 
By various particles (some enclitic and others orthotone : ca^ cand, cit, etc.) the 
interrogative without losing its own tone becomes indefinite, thus kds * who? * 
kdf ca * any one ' ; cf. Lat. ptis and quisque^ identical in form and meaning. 
Whitney, Sk. Gram. §507 ; DelbrUck, Die Grundlagen der griechischen Syntax, 
pp. 138, 145. 

* The most striking instance of this kind is an attempt to account for the 
different forms of the third person plural of the copula. It is true that the 
various forms of it, Doric hri^ Attic eiffi, Ionic kdai^ cannot be carried back to 
any one origin by any phonetic jugglery. Accordingly complicated processes 
of analogy have been resorted to generally in order to harmonize these forms. 
Gustav Meyer's view, e, g, is that a-avn is the Greek * ground-form.' From 
this form he derives kdcfi by assuming that the e was added secondarily from 
the strong forms of the root (^. g. iort) to *vT(t< for •avrt, 1. e. ♦ff-avri ; while 
Doric fvr<, Attic tlci^ are also to be derived from *dm by assuming that the 
initial vowel was assimilated to the e of the strong forms. Others employ other 
processes of analogy in order to harmonize these forms with one another. But 
Joh. Schmidt has taught for some years past that Done-Attic hrt — et<rt is to be 
referred to a form *o4vTt (= Germ, s-ind, Zend. k-^iUi)^ while kdai is to be 
referred to ♦<t-«it« in the manner exhibited above. The two forms ^'€%m and 
*a^vTi are explained as, originally, respectively the orthotone and the enclitic 
forms of the word in accor\Unce with the ideas of Wackcmagel as laid down 
in Kuhn*s Zeitschrift . XXIV. p. 457 fg.. cf. below, p. 56 fg. Of these two forms 

(T-^vn, the orthotone form, crowdetl out •<»-«»»Tt iu r)oric and Attic, while xnce 
versa *(r-arri, the enclitic form, gained the supremacy among the lonians. 
This explanation is laid down with a very slij;ht modification in the doctor- 
dissertation of his pupil, Felix Hartmann : * De Aoristo Secundo,* p. 68, while 



STUDY OF GREEK ACCENT. 2^ 

From the first opening out of the accented Vedic texts, a very 
important fact bearing upon sentence-accentuation had been noticed. 
In Sanskrit the finite verb in principal clauses is enclitic, while in 
subordinate clauses it is orthotone ; this fact lay fallow until Jacob 
Wackemagel, in the 23d volume of Kuhn's Zeitschrift, p. 457 fg., 
showed that the Greek verbal recessive accent is nothing more 
than this enclisis of the finite verb extended to all kinds of sentences, 
subordinate as well as principal, but at the same time modified by 
that peculiar law of Greek according to which enclisis cannot 
extend beyond three morae. Wackemagel's ingenious discovery 
we will discuss in full further on ; the point which is to be recognized 
here is the fact that the study of sentence-accentuation is destined 
to a prominent place in the grammars of the future, and that the 
present generation of scholars will, beyond a doubt, see this develop 
into a science ; the delicacy of the subject will call for the keenest 
penetration, but this will be rewarded by the importance of the 
results; results of comparative grammar alike valuable to the 
phonetist, the morphologist, and above all perhaps the student of 
S3rntax. 

The study of accent in these two forms (sentence and word- 
accent) has then gained a distinct place in grammar. It may be 
mentioned also that the phonetist recognizes phenomena closely 
parallel to these in the structure of the syllable. The syllable also 
has a relative accentuation, /. e, its various parts exhibit different 
d^^ees of pitch and stress, and like the word the syllable has 
usually one summit, which is a sonorous element, most frequently 
a vowel, as e. g. in hdnd ; often a lingual or nasal as in the second 
syllable of anr^rnrite, 2Xig\ngy \i^xAsinmmst That the summit 
accent is variable in position, according to the character of the 
syllable, can be readily observed in taking a set of pairs of syllables 
which vary from one another in their final consonants, these being 
in the one case surd and in the other sonant : seed and seat^ pease 
and piece ^ brogue and broke ; the syllable tone of seed, pease and 
brogue is upon a part of the vowel nearer to the final consonant 
than in seat, piece and broke. Further, there may, just as in word- 
accent, be more than one summit-accent, especially in long syllables. 

Schmidt himself has returned to the expedient of analogy in KZ. XXV, 591. 
Hartmann also emplojrs Wackemagers ideas on sentence-accent in order to 
explain the various forms of the second aorist, ibid. p. 66. And Wackemagel 
himself (KZ. XXIV, p. 470) accounts for the loss of augment in preterits by 
assuming different accentuation in subordinate and principal sentences. 



28 AMERICAN JOURNAL OF PHILOLOGY. 

If the syllable ' yes ' is pronounced in a contemplative way, e. g. in 
the sentence ' yes, that may be so,' it receives two summits with 
a decided &J1 between them. In general it can be noticed that in 
is<dated syllables &e relative accentuation of the various sounds 
gains especially clear expression ; so e. ^. in the various uses of 
the word ' well ' in such connections as ' well, let's go then,' and 
' well, are you ready ? ' The first ' well ' has filing tone, the second 
rising tone. 

The subject of syllable-accentuation so &r has not gained a very 
important place in grammar, and still bdongs to the phonetist 
rather than to the grammarian. But taken in connection with 
word and sentence-accentuation, syllable-accentuation serves to 
show that accent has been and still is a constant &ctor at worlc 
upon every infinitesimal subdivision of human speech. If we im- 
agine the course of human speech represented by a line, this line 
will be a consianify undulating one when we wish to mark the 
varying pitch of the sounds ; if we wish at the same time to convey 
a picture of the varying stress or ictus the line would constantly 
and gradually vary in thickness. Add to this the fact that this 
variation in pitch and stress is not the effect of one single kind of 
accentuation, but of a threefold one, and it will be understood how 
ddicate a subject for investigation it becomes even in living speech. 
In dead languages the difficulties are increased so as to make it 
hopeless that all the bearings of accentuation will ever be understood. 
The discussion must restrict itself almost entirely to accent in its 
pregnant sense, i. e. what we have termed summit-accent; only 
rarely will the stations for lower pitch or minor stress play a part 
in the discussion. For all the tradition on the subject, preserved 
either in accent marks or in the description of contemporaneous 
grammarians, is restricted to that, and is very fi''E^;mentary, as well 
aa vague in its terminology. 

e general phonetic bearings of this subject can at present be 
«i most convenientiy in Sievers's Handbuch der Phonetik 
ual of Phonetics), especially §§32-6, pp. I77-<I5 (word and 
ice-accent) and §§29 and 30 (on syllable tone). 



leems to-day almost a truism to state that a discussion of 
t accent must start from whatever knowledge there is on Indo- 
pean accent ; in other words, that the study of Greek accent 
be comparative. This is true precisely as much in this division 



STUDY OF GREEK ACCENT. 29 

of Greek phonetics as in any other, as for instance the study of Greek 
consonants, where one would not now-a-days presume to say 
much without bringing in the related languages. This, however, 
does not exclude the &ct that accent is, more than other &ctors in 
speech, subject to those forces in language which produce change. 
The Greek and Latin three-syllable accentuations present so fixed 
and peculiar a physiognomy even in their earliest phases that one 
would suspect that this restriction to the last three syllables of the 
word is something that was inherent in these languages from their 
origin, yet it has been proved for the Greek that this extremely 
peculiar accentuation is a development out of a system of accentu- 
ation to which such a restriction was originally totally unknown. 

The German language to-day exhibits a seemingly fixed law of 
accentuation, namely, that of the root-syllable. This seems a 
reasonable accentuation, for of all parts of a word the root would 
seem to be the most prominent and therefore entitled to superior 
stress and pitch. Yet no &ct in linguistic history is at present so 
dear as this, that the original German accentuation was not re- 
stricted to the root-syllable, but was a free movable accent, often 
upon the root, but hardly less often upon some sufiixal element 
This is proved by Vemer's law, and the accentuation of the root- 
syllable in the German of to-day cannot be due to anything else 
than the analogy of those words which, under the old free tone-law, 
exhibited the accent on the root; an analogy carried out with 
almost flawless consistency. 

This does not exhaust the variety of accentual methods to which 
Indo-European languages have arrived by various processes, often 
very obscure. The Lithuanian division of the Lithu-Slavic &mily 
consists of Lithuanian proper, Lettish and old Prussian. The last 
branch has died out without leaving any tradition as to its accentu- 
ation ; the first, the Lithuanian, exhibits a free accentuation which 
can be compared and identified with that of the Vedic Sanskrit, in 
spite of many deviations. The Lettish, which is related as closely 
to the Lithuanian as the language of Herodotus is to that of 
Tbucydides, has abnegated all historic accentuation and accents 
everywhere the first syllable. 

We need not go so &r as the Lithuanian and Lettish to find an 
equally striking and equally difficult phenomenon. The Aeolic 
dialect in Greece is differentiated from the other dialects in that it 
has given up almost entirely the accentuation of the ultimate. 
Excepting the oxytone prepositions of two syllables and a few 



30 AMERICAN JOURNAL OP FfflLOLOGY. 

conjunctions like nvnip, map, there can be no accentuation except 
that of the penultimate and the antepenultimate (Gottling, p. 39). 
This is one of the main elements in the fabled special resemblance 
between the Aeolic and Latin, and has been the cause of much 
nonsense,' and this resemblance with the Latin has also given birth 
to the equally erroneous idea that the Aeolic accent is older than 
that of the remaining Greek dialects. On the contrary, no one 
&ict in Greek accentuation is clearer than this, that the oxytone 
words in Greek are generally archaic, that they have more than 
all others resisted the recessive accent.' 

To this tendency on the part of accentual systems to change in 
such a way as to lose its original complexion entirely, the fact is 
due that the comparative treatment of accent was, until very 
recently, a method which bad not gained a firm hold upon the 



'AH these do not exhaust the varieties of seemingl]' fixed systems which have 
been built up upon Ihe debris of the old I. E. accentual ion iu the various 
families. In the Slavic tankages, the Russian has still preserved noteworthy 
poinls of contact with the accented Vcdic Sanskril, but the Bohemian has 
adopted the same system as the Lettish menlioned above, namely, the accenta- 
atioD of the fitsL syllable, while the Polish has woiked out for itself a still more 
peculiar system. All its words, excepting those bonowed from adjoining dia- 
lects, are paraiytone, and here we are again led to the only reasonable expla- 
nation, namely, that the frequent paroxytone accent of 1. E. times was here 
extended Into a law. 

We can pick a case from the modem Romance dialects which will show the 

same complete change of accentuation, and which will at the same time carry 

the solution of the chaoge with it. The words which are the representatives 

of the old abstract sutGx Idl (Lat, nom. Ids, fraltmilas) are oxytone : French 

fraltrtiiU, \ul. frattmili ; oxytone accent is a most non-Lalin quality. Asoln- 

tion for this case which is altogether probable is that the modem oiytoncsis 

has preserved the accentuation of the obliqne cases : ftaUnatitis, etc. The 

;lisb on the other hand holds to the accent of the nominative. In the same 

the French cenicription has the accent of the oblique case, -iwr. In a case 

French parlir over against Italian faridrt the accentuation of the ultima 

ies its own solution with it still more clearly. 

Umost all the important categories of noun farmation which are oxytone 
ear in their original accentuation, as can be seen even from superficial 
parison. Thus nouns of agency in -^, «op6i = Sk. bhards; but the nouns 
iction are paroxytone, ^(m; = Sk. ihdrai; adjectives in -fcf, tiHt = Sk. 
tiii.iJjixbi = SV. bghUi.Lmcbt -sSV-Ofilt; adjectives in -pit, cpvep6t = Sk. 
Vrdt; verbal adjectives in -rdt, idvTrSc = Sk, (nitiit = 0. H. G. iiilt = 
;. Imni (KZ, XXIII 133), JwmSc = Sk. faildi ; the word for father, TOTifp 
k./«i!a = Goth. /flfliir{ibid. 117); the perfect active participle eidiir = Sk, 
loiu (cf. i^HD = Sk, vi'iUff). In declension Jmbt i troiSf = Sit. f&d : fiaddt ,■ 
■ : Aiif)6{ = Sk. ifydtlt : divdi. 



STUDY OF GREEK ACCENT. 31 

minds of investigators. Parallelisms and resemblances between 
individual facts of Greek and Sanskrit tone-laws were noted very 
soon ; even large collections of words and word-categories which 
exhibited identical accentuation were made, yet this did not seem 
to impress investigators with the fact that, unless these resemblances 
were accidental — ^and that theory was not advanced — the two lan- 
guages were committed to the same original accentuation in every 
part, and that it must be shown why and how they present such 
impKDrtant differences in historical times. On the contrary, investi- 
gators were content to call in, for Greek as well as Latin, the 
recessive principle (which after all is not recessive, inasmuch as it 
stops at the third syllable) as a something gotten no one knows 
where, perhaps as Bopp has it ' because the greatest recession of 
tone expresses the greatest dignity and energy.* * 

To-day any one who wishes for a hearing on the subject of the 
accentuation of any Indo-European language must operate with 
the following principles : 

1. The accentuation of any I. E. language is a development out 
of the common I. E. accentuation, precisely as much so as the 
sounds and forms of that language, be they ever so changed, and 
be their analysis ever so difficult or even impossible. 

2. The principle which changes accent is precisely the same as 
that which changes other language matter, regular phonetic change 
based upon phonetic law. Just as an I. E. consonant is changed 
in German according to Grimm's law, so it is possible that, e, g. 
originally oxytone word-categories may become paroxytone in 
some one language,' only this must be shown to take place accord- 

' Vergleichendes Accentuationssystem des Sanskrit und Griechischen, p. 16. 

• Or we will recognize below (p. 42) as important another Greek phonetic 
law of accent, namely this, that enclisis cannot extend beyond three morae and 
two syllables. Enclisis in general is an Indo-European quality [e, g, Greek 
rz = Sk. ca = Lat. que^ etc., are all of them enclitic), but the Greek restriction 
as'to morae and syllables is a Greek phonetic law in exactly the same sense as, 
e, g, the loss of F or I. E. v. The Vedic Sanskrit knows no restriction of this 
kind ; a word of any length may be enclitic, as e» g. the stem soma * any one * 
(Greek stem d/to- in dfid^ev) is enclitic, not only in forms containing two 
syllables, but in all its forms, ^. g, ace. samam, abL samasmdt, gen. samasya. 
And sereral enclitic words may follow one another, so several vocatives, or 
vocatives with cases depending upon them, as e, g. Rig- Veda, VII 64, 3 : d 
rijina maha ftasya gop& . . . yitam: *0 ye kings, guardians of great right 
come hither.' Here four successive words are enclitic, cf. Journal of the 
American Oriental Society, Vol. XI, p. 59. 



32 AMERICAN JOURNAL OF PHILOLOGY. 

ing to a law, and this law must like all other phonetic laws be based 
upon the results of observations exercised upon extensive material. 

3. Where no phonetic law can be adduced, the influence of 
analogy must be the changing &ctor. So e. g. the modem German 
with its prevailing accentuation of the root-syllable, the significant 
syllable has been explained above ; the influence of analogy in the 
Greek ' recessive ' accent will be discussed further on ; it is perhaps 
the most striking and convincing case of the workings of analogy. 

4. The influence of foreign languages and adopted words cannot 
be Icfl out of account These usually carry their tone with them 
from home. So e. g. large cat^^ories of words in German betoken 
by their accentuation what is sdso known otherwise, namely, that 
they are of foreign descent, e, g, nouns in -/a/, -urn, etc., universHai, 
imHttiHon^ which exhibit foreign accent ,* the entire dass of verbs 
in iereUy siudieren, marschieren^ in the same manner exhibit French 
suffix and French accent ; according to Grimm words like reiierei, 
malerei^ etc., have sufiixal accentuation, although they are in their 
root good German words, because they were formed on the analogy 
of melodey (^cX^/a), abtei (abbatfa), so that this is an example 
where a distinct category of German words received both suffix 
and accent from abroad.* 

The question which arises next is : What was the character of 
this Indo-European accentuation from which the various peculiar 
accentuations of the several languages have developed ? Of course 
the question can be answered only for the smallest part; almost all 
that is known is restricted to the summit-accent, and even here 
nothing is absolutely and completely dear. We will here consider 
only the one fact which, above all others, has gained an unimpugned 
poaitioUi namely, the freedom of position of the summit-tone of the 
L E* word ; other qualities both of word and sentence-accentuation, 
which are probably Indo-European, will be discussed further on 
\\\ connection with the Greek itsdC 

The feet that the I. E, parent-language knew none of those 
rrntrictions as to the position of the tone which we see in almost 
all the laiVjk^iages that are still alive, and also in Greek and Latin, 
cR|>einally the latter, is seen by a comparisoa of the accented Vedic 

J * The ittrtwe^ce of foreign UnguAgct upon accentuation is still more strikingly 
•xhU>ilei\ in the thivcfoKl tone of the Gennan word £ ramwiati k^ namely, grdm" 
m/tfi f, jf*v»*»r»»r^rt* anii x'namm.ift k. The l*st contains the French accent (jtuM' 
9*hif/i^tH-\ Xht one prece^ling the Latin {(^AtmmaAra)^ while the first represents 
th« genuine Iterman pKvnunciatt<vn with the tone on the root. 



STUDY OF GREEK ACCENT, 33 

Sanskrit with the Greek and German. This comparison yields the 
result that the Vedic accent has preserved very closely the old 
word-accent of the I. E. parent-speech. Of course this result was 
obtained by the usual methods of comparison. Whatever in Greek 
and German accent has, upon investigation, proved itself to be 
archaic, is not only to be found freely in the Vedas, but is usually 
seen there in the form of a principle of wider scope. So e, g, the 
seemingly irregular accent of the participles and infinitives of the 
thematic or second aorist in Greek is an archaism on Greek ground. 
In the Veda this entire tense-system is accented on the same place, 
the thematic vowel, except in the augment forms, where the augment 
always takes the tone, cf. below, p. 58. In the same manner it will 
be observed repeatedly thftt the Greek cases of oxytonesis are 
usually of a somewhat disjunct and fragmentary character. Not 
clear in themselves, they do not yield up any principle until we see 
them in their full bearings in the accent of Vedic word-categories 
which accent the ultima. And again in German, Vemer's law has 
shown that the more salient principles of Vedic accentuation, such 
as the shifting of the accent from the root to the flexional element 
in the non-thematic conjugations, belong to the oldest property of 
I. E. speech, cf. below, p. 35, note ; it has also shown that appa- 
rently irregular accentuations, such as the Vedic accent of the nouns 
of relationship, piidr but maiar^ must be carried back to the primi- 
tive Indo-European language. 

No syllable, then, of an I. E. or Vedic word was, on account of its 
position or on account of its quantity, unable to bear the summit- 
tone ; no restriction, such as is seen in the three-syllable accents of 
Latin and Greek, or in the root-accent of the German, is to be 
found. Thus {ndra^ indreri^, dnapacyuia, dnabhimldtavan^a^ ag- 
ntnam, abhimaii^dhd^ parjdfiyajinviia, etc. (Whitney, Sk. Gram. 
§95) present instances of Vedic accentuation. As far as the meaning 
and value of this free accentuation is concerned, it must be confessed 
that little or nothing is known. Indeed, it may be fairly said that, 
in accordance with the more modest spirit in which linguistic 
investigation is carried on to-day, no very ardent search is made at 
present for a cause which distributes the accents over these various 
syllables. It is felt generally and justly that final explanations of 
such delicate questions are not in order. The energy of accent- 
investigators must be directed to an investigation of the simple 
details of accentuation, and the causes of these variations in the 
separate languages, before it can be hoped at all that the original 



34 AMERICAN JOURNAL OP PHILOLOGY. 

cause of these phenomena will be understood. As long as e. g. 
the restriction of Latin accent to penult and antepenult is a mystery, 
so long there can be no hope of actually penetrating into the inner 
life of the accentuation which preceded it 

Yet a noteworthy attempt to explain the I. E. accentuation dates 
back to 1847. The first one and almost the last one who undertook 
to describe, systematically, the accent in its historical development 
in the I. E. languages, and at the same time to assign a cause for its 
original character, was a French scholar, Louis Benloew, in a work 
entitled * De I'accentuation dans les langues indo-europ^ennes tant 
anciennes tant modemes.* According to Benloew the summit- 
accent was originally an accent purely of pitch, a musical accent 
without stress or ictus. In each word which consisted of more 
than one syllable, some one syllable was pronounced musically 
higher than all the others ; the syllable which was thus distinguished 
from the others was, according to Benloew, the chronologically last 
defining element in the word (le dernier determinant). That is, 
according to the theory of word-construction which ruled in Ben- 
loew's day without opposition, and which is accepted to-day also 
to a very considerable extent, a word is made up of root, suffix, 
personal inflexion, case-ending, augment, reduplication and so forth, 
and whichever one of these various elements in the word had been 
joined to the word last, that was entitled to this higher musical 
pitch. So e, g, in an augment-tense the augment, in a noun in the 
genitive the genitive ending ; when a word was compounded with 
a preposition, the preposition. As long as this principle was still 
in existence, the unity of the word in our sense had not as yet 
developed; the marked emphasis of the 'dernier determinant' 
directed the attention of both speaker and hearer so strongly to 
some part of the whole, to some special element in what afterwards 
became a unit, that it must be supposed that this accentuation was 
in force in a period previous to that of word-formation in its strictest 
sense. The cementing of the word as we have it now was pro- 
duced by an additional force. By the side of the principle of the 
last determinant there was developed slowly and gradually a logi- 
cal principle of accentuation whose piupose it was to act without 
reference, and in fact in opposition to the specializing tendency of 
the ' last determinant' This logical accent, it is assumed, affected 
the root-syllable, which, in the word as a whol^ is the ruling 
syllable. The further history of accentuation in the separate L El. 
languages exhibits, then, a gradual process by which this logical 



STUDY OF GREEK ACCENT. 35 

accentuation gains the ascendancy in the word. This in turn is 
gradually counteracted and affected by the influence of quantity, 
which Benloew, with true instinct, regards as the last factor which 
entered the arena. In Sanskrit, as far as is known, the accent is 
totally independent of any considerations of quantity ; in Greek, 
quantity, especially of the final syllable, begins to exercise an influ- 
ence on accent ; still truer is this of the Latin, where quantity and 
accent balance each other almost entirely. 

The boldness and the esprit of Benloew's thoughts on this subject 
are quite out of proportion with their sobriety, with the extent of the 
material upon which they were based. In feet they are in all 
important respects hardly more than ingenious assumptions. Yet 
his theories deserve even &-day a certain degree of consideration, 
for they gained such wide adherence that certain of his thoughts 
are even now silently accepted. So, above all, the musical character 
of the early I. E. summit-accent, which has never been proved, and 
which, if separated firom stress, is certainly to our ears an extremely 
peculiar accentuation. Vemer, in his explanation of the Old German 
accent and its influence upon the mute consonants, starts with this 
statement : * The I. E. accent was, in its nature, chromatic (/. e. 
musical), and, in its use, of unlimited freedom of position * (KZ. 
XXIII, p. 128). He then proceeds to explain his exceptions to 
Grimm's law, by the assumption that the accent became an accent 
of stress (expiratory) in primitive German, or possibly a combina- 
tion of musical and stress accent. Benloew's other important idea, 
namely, that of the 'last determinant,* has also been revived in our 
day to explain a phenomenon of the widest extent and of great 
importance, namely, the variation of stem and accentuation in the 
non-thematic verbal conjugations.' 

* In Greek this variation of stem is preserved intact only in a few cases, and 
its immediate cause, the shift of accent from the stem to the root, is lost to 
sight, owing to the leveling force of the recessive accent in verbal accentuation. 
But the variation of stem-form as well as the accompanying shift of accent is 
easily established as archaic by comparison with the Vedic Sanskrit, so in the 
following cases: 



tUfit 


rf(»tK«)0 


t\-ai{^t\-Ti) : \'^ 


Ure 


I'daii^l'avTi) 


/-mi 


/-ti 


/-ti : i-mdsi 


i'thds 


i'dnti 


foid'-a 


foia-Oa 


Foid-e : fid'fjiev 






v/i^ 


vH'tlM 


v/d-a : vid-md 







The duals, though they agree in both languages in having weak root-form 
(and accordingly are accented on the personal endings in Sanskrit), are left out 



36 AMERICAN JOURNAL OF PHILOLOGY. 

Benloew's work represents the first and also the last attempt on 
so pretentious a scale to inquire into the original character, develop- 
ment and history of I. E. accentuation. The next somewhat com- 
prehensive work we owe to the founder of comparative philology, 
Fr. Bopp, in a book entitled ' Vergleichendes Accentuationssystem 
des Griechischen und des Sanskrit,' Berlin, 1854. This work has 
really a much narrower scope, it does not profess to deal with 
general questions in any way, it merely attempts to give an exhaus- 
tive list of those words in Greek which have still preserved the 
accentuation of the Sanskrit and, therefore, in all probability the 
I. E. accent. 

Yet, incidentally, Bopp does express himself on general matters, 
and in a way that cannot be called happy, either in its method of 
treating the question or in the result reached. He recognizes as 
the principle of Sanskrit as well as Greek accentuation ' the greatest 
possible recession of the tone to the beginning of the word,' p. 16- 
17. This mode of accentuation possesses the greatest dignity and 
strength. The limitation of the summit-tone in Greek to the last 
syllables he looks upon as a d^radation or enervation of 

sideralion owing to the pmblemalic character of the endings. In this 
on of siein and accent one fact seems clear beyond all doubt, namel]', 
le weakening of the root is due to the shift of accent to the personal 
;; but the question arises, what maj' be the cause of this varying position 
accent? There has been, as far as is known, but one answer to this 
>n, that of F. De Saussore in his M<hnoire sur le Systeme Priroitif des 
es dans les Langues Indo-Europdennes, p. iSg. and that is distinctly in 
irit of Benloew's theory of the ■ Ust determinant.' Saussuie assumes 
'ciedrich Mullcr (cf. now also Fick in the ' GOttinger Gelehrte Anzeigen ' 
Bl, Vol. II, part 45, 46, p. 1461) that the so-called secondary personal 
;s of the verb are more original than the primary, not that the secondary 
: result of weakening from the primary, as has been generally held from 
1 day down. The primary endings often differ from the secondary by an 
>nal (. and it is thought that this i is the same deiktic particle which 
7, /. g. in Greek rovnivJ. Thus 

I/f. %sg. 3ig. zfliir. 
Primary : mi si ti nti 

Secondary; m s t nL 

issuming that the secondary endings first entered into verbal formation 
at thai pmonal endingi rtctiixd the Imtt, whenevtr thty ceu!d , a. reasonable 
1 is gained for the exceptional position of the three persons of the singuUt ; 
he endings are only m, 1, t, which are not fitted for carrying the tone of 
lid ; therefore the lone remains on the root and preserves in it a stronger 



STUDY OF GREEK ACCENT. 37 

Isinguage. The accentuation of final syllables or syllables near the 
end is due to the * sinking ' of the accent from a position nearer 
the beginning of the word, etc. Nowhere, however, does he indicate 
in any manner by what process of investigation he came to this 
result, though these ideas permeate the entire book and are urged 
upon the reader with an evident fondness on the part of the 
author. They do not seem to be the result of investigation as to 
the nature and quality of the accent of these languages ; they are 
in fact not offered as such. They are given merely as the ex 
cathedra opinion of the master who, if any one, has a right to 
speak ex cathedra. 

Since Bopp's book, no comprehensive treatise on I. E. accent 
has appeared, nor is it likely that any such pretentious attempt will 
be made until investigation in the separate languages has established 
a better insight into the special accentuations ; there is reason to 
hope that the now recognized importance of the study of sentence- 
accent will shed much light both upon the original history of accent 
in primitive times, as well as upon the ways in which the historical 
accentuations of the several languages developed out of the single 
Indo-European language. 

What we have gained from this discussion of Indo-European 
accent is, first, the knowledge that the word-accent was a free one, 
restricted to no special syllable or syllables of the word, and 
untrammelled by quantity ; secondly, that the I. E. language knew 
certain well-defined laws of sentence-accentuation, the traces of 
which may be fairly looked for in the separate descendants of it. 
Thirdly, that the elements which may be supposed to have changed 
this original accentuation can scarcely be different from those at 
work elsewhere in the formal life of language, regular phonetic 
change and analogy. As will be seen, what knowledge we have 
of Greek accent calls for no other factor and no other principle, nor 
is it likely that any new principle, as yet unknown, will ever exer- 
cise any important function in the progress of this difficult study. 



III. 

We turn now to the Greek itself. The literature of the subject, 
both ancient and modern, up to the year 1875 ^ carefully collected 
in the first paragraph of the book of Franz Misteli : * Uber griech- 
ische Betonung: Sprachvergleichend-philologische Abhandlungen,' 



38 AMERICAN JOURNAL OF PHILOLOGY, 

Paderbom, 1875.* Among the ancients the subject is scarcely 
touched upon in classical times. The first mention of it is in Plato's 
Cratylus, p. 399, where the terms of v? and /Sa/w? first turn up ; next 
in order is Aristotle, Poetica, chap. 20, where, in addition to the 
h^\m\^ and Papvrrfs of Plato, a fuaov is mentioned, u e, a middle-tone, 
which has been by some exalted to a most important position in 
the theory of Greek accent, as we shall see soon. Aristarchus in 
Alexandria is the next authority in chronological order ; but above 
all other works of the ancients, the source for information is Hero- 
dian : Herodiani technici reliquiae, collegit, disposuit, emendavit, 
explicavit, praefatus est Augustus Lentz ; especially the first volume 
containing Lentz's famous preface and the book n-cpl KaBokixffi wpo- 
<r<^tas, to which Misteli gives the first place among his authorities. 
In the study of modem writers on this subject one need not go 
back behind Gottling, Carl Gottling: Allgemeine Lehre vom 
Accent der griechischen Sprache, Jena, 1835 ; a book valuable for 
its digest of the opinions of the Greek grammarians, containing 
rich collections of material, but of course to-day almost worthless 
as far as theory and explanation of phenomena are concerned. 
Next in order are the books of Benloew and Bopp, which have 
been discussed in the preceding chapter. It may be added that 
Bopp*s book, while almost worthless as far as its general theories 

1 The literature which is given there is more than fall enough up to 1875. 
He omits one book which is practical and valuable for accent of nouns, namely, 
Chandler, 'A practical introduction to Greek accentuation,' which has appeared 
lately in a second revised edition, Oxford, 1881. Since Misteli there have 
appeared in addition to the many and often extremely valuable incidental 
remarks and minor investigations of comparative grammarians, a few important 
monographs bearing upon the subject : 

Leonhard Masing : Die Hauptformen des Serbisch-Chorwatischen Accents, 
nebst einleitenden Bemerkungen zur Accentlehre des Griechischen und des 
Sanskrit, St. Petersburg, 1876, valuable for Greek accent in its first half, pp. 
1-49, containing especially an exhaustive criticism of all opinions on the 
grave accent, §44 fg., p. 19 fg. 

Jacob Wackernagel : Der griechische Verbal-accent (KZ. XXIII, p. 457 fg.), 
of the greatest importance for the general theory of the so-called recessive 
accent. 

Theodor Benfey : Die eigentliche Accentuation des Indicativ Praesentis von 
ig und ^, etc., cited in the note on p. 25. Important for its solution of ana- 
strophe, and its valuable remarks upon enclisis and proclisis. 

Leopold von Schroeder: Die Accent-gesetze der homerischen Nominal-com- 
posita dargestellt und mit denen des Veda verglichen, KZ. XXIV, 101-28; 
the first systematic attempt to establish Indo-European laws for the accentuation 
of compounds. 



STUDY OF GREEK ACCENT. 39 

are concerned, is valuable as a clear and comprehensive exhibition 
of the facts which it treats, namely, the coincidences in the 
accentuation of Greek and Sanskrit words. Next, the subject owes 
some noteworthy and ingenious essays to Franz Misteli and James 
Hadley ; Franz Misteli : t)ber die Accentuation des Griechischen, 
KZ. XVII, p. 8i %., p. i6i fg. ; XIX, p. 8i fg. ; XXI, p. i6 i%. 
After the appearance of Hadley's article these essays were rewritten 
in book-form : tjber griechische Betonung : Sprachvergleichend- 
philologische Abhandlungen, Paderbom, 1875. Hadley 's brilliant 
paper was published no less than three times : On the nature and 
theory of Greek accent, by James Hadley, from the transactions 
of the American Philological Association, 1869-70; translated in 
Curtius Studien, V 407-28, reprinted in Hadley's collected essays, 
edited by Whitney. Hadley*s as well as Misteli's theories, which 
are closely implicated with one another, will be discussed below. 
Finally, much important material is contained in the four monographs 
died in the foot-note on p. 38.* 

If we now attempt to give a short general statement with regard 
to the position of the summit-tone as it appears in Greek, compar- 
ing it with that of the free I. E. summit-accent which we have seen 
established, we may best formulate the facts as follows, under two 
heads : 

1. This free I. E. accentuation has been allowed to continue in 
Greek in all kinds of formations, excepting finite farms of the verby 
when the free accent did not go beyond the antepenultima, e. g. 
icKvT6i : xKiFos = grutds : grdvas^ cf. the Germ. Mat (Ags. JUud) ; 
irow : woddff ^=^pad : padds ; Xi9r»y : tkvnov = ricdn : dricam^ vidvAn : 
vidti^i = cldttf (for older *2d«y) : Wwtd, etc. See Bopp, Vgl. Accen- 
tuationssystem, pp. 178-84. 

2. In all the finite forms of the verb and in all those formations, 
verbal, nominal, or otherwise, in which the old accentuation stood 
before the antepenult, a new principle of accentuation has established 
itself to the exclusion of the old free accent The chief trait in this 
new law is that it does not allow the accent to remain on any 
syllable beyond the antepenult, but restricts it to the last three 
syllables of the word. To this law there is scarcely an exception 
in the entire tradition of Greek ; tlie grammarians have fixed the 
accent of two Aeolic words which contain diaeresis on the syllable 

' Misteli, in his list of authorities, mentions also the most important treatises 
on Latin and Sanskrit accentuation, which do not, however, concern us so 
directly. 



40 AMERICAN JOURNAL OF PHILOLOGY, 

before the antepenult, Mijdfm in Sappho and the Lesbic im\t£Ki\Ka^ 
which are not of enough importance for a general discussion. 
Gottling, p. 20, note 2, and especially Misteli, p. 19, discuss them 
fully. There are, of course, some words in which the theoretical 
analysis of forms would lead to seeming exceptions to this law of 
three syllables, e, g, lUKaiva if we carry it back to its ^/AcXoyta, or 6vya- 
Tfjts if it is derived from *Bvyartp(s ; but this is prehistoric ; at the 
time when the pronunciation was fiAaiwi, all reminiscence of an 
earlier */i«Xavia was gone Within these three last syllables the 
position of the tone evidently stands in relation to the finer measure 
of mora, as appears clearly in the law that the accent cannot pass 
beyond the penultimate when the ultimate is long, so that the 
Greek accent is, to a considerable extent, restricted to the last three 
morae, e, g. in such types as jjdiicovv, dtdotcv, /X/yo/icv. To this there 
is in fact only one seeming exception and one real one : 

I. A seeming exception to this restriction to three morae is offered 
by such cases as e, g, the genitive ic^ov, where the acute is appa- 
rently four morae fi'om the end of the word, but where in reality 
the second mora of the long penultima has the tone, so that if we 
analyze into morae and write *ic€cVoo, it becomes clear that the 
exception is only apparent. That the acute on a long vowel means 
the accentuation of the last mora is not a mere assumption, as is 
shown by such cases as iarw contracted from cWa^y .* In such cases 
a contraction has taken place, and if the tone had been on any 
other than the last mora the result would have been a circumflex ; 
the reason for the absence of the circumflex is to be found in the 
fact that the last vowel contains two morae (*€cn-a-o6r), with the first 
of which, the toneless mora, the a contracts ; it thus leaves the 
accent untouched in the result, cWwr.' 

1 That ioTa~6^ is the old type of this perfect participle can be seen from the 
Sanskrit equivalent tastki-vin ; here the Sk. 1 equals the Greek a, as in sthi-tds 

« The circumflex cannot display itself upon less than two morae (' ^), therefore 
also this projected ♦ityra-Sdf results in oxytone iarwf. A case where this law of 
circumflex is clearly exhibited is the vocative of the word Zct«c. Zch^ (for •Atei^) 
is an old oxytone = Sanskrit dyatis. By an Indo-European law the accent in 
the vocative recedes to the first syllable of the noun, that is, the tone is as near 
the beginning of the word as possible. The result for this stem is the vocative 
Zf I' (f . f, Zlv) =. dysits (». e, d/dits). The recession has taken place, but as the 
word contains but one long vowel, the tone has passed from the last mora to 
Oie first, exhibiting at least for diphthongs the actual divisibility of long voweU 
into morae. 



STUDY OF GREEK ACCENT, 4 1 

2. The second exception to the law of three morae is much less 
easily disposed of. When the tone is on the antepenult and the 
last syllable is therefore short, but the penult is long, then it stands 
at least on the fourth mora from the end, as e, g, in a(a<rros ; and 
when both the penult and antepenult are long, apparently on the 
fifth mora from the end in a case like ^ctpof.* In both of these 
cases there is, of course, no a priori reason why the law of three 
morae should not have been kept in force by making both words 
properispomena.* The only explanation that the authorities have 
been able to bring forward is the rather unsatisfactory one which 
assumes that in such cases the long penultima received a more 
hurried pronunciation and suffered a loss in quantity. So Gottling, 
p. 27 : * the penultimate loses a part of its quantitative value because 
the strength of the tone of antepenult outweighs the following long 
syllable,' and in the same tone other writers down to Kiihner. 
The difficulty in the way of such an assumption lies, of course, in 
the metrical value of such toneless long penultimates ; they are 
just as inviolably long as any other long syllables ; the « of rfTrtipos 
differs in no way metrically from the ly of the same word, and 
the explanation given has quite the appearance of having been 
constructed ad hoc without any sufficient ground. It is not 
uninteresting that there are quite a number of cases in the lan- 
guage in which both accentuations occur in the same word, one 
having the tone farther back from the end than the third mora, 
and the other having it on the third mora. In every case the 
one which follows the rule of three morae is the older one, e, g. 
€pfifi09 Epic and in Herodotus, but Attic usually tprj^ios; Sfwios 
Homeric, Ionic, and Old Attic, later Sfioio^; rponatov Ionic and 
Old Attic, common rp&iraiov] in the same way of rroiiios and 
rroifios the first is the more archaic form. In 6fioiog : Sfioiog the 
historical precedence of Sfwiof is easily proven etymologically ; 
6fioiof is a secondary derivative from the oxytone stem ofid- = 
Sk. samd- with the secondary suffix -10- = Sk. ya- (Vedic -id). 
By an accentual law, which perhaps dates back to the common 

'Apparently only if we assume that the tone is on the last mora of ijntipo^ 
{^Hiretpo^) as in k^ov {^Ktiiroo), 

* The extent to which such accentuation is favored in Greek may be best 
seen in the rendering of such Latin names as Dentdtus^ Mod^sius, Ahenobdrbus^ 
etc, by ^kvraro^^ Mddecrroc^ kiv6^apfio^^ etc. Nothing, except the predilection 
of the language, is in the way of such an accentuation as Aevrdrof, etc. Hadley 
in Curtius's Studien, V, p. 413. 



42 AMERICAN JOURNAL OF PHILOLOGY. 

Indo-European period,' such a combinatioii as dfu{ -(~ *<> yields 
6ruMo-, i^ e. 6pflo-, c£ the case of Zm (i. e. Uv) discussed above 
on p. 40, note. We might then see in such cases the trace of a 
still more stringent law in favor of the three niorae ; possibly the 
principle which underlies the recessive accent started strictly irom 
that point 

Whatever this law of three morae is, it may be noticed right here 
that it is also the Greek law for enclisis, i. e. a Greek word can 
incline upon the preceding word only in such a way that the result 
does not produce conditions which are in conflict with the law of 
three morae as laid down above. So e. g. Ztix lun offers the condi- 
tions which are apparent in apcaa ; jcoXik ion* the same conditions 
as ^(uoToi. When, however, it is desired to mcline Tfimv, fjpa, tHtat, 
or tifiBv, ijiM, iiiii, the result is Ztii f/iBv, Zns ffw or ^^v (with a 
shortenmg of the last vowel which may stand in connection with 
the removal of the tone from the ultima), etc That is to say, owing 
to the ^t that these words contain at least four morae they cannot 
become entirely enclitic, but become so as much as possible. The 
grammars ' (_g.g. Hadiey, §232) do not understand this phenomenon, 
when they describe Ifiutu, etc, merely as optional weaker forms, 
and not as enclitic forms.* Aside from the testimony in favor of 

'The circumfiex in such casern is probably Indo-Earopean, for in Suiskiit 
also the acute vowel on the a ot samd- would be rollowed by the so-called 
enclitic tvarita on the next syllable (la), which seems to imply that the voice 
instead of sinking Trom the acute to lowest pitch without mediation, passes 
down gradually, aod this amounts evidently to the same phonetic result as the 
in ^/toio-. See Whitney in the Proceedings of the American Philo- 
lociatiou, 1S70, p. 9; 5k. Gramm. gSS. 

ammais falsely set up the paradigm ttiii, keri, laylv, etc. The words 
cs and receive this acute only when enclisis of the entire word Is 
Msible because the result would leave too many morae unaccented. 
It is therefore due to sentence-law and it nM etymological. The 
t of Jirri is preserved in orthotone lart, see below, p. 6t. The reason 
words as well as ^/f, etc., are enclitic will be discussed in fnll 
i7. 

r calls it ' eine gam eigenthUmliche Art der Deklination,' I, p. 364. 
sumption of enclisis in the shorter forms (/uh, ytrS), but of orthotonesis 
y changed accent in the longer forms (v^v, v/uic), apparently receives 
:iod of support from the Sanskrit, where the enclisis of the personal 
)f the first and second persons, being evidently of a piece with the 
the same peiaons in Greek, is also restricted to menesyllaHc forms, 
nins of the third person, >'. e. the varioas demonstrative stems which 
)at function, do, however, incline forms of more than one syllable 
r. asmei 'to him,' a{}ia 'of him,' are used both ortbotonically and 



STUDY OF GREEK ACCENT, 43 

endisis that is afforded by the parallelism of, e. g. ftoi and fiovf 
when compared with ^fiol and c^iov, we have most interesting native 
authority to the effect that in Greek pronouns, the recession of the 
accent in accordance with the law of three morae was the substitute 
of enclisis when the word inclined possessed itself at least four 
morae. Wackemagd, in KZ. XXIII 458, cites from Apollon. 
Synt p. 130, a passage, also treated by Lehrs, Quaestiones epicae, 
p. 123, which bears upon this question: ^pKtaSri 17 KyxXiats dt& rrjs 

IttToBtfrtiot Tov r6vnVi IfKovtr Ijfuev • • • r^r ra<TC«»ff furariBefifprfg Korii rfjv 
ipXtntiraP * ifiinfOTU yhp rirl rh npoKtififPOP fr/xxrcX^ciy. ThlS passage, 

from excellent ancient authority, proves almost beyond a doubt 
what seems in every other way also probable, namely, that Ijfmp, 
vfimwf etc., are cases of enclisis, and that, therefore, endisis and 
recessive accent are ruled by the same law of three morae. The 
same prindple is, of course, patent in other well-known attempts 
to observe the same law ; in (act if we take the cases which Hadley 
gives in §107 : Mpo»7r6t ris, ircMs nw, Xrfyoi nw, we have in every 
case an endisis which is rectified or rather cut short by the law of 
three morae, as exhibited in the general recessive accent ; it is to 
be noted that the position of the tone on the fourth mora from the 
end is also exhibited here, when the penult has a long vowel and 
the ultima is short, o0 ^70^ like ffntipof, \6yoi rt(ycff) like ACnaros, It 
will be seen below, p. 56, of what importance it has been thought, 
that the laws which govern the scope of endisis, and of recession 
of the Greek accent, are identical. Wackemagd's theory about 
the recessive accent, which has commended itsdf to the acceptance 
of most modem grammarians, is in the main based upon this 
coincidence. 

IV. 

If, in stating the most prominent views with r^ard to the peculiar 
character of Greek accentuation, we were to begin with Gottling, 
thb would be done in deference to a book which must still be kept at 
one's elbow in the study of this subject In some respects it might 

enditicallj, cf. above, p. 31, note 3. It may be further said that the Sanskrit 
proves nothing against the enclitic character of such forms as ^fujv by the side 
of ^fiuv, because it happens to possess different polysyllabic forms made from 
different stems by the side of the monosyllabic ones. It is not surprising that 
a language which can choose between asmdbhyam and nas for the dative plural 
of the personal pronoun of the first person, should choose tumdbhyam when it 
required an orthotone form, but nas when it desired enclisis. The Greek has 
no such choice in the cases involved. 



44 AMERICAX JOURSAL OF PHILOLOGY. 

Still be necessary to warn against it, while in others it might be 
mentioned profitably as a sdentinc curwsum of efforts in this 
direction, not as yet fifty years old. Guttling might also perhaps 
deserve a mention because he represents the last attempt to account 
for Greek accentuation, entirely out of itself, though even he 
occasionally takes a glance at the incipient work of comparative 
philology — he often refers to Humboldt and Bopp— or brings 
on some real or seeming parallelism from some other language. 
Occasionally again he sees Luther than some of his successors, as 
when he recognizes the faxx that the recessive accentuation began 
in the non-Aeolic dialects with the finite verb. The neglect of 
this fact is one of the weakest spots in the theory of Misteli-Hadley, 
which will be discussed immediately. Yet the limited space of an 
article forbids any systematic mention of Gottling's views, and as 
the views of Benloew and Bopp are already disposed of, we can at 
once tiUTi to the Misteli-Hadley theory. Misteli's theory on the 
peculiar form of Greek accentuation was based upon comparative 
studies as well as * philological ' investigations in the Greek gram- 
marians. It was firet laid down in Vols. XVH, XIX and XXI of 
Kuhn's Zeitschrift, and afterwards embodied in the form of a book, 
whose tide was given above, p. 37. In the period between Misteli's 
articles and Misteli's book there appeared Hadley's article in the 
Proceedings of the American Philological Association (cited ibid,'), 
an article which aimed to rectify Misteli's theory, and which ex- 
tended it by bringing in the Latin within the framework of the 
theory. Therefore the name Misteli-Hadley theory. 

The key to the explanation of the three-syllable or three-morae 
accent according to this theory is the assumption of a middle-tone 
(mittel-ton) which, already in the parent-language, followed imme- 
diately upon every summit-tone, as a kind of intermediate step 
which served to bring the voice gradually from the musical height 
of the summit to the lowest depth (the toneless syllable). Nowhere 
was the passage from the summit-tone to tonelessness in the same 
word one which did not involve this middle-tone. If there were 
syllables left in the word after the two which are bespoken for the 
summit-tone and the middle-tone, these — and their number is left 
indefinite — are toneless, or according to the preferable terminology 
of the German receive the * tief-ton.' This theory of a middle-tone 
is suggested in the first place by the Vedic Sanskrit. This possesses 
a mode of accentuation which distinguishes three kinds of tone, i. 
a higher (tidatta * raised ') or acute ; 2. a lower {anudddtta * not 



STUDY OF GREEK ACCENT. 45 

raised ')i i* ^. toneless or * tief-tonig * ; 3. a third, which is called 
svarita, according to Whitney §81 is always of secondary origin, 
being the result of actual combination of an acute vowel and a 
following toneless vowel into one syllable. This is uniformly defined 
by the natives as compound in pitch, a union of higher and lower 
tone within the limits of a single syllable. It is thus identical, as 
far as can be seen, with the Greek and Latin circumflex, and in all 
probability goes back with the circumflex to the common I. E. 
period, as e. g. in the case voc. Zcv : dydhs = nom. Zcut : dydtis^ 
discussed on p. 40, note 2. 

So far ever}'thing is in reasonable accord with Greek notions of 
accent But there is a further element * The Hindu grammarians 
agree in declaring the (naturally toneless) syllable following an 
acute, whether in the same or in another word, to be svarita or 
circumflex, unless indeed it be itself followed by an acute or cir- 
cumflex, in which case it retains its grave tone. This is called by 
European scholars the enclitic or dependent circumflex,' Whitney, 
§85. Misteli and Hadley then impugn the statement of the native 
grammarians that this was a circumflex, and regard it as incompa- 
rably more probable that this svarita is a middle-tone. And 
Whitney, who is the first authority in matters of native Vedic 
grammar, says (§85) * This seems to mean that the voice, which is 
borne up at the higher pitch to the end of the acute syllable, does 
not ordinarily drop to grave pitch by an instantaneous movement, 
but descends by a more or less perceptible slide in the course of 
the following syllable. No Hindu authority suggests the theory 
of a middle or intermediate tone for the enclitic, any more than for 
the independent circumflex. For the most part, the two are iden- 
tified with one another in treatment and designation.* Whitney's 
opinion with regard to the enclitic svarita, while it denies it the 
name of middle-tone, does, we can see, nevertheless support a kind 
of tone which does not lie very far removed in its nature from that 
middle-tone in favor of which Misteli and Hadley argue. 

But on the other hand the testimony for a middle-tone in Greek 
which attaches itself immediately to the summit-tone in the manner 
of the enclitic svarita is extremely weak, in fact may be said not 
to exist at all. Not that there is not mention made by the ancients 
of other accents than the three familiar ones. Aristotle, Poetica, 
ch. 20, and Rhet 3, i, 4 mentions a fiiaov in addition to the o^un/f 
and ^pvrrfs of Plato, and this, according to Misteli, p. 44, note, and 
Hadley, Cu. Stud. V 417, is probably a middle-tone, though both 



46 AMERICAN JOURNAL OF PHILOLOGY. 

admit the possibility that the circumflex is indicated by it The 
Greek grammarian, Tyrannic from Amisus, who was captured by 
Lucullus and brought to Rome, reports four accents according to 
Varro (in Servius de accentibus, cf. A. Wilmans de M. Terenti 
Varronis libris grammaticis, p. 187). Varro mentions other Greek 
grammarians who report more than three accents ; there are in feet 
those who report six accents altogether. Misteli seeks further (§7, 
p. 50) to fasten this middle-accent immediately after the summit- 
tone, in a manner parallel with the enclitic svarita^ by the aid of a 
well-known passage of Dionysius of Halicamassus de comp. verbo- 
rum liber, section XI, but in this attempt he positively faUs. The 

passage reads dioXt «crov lUv otv fU\os 4v\ fterptirai duurnffian r^ Xeyofupi^ 

dw^ ntirrtt u €. the two limits of tone in spoken speech (between 
summit-tone and low-tone) are said to be a fifth. Now Misteli 
argues that this interval must have been mediated by the middle- 
tone in passing from an accented syllable to an unaccented one, 
because the unmediated skip of the voice through a fifth would 
give to the language * einen schneidenden und widerwartigen 
character,* and because Greek * speech would move in extremes ' in 
such a case. But as Masing, loc, cit p. 23, points out, another 
passage in the same author makes this construction impossible. 
For Dionysius continues, not many lines beyond this passage, with 
the antithesis to the ftAo^ duiXc«erov in the following manner: ^ dc 

opyapiKri re kqI <^diKrf fiovtra ^laar^fiatri r€ XPV^'^ yrXciootv, ov diA frcWe i^SvWy 
dXX' carb Tov hih. natr&p dp^fUvrif Koi t6 dih nirrt /icX^ci, Koi t6 dia Tiaau' 

pcfty, <c. T. X. * Music, however, instrumental as well as vocal, employs 
several intervals ; not only fifths, but, to begin with octaves, next 
fifths, fourths, etc* It is evident from this passage that Dionysius 
recognizes a plurality of intervals only for music and not for common 
speech, and it appears that according to this author there is but 
one interval, the fifth, in use in speech. 

Moreover, this passage by no means certainly describes word- 
accent ; so e, g. Gottling, who by the way denies that Greek word- 
accent was musical at all, construes this diakticrov iamKos as a rhetori- 
cal sentence-accent Certainly it cannot be brought in as testimony 
in favor of that special kind of middle-tone which follows every 
summit-accent Hadley does indeed recognize that the testimony 
of the ancients for it, or for that matter any middle-tone, leaves 
much to be desired ; but argues that the peculiar effectiveness of 
it in the theory which he defends and extends is the surer testimony 
in fiivor of its actual existence. 



STUDY OF CREEK ACCENT, 47 

The theory is then completed by the following assumption, which 
is to account for both Greek and Latin accentuation : There was 
developed in the Graeco-Italic division of the family ^ after they 
had separated from the common stocky a disinclination to allow 
more than one toneless syllable to follow upon the middle-tone ; 
this disinclination caused a moving forward of the summit-accent 
to such a position that there was room after it, and after the middle- 
tone which necessarily followed it, for only one toneless syllable. 
Thus originated the Graeco-Italic law by which the summit-accent 
is restricted to one of the last three syllables of a word. The im- 
mediate ancestors of the Greeks and Romans, the * Graeco-Italians,' 
before their separation from one another, accented their words 
alike according to this simple law, e. g. ^(KunofuiVf *SpBpo»7roVf *gaii' 
dcreSy ^Ugendus^ i. e, all words which originally had the summit- 
tone before the antepenult simply shifted it to the antepenult, thus 
producing a very special cadence agreeable to the Graeco-Italic 
ear, summit-tone^ middle-tone^ tonelessness (low tone). In words 
which did not have the tone anterior to the antepenult, words like 
\9\v\thK^, xaKctt6iy the accent remained undisturbed ; for here there 
was no room for the violation of the law that the middle-tone should 
not be followed by more than one toneless syllable. But as Greeks 
and Italians divided off they developed their common three-syllable 
tone-law in a manner which led to pretty sharp differences. The 
point of departure from the Graeco-Italic law was the toneless 
syllable in the cadence for the Greeks, the middle-tone for the 
Italians. 

Let us first remain a while with the Greeks. They developed 
a dislike for a long toneless, u e. final, syllable, so that the Graeco- 
Italic cadence of summit-tone, middle-tone, toneless syllable, was 
modified for the Greek into summit-tone, middle-tone, and short 
toneless syllable, whenever the accent had originally, in I. E. times, 
stood before the antepenult. In order to exhibit the application 
of this law, Hadley divides the phenomena of the Greek recessive 
accent into four divisions, and one need but remember in addition 
that he regards the circumflex as a compound accent containmg 
both summit and middle-tone, in order to understand his reasoning. 

1. The simplest case. The acute cannot stand on any syllable 
before the antepenult, therefore I. E. ^IXciircro becomes Greek 

AftVcTO. 

2. The antepenult must, if it takes the accent, take the acute ; 
^Acttrcro (i. e. *cXtlir*ro) is impossible, because it leaves two toneless 
syllables at the end. 



48 AMERICAN JOURNAL OF PHILOLOGY. 

3. When the penultimate carries the accent and the ultimate 
contains a long vowel, then this must be the acute, roiavn;, not Totavny 
(= *Tot4vn7), because this would result in a long toneless syllable. 

4. A long vowel in the penultimate must take the circumflex if 
the ultimate is short, towxtto^^ not *T04ovrof , because there would be 
no room for the toneless syllable.* 

This method of accentuation in the separate life of the Greek 
also did not gain ground when it was necessary to draw the sum- 
mit-tone back from the end in order to gain the desired cadence. 
Therefore types like XcXvficW, XtTrwv, remained undisturbed.' 

Only one division of the Greek people, the Aeolians of Asia 
Minor, took ako this step completely, that is they subjected their 
entire accentuation to the law of cadence, summit-tone, middle-tone, 
low tone, therefore XcXv^evor, x^^cVwr. Where the entire cadence 
was not to be procured, as in o-ck/kJ?, they drew the accent back at 
least as far as possible, ^({(^or. 

The theory then proceeds to explain the Latin accent by assum- 
ing that the Graeco-Italic cadence-accentuation there also received 
a modification, namely, that there developed with the Italians a 
disinclination against a long middle-tone, so that the Latin cadence 
became summit-tone, short middle-tone, low tone. We will return 
to the Latin further on and see that this theory accounts for the 
Latin system about as well as for the Greek. At present the Greek 
will be dealt with alone. 

I. In the first place it has been shown that the assumption of 
this middle-tone following every summit-tone is a purely theoretic 
one, and that the testimony of the grammarians in favor of such a 
middle-tone amounts to nothing at all. Not that it is to be supposed 
that the Greek word did not possess subsidiary tones just as much 
as words of to-day ; but the assumption of a special middle-tone 
which must follow the summit, implying that the pitch of the sum- 
mit was especially high, so as to stand in need of a mediator between 
it and the low tone, is warranted by no fact of Greek grammar or 

* This ii the weak spot in the arrangement. The theory by which the expla- 
nation of the Greek accent is here attempted does not in reality claim that the 
cadence, summit-tone^ middU-tom^ low tone^ must be established in every case ; 
it makes only the negative claim that after summit-tone and middle-tone no 
more than one low tone should follow. This condition would be satisfied as 
well by ^romfrof as by toiovt(^, 

• This rule knows exceptions from the earliest times. So e, g. nouns in -t/^ 
("^^'f)* A^'^^'C, ^^<JtQ, are originally oxytone formations, Sk. srutis, citis^ and yet 
appear in all periods of the language with recessive accent, cf. below, p. 50. 



STUDY OF GREEK ACCENT, 49 

tradition. The passage of Dionysius not only proves nothing, but 
if it speaks of word-accent at all, disproves the existence of any 
interval in the hiakittrov fuKoSf except the fifth. 

2. The assumption of a Graeco-Italian accentuation (^KtivofuiVf 
Ugendus) stands entirely in the air. Not one historical fact is in 
its &vor ; it is solely based upon the fact of the restriction of the 
accent to the last three syllables. At the time when Mistdi and 
Hadley wrote, the assumption of a Graeco-Italic period was very 
generally, though even then not universally, accepted. It is to-day 
a theory of the past. In just that particular factor of form which 
stands in especially close relation to accent, namely, vocalism, these 
families are about as far removed from one another as possible. 
Further, it will be urged below that the Greek recessive, or, to speak 
with Hadley, cadence-accent, began with the verb ; it is precisely 
in the verb that Greek and Latin have diverged so extensively 
that mere fragments of the older system of formations are left in 
the latter, and it is altogether improbable that the Latin shquld 
have saved an old system of verbal accentuation for a new and 
obscure set of formations. 

3. The assumption of the sequence, summit-tone, middle-tone, 
and short toneless syllable, is after all nothing more than the for- 
mulation into a more complicated shape of the simple law that the 
recessive accent does not recede beyond three, or in one case (forms 
like aCwrm and fjirttpos) four morae. The theory does not find it 
possible to free itself from the count by mora any more than the 
formulation by which the accent was described above. While it 
appears to dispose of the case of a^axrrof better (for here it was 
necessary above to assume recession to four morae), it is deficient 
in cases like roiovrof , because it does not account for the constant 
circumflex, c£ p. 48, note i, which on the other hand is accounted 
for perfectly within the theory of the three morae. 

4. Finally, the last objection is one which more than any other 
undermines the middle-tone theory. The original I. E. succession 
of summit-tone, middle-tone, low tone, it is claimed was in Graeco- 
Italian times moved down a place or two or even more in order to 
pander to a dislike on the part of th^ Graeco- Italians to allow more 
than one toneless syllable after the middle-tone. An aesthetic 
dislike which is powerful enough to reform the accent of an entire 
language in a thoughtful, laborious manner, is a sufficiently doubtfiil 
factor in modem linguistic explanation. It cannot exactly be called 
a phonetic law, because a phonetic law acts spontaneously, and 



50 AMERICAN JOURNAL OF PHILOLOGY, 

would not be likdy to oount the sySaUes of a certain word, and 
dien, upon findii^ diat the summit-tone upon a certain syllable 
would leave too many toneless syllables at the end, move it down 
a sufficient number of morae to ward off such an event At least 
so complicated a process must seem highly improbable when it is 
compared with the workii^ of such a law in other quarters. Yet 
the explanation as a phonetic law might, for lack of a better one, 
be accepted with reserve, but for the &ct that the theory &ils to 
account for a strictly grammatical, and not aesthetic, &ct connected 
with it ; namely this, that the recessive accent has most certainly 
in Greek b^^n with the finite verb, where there is prtuHcally no 
exception to it; that it excludes, with particular care, non-finite 
forms of the verb in the same tense-S)rstem and in evident connec- 
tion with finite forms, exhibiting thus on Greek ground a most 
outspoken character as a grammatical quality of finite verbs : tkataiF, 
^XMr6firjp, Xlwut etc., but Xtirdnr, Xtirciy, XurcV^, etc Of course noun- 
formations are not spared in historical times. But here the tendency 
is not regulated by any traceable law. Certain noun-categories 
become recessive ; others, with apparendy the same claim to &vor, 
do not ; so adjectives in -w versus nouns in -t« (<nr).* It is in feet 
perfectly clear that the recessive accent in Greek, whatever its 
explanation, started with the finite forms of the verb, and thence 
succeeded in attacking nominal formations also ; it cannot, therefore, 
have been due to the disinclination of the Graeco-Italians to allow 
two toneless syllables after the middle-tone. Such a cause cannot 
have differentiated between noun and verb. 



V. 

The strength of Misteli's system as completed by Hadley seems 
at first sight to lie in the fact that it includes the Latin, which shares 
with the Greek the sufficientiy remarkable quality of restricting the 
summit-tone to the last three syllables of a word. This coincidence 
Hadley explains by the assumption of a Graeco-Italic accent which 
knew no restriction except this, that the assumed I. E. cadence 
of summit-tone, middle-tone and low tone, when it began before 

^ Both are originally oxytone noun-formations ; the adjectiTes in -(>c hare 
remained so, Opaaix = Sk. dhx^, ppadi^c = Sk. mfdtis, irXari^ =■ Sk. fTtkds, 
kXaxi^ = Sk. ragktlsy ^^X^ = Sk. bahtls, ^i% =: Sk. gunls, etc^ the nouns in 
-TIC have without exception become recessive, as in the cases of pifot^ and rlat^^ 
cited above, p. 48, note a. 



STUDY OF GREEK ACCENT. 5 1 

the antepenult, was moved down to avoid more than one low tone 
at the end of the word. After the separation of the Greelcs from 
the Italians, the two peoples refined the common Graeco-Italic 
accent; the Greek by insisting upon summit-tone, middle-tone, 
short low tone, the Lat. by developing a fondness for summit-tone, 
short middle-Umey and indifferent low-tone. Accordingly the 
Graeco-Italic accentuation, which still permitted forms like l^gen^ 
dus, gatidires, etc., was modified ; and this modification s^ain 
becomes at least superficially easy if the definition and description 
of the Latin circumflex, as given by the Latin grammarians, is 
remembered, cf. Corssen, Ueber Aussprache, Vocalismus und Be- 
tonung der lateinischen Sprache, II, p. 800 fg. According to them 
the Lat circumflex was employed upon long monosyllables (ex- 
cepting ne with the imperative), and on penultimas with long vowels 
(not, however, by position) when the ultimate was short Every- 
where else the acute was employed according to the remaining 
well-known rules. How much value is to be attached to the state- 
ment that in Latm gauderi had the circumflex, made as that state- 
ment is by grammarians who were under the influence of Greek 
grammar down to the minutest particulars, is after all an open 
question ; even Ciutius, a strong supporter of the Graeco-Italic 
accentuation, has said in my hearing that *' der Circumflex im 
lateinischen bedeutet iiberhaupt nicht viel, ist mehr auf Theorie 
g^^ndet" ' 

But the assumption of the existence of the circumflex, and the 
cadence projected for the Latin, summit-tone, short middle-tone, 
and low-tone, seemingly procure a satisfactory arrangement of the 
historical phenomena. 

The simplest case is that of types like Ugiri and Ugiret ; here 
the cadence, summit- tone, short middle-tone, and low-tone, is easily 
procured. In the type gaudire^ the same result is procured by 
dividing the circumflexed e between summit-tone and middle-tone, 
quasi '*gaudiiri. Greater is the difficulty in the type gaudiris^ 
for the first e is not circumflexed, therefore the syllable res must 
furnish the place for both middle-tone and low-tone, ^gaudiriis ; 
but who will after all believe that there was so thoroughgoing a 

' Petnis Lange is the strongest assailant of the Latin circumflex, in three 
treatises : De giammaticorum latinorum praeceptis quae ad accentum spectant, 
Bonn, 1857; in a critique of Weil and BenIoew*s Th^rie generaleide Taccen- 
tiuUion Utine, in Fleckeisen*s JahrbOcher, Vol. 79, 1859, P* 44~7i \ Untersuch- 
nngen Qber den lateinischen Accent, in Philologus 31, p. 98-121. 



52 AMERICAN JOURNAL OF PHILOLOGY, 

difference in the accentuation of the two words gaudiri and gau- 
deris, or upon what tangible fact in the life of the language is this 
differentiation based ? And in the type legindHs we are left without 
a place for the low-tone, because gen cannot take the circumflex, 
^UgimdlSy while the type legindi again divides its final long syllable 
between middle and low-tone, *Ugindli. Here the arrangement 
is weakest ; it institutes a complicated difference between the accent 
K^gaudiri {gaud^kri) and UgindHs {legindus =.), which is devoid 
of all foundation in the actual and not hypothetical life and history 
of the language. 

Of the four main objections which were urged above against this 
theory when applied to the Greek, three hold good against Latin 
also ; others can be added from the point of view of the Latin itseUl 

1. The still more complete absence of testimony in fevor of a 
middle-tone which regularly followed the summit-tone. There is 
no such testimony at all to be obtained from the Latin. 

2. The assumption of the Graeco-Italic accent, against which 
what was said above, p. 49, is to be compared. 

3. The combination with Greek recessive accent, which has origi- 
nated with the verb, and will be shown below to be due to an L E. 
law pertaining to the verb, which therefore separates that method 
of accentuation incontrovertibly fit>m the Latin, where the special 
influence of the verb is not to be thought of, and has not, as ias 
as is known, ever been suggested. 

4. The very similarity of the Latin accent to the Greek becomes, 
if we look more narrowly, reduced to the restriction of the tone to 
the last three syllables. In every other respect the accentuations 
of the two languages stand in the sharpest opposition to one another. 

a. In Greek the summit-tone is not excluded fix)m the last 
syllable, in Latin it is so entirely. 

h. In Greek the penult is absolutely without influence as &r as 
deciding the position of the summit-tone is concerned ; in Latin 
the penult is the pivot around which everything revolves, its quan- 
tity decides the position of the accent. 

c. Just as indifferent as the penult is in Greek, so in Latin the 
ultima has no mfluence upon the position of the accent, while in 
Greek it is the main factor in determining the position of the reces- 
sive accent 

5. A fifth reason agamst the assumption of the Graeco-Italic 
accent is presented by the fact that there are distinct traces in Latin 
of an accentuation which was not restricted to the last three syllables 



STUDY OF GREEK ACCENT. 53 

The law of three syllables was preceded in an archaic period by a 
freer accentuation, the vestiges of which are not sufficiently numer- 
ous to make it possible to describe its exact character, though 
enough can be seen to render it probable that it did not know this 
restriction, at least not in the form of an inviolable law. 

a. Very strong indications of a different regime in matters of 
accentuation are contained in the vowel changes which attend 
reduplication and composition. The reduplication and prepositional 
prefixes in Latin exercise an influence upon the vocalism of Latin 
roots which would remain unexplained, unless it be assumed that 
they once regularly received the accent. Thus, ^\itxijuro becomes 
in com^ositiovi pi'jiro,f(uio becomes ^cdn-Jicio^ gndius (with very 
old vocalism = Greek yyam^r = S\i,jildids) becomes cd-gnitus ; it 
is necessary to assume that the accent stood originally upon the 
preposition at a time when the root-vowel was not as yet weakened 
(^pi'juro^ ^Cihgndhis), and therefore accented in a manner 
thoroughly different from the laws of accent in historical times ; 
for it would be incredible that this weakening of the root-vowel 
should take place under the summit-tone i^pe-jirOy etc). This 
accentuation of the preposition with the finite forms of the verb 
inclining upon them is Indo-European, and at any rate an accen- 
tual condition which must be admitted for the Latin at some remote 
period. On the same principle con-ficio must have originated fi-om 
a prehistoric ^c&n-faciOy with the accent on a syllable anterior to 
the antepenult And, further, in the perfects, ieiigi, pepigi, cecini, 
fefelliy cecidi (: cado)^ cecidi (: caedd), the weakening of the root- 
vowels is due to the accentuation of the reduplicating syllable ; 
this leads to forms like ^Utigimus^ etc, which again have the tone 
before the antepenult. Moreover, certain Italian forms not Latin 
support this view. E, g, the Oscan forms fe-fdc-id (perfect opta- 
tive third singular), or fe-fdc-ust (future perfect third singular), 
when compared with Latin con-fic-io^ or with an ideal reduplicated 
^/i'Jic'i fi-om ^f^-fdc-i^ show that this regular weakening of the 
root-syllables is a special Latin phenomenon; so also Umbrian 
JupaUr is probably the common Italian predecessor of Latin Jupi- 
ter, If this weakening of the vowels, as would appear from such 
examples, is not common to all Italian dialects, but belongs espe- 
cially to the Latin branch, and if it is assumed correctly that these 
weakenings would be impossible under an accent Vikt^/efdcust, 
fefdcidy we have an historical corroboration in actually occurring 
Italian words of the assumption that the three-syllable accent is a 



or rHJ^01,DGY. 

cvcx iair of the ^ -»*""* 
r:rti X jnt wnhm tut last liiree syDahleSb 
in A.jiitr annnjii tc> cstabl^i LnKJiip 
. *3r a: tilt twc« ian^ruacres? Therefore 
rir j: Tm- Lare^ and I jrrm sv^bseds of 
r^s wiAC£st pQssiDit gronnd, and an 
.?255:" t armrr wiiich irrnares tiie exter- 
u:t nii«r bt TOpmarfaisd with reascmable 



TL 

:r>= Zin?ek rerrssrvt accent must start ^txn 

•= 7--T wiiere aicme it is evidenthr at home. 

> .^jr^-tiSL TiMr T'erb and tiie accent is not iKidoed 

, ^ - _ z itnd has been pointed out abcn'e as the 

. . -c .: < icnii. Ye: the ma had been nodoed and 

-. .: . rrrr:: rx^ec br G^iTTlrng. niio puts the verb oa 

, . : , r: Atr.cir accem in liiis respect. It is Wadc- 

. ; .. > *,>r rt-asar of his success that he began his 

. „: ." > sas.T as the basis. And he has sncoeeded, as 

. X.V*.-., :* fVTV;riTnrng the Greek accent as ^ as the 

. . -vv-. :> » A 5«r»es of quaZities or laws of tneatrocnt to 

y . ^. ^' :^ • J6> 3^ :y>ecicd in srcimre-n^jms in L El. times, 

V , -V V <v^:>^^.^e accent apf>eais to be a de^^dopinciit of 

s .^ ^ ?v^ .vrtrr^v-acoent in distinction from word-- 

, ..>. i,^c!? the start let the etymolo^cal accent of 

,,.« >r iirvst, or better, keep in mind that die ety- 

^ . ^ wv , V * *vx-d may under certain drcumstaiKes vanish 

^ ^^v.vv ,^: 3<ra:fnce-accentuation. 

^,v x..^ js %-:h the observ-ation that both in Greek and 
\ K .:v xt^b is occasionally subjected to endisis, of 

V V vvx;v>t possible differences in other respects. In 
. . .;v v^cb becomes enclitic under certain conditions 

^ V ,vv v.v :i laws (see Whitney, Sk. Gramm. §592 %.)• 

, \ >i c \ vi bci in the present indicative, cuu and ^/u, 

V Av. ^explanation, according to which this enclisis 

V , vvv. >^ meaning, he rejects justly, because ^/u is 
^ . . , He assumes, then, that this restriction of the 

^ , ^^.. ^lu^i tnic that 4»HMI is, and continues to be, the 

,. vv N. vA,\ »u^. often meaning * aver,* * asseverate,' and some- 

wx t A verb of swearing. Xkyo in Homer is not yet a 



STUDY OF GREEK ACCENT, 57 

enclisis to these two indicatives is due to the Greek law of enclisis, 
according to which an enclitic word may not contain more than 
two syllables and three morae. This, it will be remembered, was 
exhibited in detail above, p. 42, where the examples Zcw ^/ui^, etc., 
with enclisis of the orthotone ^/iw, was shown to be the substitute 
of the enclisis which is exhibited in Zcw /km. Of course these are 
not the only individual Greek finite verbal forms which, in spite of 
this restriction to three morae, could be inclined, but here Wack- 
emagel recognizes with consummate acuteness that the present 
indicatives of these roots represent the only cases in the language 
where the entire paradigm of the tense or mood would allow the 
enclisis throughout. A form like Xf'yw, ttcI^c, ^itrav would by itself 
be capable of enclisis, but not Xryofw, Xtycrc, TrcidcTc, H<rfi\v ; therefore 
enclisis could not sustain itself in the paradigms to which these 
words belong; on the other hand, the undisturbed capacity for 

enclisis of e//u, (e?), core, eorroi', tafievt core, tiai ; </>i7/u, (<f>jjs)t^ 4^^^i ^tov^ 

4>afuVi ^are, (ftaai, without a single interloper that would be debarred 
from enclisis by containing too many morae, is the secret of the 
preservation of their enclisis. The test for other tense or mode- 
systems is easily made and will always bring up some form con- 
taining either more than two syllables or three morae. The enclisis 
of these two present indicatives is then identical with the enclisis 

in Zfvs fiou 

The question now arises : What has happened to the other verbs 
which were debarred from enclisis by containing too many morae ? 
Precisely the same treatment that has happened to an enclitic pro- 
noun of too many morae. They were inclined as much as possible, 
in accordance with the principle exhibited in the change of ortho- 
tone 17/iiv to enclitic ^/uv, and orthotone Jifiav to enclitic ijiuov ; just as 
Zcv£ fjiuAv contains orthotone riyMv changed to rjyn.v, just so does Zevr 
doii; contain the prehistoric boif) = Sk. deyat; however, not in its 
orthotone, but in its enclitic form, for hoii\ is the enclitic to *^o4^ just 
as much as t^iuav is the enclitic of t\\imv. This may be formulated 
in the following proportion : 

The recession of the Greek accent in the finite verb is accordingly 
everywhere not due to a process of accentual change within the 
word, but to a secondary accentuation which is a substitute for 
enclisis. It is false, therefore, to compare directiy the accent of 

* BI and ^c will be discussed further on. 



t. Tfaos, 



xwwludi 
S591 fe- 
: cImM t es , 



tire finite 

ded «sa 
aiticiples 
ect r^u- 
ve there- 
■ed, even 
r rdation 
« accent. 
•r, riedn 
^wt; so 
diematic 



inner the 

[plained. 

(written 

on the 

theSan*- 

irognupme 
le ea tilled 

poD every 



13s, note), 
inking his 



STUDY OF GREEK ACCENT, 59 

Other hand, here again the forms which are not enclitic when un- 
compounded retain their accent, and the preposition loses its accent 
both in Greek and Sanskrit, irfroXa^a»v, rVcttv ; in the same manner 

KoBrjTai and Kordicfirait but KaBrj(rBaif KoraKtiarBaif cf. Whitney, Sk. 

Gramm. §1083. 

Wackemag^el turns next to the second persons c? and <t>ns, which 
are orthotone, and would endanger his entire explanation unless 
their orthotonesis is explained. The explanation which is proposed 
is a totally different one in each case. 

For ft an etymological explanation is attempted. This word is 
Attic and Ionic, but post- Homeric ; it is a form, then, which is 
later than the period in which the enclisis of the verb was fixed. 
Possibly it may be restricted even to Attic alone, inasmuch as it 
has been removed by Stein from Herodotus.* In order to explain 
this late and contracted eZ, Wackernagel assumes that it is a middle 
form *t(r€(rai to ^o-o^im. Such a word, containing as it does three 
syllables, would, owing to the limitations of enclisis, not become 
toneless, but would appear with recessive accent as a substitute for 
enclisb in the usual way, and this *l^(<r)f(<r)ai, *Uai would then con- 
tract to f^ as *irou(_y)e(o')a», iro»«eai becomes wotci. But there are at 
least two objections to this explanation. First, the natural expla- 
nation of c?, which seems to be almost unimpugnable, is a totally 
different one. The word, whether restricted to Attic or not, is 
evidendy old ; it is *f(n = Sk. dsi = Zd. ahi = Goth, is = Lithu- 
anian est and Old Bulgarian j-esi ;* the assumption of a ground- 
form *f<rf<ra» is therefore unnecessary and improbable. Secondly, 
Wackernagel has assumed with indubitable success that within one 
tense-system, forms which by themselves could have been enclitic 
became recessive by the attraction of the rest of the system ; why 
has not the analogy of the enclitics in the paradigm of c/^ succeeded 
in overcoming this single recessive example in its turn ? It seems 
therefore much more probable that the lack of enclisis in cf is due 
to the influence of contracted forms in general. At the time when 
*/((r)i contracted to c?, other contractions taking place at the same 

'According to Veitch, Greek Verbs Irregular and Defectiye, Stein and Abicht 
read elCt while Becker and Dindorf read el. 

* The In^P- European form of the second person singular was *esi^ e, g, the 
two /s coming together from the root es plus the -si of the second person singular 
were simplified into a single s by some I. E. law of sound, before the separate 
existence of the languages of the family. Neither in Greek nor in Sanskrit 
would the theoretical */m lose one of its s\. For the Sanskrit, see Whitney, 
Sk. Gramm. gi66. 













a^ct^-r-y.^a^ >^ ' / -v *r.' T liT "In? ^i.sr * iLmsil" Hni 3i:t idir 



U:t^t^ ft^^j. •f^ 4 •//*>' />f p'-rf^f/i ratb^ ba-.-e dser carjral ety- 
H,>yi/vt^u' -4 f/y<#^, J*4 ,//.♦/ r»^/J ^.\\ fjrtrjsr by the rhetorical tone, 
fj>ii.if**4 »## /|iM^^///r*, ^<, Whitf»*-y in Kuhn and Schleicher s Beitrage, 
I pMj. (t i^ / li'4/ tfi^t tft^ criticism made above against the 
ai,fi>ij^/i(/fj^/i* f)i4l i<o ol/l ^ittttftu which functionally was not different 
Iruni /</^ e»l*oM|/J fMimiM ^^Uu/t/mc and resist the analogy of die 
rfttt mI the Uiww^ tti jIm' |mni(Ji>i(rn d(Kr» not hold good here, because 
thcicr m a ihunmy,\i^n\im funrtional difference in ^^ which might 
wril hoUl it aliovt' itiei Uitt vn of aimimilation; especially true might 
thiti hf U\ the t»*b»M»l ihr |»r(Miliar interrogative tone, which this 
word i« ttuhjtH tt^tl III wllh rftpm iul frr(|ucncy. In the later literature, 
AN rrpit^ht^iUativtb ill whh h W»u'krnm^cl brings on Plato and the 
CiMUoihaUH, thia tuiiHUMur nl \)\f\v In inlorrogiitive and subordinate 
cluwrs* ij» not tiii htiiillv mlhrHnl to; hr tinds in 140 passages 18 
nut intcno^ativt^rtmt \\\\\ n\\\\\s\\\i\u\Wsh\\X these passages are made 
to y u'kl strtM\^ ^uppoi I tu the v>^^ u^ t»\exV< of his method in bringing 
on the Veiiic uocent tor auv^t>^\\t vau^^jvu\!*vm\ ; they also are explained 



STUDY OF GREEK ACCENT. 6 1 

by Vedic analogies. The word ^r occurs in these i8 passages in 
the first one of two paratactic clauses, e, ^, Plato Gorg. 491 B, o^ 
\uv yap 027^ . . . rya» dc aov rovvavriop. Compare with this Whitney, 
§596 : * The verb of a prior (principal) clause is not infi-equently 
accented in antithetical construction. Sometimes the relation of 
the two clauses is readily capable of being regarded as that of 
protasis and apodosis ; but often such a relation is very indistinct** 
Of course the Greek example comes under the head of antithetical 
construction ; in the same manner the other 17 examples of Plato, 
etc., are readily disposed of. It seems that Wackemagel has 
beyond peradventure pointed out the correct reason for the pecu- 
liarly isolated position of the word ^'^ in accordance with the rules 
of Vedic and Indo-European accentuation. 

He turns further to various minor specialties of the recessive 
and enclitic accent, and explains them again in accordance with 
well-known laws of Vedic accent. Only the most interesting of 
these, the orthotonesis of Han, will be mentioned. The older Greek 
grammarians, according to Lehrs, Quaestiones epicae, p. 126, know 
of no functional difference between ?<m and -^ fOPr», but teach that 
the orthotone word stands at the beginning of the sentence and 
where certain particles, etc., immediately precede the word. Ac- 
cording to some, only ol has this effect ; according to others ov, koI 
and its ; c^ dXXdf and rovro are also added by a few. With the 
exception of tovto these words are either too weak to allow inclina- 
tion upon them, or, like kcU, are not real members of the sentence 
which they introduce, so that the tlcm which follows stands in reality 
at the beginning of the sentence. This peculiarity is again explained 
by a rule in Whitney's Sk. Grammar, §593, 'The verb of a 
principal clause is accented when it stands at the beginning of the 
clause,' e. j^, syima id {ndrasya gdrmar^tf * may we be in Indra's 
protection.* 

Other details of Greek accentuation, which need not be repeated 
here, are successfully explained, and everywhere Wackernagel's 
results are strictly in accordance with the principles which have 
been stated above for all kinds of phonetic investigation, and they 
are themselves new proof of the success of such investigations when 
carried on with these principles. In the first place every line of 
his investigation is permeated with the thought that it is not allow- 
able to discuss the accent of the separate I. E. language without 

*£, ^. ^d'prd^nyi ydnti^ pary anyddsaU: 'some go on and on, others sit 
about.' 



62 AMERICAN JOURNAL OF PHILOLOGY, 

taking for a basis the reconstructed I. E. accent. Further, this 
I. E. accent could only change by regular phonetic law or by 
analogy. Both factors are shown to have been at work. The 
phonetic law is the Greek law of enclisis by which real historical 
enclitics appear accented, though in manner clearly enough a mere 
compensation for enclisis; the reason for this phonetic law lies 
within the province of phonetics just as, e. g. the rhotacism which 
changes in so many languages an ^ to an r. 

The workings of analogy we saw in many ways ; above all this, 
that the enclitic character of the verb in principal clauses has been 
extended to the verb in subordinate clauses. It would be interest- 
ing in this connection to count the number of principal and 
subordinate clauses in Homer; no doubt the principal clauses 
would preponderate, as they most certainly do in the Rig- Veda. 
Wackernagel is the first one who has clearly established any kind 
of law as regards the sentence-accent of the I. E. languages, the 
leading &ct being the enclisb of the finite verb in principal clauses. 
His results prove completely the feet that the study of accent 
cannot be carried on from the point of view of the word alone, but 
that it must also consider the larger speech unit, the sentence, and 
perhaps ultimately also the smaller, the syllable. 

Wackernagel does not carry his results beyond the finite verb, 

but he leaves no room for doubt that the nominal accent in Greek, 

so fer as it is not archaic and etymological, is enclitic and recessive. 

No doubt the noun has to a large extent followed the verb in its 

enclisis ; the Vedic accent leaves us here almost entirely, but yet not 

altogether. In the Veda the vocatives are accented only when they 

stand at the beginning of a sentence, or clause, or verse, elsewhere 

they also are enclitic ; see Whitney, Sk. Gramm. §314, and Haskell 

in the Journal of the American Oriental Society, Vol. XI, p. 57 %. 

Further, an adjective or genitive qualifying a noun in the vocative 

constitutes as fer as accent is concerned a unity with it. Thus there 

arises in the case of a vocative in the middle of a clause a group 

of two or three, sometimes even more, unaccented nouns, cC above, 

p. 31, note 2. The quantity of enclitic vocative material cannot 

have been very great at any period in any language of the femily, 

yet it may have at least helped on the analogy of the verb in its 

inroads upon the noun. Possibly fiiture investigations may 

succeed in pointing out the details of this process in an acceptable 

manner. 

Maurice Bloomfield. 



III.— ETYMOLOGICAL STUDIES. 

II. 
Liceo^ liceor. 

These words are generally brought together on the supposition 
that the first is used as the passive of the second. This supposi- 
tion I hope to show is mistaken. 

To b^^ with, it is most improbable that the relations between 
active and passive forms should be so entirely reversed and the 
consciousness of those relations so entirely destroyed that the 
same word should exhibit the active relation expressed by the/oj- 
xn^^ form 2xA ^<t passive relation expressed by the active form. 
It has not been observed in discussing these words that, where the 
same voice has been adopted to express both the active and the 
passive side of an action, the verbs so used either come fi'om dif- 
ferent roots or else are differentiated in form. Thus we have : 

Active : hpa» or iroif ly / /SoXXciv ; Xc>«y / verberare. 

Passive : wdax^^ / ircirrciv y xXvciy, oKovtiv / vapulare. 

On the other hand, iacire, iacere; pendire^ pendere, &c., cf. 
Curt. Gr. Et* No. 625. 

This sXxoug prima/cuie probability against connecting these two 
words is strengthened by an examination of their usages. 

First, if we are to assume that these two words have reversed 
the active and passive functions in this most extraordinary manner, 
we ought at least to be able to find some traces of the reversal. 
If we cannot discover an active use of liceo^ the fi-equent use of de- 
ponent verbs in the passive at least entitles us to expect a passive 
employment of licear. Now (i) liceo is said to be used in an €u:tive 
sense. But of the instances that can be cited. Mart 6, 66, 4 rests 
on a sheer blunder, Diomedes 398, 25 is wholly indecisive, and 
the sole evidence remaining is Plin. N. H. 35, lo, 36, § 83, percon- 
tanti quantf liceret opera effecta parvum nescio quid dixerat, where 
Kcereni for liceret is an easy and probable correction, already pro- 
posed by Sillig.* Even if this doubtful sentence be admitted, it 
will be no evidence for writers of greater antiquity and fewer pecu- 

1 1 may add that I hare investigated the whole lexicology of liceo and liceor 
in the Jonrnal of Philology, (English), Vol. XI, p. 33a. 



64 AMERICAN JOURNAL OF PHILOLOGY. 

liarities than Pliny ; and the active use of liceo may with justice be 
regarded as a later development, owin^^ to liceo being thought a 
more rational form than Hceor to express */ put a price <m^ 
(2) Liceor^&neveryx&tAxTi^passive^^Tss^. In the second place there 
is a marked difference in the meaning : licere is * to be knocked 
dawn' the result of the bidding ; liceri is simply to bid, to offer. So 
long as liceri is going on, licere is impossible, so that the one can- 
not possibly be the active side of the other. To take two examples 
of licere. * Omnia venibunt quiqui licebunt^ Plant. Menaechm. 5, 
9> 97i is * everything wiU be sold to the highest bidder,' or * for 
what it will fetch.' So Cic. Att. 12, 23, 5, quanti licuisse tu 
scribis, (if not from licet) means * what they fetched' So even in 
Pliny /. c, quanti liceret is * what price he would put on them,' or, 
in other words, 'what was the final, the selling price,* not what 
he would Ind for them ; and Mart. 6, 66, 4, parvo cum pretio diu 
liceret, * when the price stuck for a long time at a trifle,' * when all 
he could get for her was a small price.' Licere in feet is used of 
theyJna/ offer that concludes the sale or bargaining ;^ liceri of any 
bid, as I need not adduce passages to show. 

What then are the two distinct roots from which liceo and Hceor 
come? 

Curtius /. c. has given that of liceo. He compares it with licet^ 
Greek XctVccy and Sanskrit rid. So that licet res tanti will mean 
* an article is left, the bidding leaves off 2X z. certain amount,' • 
tanti being a locative; see Roby, Lat. Grammar, Vol. II, § 1186, 
and compare the use of stare, constare. 

Corssen* supposes the root of liceor to be RIK, reach out It 
IS seen in por-ric-ere, etc., pol-liceri ; Old High Germ, reihhan, 
Goth, leihwan, O. H. G. lihan. Germ, leihen, Eng. lend. And 
an exammation of the original meaning of the German Helen, to 
bid, which was to hold out, as in beut den Finger, Keisersberg, 
inclines me to believe that this suggestion is probably the true one. 
The persistence of the middle form liceri in the sense * to reach 
out' is very noticeable. Compare in Latin polliceri, licitari, 

» In other words licere\% the result of the HcitaHo maxima. Suet. Cal. 22 : com- 
pare the passage quoted below from the Di£est. 

• The meaning and derivation of licere are well illustrated by PauL Dig. 10. 
3. 19. penes quem licitatio rtmansit Another conjecture may be haxarded! 
The personal use of Hctrt may be a development of the impersonal. The 
auctioneer may have said licet, • you can have it,» when he knocked it down • 
then the article iUelf was said Hcert, So pretty nearly Curtius Gr Et » ' 

» !• 500. 



ETYMOLOGICAL STUDIES, 65 

which seems generally to mean ' to reach or lunge with a weapon ' 
in fighting, and in Greek dpeyioBaij and for the usage dtgiio licerixh^ 

Homeric xnpafiv opi^aoBcLi} 

From the same root RIK come besides pol-lex the thumb as 
standmg out fi-om the rest of the hand ; and not improbably /^A 
Hng'O, to lay or stretch out a corpse for burial, and pol-linc-tor an 
undertaker. 

Trio^ Sepiemtriones. 

These words have been discussed by Prof. Max MttUer.' In 
his discussion there are some points to which exception must 
be taken. 

The first of these is his summary treatment of Varro's authority. 
Varro says ' triones enim boves appellantur a bubulcis etiamnunc, 
maxime quom arant,' etc. On this passage Prof. Max M tiller 
observes : "As a matter of feet trio is never used in this sense 
except once by Varro for the purpose of an etymology " — this is 
a gratuitous insinuation — " nor are the seven stars ever spoken of 
elsewhere as the seven oxen, but only as the oxen and shaft — 
boves et tetno — a much more appropriate name." It is not likely 
that any one will follow Max Mailer in attributing fraud to Varro, 
and in a matter like this it is impossible that he should be mistaken^ 
especially when he speaks with so much circumstantiality — a 
bubulcis — etiamnunc — maxime quom arant terram. Max M Uller 's 
reasons are of the lightest. His argument from the feet oiirio not 
occurring in this sense elsewhere would put in jeopardy all Sma^ 
\€y6fupa ; and his appeal to ' appropriateness ' is not more convin- 
cing. Different views are held by different persons about the 
appropriate, and 'the seven oxen' seems as appropriate a name 
for seven stars as, say, Kwntv Canicula is for one star. Accepting 
Varro's testimony means rejecting Max Mailer's etymology, which 
indeed is improbable enough in itself. He derives trio fi-om an 
uncertified form "^stru^ which he supposes to be an extinct Latin 
word for a * star.* But not only the word but also the root, with 
which he connects it, STRI for STAR are devoid of authority. 

We must start then with the form trio and the meaning * ox ' 
and look for some more satisfectory derivation. Max Mailer, 

^LUUari machaera^ Caecil. ap Non. 134, 16 = lyx^^ bpi^aaSai, Horn. II, 4, 
307, etc. Cf. licitatar gladiator, apparitor, occisor cui multa licent (!) Gl. Isid. 
(Dncange). 

'Science of Language, Series II, p. 804 and foil. 



66 AMERICA17 JOURNAL OF PHILOLOGY. 

though preferring his ^strici\y has suggested that it comes from the 
root TRI, rub, crush, another or a cognate form of TAR in tero^ 
tritus^ Tpi-^-«, etc, the oxen, I suppose, being regarded as crush- 
ing the clods. This derivation is phonetically unexceptionable; 
but it seems to be a somewhat artificial way of naming the plough- 
ing oxen. At all events I think I can suggest a better. 

Two words throw considerable light on trio by the phonetic 
changes which they evidence ; they are lien and vicu Lien is for 
(sp)/i(gh)^«, original form SPLAGHAN* ; it thus shows a weak- 
ening of A to i and a loss of GH. Precisely the same changes 
are shown in via^ a word which has not had its rights from philo- 
logers." Via is for vigh-a from root VAGH carry in veho, etc. 
It is formed straight from the root, like the Goth. vig-Sf and not 
from a form veh-ya with suffix j^a. For the^ which only appears 
to disappear is by no means necessary to explain the x, and the 
Oscan veia^ carriage, is no evidence for the Latin. 

Trio then is for "^triho *trigho from root TRAGH in traho 
trag'Ulum = Eng. drag. It means * the drawer of the plough, 
etc»/ the ox, especially when engaged in the act of drawing it 
(maxime quom arant terram). The suffix -on is frequendy used 
of persons or things regarded as repeatedly performing an action.' 
Comjx^re palpo, combibOy ca/citro, etc ; and trio, like all these, is 
closely connected with a i^rbal stem \^trah). 

SnJns^ Siti/$im^ Sfrcsct\ scrcH$is^ <r:Wov. 

The first of these words, shJhs, is from a root SUR SVAR, 
shine» biurn, wi\!ch \^>? see in Sonsk. svar heaven, root sur shine, 
rule* Ctr, s«%V>Ho-(, l^\u st^r-r-HH-s, Soracte. If so, it will be for 
.N7>r\/AxN\ The K^^ss of r before consonants is discussed by Corssen, 
In this c;isc» as in that ^^f /v'J*» v*^^ perd-o = &. W/Ai) it is partly 
duCs 1 think* to An cndoa\\\r to a\\>id confusion. It was felt that 
.v\%^»\}V<.>r, hriviht* shvHiKi Ix* didcr^nnated off from surdtts, deaf, 
iuM as A\* V,\ r^.sV frv>m /h.^»^m;V\ ^oar^*. 

FvM th<^ «u\uu»\< ' lM;<hi, hv>t * sec Vir^r. Acn. S, 52S, anna inter 
nulvin i^^cli rx\iHMK^ .wr^a jx^ jf*iji«» rudLire x-ident, and Non. 
\v so*, .vxA*«i vlu iivir >>\ ^y^A** uiule et Tvberiinus' ait *aureos sub- 
lUu \t v^>;t\t^ sVA^iV^.x" ^\t \ ucivr/ Favn ;he s^ense of * burning, heat- 
r,\>i ' wv <\^M*v ^\H that v^t' 'vii\v.\j;.' In Latin srresco shows this 



* V *u;. Nvv ^\\ «5:x- ftvm Corssen I 46a 



ETYMOLOGICAL STUDIES. 6^ 

sense most clearly ; but it enters into sudus and serenus too. This 
is what makes pelago sereno (Virg. Aen. 5, 870) such an odd 
expression. With this 'dry sea* may be contrasted Statius' 
thoroughly appreciative use of the word, Ach. i, 120, properatque 
dapes largoque serened igne domum. The sense of ^drying * is 
perhaps preserved in Anglo-Sax. sedrian^ dry, sere, Old High 
Germ- sdritiy dry up.^ 

The general sense of reducing volume by heat probably appears 
in aifxuop for (rfip^to-v which is used in the same sense as the Latin 
defnUum ; unless indeed the name has reference to the bright 
look of the liquor. In this case the Sanskrit surdy wine, vinous 
liquor, but also water ^ may be compared. 

Are we obliged to derive this word from a colorless pronominal 
root as Curtius does ?" Is it not better to take it from the root AV 
to breathe, which we find in 5« (o/«) Skt root v& blow, Latin 
venhis = Eng. wind^ etc ?• It will then mean the * livings breathing * 
man himself. 

For the superlative suiEx /a as in wr-a-ro-r, etc., compare the 
German selb-st by the side of selb-er, Eng. self, and the Plautine 
ipsissimus. For the transference of meaning compare the Sanskrit 
dtman, breath, used in the oblique cases for * self* and the Hebrew 
Tu/esh, breath. This representation of a difficult and complex ab- 
stract idea by an analogy from the concrete world may be illustrated 
by other examples. Eng. self, Germ, selb- has been compared with 
M. H. G. sin Hp (leib), his body.* Hebr. etsem, self (originally of 
things and then of persons), meant properly ' bone.* So gerem in 
later Hebrew. In Dinda, a language of Central Afiica, yi guop is 
yoMTself (lit. your body).' If the original meaning was such as we 
have described, the consciousness of it was lost very early, as we 
might expect. Compare Hom. II. i, 3, 4, irohXhs d* l<f>6ifiovs yjtvx^f 

'Aid« irpoiayjt€P rfp^av ' av roi/f dc iXoapia T€vxf Kvv€crcri, which Contrasts 
very curiously with Arist Pol.V6,l6, alrovs re • • • ical r^ crafiara. 

* Taken by Curtias, No. 600 ^. from a root SUS. Prof. Skeat has suggested 
to me that sudus for su{s)dus is from the same root. The possibility of this is 
not to be denied, though the other words point to a root SVAR. 

* Curt.* 543 Eng. tr. II 161. » Curt. No. 587. 
« Grimm, Deutsch. Gramm. Ill, pp. 5, 647. 

* Pott, W. Von Humboldt und die Sprachwissenschaft, p. xx. 



68 AMERICAN JOURNAL OF PHILOLOGY. 

ftoi^d^f bulbuSy egula and the tests of a borrowed word. 

Curtius^ brings these two words together as examples of the dis- 
puted Indo-European B. He does not however further determine 
the root, an omission which I shall endeavor to supply ; and he 
rcj;ard3 the Latin bullms as * not borrowed ' fix>m the Greek, * on 
Aca>unt of its derivatives ' bulb>6sus-atio>ulus, a statement which I 
shall cndea\*or to correct The root would seem to be /Sop, origi- 
uaI gar, eat, with X for p ; the second /3 is either to be explained 
AS a * mutilated reduplication,' compare yop-y-6t quoted below and 
the I^tin ^«-4-or; or dse -^ is the suffix, for -/o ; compare cpv- 
(tiAftc v^^ ^^^ «yM^ «yM»^o). The onion is conceived of as an eai- 
0KV r\x>t, c\Mn|\are >«y-ifar«ct * A^^mm Hesycfa. firom the same root 
1%AR, ,AA3»w then mill hax^e nothii^ to do with the Lettish bum- 
h»f<s kuoK with whidi Curtius cvMnpares it, and which can hardly 
W ^)vxr.iixxl ti\\n) hamfml^: bubble placed by Cortius with iulia 
aih) ^'ivi'^A.^N Wuh rei^^jud to Cortius^ second statement that Aul- 
^^< \s W'A K\n\>m>p\i m>ai the Greek, it is to be observed, first, that 
it i> )\^M th;« oUss of word's njunes of regetiblcs or vegetable fffo- 
\hu^t5s iH^U Aix^ K^TTomvd by Latin in greatest profoaoo, and 
*\\\M>x\'\ tS^t th^ AT>:v,:^,>«« 6\>:a the ncmber of the derivatives is 
u'V^x^xMkv Txr^ it :h<^ ^x^^a Aiix^es t-oca h^Jta had been in early 
^\>x< ji\^;s\ V, \t?\'^. A> ;>>cv Are U^c x:>3 wiTacaL* aad e\xai if tbcy had 
Kv^\ Nn\ r ^n-^ as r.xi^r.v *$ tijey are, iber wc<ikl baine proved 

, i *vWn H'-i / .A wA. i.: ji. wtA ^rc».a« isjcc r^KjueEry in aH sorts 
>v a,n\\n ;avv ;S^ >vX'^--X ->* I JCit irreninre: vet x>o ooe ever 
>*ss Nx As.t w^A #-*»u »t A Kv-rxN^TC w.TO. 

VN^v >>,v^vx V V ^i.vrxr cv'm^^so.M' ir. rbe nucxr whki it will be 

v\ , .X Kh> >aA>^ ,'\t»: i jif X mvn-,": xt i ii^iKr^iasiT 5C£r>ds isolated 
*^v^"^ ^V v^ ^>sts tw^\ %K^r '.v\n\huv\£ wrtccbir^ss 
vs- \^ NX- .\^ V c^ K\*\^%vv* TUv ht iuij^ UTTC^-CMX due l! 

^V^'\v \s* s .»x.^\ Kxxv KvM- 4t<v ir ^*\r.^ t^un^'t; jn5Cir3:>e rhxt has 
^\\..s, ^^ ^ .. »^v ,5^ ^,^. ^^y J ^^, , Sf 5t>v"i«T i^ oeiD^isscrxbon to 
>s >x .v.v \>, ^: -Y ^v nM :V ^vK*' li;in*'. il wcfi xr4ZT>lm«tbe 



ETYMOLOGICAL STUDIES. 69 

largest family of derivatives in the language and be as demons- 
trably an alien. In &ct it does not matter at all how many words 
are derived from it, but whether any are connected with it The fer- 
tility of a borrowed word is only a question of use and time. As 
soon as its strangeness has worn off and it is not distinguished by 
the linguistic consciousness from the rest of the language, it will 
resemble them in having derivatives. But however fertile it may 
be of offspring, it cannot * beget ancestors ' so to say. Philology 
will observe that only its descendants have any resemblance to it, 
and that they and it stand alone without other relations in the lan- 
guage, and will thus convict it of foreign extraction. This is true 
isolation, and the isolation of bulbus. I will illustrate apparent 
isolation from a single but very striking case, a word which, so far 
as I know, has not hitherto been derived. 

Egula is a word once found in Pliny^ as the name of a particular 
kind of sulphur. It is derived from root AGH to choke, which we 
see in Latin ango, Greek &YX!^i^ ^tc. It is the only word from 
the root AGH with an ^ in which the original physical mean- 
ing is preserved, eg-ula being the * choking * or * stifling ' sulphur. 
In all the other cognate words eg-enu-s eg-eo, etc., which show the 
^, the meaning is the same as in the Greek dxqPf viz. the * res an- 
giisia domi,' the pinch of poverty. 

This word, which is explained by Hesychius as €irrpagl)ris, is to be 
added to the derivatives of root GAR (Curt No. 643.) It shows a 
mutilated reduplication and a meaning yo/, big, which is often 
derived from that of ^ feeding^ e. g. ob-esu-s by edOy Tp6<l>i 
(Homer) by Tp€<f>€iy. And I see no reason why -yopycJff, fierce, 
grim, and ropym, should not be the same word in the active sense 
of * devouring.' 

If we may trust Greek sound laws, this word has lost a spirant 
between the o and the ». As we have no other evidence as to what 
it was, we must at once resort to the meaning. The following 
usages in Homer are significant — (i) that of anxiously expecting^ 
of being pamfully intent on a thing: Od. 2, 351 mivov 6iofi€Pfi t6v 

> Pliny 35, 15, 50. 'Curtius No. 166 (a). 



^0 AMERICAN JOURNAL OF PHILOLOGY. 

d{f<rfu>f>f>u #T no$«v tXBoi'f II. 1 3, 283 /f At n ol mpaiUff fuyaka aripwoun 
ifar&acMi Ktipat 6iofi€v^', Od« lO, 248 iw hi ol Zfra'€ daKpv6flHw m'/arXavro* 

y6w V i»ltro 6vi»6t. Again, (2) that of ominous presaging, Od. 9, 
213 Ovfiitf 6i<rar6 fioi, and even impersonally, Od. 19, 312 alXXd /uh, 
tfff Avh Bviihv 6i0raiitf tarrai ir€p. (3) Lastly that of sure comnctian^ 
o{ prophetic anticipation, whether of something within or something 
without our own range of power, II. 13, 262 ow yap o»« nokt/uC^uf^ 

II* 6, 34' Mxh^*^^^^ ^' 3/«». 

Thi» straining and watching, this fore-boding and this absolute 
conviction and confidence of prophecy point us to the divining art. 
The word, so to speak, gives us a complete picture of the ol»poaK6^os 
in the various phases of his art. We see him waiting with straining 
cyc» for the interpreters of heaven's will and trembling in a sus- 
pcHHc of hope and fear. The message come, we hear the mysteri- 
ouB tones in which he announces destiny to the people, and we 
ttpprcc iatc the confidence of prediction with which he meets and 
crushes nil doubt and disbelief So that it is not without reason 
that wc find two glosses of Hesychius dose together : 

A^\Tptin^: this clue» we shall take 6l%i to be for 6fi» and to be con- 
noctrd with the I^tin ati-s and the Greek 6(^f)i'-<ap6's and to have 
xwvAWi orl>iinrtlly to ropisu/t the birds : being related to ^o/w, a bird, 
AU i^Knolclo CijTck u*i>rd, as pa^tiu is to /i^nr. Nor shall we wonder 
that A won! expressing confidence or conjecture about the fiiture 
wluntUI h;ux l>rcn dcrixTd firom 'bird* when we recall passages 

liko Av^^l\^ph* Av, 7>3 /y^wr re ro^iCrrr wd*^ oaaanp vtpi parreiag dm- 

fc,M»v« •t> r. X.; or ihAt a xixmxI proj^er to the diviner's art should have 
Imssmuo jvut of the c\Mt\nion stock of the language when we think 
ol \\w \,M\\\ APtsp9Viin\ a^^gurari, dizinare, ominan\ autumare. 

J. P. POSTGATK, 

^ Vt\f' i;mw m >»huH M. Sohm^H pv^s tli« first gloss (which, according to 
^>v\N, ^* \Nsnv>^M \w <W MSS^ iss h*iv.ly s*ti&f*cloiy. If t^oc^e^/c is a future, as 
\} A^^p1''^\\ ^>^ Iv^^ x^-^o ^>t^v,u\ oTOMdxi iv/v^4» [tMtiif /{- vuij be for ou^xtlc and viovt^et 



NOTES. 

Grandio, Gradio; Grandivus, Gradivus. 

In Aulularia, vv. 48 and 49, we read : 

Si hercle h6die fustem c6pero aut stimulum in manum 
Testddineum istum tlbi ego grandib6 gradum, 

but BDEJ according to Goetz have gradibo. The passage is 
quoted by Nonius Marcellus, p. 115, i, where the codices give 
grandivo. The connection with grandire^ is of course undeniable, 
but the pun is much more effective if we read grddibo gradum, 
and it becomes a question which none of the editors seems to have 
entertained, whether the spelling of the MSS. ought not to be kept. 
Plautus seems to be fond of puns where there is a difference of 
quantity in the vowels of the words played upon. To cite but a 
few examples, Amph. 318 exdssatum os, ^2 ds exdssas; Bacch. 
362 Crucisalum — Crasalo, 687 cr&ciatum Crasalum; Mil. 325 
ludo luto, 1425 miitis — mltis ; Merc. 82 dmens dmans, 643 mdlis 
— vidlum; Rud. 122^ licet — infellcet — Itcentia, 

That n had a weak sound and a tendency to disappear before 
certain consonants is a well-known fact Some interesting remarks 
on this phenomenon by Buecheler may be found in a fecent number 
of the Rheinisches Museum, Bd. 37, 1882, pp. 525-9. The dis- 
appearance is most frequent before 5, but occurs also before gut- 
turals and dentals. It may be well here to give more fully some 
&cts relative to the tradition of the Plautus MSS. in this particular. 

According to Ra3SOW,de Plauti Suhstantiwis,pollutor (or pollmcfor 
occurs Poen. Prol. 63 (Codd. pollecior, Fulgentius pollincior), and 
Asin. 910 {^pollictorem B*DJ .poUictore E). Nonius Yiaspollinctorem, 
keeping the nasal as in the case of grandibo (Aul. 49), and he is 
followed by Goetz and Loewe in their edition of 188 1. In Asin. 
276 the same editors follow BDEJ in leaving out the n oi praegnaiis. 
In Aul. 163 BDEJ hacvepregnantemy and Goetz x^aA^ praegnantem. 
In Amph. 723 B has pregnaii, which Goetz and Loewe adopt 
against pregnanH of DJ. True. 390 Schoell reads with A and B 

* The adjective grandis is found associated with gradus in the following pas- 
sages: Cure. 118, grandiorem gradum; True. 286, grandi gradu ; Epid. 13, 
gradibus grandibus ; Pacuvius 37, praegrandi gradu. 



72 AMEJtICAX JOURXAL OF PHILOLOGY. 

fraet:JULiem {pregntmUm CD\ and so in 8ii where B has preg- 
nAii^i^T^AprtgmtmiewL. In all these cases, as Boedider remarks, 
the influence of the nominative /r«^^pMj in bringii^ about the loss 
ot'« must be taken into account: but that « itself had a weak sound 
before / is proved by numerous examples from inscriptioas. I 
think, therefore, that SdioeO is jusdted in reading Tariimas True 
64<>^ with essential s^ireement on the part of the MSS. and I see no 
gvx>l reason why gTMiih^ AuL 49 should not find a place in the 
tx^xt. Degenerate punsters of the present day, when hard pressed, 
tAke the most shocking Hbcrtxs with the pronandatioQ oi their 
RK>ther-toR^::ue, and whv mar not P^autos* miser, for the sake of a 
rtvjilly gvxxl pun, be allowed to yield to a rclgar phonetic tendency. 
Hut there was a teodeocy to drop m also before d^ the following 
exjcnjves^ v:i\x« by Corsaau Voc I 257. and Schochardt, VulgarL 
I KXv prvn^c : .' j^-ir^ •Wi/Ai'y;' KaUass. Setrndtms^ Secudo^ clades- 
r, vxjc. .-f .VxjjcW. ,*»^ia/. 4\A£rm ix csMltm. To these Boedider 
AvUfe mjtS^ K>r wf^xx.' tr-oc: a EVtwocv He says* however, L c p. 
5Xv '' Kc5<xrr wjir Arr Nj&suI vcc i": dis Hochafem duldet so viel 
ich mxr-:J2k aa k^,n^tn Woct ein S±w-jckcaL u::«a n>ie lb a r grando mit 
^ ^v^xru x'aW*.'^ Bu; i' jrmcnx^ dlvi ace becocae ^rmdCy the drop- 
|v;^ vHi; vV « Kvxy. 1 thlrk. t«e escirC^ijed Ir Grsdssms. For this 
i^o.t^unK' of Mj;:^ tbe dxtvoines gni? ro ^r^Th-^o ry etjmology. 
V^y>:v* v*^^ 5:^A<s 5c=:riV *^K?^^crr9^ ^^^siz^ der {in den 
KA;vx^:^ \vv>^chrxr*.^fcc<.'*' Kj;rr«^' =r>ro? octktasiy " perh. firom 
^*'^ ,•)»,"* 1^:2? ^r\^^ijLajcxc ii'.ccfNi by P;r«£jer-Jcrda2, Romisdie 
V^xthxVvxx'v Of A<^ I. r^ ^i5< 2? ii?o c--*«:^ bv rxBt ci the recent 
vv ^y^ N>K^ XVv^'I liK" exyr«5? Jt= v'xrur.cc c£ Aea. 3; 3^ Forbiger, 
iv\\\\ vk^>*x.vl\^ Oir^^x'^i:* :s>^\ Vxnotk- bcwever I1881) 
^ \v^ ^j\ x^^ ^*/*xcf^ V^^^^^**-^ v^'*^-3j^ oer gewalrE^ Gott," 



X vv^.fr K'is^'K Vt.v V. v$A^ •'^ ^'^^^^ ^'* **t> ii^ ^ UMMW ist |>(:cxoc^ Gos end 
■...\ >*.,^V. S* 'II,; K,>v^*** -v'^ K^.m,*- :>x\ tt^ TTbeC 7. r^l. says : 



NOTES. 73 

To the derivation of Grddiims from grddior the quantity offers 
a serious objection. As the dictionaries give no fair idea of the 
frequency with which the word occurs, I give the following statis- 
tics. The foUowing poets always have Grddiims ; after each name 
I add the niunber of occurrences,* Statins (20), Qaudian (9), 
Juvenal (2), Seneca (2), Vergil (2), Lucan (i). Grddivus occurs 
six times in Ovid, five times in Valerius Flaccus, and twice in Silius 
Italicus. I give the verses where Grddivtis occurs : 

Orid, Met. VI 427, Et genus a magno ducentem forte Gradivo, 
Val. Flac. V 651, Ramperet irridens strepitumque minasque Gradivi : 
Sil. It XV 15, Qui consulta ducum ac flagret meliore Gradivo 
XV 337, Moles ilia viii, calidoque habitata Gradivo 

It will thus be seen that Grddivus only occurs at the end of an 
hexameter, while there are forty-nine examples of Grddivus. 

The verb grandire is used of the growth of plants. So Nonius, 
p. 115, explains grandire as grandemfcLcere^ and cites from Varro, 
" Quum aut humus semina concipere non possit, aut recepta non 
reddat, aut edita grandire nequeat," from Attius a similar use of 
pergratidescere, ^^Fruges prohibet pergrandescere!^ For grandire 
used as a neuter, the dictionaries cite Cato, R. R. 141, 2. As this 
passage, in connection with the Aulularia verse above discussed, 
first suggested to me the possibility oigradivtis being derived from 
grandire, I will give it in full It is a formal prayer to Mars. 
" Mars pater, te precor, quaesoque uti sies volens propitius mihi, 
domo, familiaeque nostrae quoins rei ei^o agrum terram, fundum- 
que meum suovetaurilia circumagi iussi, ut tu morbos visos invi- 
sosque, viduertatem, vastitudinemque calamitates intemperiasque 
prohibessis, defendas avemmcesque. Utique XyifrugeSyfrumenta 
vineta virgultaque grandire beneque evenire sinas, pastores, pecua- 
que salva servassis, duisque bonam salutem valetudinemque mihi, 
domo, familiaeque nostrae." Now, as we have from the verb aver- 
runcare AverruncuSy u e. an averting deity, so from grandire we 
have Gra{n)dvimSy a deity promoting growth. Preller, Romische 
Mythologie, I, p. 340, has the following note : "Auch der Deus 
Avemincus bei Varro VII 102, Gellius V 12, 14, ist hochst wahr- 
scheinlich Mars." It seems to me evident that in Aen. Ill 35 f. 
Mars is invoked in both capacities as Gra(n)divus presiding over 
the growth oivirgulta (c£ v. 23), and AverruncuSy averting threat- 

> I have looked up the passages according to the best accessible indices. 
Harpers' cites but one case of Grddivus, 



zz:r c ' L arraTrhg" 



■a^ 





i -'er •' --*' • "-"-^ ' ~.*j=- irt rii^nnmsian 
















::v -stftpiC teat iST' utT^ arziiciniL; rcsd-ruiace ot 
^ ^Z/MT^. 'tft ,*'.iC Z-^tr-nni II 2:. r* -sz:;*";? rs >fa^ as 
, i^ ,V4^^^ //^ jorur^.U V- ^ iut ^-^irrrrgxirT ic lie a.r»rre passage 
S;i " -^3^ >i/ > T!:*I''. V'X I p, 5t2^ '^j'y utrrajfK eavjv 'JL;i«a. ad est 
V..V..XULMU .a 'jfv^ila, ^i uA. isi vtHajr-T'.cs sit ?^ tf' < ' y '« ^ est: mmi gra- 
It//* j'./i//* . . • alti j^radit^unif qw>d ^adujm vjifcraxt qid pugnani; 
.uu ^lu/d uip^i^re gradianlur^ falii a graditudziu quod hue et Ubu 
^fu^iuiur . . . aid gradtvuMf quia numquam eqtusitr ; aui a 
^ ' .;u/* dicttim (cC aLi/j i\\*t jul/jitlotw in D and T as given by Thilo, 
I..U Isiuor, Urig. VIII II, 52;, The Epitome of Festus (p. 97, 
Vu.ci • gives this explanation; **Gradivus Mars appellatus est a 
;f.u.u'Huo in bello ultro c\ix(H\\\t\ aive a vibradone hastae quod 
\,.Mv.vi Uicant <t^da/w4v vel ut alii dicunt quia gratnine sit ortus, 
v;aoJ iuLcrprctantur quia coroim jjmminea in re militari maximae 
V .1 ^vuioiaiiouis." It should br ntiltnl that according to Lindemann, 
Vau I h.u> Grandrvus. St*. \k\k\ \\\ Scrv. Comm, in Aen. I 292 
1^ Uuks p. io8\ "Mara eniu\ luiu Mevit Gradivus dicitur, cum 
u.uKimlUis est Quirinus/' C\hU\ lUiuburjjensis hos Grandwus. 
I'll... Mil i^Keil, II. p. 146) U'vc* (*^iAJHH.< AS epithet of Mars, and 
hvu ihc Codex Sangalleu:*!* ha» (•>uv^Vc/^(.v\ IVrhaps when the 
Jvu\atiua from gramen m\^ tvi^^ pu^Kv^txl there wus still a nasal 



NOTES. 75 

sound heard in the first syUable of Gradivus. Gram*ndivus pro- 
nounced with the e slurred would not be unlike Grddivus. If the 
view above set forth of the origin of Gradivus is accepted, and if 
gradibo is received in Aul. 49, it would follow that the a oigrandts 
is long by nature. Were it short, it would remain so after the 
extrusion of the «, cf. Taritinas^ KoXedar . As to the etymology of 
grandis itself, whether it has any connection with Ags. great, 
Ahd. gr5z. Urdeutsch *grauts, as Johannes Schmidt with others 
assert, I do not feel competent to pass any judgment. The con- 
nection assumed by Vanicek of grandis with grams, Skr. guril^ 
does not seem to me to be clearly established. 

MiNTON Warren. 



The Bucolic Caesura. 

In the Hermathena, No. VIII, Mr. Tyrrell follows Dr. Maguire 
in throwing doubt upon the commonly accepted theory of the 
bucolic caesura, summing up his conclusion as follows : " The only 
expression of the rule, as &r as I know, which really colligates the 
phenomena is that of Dr. Maguire, Fellow of Trinity College, 
Dublin, and it runs thus : * When the fourth foot ends with a word, 
the fourth foot must be a dactyl, if there is a stop after the fourth 
foot: " 

Mr. Tyrrell mentions Marius Victorinus and Terentianus Maurus 
as the authorities for the existing rule, but he does not quote them, 
nor does he allude to a passage in Servius, which to my mind is 
very important as setting the matter in its true light Before 
considering what is the real import of the ancient grammarians' 
testimony, it will be convenient to quote them in full. 

Servius on Eclogue i init. Carmen bucolicum, quod debet quarto 
pede terminare partem orationis. Qui pes si sit dactylus, meliorem 
efficit versum ; ut * nos patriae fines et dulcia.' Primus etiam pes 
secundum Donatum dactylus esse debet, et terminare partem 
orationis ; ut * Tityre.* Quam legem Theocritus vehementer 
observat, Vergilius non adeo. The Pseudo-Probus gives the rule 
in a much shorter form. 

Terentianus Maurus, p. 389 (Keil) : 

Pastorale volet cum quis componere carmen, 
tetrametrum absolvat, cui portio demitur ima, 
quae solido a verbo poterit conectere versum. 
bucolicon siquidem talem voluere vocari. 









■^' ■■ "^ 1 —11 < * "m. -w -~" 





N:^ this is 

TTTCLT "frar T':ier:cr3i» Jt is lutmic jirrls ajf^ns -ncLiaes it ; tbe 

Tba» 3esn3 3C reascii ±ist Dcr dnicc rat rbe ti^ of hucoHc 
Cite'ntTa W3S rj^bdj j^rrei, bj an jjrrimanTm^ 3D liae cadence of whidi 
LacTttsus and Car: » -is are so xni ibiTt rLLnti ^tqmjrm, panJtu 
Tbtrxrfna ai u^dc^ts^j partial to x cmch more partial than 
Hotaer, E'Jt Isuspeadiatrtere^ocwhythetergj^iu-A'irrtf^^icra 
came thus to be applkd was this : that the Greek gTammarians 
had b^rgijn by characterizing a particniar kind of kexamrter as 
bucolic And the particular kind of hexameter I suppose to have 
been such a one as a^ixm ^Kx^uxm, UoUnu <t>^Xah opxcr aoMtf or 
DuciU ai urbe domum^ mea ccarmtnOy duciU Daphmn. This, I 
think, may be inferred from tbe expression carmen bucolicum in 
the note of Servius, and the theory which he quotes from Donatus, 
that the Arst foot ought also to be a dactyl and end a word. When 
the name bucolic had been attached to a hexameter of which the 
first word formed a dactyl and in which the fourth foot also ended 
a word, the phrase bucolic caesura may easily have been attached 
to the particular cadence in the fourth foot 

Henry Nkttleship, 



NOTES, 77 

On a Transposition in Seneca. 

The method which I recently applied to the case of a trans- 
posed passage in the New Testament has an interesting illustra- 
tion in Seneca, ad Marciam. Madvig (Adversaria Critica, p. 355) 
pomted out that two passages had been transposed in c. 17, and 
that the words *' Didt omnibus nobis natura • . . qui tibi nihil certi 
spoponderunt " (c. 17, 6, 7) should in reality stand after the words 
" sed humanum est " (c. 17, i). Almost all such cases of transpo- 
sition arise from the misplacement of a sheet or sheets in the MS 
or papyrus roll of which they form a part. And it follows at once 
that in all such errors we must have an integral number of pages 
for each of the two passages concerned in the transposition, and 
also an integral number of pages for the portion of the book ante- 
cedent to the disputed ground. Let us apply this test to the 
passages of Seneca just referred to. 

Taking the Teubner edition as our standard, we have to carry a 
passage 12.6 Teubner lines in length to a place 38 lines earlier. 
Obviously 38 is 12.6 X 3 very nearly. Assuming 12.6 lines to 
represent a single page of the manuscript, the space through which 
it has to be moved is three pages. The previous part of the book 
is 568 lines = 45.0 X 12.6 very nearly. 

The error, therefore, arose at the 46th page of the MS, and 
consisted in placing the 47th page after the 50th. Madvig's criti- 
cism is therefore completely confirmed. 

In the next place, we may enquire into the stichometric size of 
the page in question. The average Teubner line being found to be 
46.9 letters, or somewhat less, and the average hexameter being 
about 36 letters, a page of 12.6 Teubner lines is about 16 hexam- 
eters. ' 

Finally, the remainder of the book being reckoned, we have for 
the MS in which the error was made a roll of very nearly 83 com- 
plete pages. J rejji^el Harris. 



"Is BEING Built." 

The earliest examples of the use of our passive progressive form, 
is being builiy etc., that have thus far been noted, date from 
1769-79; see Dr. F. Hall's Adjectives in Able, also English 
Rational and Irradonal, Nineteenth Ceniury, Sept. 1880, by the 






Tr^ii 



^ liiz: rr ^r: —at Z 



T — :: i— — — n ^: — ^- l-r-i. i* 3. s nx sr irei as 



:=. ^ ^tFT' 



REVIEWS AND BOOK NOTICES. 

Franz5sische Studien. Heraasgegeben von G. KOrtino und E. Koschwitz, 
III Band, 3 Heft Heilbronn : Henninger. 

Die Wortstellang in der altfranz()sischen Dichtung "Aacassin et Nicolete," 
von Julias Schlickum. 

Prof. Suchier's edition of the Aucassin et Nicolete text, Paderbom, 1 881, is 
taken for this special study, and in the arrangement of the material for it the 
reader will find much that reminds him of MorTs paper, Die Wortstellung im 
altfranzOsischen Rolandsliede, Roman. Studien, Band III, S. 199-294. A short 
review of the Suchier edition of A. and N. was given in this Journal, Vol. II, 
pp. 234-36, in which mention was made of the great importance of this work, 
both for the study of the morphology and the syntax of Old French, and espe- 
cially for the latter, as its peculiar form — the Chantefabk^ a mixture of prose 
and poetry — enables us to examine these two species of sentence as given by 
one and the same author. This advantage is manifest from the outset where 
we find differences between the poetic and prose construction, and where the 
former frequently prefers a certain set form varying from the normal one simply 
to produce by it some psychological or rhetorical effect. Metre and assonance 
come in here also as important elements to cause the poet to change the position 
of his words. This is seen particularly in Old French, where, in the relation 
of object to verb, the strictest rules of S3rntax are overthrown, while in the 
modem language this relation is subject to fixed and rigid law for both poetry 
and prose. 

If we take the simplest phrase-elements — subject, verb, object — whose rela- 
tions to one another are treated by the author in the various kinds of sentence- 
declarative, interrogative, imperative, and exclamatory — we find deviations of 
usage from certain other models of the Old French literature that are at once 
striking and instructive, and which, when compared with the canons of modem 
syntax, show strong tendencies to a development of the present inflexible system. 
For instance, though there is still a certain freedom with reference to the order 
of subject and object, yet the predoihinant tendency is found to be in the direc- 
tion of the rigid law of the modem language where the subject is required to 
precede the verb. The writer finds 312 sentences of this kind to only 70 with 
other positions, thus showing how this expression of the logical sequence of 
ideas had gained the upper hand in the early period of the language. Inversion 
is not arbitrary, as Diez and MSltzner have maintained, but subject to fixed 
laws that scarcely know exception. These results compared with compositions 
of a century earlier, e, g, the Chanson de Roland at the end of the XI century, 
present a striking contrast, in that the latter gives us numerous exceptions to 
its rule of inversion. Here, with verba dicendi^ inversion of the subject was 
regular, and this is also found to be the case for the most part in the poetic 



REVIEWS AND BOOK NOTICES, 8 1 

should also hare been drawn upon largely to show the general character of 
many of the sentence-forms ; bat this is not done, nor is there scarcely even a 
reference to the Latin construction, which, in certain cases, would have illus- 
trated the force of a given order of words much more clearly than any possible 
explanation according to general principles. On the whole, this little pamphlet 
may be used with advantage by those studying the text for the first time, and 
it forms an important factor in the line of special syntactical studies which 
must be the basis for any future general syntax of Old French. 

4 Heft. 

In VoL III, p. 434 sqq., the editor of this Journal, in his "Studies in 
Pindaric Syntax," designates the four principal forms of the Conditional Sentence 
which the Greek holds sharply apart as ' logical,* ' anticipatory,' * ideal,' and 
• unreaL' The second of these, the • anticipatory ' — idv ri ixf^fttv^ d^aofiev-^ 
was turned over by the Latin from the region of objective possibility into that 
of reality (* logical') and both represented by the indicative, thus reducing the 
number of its leading conditional forms to three. For the two remaining Greek 
species, the ' ideal ' and the ' unreal,' the Latin kept its own peculiar mode of 
treatment with a clear distinction of mood and tense in each. In the * ideal ' 
type where the Greek has ei with the optative in the protasis, and optative with 
&v in the apodosis, the Latin uses the subjunctive (pres. or perf.) in both clauses 
(Haec si tecum pairia loquahtr nonnt impetrare debeat ; Si me suspendam^ meant 
pperam luserim\ while in the 'unreal' condition — *the h3rpothesis contrary to 
fact ' — it employs a past tense of the subjunctive to express the double point of 
▼iew of present and past time. For the domain of the present the imperf. 
stands in both clauses (sapientia non expeteretur^ si tdhil efficeret = el ri el^ov^ 
t6i6ow dv) ; for that of the past, the pluperfect (si volmsset, dimicasset = el ri 

In the representation of these phases of the * unreal ' condition, Modem French 
syntax presents a striking contrast with the Latin by the use of the indicative 
imperf. in the protasis and the imperfect future (conditional) in the apodosis 
i/ele femis encore^ si f avals ^ le /aire — Comeille) for present time, and the 
pluperfect indicative and pluperfect future (si , , , les UgiskUeurs avaient /tabii 
la eessiim des biens, an ne serai/ pas tomb^ dans tant de s/ditums) for past time. 
This catting loose from the traditions of the mother language and nearer 
approach to the Greek type of construction is most striking, and it is the chief 
merit of the paper before us — Historische Entwickelung der syntaktischen 
Verhaltnisse der Bedingungssatze im Altfranz6sischen, von Joseph Klapperich 
—to have followed up the traces of the Latin sentence construction in French, 
and to have shown, as might ^ priori have been expected, that this passage 
from the domain of subjective to that of objective representation of thought 
was not a process that took place suddenly. 

Just as we saw a reduction in the number of conditional forms in passing 
from the Greek to the Latin, so the author of this treatise discovers that from 
the very earliest period of the French the ' anticipatory ' had been merged into 
the * logical ' condition, which always takes the indicative, while the type of 
subjective possibility — the * ideal * condition — ^has been pushed forward and, 
for the most part, identified with the * unreal ' condition. This leaves us, then. 



82 AMERICAN JOURNAL OF PHILOLOGY. 

aaif two leading species of condiliooal pbiase for the modern language, tie. 
the ' logical ' and the ' unreal,' instead of the three of the mother idiom. Of 
these two main sets of the hypothetical sentence, it is the first -which is charac- 
terizcd by the almost exclusive use of the indieatiTC present and future in the 
Bubordinste and principal members, respectively, of the phrase. The substiln* 
tion of the fature for the present in the protasis is very unusual, though we do 
find sporadic traces of it as far down as the XVI century, and it is doubtless 
to the Low Latin that we mast look for the model, according to which, with 
time, the rigid Modem French rule was built up of excluding the future from 
the conditional memberof a hypothetical clause. Draeger, in his HisL Syntax, 
II, p. a86--S, notes the use of the present for the fature in this case as a cominoa 
phenomenon io the Folks l^tin, and the usage has simply been confirmed by 
a farther development of the modem syntax. The French here differs ^eiy 
materially from otber members of the Komancc group of languages which admit 
the future as the legitimate type of the protasis. E.g. Ital. Dante, Inf. I rM, 
alUqua'foi tc In vatrat laiire, anima JSa a eii di mt fiii dtgna. Port, Camfies Os 
Lcsiadas, IV 18, Rei lendti tal, que se e vahr Hverdit Igual iw Rri, que agora 
alniantaiUi, Detiaralareii tndff oqut qiditrdtj. The same construction prevails 
in Spanish, while the Provenfal, on the other hand, agrees with the French. 

Very few examples are found by the writer where, according to the Latin 
arrangement {in ituidiii hie m>, n quid dtjuiai), the subjunctive is used in the 
protasis of an 'ideal* condition, with the future in the apodosls. 

For the ' unreal ' condition the Old French used the imperfect subjanctive 
in both classes, corresponding exactly to the Latin usage. In the eirlieat 
period of the language, however, this imperfect frequently represents the Latin 
pluperfect in meaning, from which it had taken its form, and as this constrac- 
tion prevails throughout the whole of the Old French period, it cannot be 
reckoned as a rare phenomenon as Miizner does in his Syntax, I, p. 37. In 
fact its use is so common that it has usurped the legitimate field of the pluper- 
fect subjunctive in both clauses of a conditional sentence, so that this latter 
: all in this capacity in the oldest texts. The oldest documents 
thing of the Modem French order, pluperfect in the protasis 
'e (conditional) in the apodosii, of which the earliest examples 
eWace'sRomandcRouof the second half of the XII century. 
, however, the pluperfect subjunctive held eiclnsive sway in 
n the condition bore upon past time, and it was not till a 
;ent period that it split up into the Modern French type of 
bjnnctive pluperfect in tbe suliordinate clause. Thesubjunc- 
in such cases is now dying oat, according to K.'s investigation, 
continuance in use so long from the fact that the compound 
ed here at an epoch when the conditional phrase that bore 
or future had already begun to go over to the Modem French 
his I bold to be, however, only a specious cause for its long 
reason for it lies in the conservative tendencies of the language 
hold fast to the old models of expression which we see strongly 
: exclusive use of the subjunctive element up to within a recent 

I of the present and fulnre, where the Old French characteristic 
le imperfect subjunctive in both clauses, the modem form of 



REVIEWS AND BOOK NOTICES. 83 

imperfect indicative in the protasis and imperfect future (conditional) in the 
apodosis appears at an early date. The first examples cited belong to the 
Comput of Philippe de Thaon — beginning of the XII century — and yet there 
are cases of a manifest tendency to it as far back as the Chanson de Roland 
(XI century), where in v. 1804, Se veissum RoUant, ainz qtCilfust mors^ EnsembV 
odhd i durriums grang colp^ the only variation from the modem type consists 
in the use of the imperfect subjunctive in place of the imperfect indicative in 
the protasis. This and other examples of similar construction prove beyond 
doubt that the emancipation from the Latin mould took place originally in the 
apodosis. This construction cited from the Chanson de Roland has died out 
in French, but it still lives in the other Romance languages, e, g, Ital. s'egli 
vtwisse, h itwerebbe ; Spanish, siyo U viese^ se lo daria. This is, then, the bridge 
by which we have passed from the Old French subjunctive in both clauses to 
the modem imperfect indicative + the future. From the beginning of the 
XII century this construction is constantly gaining ground, until towards the 
end of the same when it becomes the predominant type of hypothetical phrase 
for present and future time. 

The modem construction with imperfect indicative in both clauses when the 
condition bears upon past time, is unknown in the earlier documents. Here, 
too, the principal clause became the transition link by which the present typical 
form was developed out of the old one. 

The original Old French subjunctive-protasis lived on up into the XVI 
century, when it finally became folks style ; and just as the subjunctive kept its 
place in the protasis longer than in the apodosis in the pure condition, so in 
hypothetical constructions with the comparative particles comnu and que^ the 
principal clause was the first to yield to the new conception of time relation 
and pass from the subjunctive to the indicative type. 

In conditional relative phrases the writer finds the same construction pre- 
vailing as in the conditional with j», except that they do not entirely exclude 
the future from the subordinate clause. 

Several other less important kinds of condition are examined in this inter- 
esting paper, the principal one of which is the hypothetical subordinate phrase 
used as a formula of conjuration. Diez' ellipsis theory is here stuck to by 
the writer in opposition to Bischoff, who, in his Conjunctiy bei Crestien de 
Troies, regards it as a mixture of two optative constructions. The Roman de 
Rou contains the first use of se in these formulas, e. g.W 670 Gentilz ber, dist 
H neist Se Deus mi beneit^ Tum sid prez. 

In the sequence of two conditional subordinate clauses the omission of se in 
the second member is the common rule in Old French, and it was not till in 
the XV century that the modem law of substitution of que for se in this case 
became general ; however, we do find occasional examples of it as far back as 
the middle of the XII century, e, g, Rou III 8943, Et se Deus le velt consenHr 
E que a lui vienge aplaisir^ Bien le feron (tore en az/ofU, 

On the whole, I think this treatise, of sixty-five pages, altogether the most 
comprehensive and the best that has yet appeared for this department of syntax. 
The writer has evident control of his material for the Old French and gives us 
frequent references to the Latin, but as is usual with all such works very little 
account is taken of parallel or identical phenomena in the cognate languages. 
This lack is especially felt for certain phases of construction which at one 



REVIEWS AND BOOK NOTICES. 85 

of which K. discerns in Be6wulf); Edinburgh Review for Oct. 1845; W. 
Wagner's Deutsche Heldensagen, Leipzig, 1881. General references and 
discussions that elucidate particular points may be found in Weinhold's Alt- 
nordisches Leben, Wackerbarth^s Music and the Anglo-Saxons, Bouterwek's 
Caedmon, Lappenberg's History of the Anglo-Saxons. It may be well too to 
mention Grater's Suhm*s History of Denmark, Baldwin's recent work on English 
Literature, Hammerich's De episk-kristelige old-kvad hos de Gotiske folk 
(Copenhagen, 1873), W. Grimm's Die deutsche Heldensage. The reference 
sub Lumsden, "Academy, Vol. XVIII," should be Academy, Vol. XIX ; and 
read, in the next line, " by WQlcker in Anglia, Atuteiger^ IV, 69."^ It has not 
been possible for the reviewer to make exact references in every case, the 
volumes in question not being always at hand. Doubtless Dr. Garnett has 
many of them already on hand for a future edition. 

The peculiar feature of this translation is its line-for-line literalness, with 
alliteration as an occasional grace. Undoubtedly such a theory must result in 
much distressing involution of phrase, which the translator has foreseen and 
done his best to overcome. As a translation for popular reading it fails precisely 
in this point. To the Anglo-Saxon scholar the involutions are quite intelligible, 
for he has grappled with them from the beginning of his studies in Anglo-Saxon 
poetry; but to the general reader they are obscure and tormenting enough. 
In so far then as the translation is an exact and faithful reproduction of the 
original addressed to students, it is a success ; but in so far as it is intended to 
popularize a most difficult poem, it cannot be called such. It cannot be said, 
moreover, that the new version is perfectly free from those Unwdrter^xX^vAix^ 
Ettm Oiler's alliterative version was reproached ; or from monotony, as the per- 
petual recurrence of such words as*' victorious," '* jewel," *' treasure " (translating 
A. S. words in which a subtler insight would have perceived picturesque shades 
of meaning) ; or from inexactness here and there in the translation of particles 
(11. 369, 182, Mtru omitted ; 1. 735, ^gen, rather := " any longer " than ** not yet " ; 
L 862, " now '• for " nevertheless "; L 1353, " unless " for •• except "), or from 
neglect of certain points, as of the duals in several important passages (11. 1707, 

1783). 

In other passages the translation is a distinct advance on Heyne's text, and 
throws light on obscure points, as 11. 1143-4, 1213-14, 2051, 2860-1. LI. 2522-3 
and 31 17 take a liberty with the text (no notes explaining the variations). Um- 
brage might be taken at what appear to be verbal slips or inaccuracies, as 1. 
1 861 , " swan's bath " for ** gannet's bath "; 1. 236, •* weighty words " for " words "; 
L 293, " horse-thanes " for " kindred-thanes," " comrades "; 1. 307, " went " for 
••went down"; L 435, "renounce" for **scom"; 1. 498, **band" for "joy" (see 
Toller-Bosworth, 2x8, for numerous references, though this one is omitted); 
1. 1043, "of" for "over"; 1. 1175, "would" for "wouldst"; 1. 1191, "by" for 
"'twixt"; L 1285: query: can ktoru bunden mean " twisted sword "? ; 1. 1537, 
"cared she not for the contest" for ** cared he," "shrank he," etc.?; 1. 1616, 
" twisted " for " drawn " ?; 1. 1736, " sorrow " for " remorse " ? ; 1. 1793, " pleased 
to rest " for ** longed sorely to rest " ?; 1. 1943, " any dear man " for " leman " ? ; 
L 1980, " with mighty words " for " with formal or courteous words " ?; 1. 2029, why 
" courtier " ? In our opinion oft here belongs io gesette; n^ is wrong, and the trans- 

1 Mr. Brii^t. 



86 AMBRICAlf JOURNAL OP PHILOLOGY. 

Ifllion is " seldom after s leadei's fall restt the deolh-spear [eren] a little while "; 
1. 3i45,"by" should be omitted; 1. 3175,'' saddle-bright" for "bright-saddled" 7; 
1.9199, ""t limes" is omitted; I.3449. "with" for "Ihrongh," "00 account of; 
]. 2^76, "fearful" for "grlsly-hued"; 1, 3S77, itugt Idfi \ttiat to mean "with the 
edge of the iword," oot "vith the HR^fA^ relic"; 1. 3640, " thought of honors for 
as " for " exhorted us to deeds of glory"; 1. 9750, "on account of" for "after 
seeing"; 1. 26x0, the ambiguous " doom of the saints" for "realm of the saints." 
Quotation marks have been omitted 1. 687. Many of these corrections would 
naturally spoil the rhythm which, though rugged, is based throughout upon the 
consistent intrt>duction of two accented words to each hemistich ; but the 
translation would gain in accuracy. 

In judging a work of this nature, however, one may easily be led to be OTer- 
censorious. The difficulty and corrupt state of the leit must be kept catefolly 
in view; the inadequacy of the lexical helps to a thorough study of Anglo-Saxon 
is another point to be remembered ; and the stilt very imperfectly understood 
canons of A. S. poetic syntax may well admit a variety of translations in pas- 
sages that seem at first perfectly clear. Dr. Gamett is modest in everything 
that he advances. Though his translation cannot be called poetry as, in some 
senses, the translations of Simrock, Heyne, Wackerbarth, Conybeare (partial) 
and Lumsden may be, it is rhythmical and vigorous, now and then feticilons in 
single epithets, now and then dramatic when it grapples with the memorable 
episodes. It is worthy of extended notice; it deserves, as it has received, the 
approbation of Prof. Child and Henry Sweet ; and it need not fear the criticisms 
of chnrch-mice or of academicians in a comer. 

J. A,H. 



ed with Introdncloty Dissertations, Critical Notes, Commentary 
on. By W. GuKiON Ruthbkfokd, A. M„ of Balliol College, 
.«ndon: Macmillan & Co. 1883. 

tulherford's English reviewers. Id a highly commendatory notice 
irynichus, says that Mr. Rutherford 'has quite succeeded in 
msing though somewhat aggressive dogmatism of style of Cobet 
ritics.' It is much to be feared that praise like this has encour- 
erford in his irapp^ia, for his edition of Bafarius is studded with 
narks which will ceriainly earn for the editor the close, if not 
Ention of sundry Greek scholars. It is not pleasant to be told 
rather from want of knowledge and judgment than of native 
4 a), that ' the abundance of [one's] work has given [one's] 
ninance to which, if linguistic tact and careful scholarship are 
IS little right' (p. 79 i), that one 'has flaunted his ignorance 
>. 99 6), that one shows 'his nsual absence of appreciation of 
be Greek language * (p. 109 i), that a certain form is ' a paltry 
it has been accepted by Liddell and Scott '(p. ism), that one's 
making conjectnres ts only equalled by one's boldness' (p. 13 j 
wn proposal is hardly a conjecture ; it is a correction.' This is 
eanlng of passages that have caught my eye in turning over the 
re are more of the tame order. This is not so bad as German 



REVIEWS AND BOOK NOTICES. 87 

criticism at its worst, and there is perhaps too much deference both in England 
and in this country to official ignorance and incapacity, but Mr. Rutherford 
goes perhaps a trifle too far. But if he has shown somewhat more acerbity in 
this book than in the New Phrynichns, he has, on the other hand, tempered the 
fervor of his style, and the average philological mind will not be so much 
disturbed by his rhetoric as not to profit by the valuable work he has done for 
and about Babrius. 

A sharp, clear knowledge of Attic Greek ought to be insisted on, not for the 
purpose of sneering at subsequent developments as so many morbid growths, 
but for the sake of getting into full sympathy with the finest type of the Greek 
mind ; but in whatever spirit that sharp, clear knowledge of Attic Greek is 
promoted, we ought to rejoice even if we cannot agree with the temper of Cobet 
or his admirer, Mr. Rutherford. Later Greek when read with college students, 
if read at all, ought to be read with an incessant reference to the model lan- 
guage, and it is only from a sense of the usefulness of such a process that I 
gained my consent to edit the colorless and lumbering apologies of Justin 
Mart3nr. Indifference as to the period and the sphere of Greek vocabulary, 
Greek formations, Greek S3mtax, is one of the great evils with which an honest 
teacher of Greek has to contend, and this indifference is systematically encour- 
aged by the scrappy readings of early youth ; and I can almost forgive one of 
my own pupils who has steadily declined to teach anything but model Attic 
Greek, leaving the boys to pick up Homer as they would Chaucer in after years. 
Now Babrius is excellent practice for the exercise to which I have adverted, 
and I have often used his fables for the purpose of testing knowledge as to the 
history of Greek words, forms, syntactical rules. As a special student of the 
Greek verb, as the editor of Phrynichus, Mr. Rutherford has been able to do 
good in pointing out late forms and words of recent origin, but this has not been 
done systematically either in commentary or in lexicon, and the educational 
value of Babrius from this point of view has not been fully realized. On p. lix 
of the introduction he gives a short list, * which,' he says, ' every reader of 
Babrius will be able to increase for himself.' His reason for this limitation 
was not only to satisfy his own sense of proportion and to avoid * insulting the 
understanding of [his] readers,* but because questions of percentage are involved ; 
and while * it would be possible to represent numerically the differences in the 
frequency of such violations of usage between a t3rpical Attic writer and 
such an author as Babrius,' it could only be done *at a cost of labor quite 
incommensurate with the advantage.' Unfortunately one must run the risk of 
insulting the intelligence of some readers if one wishes to be useful to a large 
class, and the wearisome task of ascertaining proportions must be undergone, 
if such work is to be considered final. Of course it requires judgment to know 
when statistics will pay, and as Mr. Rutherford has decided that they will not 
pay, nothing more is to be said. If I were editing a post-classic Greek author 
I should not trouble myself to count all his articular infinitives, but in a com- 
mentary on Philostratos, for instance, I should not fail to notice the familiarity 
with which he emplo3rs the most daring constructions of this class, constructions 
which go back to Thukydides and Demosthenes, and I might be at the pains 
to count the rare combinations. Mr. Rutherford's disdainful attitude makes 
it hard to criticise the omissions of his commentary, and I may insult the 
intelligence of my readers by missing a note on 50, 6 : 6 (T ov irpcMaetv dfjnn^ which 



;<i 




'lar '^■w^ WM .1 -r.Tt<Mi: x -"Xi.'ci 

-.- < ■-^-'-r-'. :;*^ 

A'^rr '.i At .< .*i<*Tfnm', ;gni!Tai ^ ii i mn iicsi 
-irr -f . r - .-r-vher -wn^, l ecil lie *wt—nag»« -pnicii '-re 

'f^r^r -^ -,^ ^frm% *o .ma^Tif *tat 

y^rfr." -^f^xtr, ifwt aot i ^rcakin^ iotm if 'lie 'isoes. 

*^at ^Tr .-w ::nn*iii«it »tt1i x t^ioi s«iw.j « die tacts 3 :mc which I luEve 
#?U<^»'i^ri» lUrxk^M It. Mr. 3..ithcrtarri anrs. x Itt. ' esBCcpt in the seme of 
«r/^ /'V 'he -wrrh imv rpfhses in .%it:c; any .•tif i si i u cu o n but ihmt with ^rt or 
/if. r lave no 'M-'erticyn "o 'his is x mle far Greek TTOie cantpontian, a mle, 
hy be v-iv **itch Mr Si/i^nrck iaipTmtlv vroiazes: I reca^ize die fiict tfcat 
<rrc^*>>n* »re cnmparanveiy rare, and, uideeti, I ha^e tried to explain tlie 
phTirtmenrm in my commentary <m Jostin Martyr, ApoL r, 12, ja* which I 
w'll 'nUe the liberty af qaotin<jr •• 'is-piv in the sense of 'say' not * order* 
4V>mTtu>nly 'ak«s 'W nr ci^ in classic Greek; bnt the exceptions are Sar more 
num«*ron4 than one ^bfmld mppose from the way in which the mle is stated, 
an Thuc, 7, 35, a; Hdt, r, 39; 2, 30; Andoci, 57; Xen. HelL i, 6» 7 ; 
a, a, rj ; Cyr. 5, 5, 24; Pl^t- Gorg, 473 A; Legg. 2, 654 A [to which mcy 
he »f*>(ied 673 B]; Clitoph. 407 A; 460 A; Aeschin. 3, 37; Lycnrg. coatra 
J^ocT. 50, to «iy »othing of the poets soch ss Find. OL 7. 6a; Soph. Aatig. 
755, f tc. for later Greek, examples are not necessary. The rale, howerer, 
U not withoot iu resMm, V^irtlv ori^sally gires the exact utterance (iirof). 
5(o in (f omer (r/i/T iisrev). When the hrt form di orutio obliqwa became commott, 
\X WS4 nataral that this form, which is nearest to tfmi^ nnriSs, should be retained.** 
Hfym« of these examples have found their way into the new edition of Liddell 
and H^ Mt^s I^exicon, and others might be added, such as Lys. 10, 6 ; 9, la (M>), 
and l«flio« 3, a9« Bat I know what Mr. Rntherford will say. He will say as 
he \\%% «ftid time after time, that Xenophon does not connt, that Thukydides 
\\ti% to u«e nn * Immature Attic' (p. 8 a\ that Lysias 10 is questioned by Har- 
poliffttion, thotigh Blsss does not know why. Nay, if Ljrsias 10 be prored 
gf-niiifif , Mr Riitheifoid will be able to point triumphantly to a remark, p. 36 
A, In which place he wys " Mr. Gow, Fellow of Trinity College, has drawn my 
attpiuhiu to ths fact that French critics saw in Victor Hugo's works, written 
eUirlt»K hU ii(»)nurn In Jemcy, an absence of • la malice et la ddicatesse Parisi- 
^niip/ niHl n «<linll«r nmh ^mid I have always felt the want of in Lysias' Attic." 
T««tr *o irlinrd m«»t l>i» a punitive curse, and the attainment of it can hardly be 
ton^lttfiftl d«'*lt«l»U. Moi»t pvrsont will think that what was good enough for 
?4ii|*hoKlri, \^hAt Wrts ^mhI tnough for PUto, was good enough for Babrius, and 
VrlU not voyna It a npfilnl fri^th^r in the cwpof the late fabulist that he slipped 
onW vMu f \\\ thi. trjfiud iKaU. g7» 4V Uut I do not wish to go into deUils which 
\\xn\UI tnw^Ut UM\t iU*i ««*UMU of ih«» KmvUmentals of Greek syntax. There are 
^^h^^ ^ vMnu Itt \^huh the exUttvMt vU»e(Y^ attention and commands respect. So 



REVIEWS AND BOOK NOTICES. 89 

we have an elaborate and interesting introduction in which the person of 
Babrius, the history of Greek fable, the language of Babrius and the history of 
the text are discussed. Mr. Rutherford's style is perverse and does not deserve 
the commendation of conciseness which a friendly critic has bestowed upon it, 
but at all events it is not dull, and the collation of the Athoan MS preserved in 
the British Museum gives a special scientific value to this edition. Between the 
Athoan MS and the Vaticanus Mr. Rutherford thinks there is not much to 
choose, nor does he consider Suidas much better authority than the two sources 
mentioned. For his recension of the text he claims the character of conserva- 
tism, but when he does introduce his ' own tentaraina ' he does so with the same 
confidence that marks every line of his work. They are not numerous and few 
of them commend themselves irresistibly. 

The edition has four indexes : i. Index Fabularum ; a. An English Index ; 
3. A Greek Index ; 4. Index Scriptorura and a welcome ' Graecitatis Babrianae 
Lexicon,' due in great part to Mr. H. Duff, Fellow of All Souls College, and 
* intended as an aid to the work which sooner or later must be undertaken, and 
to which so little has been done — the scientific Lexicography of the Greek 
language.' 

Mr. Rutherford's * New Phrynichus,' it seems, has already become a standard 
work of reference in England and his Babrius will extend the reputation gained 
by his previous labors. A little closer study of Chandler's ' Greek Accentua- 
tion ' would have been of service to him, but on this point also an improvement 
is to be noted, though he writes yovpff in the text itself (95, 21) and emends a 

passage (107, 7) with ifui^ov, 

B. L. G. 



Geschichtliche Entwickelung der Constructionen mit irplv. Von Dr. JosBP 
Sturm. WQrzboig : A. Stuber. 1882. 

Dr. Sturm has done excellent service in his presentation of the historical 
development of the constructions of irpivt and while I could have wished that 
he had made some use of the article on irplv which appeared in this Journal, 
Vol. II, p. 465 foll.,^ the coincidences are all the more gratifying to one who 
cares more for truth than for originality. 

According to Dr. Sturm there are two principal periods separated from each 
other by sharp lines of demarcation. In the first period the use of irpiv is 
developing, in the second the development is completed. 

The first period, which means Homer and Hesiod, shows us irpiv in its begin- 
nings. The original construction irpiv with the infinitive was already developed, 
but as subjunctive and optative were just reaching vitality in Homer, no fixed 

> I refer only to the theoretical discussion, for I have since discoTrred that the sutistics of 
my collectors — ^toexpeiienced young men — were fiur from complete. This defect I had hoped 
to make good erenow by the dissertation of Ltlth, Dt usm pariicuUu vpcV qualis apud ortitorei 
Atticos fuerU, Rostock, 1877, but I have not been able to procure a copy. It is some cobs<^- 
tioQ to know that Sturm had done the work over again before Ltlth's disserution had reached 
him ; but he has only given a summary and not the references to the passages, so that I cannot 
make good the deficiencies in Demosthenes, which excited my surprise, if not my suspicion* 
my article (1. c p. 483). 



90 AJiERICAX JOCRXAL OF PHILOLOGY. 



of inficiijre &se as contndistiBgcisbed from sabjimctiTe and optatrre 
«ses cxnld be attained. Hence tbe infinitrre coald be used tndifferendj after 
Begathre and afto* affimative ^-Vini^* ]Ii,3X is nsed vith tbe in£. as well as 
cpi^-, bat T-ocir is eying oat, is nerer nsed viih tbe sabjancU^ve, and does not 
appear in Hes:od. Tbe vefy care ssbjanctrre cotistractions betray tbe old 
parat^TW In Hocner rpiv nerer takes a* or sir ; it is stiU pvelj adTerbiaL' 
Hence it is not ret saited to introdnce a dependent dansc in tbe indicattre. 
The req;iired sense b reacbed by a^f or by rp^ j' fcr (H, tbe latter fonnnbi not 
appearing in tbe sabseqaent period. 

Tbe second period esnbcaces all tbe otber aatbon of tbe classical time to 
Plato inchLsrre. Tbe conditSooal reladon aras felt nkore and nore as tbe snb- 
janctire was derelopcd more and nore. Conse-qaently tbe infinitire was 
restricted miiclv to the aSnnatiTe relation, and oc tbe ocber band tbe connexion 
wiib tbe co=i:tk>nal sentence became rety close. Parataxis ranishes; Tp<jr 
like other cocjzsctions takes tbe panicle ij-, becomes a fall conjunction and 
admits the icdicatire. Tbe present infinitire is ased more freely tban in tbe 
first period, in vbicb tbe aorist infinitire was almost excIiisiTely employed. 
Tbe peifect also comes in. Tbe dawm of tbe new period is seen in Hesiod. 
In comparison with Homer tbe scbjanctiTe b mote freqnently employed, and 
on tbe otber band tbe sbadow of tbe old period fills bere and there on Ecripides 
and Herodotos. 

In tbe second period Dr. Stnrm distingaisbes three gtxjcps. Tbe first is 
repcesented by tbe writers of tbe New Ionic dialect. VLsj t and T^^rrtoav 9 
are ased not only with tbe iafinitrre bat also with tbe indicatrre and sab'onctiTe 
aorist. Tbe opcative is not foand ; tbe indicatiTe is nsed only after negatire 
sentences, and then tbe conjanction tat is always strengtbened by tbe particles 
7r <^ or <*T. Tbe present infinitire is rare. 

Tbe second grocp embraces the poets and Thckydides. O ^ has ranisbed, 
not to reappear in oar field of obsGrration, nor do we find it in Attic inscriptimis. 
Tbe indicatrre is ased after aSnuatrre as well as after negatrre sentences. In 
Thakydides the particle <V is ased only after amrmatrre sentences^ except ooce.' 
Tbe aor. opt. reappeais. Theo^ts is the first to ase it in assiaailatiolu Tbe 
snbjanctire present occaxs once in TbokydideSs once in tbe fra±:ments of the 
comic poets. The present and tbe perfect infinitire becoaae relatrrely more 
comaaoo, the latter especially in Earipides and Aristophanes. 

Tbe third grocp is made op of Xenophon. the orators and Plato, and shows 
tbe foUowiag pecdiariries : \S\ t^v*- with the indicatrpe b nsed only after nega- 



TW panactic oH^ ti^pir vtt& scbj. 'nifH W <ic3aed. See ^ke |miii>^i i cited ta L. aad 
'% LexiccB *t& e>i~V Bat h is hard to s«e hov dw cc«stra(Ctk« cas be p*ve}y x l i obiU 
tScrsB^rnt, WbereT«r v^^ tt pre ce ded by v«u>, mpmm^**, W n», U b o« its way to tbe 
aad t^ d: Jereace faere as dsevbere l i eta e ta Hcoacr aad Uier Greek is tibe diJ 
aad xssTCCsaliXj^ D^^ vit^ tbe taT, tbe crr^auit cottstroctsco, is ii 
taoL 

* Tftere b ao fa*. eitt-Ue 1 1 ai rn far this, aad besides ^« imiMiM rests oa a sad tJ ttwfa, ^e 
S>Me bi^aiier tbax RAhaer Bade, as I potaaed act L c p. 469. ».$«. x; i,tit,s: 3^ >9. »; 3. 
7 are aegar^Tc asd aoc adtfataxirc:. 7, ]^, t awi 7, 71, 5 sbcv persistcscy vtucb br i ags oat 

;1 ' wan. S«e tbe pm^iray fircaa Aasckiaes cud bc&cv. Wh£> cc Tte c t j^ cbe faalts cf 
I B3R ao« ooiit to o wre ct my c«a iaadvcrteaciesL. la tbe amde cised p- 459^ rj L i 
r'Tpr. 5- • -*****^*««d«3, WH. 7- • ««*»«^^»*^* T^ feise ciiaac« 
■»r*<^VP-4T9.L3 



REVIEIVS AND BOOK NOTICES. 91 

tive clauses (except Aischin. i, 64), and never takes a particle (except trpiv ye 
X. Oik. 7, 7). (b) the indicative is most frequently used by Xenophon. In the 
Isokrates it is chiefly employed in ov irp&repov iiraixjaro irplv and similar phrases. 
In Plato it is dying out before eoC' (0 In unreal sentences the indicative is 
found only in the orators and Plato. Euripides satisfies himself in two pas- 
sages with the original infinitive. The present infinitive occurs with special 
frequency and reaches its height in Xenophon, and the present subjunctive 
and opt. are comparatively more common, (d) Instead of irpiv we find for the 
first time irpd tov with the infinitive. Jlpdrepov ^ seldom does duty for irplv, 
and is limited to the infinitive. {<) 1^ irplv, a new combination, emerges in 
Xenophon. 

So much for the facts. As for the theory, Dr. Sturm argues against the 
explanation of rrplv with the inf. as arising from parataxis, and well he may, for 
it is sheer nonsense. The infinitive must be dependent, but how dependent ? 
To the prepositional theory he is utterly opposed on the ground of the historical 
sa//us. Such a construction would require the article, an old objection. Besides 
neither nplv nor irdpo^ is used as a preposition in Homer. Sanskrit analogies, 
such as purd with the inf., suggested by Wilhelm and taken up by Monro, are 
made doubtful by Jolly's refusal to consider these Sanskrit genitives and abla- 
tives as true infinitives. The omission of 17 is extremely hazardous in view of 
the fact that Homer uses irpiv ^ only twice and nplv with the inf. 79 times. 
Schumann's parallelism between irplv with inf. and f^ &are with the infinitive 
is condemned as unhistorical. Uplv is fullblown — Chtts at best emergent 
Passing by other theories of which enough has been said in the article already 
referred to, we come to the one on which Dr. Sturm builds. The infinitive is 
the limit of irpiv, Uplv yevia^ai is * sooner with reference to.* This is the 
explanation given by Wagner, the explanation adopted by Holzweissig, as one 
of the certain results of comparative grammar, and there is no denying that it 
does not require so wide a leap as the prepositional theory. And yet the 
parallels are not altogether satisfactory. In 642: dfulvov izavroia^ aperdc, 
fjfUv irSSac v^k fidxec^at, in A 258 : ot irepl fihf pov^^ Aava6v, irepl 3* karkfidxeo- 
^oi the preliminary accusative saves the construction ; C 230 : fieli^ovd t' elatdietv 
Kol irdmrova is nearer, and so is v 33 : dairatrlu^ d* &pa r^ Korkdv tp&o^ fje240to 
66pirov eiToixf<r&ai^ but the position of irplv so far away from the verb, to 
which, on this theory, it really belongs, is unexplained. I cannot help 
thinking that Schumann, however wrong historically, was not so hopelessly 
wrong grammatically in his parallelism between irplv and Ijote, 'Qtrre as Sturm 
himself has pointed out is in the same line of development, though later. We 
must always start with the final use of the infinitive, and if irplv with the 
infinitive is to begin as 'prevention' and end as 'priority' we shall be 
nearer the truth than if we begin with some such abstraction as * in Bezug 
auf.' But whatever the origin, the question of the prepositional feeling re- 
mains untouched. Nobody considers <^c with the ace. a preposition, and yet 
it is in feeling a preposition. Nobody considers 'than' a preposition, and 
yet it behaves as such. We must learn to respect the conceptions of the users 
of language. 

The combination irph / bre S^ would seem to postulate a quasi-prepositional 
use of irplv, and one might be tempted to compare the history of ecjCt fJ^XP^ {^XP^) 



93 AMERICAN JOURNAL OF PHILOLOGY. 

and perhaps even lert. Dr. Slurm sticks to the original parataxis, irpiv yt 
belongs (□ the first pait, Arc ifi begins a new sentence. So M 436; ^ ^ rCiv 
fir! laa ji&xl Ttraro irW*^ Tt | irpiv y irt i^ Zevf miiJof Smiprtpov ''Emopi iiuav. 
mast be inteipreted ' The battle hung in the baUnce— at leut before : vhen 
now Zens gave Hector [he Tictor7 ^ until Zens gave H. the victor;.' To Ihii 
it majr be objected that if npiv bad already become almost a fonnnla with the 
infinitive, there is no reason whjr ihe ^nalogj should not have been extended 
to the finite constructions. 

Ai to the prevalence of the aorist infinitive in Homer, Sturm limply accepts 
what Cavallin had said about the tenses of the infinitive. We do not need to 
be told by Cavallin that the tenses of the infinitive have to do primarily only 
with the kind of time. That has been a common possession for aeveral decades, 
and my objection to this statement is that the student is put off with a foimnla 
which he does ifot always know how to apply, and I think it well (o give 
emphasis to the negative element of irplv in order to bring the use of the 
aoiist inf. more clearly to the consciousness. The prevaleace of the finite 
aorist after the negatived irpfv ii simply in accordance with the general needs of 
the temporal sentence. Overlapping ' action is less common than clear priority 
and posteriority. Hence irplv dv with the pres. subj. is rare, and Dr. Slurm 
has actually denied its existence in the tragic poets in spite of Sopbokles, Phil. 
1409. 

Dr. Sturm defends the passage in Solon (36, 11) : oOr' av narfaxr i^/av otr' 
iirataoTO, | irplip &v ropifof iriop ifWj jiXa on the ground that the author had 
in his mind a familiar proverb with the future or the optative with Av in the 
lead (»{> wabaerai or obx fiv nalntairo). This is a kind of npraatnialu) and is the 
only possible explanation, but not satisfactory in an unreal sentence as I have 
said (A. J. P., I 458) where I suggested iipiv iwTupdfof iriap iftiAtv fi^, or 
better h^tixn ^dXo, referring to the very passage in Eur. Ale. 373 which Dr. 
Sturm has cited. I am glad to see that he bai accepted FOrstei's hito^at for 
luTnu in Simon. Amo^. i, la, a verse treated at length in my article cited (p. 
468), though neither FOrsler nor Sturm has tried to account for the error. 

I wonid add that Ihe spread of rplr ^ in late Greek seems la be due partly 
to the mechanical grammar of the post-classic period, partly to the influence of 
Herodotos. The final step, which we find perpetuated in modern Greek, the 
Bse of Tpiv 7 with subj. in all classes of sentences, aSBrmative and negative, 
■■ ■"'• -noticed by Dr. Sturm, although it might fairly be considered to lie in 
of developmenL 

have previously intimated, the practical results of Dr. Sturm's treatise 
the main been anticipate)), but it is one thing to have laid down the 
lines of usage, another to show the history of the construction with 
ive proofs, and I should be the last one to withhold from Dr. Sturm the 
: praise for his laborioas and in the main careful piece of work. 

B. L. G. 



REVIEWS AND BOOK NOTICES. 93 

Lysiae Orationes XVI. Edited by E. S. Shuckburgh, M. A. Macmillan & 
Co., 1882. 

The orations comprised in this edition are those which are numbered 5, 7, 9, 
10, 12, 13, 14, 16, 17, 19, 22, 23, 24, 28, 30, 32. The selection is probably as 
good a one as could have been made. It nearly coincides with that adopted 
by Frohberger in his school edition » which, however, Mr. Shuckburgh does not 
seem to have used or to have known, llie text used is substantially that of 
Scheibe in the Teubner series; but occasional variations from it are made, 
which are for the most part noted at the foot of the pages. The editor thus 
describes his own views in preparing the book : " My object in the commentary 
has been to bring before the student, as far as possible, the circumstances, 
social and historical, in which the speeches were delivered ; and at the same 
time to direct his attention to an accurate study of the language." Of these 
two purposes, it will probably be thought that the former has been most suc- 
cessfully accomplished. The editor has adopted an excellent method for 
enabling students to follow with interest the arguments of the speeches. An 
account of the circumstances under which each was delivered, so far as these 
can be ascertained, is placed before the notes in each case ; but in addition to 
this, throughout the speeches themselves at frequent intervals, the editor has 
inserted spirited summaries of the argument of the adjacent sections. Such 
historical or antiquarian information as is necessary for the understanding of 
allusions is supplied in the notes with succinctness and accuracy for the most 
part ; and there are five useful appendices, the most elaborate being on the 
usurpation of '* The Thirty." On the whole and notwithstanding all the short- 
comings which have been noted, a few of which will now be referred to, this 
book may be cordially recommended to instructors as very much more useful 
than any edition with English commentary which has heretofore been accessible 
to American students. 

In the text itself, set up as it was from Scheibe*s printed pages, the editor 
seems to have trusted too much to the diligence of the proof-reader ; for such 
words as xretJi^, abfiaroCt iirerdAfifiaet n;y;tdvc^, occur not with extreme rarity. 
But the same scapegoat can hardly bear the blame of *C. Scheiber in the 
Tmubner series ' of the preface, or of the tt which, on p. 8, Baiter has assumed, 
perhaps in emulation of the /^ which Sauppe enjoys. Indeed, there are too 
many marks of hasty work in both text and notes. The citation made in the 
very first note is rendered unintelligible by the omission of a word. On p. 194, 
in a note on the fiiroiKot, the writer tells us that they were " subject to military 
service, though they were not admitted to serve as hoplites"; and in confirma- 
tion of this we are referred to Xen. Vect. 2, 2. If that passage is examined, it 
will be seen that Xenophon expressly asserts, and at the same time deprecates, 
the liability of the fitroucoi to serve as hoplites ; and Boeckh also, who is referred 
to in the same note, says that they served as hoplites, at first only among 
the garrison -soldiers, but at a later time in campaigns. The statement, also, 
that they ** were liable for any offence against the various enactments concern- 
ing them to be sold as slaves,*' is not supported by the passage of Boeckh 
referred to, where we are told they incurred such liability only if they failed to 
pay the fieroituov, and is denied by Hermann (p. 226 of the London edition. 



94 AMERICAN JOURNAL OF PS/LOLOGY. 

1836, which U the tne Mr. ShnckbsiEfa hu mule ase of ), who sajrs lb>t thej 
" weie (otd as slaves only whea (her usnmed the pecnlUr priTileges of actaal 
citizens, onitted to p>y the tu, and, pnbablr, if thejr neglected to choose a 
patron." The note on the Siair^rai, on p. 331, it eqnalljr insccnrate. The 
question as to their nomber cannot be regarded as entirely tcttled as jret, bnt 
probttbly no one will be fband to snpport Hr. Shnckbui^'s statement that 
there were only forty, four being chosen annoally bj lot from each tribe. See 
on this matter Perrot Drrit FuNU i'Alkhui, p. 1S9 £ The assertion, too, that 
"before the time of Demosthenes all cItiI saits were heard fiisl before one of 
them," is mnch too brocdljr Mated. Hemum, §145, goes fully as far as the 
anthorities warrant, in saying that the system of employing arbitrators " freilich 
spltei nm der damit Terknapften geringeren Koiten and GeEahren willen eine 
solche Ansdehnang erhalten hatle, dais dieselben Rtrmlich als eine erste 
Instanz in den meisten PiiTatprocessen betrachlel werden dUrfeD." On p. 303 
we have a strange inteipretation of ifUutv saaot ytymhni • tbey seemed to be 
of a somewhat inferior character.' The true explanation is quoted front Bremi 
(cf. Cobet,V. L. p. 158), that the phrase means 'worse-bom, not true-bred 
Athenian,' but is delibenitely rejected, and we are told that we must suppose 
"a phrase nmuc ftyvta0<u equivalent to xoKOi Ix"*, ' to be ill,' i. /. in beha- 
vior, reputaiion, etc" In Or. XII 31, where the ontor says that Eiatosthenes, 
though sent, as he asserted, by the Tbiity to arrest Polemarchus, might easily 
have declared that he did not meet him or had not seen him, rovra yap oir' i>E;'j[iiv 
obrc ^Aaavav tljcti", itrre fO/J" inri Tim i;fipini ^oiAo/iivuv o'ldv f thitu i^iityx&^vai^ 
we are told, p. 337. to translate the last words 'did not involve or admit of 
Tcfulatiun ot examination by torture.' On p. 246 we read : " for the name of 
Aristocrates as a leader of the moderates we are indebted to Ljsias, not Thn- 
cydides"; but in ThucVIII 89,3, we find 'A/wmNvdr^vritf ZwUioti mentioned 
alons with Theramenes, and yet Thnc VIII 90 is referred to in this very note, 
: told to translate rdv ffarpof ir/iof pijTpbc vijmni, • his great- 
on his mother's side.' This error is due to the mistaken 
i just before Co the elder instead of to the younger Alcibiades. 
}me confusion in the passage of Isocrates referred to, or the 
II as the father of Cleinias must have been named Alcibiades. 
i"enneirs Pindar, Pyth, VII. ' 

e is left to speak of the grammatical notes. Keference is 
o Prof. Goodwin's books.and occasionally to the Greek syntax 
Clyde. Acoasiderablenumberofpoints have been marked for 
ire chiefly faultsof omission. As a single instance we may take 
) troUy ^Stov iym/iiu tlvai vxip uv i/itif iitiaxm avrnirciv, ) 
joJiHWiv diroloj-^riott&M, where jt^Sun — j is passed without re- 
wben a note is given it is sometimes not as dear as coald be 
; XII 83, ri yip dv iraBAirret fuapi 1-^ ifiov ct^tfav Tin Ipyuv 
e (p. 349), " ' will they have fully paid the penalty they de- 
periphrasis for a perfect optative see Madv. giSo d. It refers 
itlon as to things that would tien be past." 

DIM, tftet A/ciUMJti u Ihc nunc of Ihc gnnd&thtr of CIriniu, leSTinf 
Bbi ihil /K Ikh/amlfy then li no Improtabilily that tlic uD> of both 
HFsbytbc &CI thu iba Aldbiado of ilig tpecchwu ibeKKiafihcbBoiii 



JiE VIEWS AND BOOK NOTICES. 95 

Mr. Shackbargh*s opinion of Theramenes (p. 245) may be quoted to conclude 
this notice. " I think it is clear, from a careful review of our authorities, that 
Theramenes was an honest man. But he was a philosopher and a doctrinaire, 
and had a Socratic ideal of a perfect state which, both in the time of the Four 
Hundred and in that of the Thirty, he thought he saw his way to realise, but 
was quickly undeceived by the development of selfish aims in his colleagues. 
As, therefore, he S3rmpathised neither with the prejudices of the Democrats, 
nor with the self-seeking of the Oligarchs, he came to be trusted by neither.*' 

C. D. Morris. 



Gorboduc, or Ferrex and Porrex ; a tragedy, by Thomas Norton and Thomas 
Sackville, A. D. 1561. Edited by L. TouLMiN Smith. Heilbronn : Ver- 
lag von Gebr. Henninger. [Englische Sprach- und Literaturdenkmale 
des 16, 17 und 18 Jahrhunderts ; herausgegeben von Karl VollmOllsr, I.] 

This series, which an enterprising German publishing house has undertaken, 
and of which Gorboduc is the first number, will meet a real want of scholars, 
in England and America as well as in Germany. The series could also not 
have been better begun than with this piece, the first English Tragedy. To 
many it was not accessible in Dodsley's " Old Plays," and was unfortunately 
omitted in the last (1874) edition of that collection. The other editions are 
somewhat rare. Arber announced it in 1869 and subsequently, as among his 
forthcoming Reprints, but it did not appear. The edition now issued is not 
only handy and inexpensive, with clear type and on good paper, but has also 
an additional value for scholars in the full collations of the editions of 1565 
and 1590, appended at the foot of each page. The text itself is that of the 
(authorized) edition of 1570. The editor, Lucy Toulmin Smith, a contributor 
to the Anglia, and joint editor for the Early English Text Society of English 
Gildsy has also prefixed an English Introduction, and added explanatory notes, 
also in English. The work of editing seems to have been well and carefully 
done ; the collations and Introduction are especially good. The notes are in 
part less valuable. On the one hand explanations are given such as no English 
scholar, and in many cases no intelligent reader of English literature needs, 
and on the other, interesting questions of grammar, phonology, etc., are only 
slightly touched upon or omitted altogether. Where is the need, for instance, 
in an edition intended to supply to scholars the materials for a critical text of 
the oldest English tragedy, of explanations like these : reek^ to heed, to care 
for (390). marches^ borders (414), avowed^ a-vowed, promised on oath (574), guer- 
don^ reward, recompense (1437), in fine^ in the end, at last (1539), want^ lack 
(1 71 5)? There are surely enough interesting forms of speech inviting discussion, 
and questions of all sorts connected with this play, to make us regret such a 
waste of valuable space. 

L. 465 to reue me halfe the kingdome^ the editor is hardly correct in saying : 
** mte me, the preposition is suppressed, compare 1. 513 /^ reauefrom me my native 
right** It is the new use with the preposition that is * suppressing * the time- 
honored dative of interest, cf. 809 To reaue me and my sonnes the hatefuU breath, 
\bqVyCourage is said to have been ' brought in by Chaucer.' No doubt he helped 
bring it into vogue, but he did not introduce it. The word occurs in Early 



96 AMERICAN JOURNAL OF PHILOLOGY. 

English Alliterative Poems, ed. Morris, and in the Ayenbite of Inwjt, both of 
which were written before Chaucer was at work on the Canterbury Tales ; and 
Robert of Gloucester's Chronicle (end of the 13th century) has corageus. 

On p. xxY the editor has some good remarks on the instructive differences 
between the earlier and later edition in the grammatical forms. It is to be 
regretted that such differences are not oftener pointed out and explained in 
the notes. For instance, where the two earlier editions have forms like wast 
(919), or hast (935), the edition of 1590 usually adds /, waste^ haste^ etc. ; 1032 
only the first edition has wast^ the others waste. For the significance of such 
final silent /, as regards the lengthening of the stem vowel a, ' a feeling which 
perhaps came in towards the close of the 15th century/ see Ellis, Early English 
Pronunciation, p. 567. 

P. XXV, the d which represents Anglo-Saxon IS in many words is spoken of 
as 'the old d^ and 383 (should be z^2) furder\% referred to as an instance where 
the ed. of 1590 has further. Under line aio we find the note: " Furder^ the 
A. S. "5 was often retained by the early printers as d. It frequently so occurs 
in the ed. of 1565.'* This seems to misapprehend the facts of the case com- 
pletely. Northumbrian Anglo-Saxon, Orm, Hali Meidenhad, La^amon, and 
many other early monuments exhibit frequently d for "5 (p). It is extremely 
common in Middle Scotch also, and Murray (Dialect of the Southern Counties 
of Scotland, p. I2t) says: "the d (before r, as in furder) was pronounced, I 
believe, neither as in ^are nor in Mere, but with an intermediate sound, the 
front or dental d (formed by touching the teeth with the tip of the tongue), still 
used in the same words in the Northern English Counties." 

The references in our edition to Anglo-Saxon forms contain a few errors : 
behatan (166) should be hehdtan ; 762 an A. S. form seyl^ reason, is cited. Both 
word and meaning are incorrect Bosworth does indeed give scyle^ difference, 
variety, distinction, on the doubtful authority of Somner,but there is no accred- 
ited A. S. form of the word. It is derived from the Old Norse j^7, distinction, 
discernment. The meaning attached to vnskilfuUi^ti ), * wanting in knowledge,* 
should be * wanting in discernment,' and the note to 201 should read not 
'reasonable* but 'discerning.' The instances in Ancren Riwle and else- 
where, in which skill has the derived meaning ' reason,' only confirm this. 

1002, the form abycyan, nearly as common in A. S. as the abicyan given in the 
note, would have been a better illustration of bye, to pay for, since the y repre- 
sents original u, 

u6o, an A. S. verb racan^ to scrape (rake), is referred to. It should be raccian. 
The form raken (without 1) is found in very early English, however, though 
connected with Old Norse raka. Compare Matzner, Altengl. Sprachproben, I 

62, 1. 2132. 

In addition to the editor's remarks on alliteration, pp.xv-xvi, it is worth 
while to note that in two cases, at least, the edition of 1570 preserves the allit- 
eration complete where both the other editions miss it. This can hardly be 
an accident, and tends to show that this authorized edition was prepared for 
the press with some care. The edition of 1590 is based upon that of 1565; 
which accounts for the recurrence in it of the non-alliterative forms of the lines 
in question (350, 538). Bodenstedt remarks of Shakspeare's Macbeth that the 
words bhcd and bloody * reappear on almost every page, and run like a red thread 



REVIEWS AND BOOK NOTICES, 97 

through the whole piece.' A far more surprising frequency of the words is 
noticeable in Gorboduc. In the 4th and 5th acts (Sackville*s part) they 
occur 33 times. The words egal, egalntsse = equal, etc., common enough in 
Middle English, but rare in Shakspeare (cf. Tit. Andron. IV 4, 4), occur very 
often, e, g, iii, 220, 250, 255, 270, 336, 412, 516, 855. 1139, 1159. Even Chaucer, 
who uses both forms, egal and eqtials^ is not nearly so partial to the former as 
the authors of this tragedy, nearly 200 years later. 

In conclusion we heairtily recommend this, in the main, excellent edition of 
Gorboduc to all scholars and lovers of English literature. 

H.W. 



Sammlung FranzOsischer Neudrucke. Herausgegeben von Karl VollmOller. 
Nos. 3, 4, 5. Heilbronn : Henninger. 

This collection is one of the most important that have yet begun to appear in 
any department of French philology. The difficulty of getting texts at a 
moderate cost to work from has been the chief drawback to the progress of 
middle-French study. As the transition period from the old to the modem 
luiguage it is now beginning to claim the attention of scholars that it deserves, 
and in a few years, with these new facilities for investigation, we may expect 
rich results drawn from them for the department of grammar, and especially 
for the historic development of French syntax and versification. 

Numbers I and 2 of the collection appeared in 1881, the former containing 
Lc Festin de Pierre ou Le Fils Criminel, by de Villiers, and edited anew by W. 
KnOrich ; the latter, Trait^ de la Comedie et des Spectacles, by the celebrated 
Armand de Bourbon, Prince de Conti, and newly edited by Prof. VolImttUer 
himself. We now have before us the three following numbers of this interesting 
series, which, together with the sixth, are edited by Prof. Wendelin FOrster 
of Bonn, and consist of Les Tragedies, de Robert Gamier (1534-90). All 
three of these volumes are faithful reprints of the third general and first com- 
plete edition of Gamier's works, published at Paris only five years before his 
death. They are furnished with the variations of all preceding general editions, 
mod are to be followed in the fourth volume (No. 6 of the series) with a short 
glossary for the whole set, containing all words not found in Sachs*s French 
Dictionary. 

It was the year in which Ronsard, the most celebrated French poet of the 
sixteenth century, died (1585) that the first complete edition of Garaier's works 
was brought out at Paris by the then celebrated publisher Patisson. It is this 
edition, as annotated and revised by the author himself, which Prof. F5rster 
has made the basis of his text, adding thereto not only the various readings of 
the general collections, as just mentioned, but also those of the single issues of 
each tragedy — Porcie i568,Hippolyte 1573, Coraelie 1574, Marc Antoine 1578. 
La Troade 1579, Antigone 1580, Bradamante 1582 — with the exception of Les 
Ivifves, which, so far as is now known, was never published separately. 

In the edition of 1 585 these tragedies are not arranged chronologically, but have 
the following order: Porcie, Cornelie, Marc Antoine, Hippolyte, La Troade, An- 
tigone, Les Ivifves, Bradamante, and VollmOller in re-editing the text has divided 
up his material so as to give it to us in as nearly equal parts as possible. He 



98 AMERICAN JOURNAL OF PHILOLOGY. 

therefore puts Porcie, Comelie, and Marc Antoine in VoL I ; Vol. II comprises 
Hippolyte and La Troade ; Vol. Ill, Antigone and Les Ivifves ; while VoL IV 
will contain Gamier*s chif-tCctuvre^ Bradamante, with a short biographical 
sketch of the author and the vocabulary noticed above. To the first of these 
pieces (Porcie) all the orthographic variations will also be given, that the 
student of mid-French may be better able to appreciate the fact that there did 
not exist in the XVI century a regular, uniform mode of writing. 

Gamier wrote eight tragedies in all, and of the seven to which the general 
reader here has access he will find Les Ivifves the most original, the most 
interesting and altogether the best It is here that the author cuts loose from 
his servile imitation of Greek authors, which so strongly characterizes his pre- 
ceding works, and stimulated by a noble sentiment, draws more upon his own 
individuality for the treatment of his subject, which, as he himself thinks in his 
dedication, ought to appeal in a peculiar manner to the best impulses of the 
human heart. He was an intimate friend of Ronsard, who was most lavish in 
the praise of his works, as may be seen in his letter prefixed to La Troade : 

'* Si Bacchus retoumoit au manoir Plutonique, 
II ne voudroit Eschyle au monde redonner, 
II te choisiroit seul, qui seul peux estonner 
Le theatre Fran9ois dc ton Cothume antique." 

Marot, Du Bellay, Rabelais and Montaigne, all contemporaries of the author 
of Les Tragedies, highly appreciated his literary productions, and that they were 
extensively read is shown by the fact that in the first two decades of the XVII 
century they passed through about thirty editions. His tragedies mark an epoch 
of remarkable advance for the French stage, which has been justly characterized 
by Adolf Eberts in his excellent Entwicklungsgeschichte der Franz5sischen 
TragOdie, where he has devoted to this poet a brilliant study with reference to 
his paramount importance for the development of the French drama. 

A. M. E. 



GULIELMO Studemund. Due Commedie Parallele di Difilo. Torino : Ermanno 
Loescher. (21 pp. with an apographum Codicis Ambrosiani G. 82 sup. 
rescript! paginae 244.) 1883. 

On the 28th of September, 1882, at the thirty-sixth meeting of German philo- 
logians and school-teachers in Karlsruhe, Prof. Wilhelm Studemund delivered 
a most interesting address on two parallel comedies of Diphilus. An Italian 
translation of this address by Dr. Aristide Baragiola, prepared for the Rivista 
di Filologia ed Istruzione Classica, forms the subject of this notice. 

After showing that the fertility of production of the famous Greek poets of 
the new comedy, Menander, Philemon, and Diphilus, was due largely to the 
employment of similar plots or motives, spiced with a variety of incidents and 
characters, Prof. Studemund calls attention to the frequent role which shipwreck 
plays in these plots, and the final disentanglement of the knot by means of an 
dvayv^pujig^ often brought about by means of toys {crgpundia) or trinkets belonging 
to the person in whom the chief interest centres. Of this sort is the play 
entitled Rudens, which was composed by Plautus, probably in the last decade 



REVIEWS AND BOOK NOTICES, 99 

of his life, in imitation of a Greek original of Diphilus, the Greek title of which 
is not however given in the Prologue. Prof. Studemund presents a brief outline 
of the play, and shows that it must have followed the Greek original much 
more closely than the burlesque Casina, which is based upon the KAJfpohfievoi, 
also a comedy of Diphilus. Now the Vidularia (Wallet-comedy), of which 
considerable fragments are found in the Ambrosianus and scattered through 
the Roman grammarians, shows a striking resemblance in its situations to the 
Rudens, so that the latter might just as well have the title Vidularia, and the 
very fact that it is not so called would seem to indicate a later date of compo- 
sition. The Rudens it will be remembered gets its name from the rope wound 
about the vidulus which the fisherman Gripus has hauled up from the sea in his 
net The scene of the Vidularia like that of the Rudens is laid on the sea 
coast. There is a fisherman, Gorgo, a shipwrecked youth of good family named 
Nicodemus, an evil-minded slave, Cacistus, who fights with Gorgo for the pos- 
session of a vidulus dragged up by the latter's net. The vidulus contains a 
ring which eventually brings about the recognition of Nicodemus by his father 
Dinia, who fortunately lives close by. Other details, for which there is no 
space here, by their surprising correspondence with the Rudens, make it prob- 
able that the original of the Vidularia was also the work of Diphilus. But Prof. 
Studemund has raised this probability to a certainty. With a patience and 
indefatigableness of which few men would be capable, he has succeeded after 
repeated efforts, which in all consumed quite a month, in so far deciphering a 
page of the Ambrosian palimpsest as to make it clear that it contained a Pro- 
logos to the Vidularia. The seventh and eighth lines, although by no means 
every letter can be read, have been reconstructed by him with great acumen 
and great probability as follows : 

Sc(h)edi[a haec] vo[catast a] g[r]ae[co com]o[ediaJ 
[P]oeta ha[nc] noster f[ecit] V[idularia]m. 

Now comes the interesting part, ^x^^^ ^s a comedy-title is attested but for 

one Greek poet, and that poet is Diphilus. The Etymologicum Magnum has 

preserved one verse of the play (cf. A. Meineke, Fragmenta Comicorum Grae- 

coTum, I, p. 456; IV, p. 410; V, p. cccviii). The appended apographon of 

the page in question of the Ambrosianus is a marvel of painstaking accuracy. 

The discovery itself is one which will interest all Plautine scholars. The labor 

which it has cost will probably be appreciated by few. If Prof. Studemund 

could live always, with his keen vision unimpaired, we might hope for many 

palimpsests to yield us such surprises. Meantime we should be only too grateful 

if the punctilious * Thensaurochrysonicocrypsides ' would vouchsafe us at once 

his apographon of the Ambrosian palimpsest, and give us the fruits of his future 

vacations in Milan in the form of addenda or corrigenda. 

MiNTON Warrkn. 



AltfranxOsische Bibliothek. Herausgegeben von Dr. Wendeun Forstbr. 
Dritter Band. Heilbronn : Henninger. 

Three volumes of this important collection have already appeared, viz. Vol. 
I, Chardry's Josaphaz, Set Dormanz und Petit Plet, an Anglo-Norman poem 
of the XIII century, edited by }. Koch ; Vol. II, Karls des Grossen Reise nach 



ICO AZlESrCAN fOrsy^AL OF PHILOLOGY 



Jerc^lem mri CinstantnitTreL 211 'TId French poem of the XI ccntarr. edited 
y T^rt. E.iward Sosch^intz if jreiiswaid . VoL IV, Lothringischcr F-a^rer. in 
Cii French fnnslaticFn it the XIV ueiuury, *tiited br the ^at^. r-w^^ lamented 
F-edrc^ Arteistedt. We 'lacre jcre aclore as the '■him volmiK of the series^ 
C c^avran, u. tfmn r<'>si3cher Roman, nach icrOxiord«' Handachrift» HodJ. Hattan 
rco Her:ni5i^e^hea vtm Prof. K.Tii V-jilmt^Uen <5f the Uni-reraty of GtHtingen. 
T'lis l(tiile:an -odefc .s In -Tnail ocraxo fomu oontains ro3 leares, and both 
"'mm "he lan]^.-u^ ind trom iiatonc rgferfftres fotmd in it nrast be assigned to 
*he -T^rch :n the thirteenth jenmrv when ferosalem was in trossession of the 

I- * ^ t 

C'lnstions, "hat is, between r22q and C244> In a portioii of the MS oubiished 
n ^a^Land as &ar back as rtkx>--+J. J. Cjnybeare: The ronsooe of Octavian^ 
E*Tipt?mr jf Rome, abnd^^ from a iiS '7f the Bodleian Library. Oxford, tSo9^^ 
•^he eriiror attempted to show that the author was an Angio- Norman and that 
•he po«n was written in England, but according to the in v esti garian q£ the 
!an,7aage carried mt in the jies c at edition there can no longer be any doubt 
j>iint 'ts be:ng onginally a Pi card production, with certain Tariations of fonn 
"hat have naturally crept mto :t throngh copyists. The author is net known,, 
^ut the poem is we have it here is the work of an Anglo- Norman copyist, who 
hr.s mixeri ap Norman with Picard fbrms» in such a way. however, that the 
r'-^.nal iiaiect is easily recoguizabie. It is compcscd in regular eight-syllable 
verse, Thich has been frequently tampered with by the copyist, who has seen 
lit vimerimes to increase, sometimes to cut short the legitimate number of 
5v i.a^^ les. 

There is a -- hmwm Je ^'mt, Florent et Octarien, belonging to the fourteenth 
century, and yet onpubiished. and of which three MSS exist in the National 
LL>rary at Paris. With this later treatment of die same subject ProC VoUmOUer 
compares the Oxford teat and nnds them to agree in all essential points up lo 
near the close of the iarter, which rapidly comes to an end, while the chanson 
g:es on, according to the style of that time, and brings in all sorts of extraneoos 
matter connected with secondary personages^ repetitions^ long drxwn-oat tales, 
ctc^ etc From this striking agreement in the main line of the story the editor 
crncludes that both documents are based upon an old ckoMson de grste, prob- 
a:/iy belonging to the twelfth century, and which is more CliLhfally reproduced 
in the Octavian, published here, than in the later chanson of the fourteenth 
century. ExtcnsiTC extracts from the latter are to be foond in the Histoire 
litteraire dc la France, VoL XXVI, p. 334 ct seq. (i373), where oar present text 
is simply mentioned as "^quatricme manoscrit de Florent et Octarian." 

Dr. Sarraain of Marburg is now at work on an edition of the middle-English 
Octavian poem, in which he will dlscass its relation to the French version. 
This will be likely to throw much light on the origin of the different sets of 
MSS of this faroritc theme of the middle ages, and thus be of great interest to 
scholars of the Romance field, besides famishing them with more material for 
the stndy of the close linguistic relations of the English and Noimanized version 
before ns. ^ j,I_ £_ 



REPORTS. 

Germ ANiA. Vierteljahrsschrift far deutsche Alterthumskun'de. Herausgegeben 
von Karl Bartsch. Wien, 1882. 

Soon after the publication of P. Erasmus MttUer's Sagabibliothek, it was 
admitted by most scholars that the Thidrekssaga was indebted to a mixing of 
several sagas for its origin, some of them but loosely joined to the central 
figure, Thidrek of Bern, and in this respect none more so than the story of the 
jarls Iron and Apollonius. 

Friedrich Neumann now opens the first number with an article in which he 
proposes to show that this story consists of two separate sagas blended into 
one in the Thidrekssaga. Furthermore, to reproduce as nearly as possible 
according to context their original form, trace the connection of the Apol- 
loniussaga with the Kudrunsaga and explain the origin of chapters 245-75 in 
the Thidrekssaga. The article brings to bear a good deal of the light of 
common sense upon the subject, and the conclusions arrived at by Neumann, 
we think, are sound. The separate sagas of Iron and Apollonius were rather 
nnskillfully united by the writer of chapters 245-75. The two jarls were made 
brothers, and to connect them with the heroes of the Thidrekssaga the names 
of Attila, Diefiich, Ermanrich and others were arbitrarily introduced. As 
regards the points of similarity between the ApoUoninssaga and the Kudrun, 
we must confess that they are often as striking as are found in other sagas 
quoted by editors of the Kudrun, ** und selbst wenn wir bei dem Muhlenhoff- 
schen Kudrun text schwOren soUten," which we do not, having always looked 
upon this reading as an anthology of * ^ventiuren ' from the epic which, accord- 
ing to the taste of Muhlenhoff, were the most beautiful. 

C. Marold continues his article ** Kritische Untersuchungen Uber den Einfluss 
des Late in. anf die Gothische Bibel Ubersetzung (cf. American Journal of Philo- 
logy, VoL II, 7), and C. Mogk prints eleven fragments of one of the many 
versions belonging to Rudolf v. Ems* ** Weltchronik." The original leaves 
(twelve) are in the Royal Library in Kopenhagen, and according to Edzardi 
were found attached to the covers of some law books in 1862 by assistant libra- 
rian Weeke. They seem to have been written about 1300, and represent a new 
phase of that favorite work of the middle ages. 

Fedor Bech in a brief paper draws attention to a verb dougen^ tougen, occur- 
ring with Middle German writers. The places where it is found are either not 
mentioned by Lexer in his M. H. G. WOrterbuch, or, following W. Grimm*s 
and Pfeiffer's explanation of the word, placed with the wrong verb. Earliest 
and most frequently it is found in writings that originated in the west of 
Middle Germany. 

Ich dougen des scharpfen swerdes slach (25, 7) 

Ow6 wat wir dougen (29, 6). — MarienUgenden^ ed. W. Grimm. 



I02 AMERICAN JOURNAL OF PHILOLOGY. 

Grimm in a note says dougen = Terborgen tngen, Teriieimltchen, and Lexer 
accordingly places it under tougen -= yerheimlichen, with which it has nothing 
to do, but eridently belongs to the Low German dSgtn = leiden, erdalden (cf. 
Schiller-Labben, M. N. D. WOrterbnch. I 532). Pfeiffer in his glossary to 
Nicolans t. Jeroschin (east of Middle Gennany) identified dougen with douwen 
= rerdanen, and Lexer notes cbugen as a secondary form of dduwen^ Vol. 1, 455 
(cf. Weinhold Gramm. S206. MQller-Zamcke M. H. D. Wsrterbuch. I 386a, 
Schade Altd. Wdrterbnch, p. 97). On p. 14S0, VoL II. Lexer only has the 
noun Umfe as a stf. In the " Wartburgkrieg," 116, 6 ed. Simrock, occurs wtiider 
tou/en, and in the " Parzival," 43, 6. starber dtte toufen stt (cf. Lexer Nachtrage, 
375, Untftn stn.). Bech suggests a nominatire toufefu^ loufen^ as a secondary 
form of Umfe (cf. Grimm Gram. 3, 171, Weinhold Gram. § 256). 

Vemaleken furnishes a parallel to Grimm's fable ** Das wasser des Lebens.*' 
This version of the ancient and popular myth which he gathered in the Schrat- 
tenthal in Lower Austria differs in the main but little from Grimm's story. 

Reinhold Bechstein reviews Alwin Schultz*s work *' Das hoBsche Leben zur 
Zeit der Minnesinger," I VoL, Leipzig, 1879. In the course of his eulogy on 
the book he says that he cannot help being in a measure vexed that an art 
historian should have taken the initiative in a matter which so entirely belongs 
to philology. But when he looks about among his fellow-philologists he can 
not find one — ^Weinhold excepted — ^who would be able to write a book like 
this one. This is a lamentable fact in the midst of a science that otherwise 
does such good work. *' Thus it cannot remain,'* he exclaims, '* Wir mOssen 
den realien fortan einen grOsseren einfluss einrfiumen. Die realien mQssen 
unsere kritischen und hermeneutischen, selbst unsere grammatischen studien 
durchdringen und befruchten, sonst bleiben wir im leblosen krame stecken." 

The Book Notices of the first number close with favorable criticisms by K. J. 
Schr5er and Felix Liebrecht of Amelie Sohr's book **Heinrich Rttckert in 
seinem Leben und Wirken ," Weimar, 1880, and Eva WigstrOm's (Ave) Folk- 
digtning, samlad och upptecknad i Sk&ne, Kobenhavn, i88a 

The Miscellany contains a rather spirited reply from G. Mikhsack to Anton 
SchOnbach's adverse criticism (Anzeiger f. d. Alt. 7, 1881) of M.'s edition of 
the Heidelberg Passion play and two fragments (Passional and Konrad's Tro- 
janerkrieg) by Bartsch. 

R. Springer contributes the first article of the second number '* Die legende 
vom Judenknaben." Of the popular legend of the Jewish boy who, with his 
Christian playmates, goes to communion and is punished for it by his father 
by being thrown into the flames, but saved from death by the Virgin, we 
possess not less than 5 Greek, 14 Latin and 8 French texts (cf. Bibliotheca 
normannica, ed. Suchier, Vol. II, Halle, 1879). Besides the versions in Spanish, 
Arabic, etc., we have two in German, " Das Judel," published by Hahn (poems 
of the 1 2th and 13th centuries) and Mtthlenhoff (Altdeutsche Sprachproben), 
and " Der Judcnknabe,** printed by F. Pfeiffer (Marienlegenden), which differ 
materially in the treatment of the legend from that in the other languages. 
Sprcnger gives a critical text of the " JUdel," places it with the classic period 
of M. H. G., the 12th century (cf. Wackemagel Lit., p. 205), and makes Konrad 
v. Heimesfurt the author. He considers it the older version upon which the 
younger " Der Judenknabe " was partly based. 



REPORTS. 103 



From Sprenger's text : 



hende winden unde klagen 
unt weinen was dH wider strft 
unz n^en ze complete zit 
331. daz man d& tavelte in der stat. 

we note the following correction for Maller-Zamcke M. H. D. WOrterbuch, 
Lexer Handw5rterbuch, and Weigand Deutsches W6rterbuch. Muller, Vol. Ill, 
19 taveltn swv. = tafel halten, speisen (quotes v. 331 above). Lexer, II, 1410 
has additional explanations of the word, among them ** durch Anschlagen an 
eine h6lzerne Tafel eln Zeichen geben (statt des Lautens),*' cf. idvem^ Schmeller, 
I 587. This meaning should be taken in v. 331. On holy Friday no bell is 
rung in Catholic countries, but the sign to begin service is given by knocking 
on a wooden tablet. Tafeln = speisen is New-German, emend. Weigand 
IP 871. 

F. PfafT supplies a Middle German fragment (240 verses in the city archives, 
Frankfurt a. M.) of Reinbot*s Georg, to which Bartsch adds a list of the known 
MSS of that poem, and K. G. Andresen sends an almost exhaustive list of 
family names derived from dUU thiuda, 

Fedor Bech " Zum Wortschatz des Chemnitzer Urkundenbuchs." From the 
lai^e collection of words we note Grimm, D. W., II 426 brueling = frischling, 
wie cs in den BrQl (= wiese) getrieben wird; Sanders, D. W., I 228 brueling 
itooibruch = sumpf; Bech connects it with brUfun^ Low German brojen (Schil- 
ler-Lubben, I 427^). Bechen swSu, i. e, das zu Speck und Schinken bestimmte 
Schwein, incorrectly printed in Lexer's Handw6rterbuch beckenswin and trans- 
lated by bdckerschwein (Nachtrage, 46), cf. bachen in Lexer. Derjenige^ pronoun, 
according to Grimm and Weigand not before the i6th century ; Bech finds it 
in the 15th. Durchaus^ adverb, Grimm, WOrterbuch, II 1583, not before the 
i6th century ; Bech quotes it from the 15th. ffem^ Grimm in W. Ill 52 says 
** es ist ein unverstand schon dem nominativ ein obliques n (h)ehrn, (h)ehren 
beizttsetzen wie Biirger thut : 

hierauf sprang {yC)ghren Loth herbei 
mit brausen und mit schnarchen." 

Bech shows this usage in documents of the 1 5th century (cf. Sanders, I 344a). 
B. continues with two minor communications ** Vom Eichhom als Wildpret,'* 
and ** Das wort tinne.** The first article furnishes Alwin Schulz, the writer of 
** Das hafische leben zur zeit der Minnesanger," with more material to enlarge 
in the next edition of his work, the list of the different kinds of game used at 
the table of prince and knight during the middle ages. The second communi- 
cation maintains his reading Hnm = schUfe, instead of Uime = daumen, in F. 
Pfciffer's "Arzneibuchem," II 4, as defended by Sprenger. The Komenburg 
fragment which Blass published in the Germania (26, 380) seems to confirm 
titni€ as the proper word. 

Most mediaeval epics have received for years a close and scholarly attention. 
This attention, however, has not been uniformly spread over the whole field ; 
favorites like the Nibelunge n6t and Kudrun have received more than their 



IC4 AMERICAN JO CRXAI^ OF PHILOLOGY. 

fair ihore. -yther less lonsricxtons Tot ^erimxjs as mnrh :a wont of elacidotiaD 
aave zsrrzxz.^j recenred less. It las stmckas as >rnn.irirah le. considerrii^ tiie 
connect' I. n of 'he Ortnit-WcifiietXTchsa^c; with. 90 anixy erics, loir ssraH a 
part rf he Tnrk if phiioi'-'i^isis has ':e«i iemted to ir. Fnednch X< 
:n X parer, "Die Enrwrcitcian^ ier Ormttcicr itmii^ md icr Cmntsa^e.'' 
mhiires a ^earchin^ maiysis rt the poem, -winch ^i te s pronnse "har the subject 
wtU ae ieait with store awaejitiT. He iTmrtTnifn Lhat die C'rmit-W'jifdie- 
tnch^a^e in^nated in die combmin^ 3t roro incepemimr ^tibiects, 'Drtnit the 
dingon— layer brir.% x iicerent ^'crsoiia^e from Ortmt **der Riuxe" who sails 
beyond :he 5«u Scon alter "he raking^ rf Tynxs in the year tiii a aew^eisiaa 
appears hat iiaif«n Suiiers the ^acitai af the cbcsbt, and throuijh the mducBce 
if he battles roimd Vans Taoor :n 017 a later poet sabsntutes Montabure 
for Suiiers, ^.acI^^ Albeich. hKetcfore of littie inrport, as the central ligiize. 
Tae poet jf 3ar text ouxes rwo Tersions, the aider hanng Saaers as the object 
3f the expetiition, the younger, Muntabure. He tnes to iisgnise the cunfra- 
dictions incident to this process, bnt iocs not ^rrrrrf. 

F. Vetter sends ^ome Tninn r cm i imiini ra nfm s, anri Bartscfa pr nts hre folksaags 
if the XV jentury. 

Feibc Lebrecht pmnoitnces ctvonbiy apon Lss litttsaturcs poculaires cfe 
tontes les nations : Tott r. Littenstttre omle de la Haute Breta^ne, par Fkol 
Sebillot. Ports, r33i ; and FTfrmann F-scher reriews H. Paul's **Ziir Nibeiungoi- 
fra^e," Ho-ie, 1^77, reprinted in PanI and Bnnme's Beitrage, VaL IIL 

A folksong of the time of the T!iirty Years' War on ** Gostar AdoITs Tod," 
sent by F. Pfo^, and a comnmnication from Bartsch regarding the irrtf Af; <}ao« 
m die >rbelangen printed by B. Symmxs in his le^iew in Literatttrblatt No» I, 
iSSo, of B.'s Nlbelnngen W^rterbnch, close the second nnmb^ 

la an annonncement of W. H. CarpGita's *^ Grondriss der nenislandisdif 
Grammank,'* Leipzig, lS3t, in thelr ir <*Tn n rrfaiartfUr ^ei U i . and rooas. Philologies 
iSdi, 3, Finnor Jonssoa sharply criticised the glossary and leading matter 
accompanying the work (c£. Amrr Journal of Philology. II 5V and ByOra 
Magnnssen OLsen now follows with an article, ** Zur neuislindischen Gram- 
matik," first in third number, in which, he Tiolcntly attacks the ** Laut and 
Flexion >lehre ** of Carpenter's grammar. Olscn seems to us to hare dealt with 
the bock somewhat in a spirit of rrvamcit^ * Xeben der all^^emeiaen Cwisscn* 
schafdichen) habe ich andi eine besondoe peisCnliche Aucbrdening, gesen 
dieses werk einspruch xa crheben'* (des Pudels kern ?V The book was written 
conjointly in Rykja-rik, Iceland, by Olscn and Carpenter, but finally augmented 
and published in Gennany by the latter. Olsen finds the " Flexionslehre " — 
the outlines of which he furnished entirely — intcntiocaily changed and inter- 
polated by awkward additiotis, and his communicatioas were in many instances 
misunderstood or not understood at all by Carpenter. The sketch q€ the his- 
tory of the language in the introduction of the work O. pronounces ^ aus ihrem 
zusammenhange losgerissene Literatumotizen ^ which the author mainly copied 
from Vigfusson's Icelandic reader and M6bius' essay '• Cber die altnordiscbe 
Sprachc." A " Lautlehrc," originally not intended for the work, was added by 
the atithor. It offers little of new matter, O. continues, but many errors, and 
as a rale the author plagiarises Gislason and Wimmerj whom he, however, 



REPORTS. 105 

often misunderstands. *' Sehr zu bedauem ist, dass der erste versach, eine 
wirkliche neuislandische Grammatik zu schreiben, so Srmlich ausgefallen ist." 
But we pass from these personal compliments. That Carpenter's book needs 
corrections is acknowledged, we presume, by himself, and for this an unbiased, 
disinterested criticism is wanted. Olsen does not give it to us. With the 
knowledge of the importance to philology of a grammar of the living Icelandic 
tongue so long existing, it seems odd that we should so recently have only our 
first book on the subject. 

A minor communication from R. Sprenger, '*Alber von Regensburg und die 
Eneide," seems to establish the fact that Alberwas acquainted with Veldecke's 
Eneide and the description of hell in that epic. A comparison of quoted pas- 
sages from Alber's Tundalus and the Eneide shows a remarkable resemblance. 
We note the expression "Ein ovele nigebflr," Eneide 3238; " Ein tlbel 
nachgcbfire,*' Kudrun, Strophe 650, 4. Sprenger thinks the Eneide probably 
prompted the saying in the Kudrun (cf. Martin, Kudrun). 

The only text heretofore known of the poem " Wigamur " was the Wolfen- 
bQttel MS of the end of the i6th century. Lately the Salzburg and Munich 
fragments were discovered. F. Keinz prints the Munich text and assigns it to 
the middle of the 13th century, not long after the writing of the original. This 
completes the publication of the known Wigamur MSS. The WolfenbUttel 
MS was published by Btlsching in the Deutsche Gedichte des Mittelalters, 
and the Salzburg fragments by R. M. Werner in the Zeitschrift ftlr deutsches 
Alterthum, XXIII 100. 

Edzardi, in an article,'* Fensalir und VegtamskvitSa," dissents from S. Bugge's 
explanation of the word meyjar in the Vegtkv. la, 5-8 : 

hveijar 'ru paer mejrjar, 
er at muni grita, etc. 

Bu^e finds the key in the Homeric Kovpai 6Xioto ykpovroq — the mermaidt 
(meyjar) lament the death of Achilles — Baldur being Achilles. . Edzardi dis* 
courages the tendency oC explaining the northern sagas by Greek myths, as by 
such a process they are hardly ever reasoned out to any satisfactory issue. 
They may generally be interpreted much more simply from Germanic mythology. 
(We agree with E.) Edzardi makes nuyjar to refer to Frigga's eyes, and illus- 
trates this from the Wodan-Baldr myth. Corresponding to the passage in the 
Vegtkv. the Vsp. 34, 5 has : 

" en Frigg um gr^t 
\Fensolum 
^k Vaihallar." 

Bogge explains fensalir =r meersale (hafsalir). Frigg i Fensolum = the nereide 
Thetis beneath the ocean. Edzardi here considers fen -=. teich, sumpf, and 
fensalir connected with the popular belief that certain swamps and ponds were 
the entrances to the abode of Holda (Frigga), the lower world. 

C. M. Blass prints " Deutsche Randbemerkungen " of the 13th century, found 
with a Psalter at present in the city archives of Komenburg, and Bartsch and 
F. Keinz supply similar matter from Erlangen and Munich libraries. 



rt' 



t . 



I. I .. B 






^^Jw - A ^ ^^^^ .^^ X- ^ •, .^i^MK. ■m^^^mm' - - I _ *. 






-T^^ " ■^• ^ .fc^ ,^ ; ^»^ ■ »* ~f» 







^w cr :crr*=r.:^j: 




' «. 



eaiVi^ jj t:*^ lun x * 'rnnir jsui. fall Tcunmt "* af iis X3 J^rr xs x riit-rilogist 
txid v.T.c He in':c». xcw-rwer. rue 3«2Kre :3i5 ■ ■■ i^rt.,..,. ^ ^^ * E^Jsxit ** 
1^ 'v;ll law* LearTij::! acme ncrs GcmmEac. •m"~-Jt xmi .zrmcal seiii>£. 



exxaxvoes ir^ine vtnei z€ IC-airiif t, Fissescrmaeai "• £.ailieiz lesa,' 

!u» th;r.ij ar» pricalily onirx-iccs sf ^maj^ g s ia VeLiecke'* Eaei-ie, Ulrick t. 

Zft X t kiv :>fea't M.ariT^'.iTT and Gccfrosi's Tracxa- La x acce je ». 7120 i» Hxit- 

hx.*«i:Tr:a*, Ii4tT*'cf'>re trxiulxred by *• Wbidixad i. Jrjta ier Hxsea.* Th« 
c/>tiii«c..'^c w.th the o'liiT ▼enes rxther pciaa to xrxx^-v = x kind of vexpon 
n*ftd in h3n*.r.(5 hare*, xci Xhis Tiew seecu to be sapcctted by x pASSx^e torn 
Oerhxrd ▼, Min'icii: 

•* ok ko«iiet jegCTC xl her getrekct 
mit tnmden^ pxnden xxui mit ksndcxi.'* 

CI ftchillCT-Llbben M. N. D. Wsrtb. V, 724. 

In X note on " Hartmxnn's 2 BQchlcin," Sprenger finds the verses 670, 71 
almost repro'luccd by Konrxd Fassesbmnneii in his ■* Kindheit Icsu *• i642. 



REPORTS. 107 

From it he draws the inference that Hartmann, and not one of his later imita- 
tors» really was the writer of the *' 2 BUchlein.*' The time which is assigned 
to the origin of Konrad's poem contradicts the opinion which would place the 
authorship of the " 2 Bachlein '* to another and later writer. S. thinks the 
change of the word gtwantiy, 1681) to gehnt (Lachmann) unnecessary, since 
the former perfectly suits the sense qf the passage. 

Felix Liebrecht reviews the 3d Vol. of Eugene Rolland*s Faune populaire de 
la France, Paris, 1881, and the Miscellany contains a contribution from A. 
LQbben, " Zum Sachsenspiegel/' in which he questions the assertion of Richard 
Schr5der (No. 9 Literaturblatt f. germ. u. roman. Philolog., 1880) *'der text 
des Oldenburger Codex sei die niederdeutsche rtLckUbersetzung eines hoch- 
deutschen textes." 

The fourth number opens with an article of Fedor Bech, ** Zu dem Pariser 
Tagezeiten," in which he points out the passages in the work that were directly 
modeled after Frauenlob ; and Edzardi has ** Kleine Beitrage zur Geschichte 
und Erklirung der Eddalieder." In the 23d and 24th Vols, of the Germania 
the Gripisspa was treated by Edzardi. A further examination of the text has 
convinced him that in one place — Strophes 33-44 (Hilderbrand) — it is even 
more incomplete and perplexing than was generally supposed. E. prints the 
strophes in the order which he considers the original. 

E. Steffenhagen furnishes a " Kieler Bruchstttck aus Berthold's von Halle 
Demantin.'* The writing proves it to be of the 14th century, and the context 
corresponds to the verses 1 287-1438 of Bartsch's complete edition. 

Ferdinand Vetter follows with some minor communications, and C. v. Her- 
denberg prints a paper MS of the 15th century, '* Die vier Temperamente." 

A series of remarks and questions made by Otto Behagel, which, he says, 
were suggested to him by his edition of the " Heliand," will prove of great 
value for any future edition and the study of that work. 

R. Sprenger reads Erec, 2265 : 

swaz aber im des gebrast 
{doM meinde daz er was di gast : 
stn lant was im verre), 
Artds der herre 
gap im swaz er vor sprach. 

Meinen has here the meaning of to cause. Thus it is also used by K. v. Heimes- 
fnrt, " Urstende," 113, 41 : 

** daz er des ersten genas (mit dem tode ftLrs erste verschont blieb) 
daz mHnde deiz was sp&te." 

A list of recent publications in the field of Germanic philology by the 

editor, Karl Bartsch; J. H. Gallee, of Utrecht; K. Gislason, of Kopenhagen ; 

K. F. Sodervall, of Lund, and an index to Vols. XXV-VII of the Germania, 

close the fourth number. 

C. F. Raddatz. 



I08 AMERICAN JOURNAL OP PHILOLOGY. 

HEums. iSSa. 
No.L 

E. Fabricias, of Stixssburg, writes The Build ine-contract of Delo», C. I. 
G. aa66. This inscription is one of those first published by Chandler in 1 763, 
and ii now preserved in the Ashmolean Mnseoin at Oxford. Recently tbeie 
were found at Lebadea and Tegea important inscriptions dealing with kindred 
matteis. Thns many new poinls of analogy have become available, and 
Fabricins has been prompted to discoss the Delian inscription anew. The 
time of the inscription may be gathered from the form of the letters and from 
other data ; Fabricini assigns it to the latter part of the third centnty B. C., 
when Delos enjoyed aotonomy. 

The contiaci relates to the bnilding of the floor of the temple of Apollo. 
The snccessfnl bidder for the contiact, his bondsmen and the official witnesses 
to the contract are given, these latter being partly officials (the cleilc of the 
conncilithe clerk of theicpoiroioiand of the market-masters), and partly private 
persons. This contract is preceded by a general outline of the conditions of 
the contract fur si with appended legal points. Any unsuccessful bidder 
could bring a dfjo; y>c{idov[ against the actual contractor, >. e, accuse him of 
malfeasance in the execution, fraud, etc. It seems that the compensation 
agreed upon was given to the contractor [feflai^ in two instalments, ten per 
cent, being held back until the entire work was concluded. The commissioners, 
if they failed to make their payments on the date agreed apon, were bound to 
pay an cn-i^pd, and the contractor likewise had to pay a fine for any tardiness. 
In accepting the completed work the commissionera were aided by the 
ipXtTirTur who acted as their professional coaniel and official etpert. 

C. Galland discusses interpolations in Arcadins, a late gtunnurian who drew 
largely on Herodian. 

W. Ditlenbe^er contributes notes on Greek noun -inflexion. The form 
Irwli); occurs in an Attic inscription of 394 B. C. This form D. believes to be 
if iimff, denying the claim that Ifnrw or IsTrier are the prior 
ing analt^oua forms from inacriptions. 

cusses Die unCergegangenen Ortschaflen in Latium, basing his 
N. H. Ill 5, GS, 69. This list of communities mostly refers to 
tt their existence as civil corporations before the beginning of 
; a few were razed by SuUa. Most of them were of the Priici 
ing to Mommsen, Pliny's information came from the old Roman 
t directly; probably he derived it through Varro's AnHqyUatet 
list as edited by Mommsen includes Ameriola, Amitinum, 
ttina, Cameria, Collatia, Comicnlum, Crustamerinm, Ficana, 
torium, Pometium, Satricum, Scaptia, Tellena, Tifata. 
e Episode of Chryseis in Homer. The author of this somewhat 
jf opinion that modem Homer-criticism is unduly barren, and 
rt is to relieve this unproductiveness. According to Hinrichs, 
artistic and chronological points has run its complete course, 
he thinks may still be done by verbal analysis. The return of 
:cording to Hinrichs, is the work of a wretched " Flickmeister." 



REPORTS, 109 

This wooden person went to work in a manner worthy of a lazy and mechanical 
schoolboy ; culling a phrase here and two words there, a couple of lines plun- 
dered from 7, some phrases and general situations filched outright from the 
hymn to the Pythian Apollo, also from /?, 1^, r, and from more than a dozen 
different books of the Iliad. Hinrichs professes himself a disciple of Lachmann. 
That critic, indeed, considered this episode as " sehr geschickt,** and " an sich 
Tortrefflich,'* but Lachmann left much laborious detail to later workers. The 
whole paper seems to the reporter nebulous and unsatisfactory. 

E. Petersen (Prague): Der Streit der Goetter (Athena and Poseidon) um 
Athen, discusses anew the St Petersburg vase, and pronounces and explains 
his dissent from Robert's interpretation of it (Hermes, 1881, p. 60 sqq.). 

C. Robert: Die angebliche PyrrhosbQste der Uffizien und die iconographi- 
schen Publicationen des i6ten Jahrhunderts. In this archaeological discussion 
Robert produces some interesting notes which throw considerable light upon 
the later renaissance in Italy. The enthusiastic desire to identify portrait 
busts, Hermae, etc., with great men of old led to much falsification of inscrip- 
tions on the part of connoisseurs, collectors, authors and publishers. This 
became evident by inconsistency in successive publications of the same art- 
objects, by gross blunders of the falsifiers, and by covert admissions of authors 
and publishers. The Anthology very generally served as the source from which 
were drawn the epigrams put on bases. 

F. Blass : Neue Papyrus-fragmente im Aegyptischen Museum zu Berlin. B. 
publishes and comments upon the second of these Greek papyri found in the 
Fayum. It is exceedingly fragmentary ; the date is of the V century B. C. It con-* 
tains four articles, all of which pertain to and explain the text of Demosthenes 
contra Aristocratem, e, g, on Miltokythes, the Spartan mora, 6 KaTu^tv vdftog, 
Harpocration, as Blass shows by parallel quotation, contains the gist and often 
the words of these alphabetical scholia, but much more briefly abstracted. Blass 
makes some sensible suggestions as to the probable history and the successive 
abstraction and condensation of scholia, as in the present case. The probable 
fountainhead was a regular commentary rather than Atticist collections. 

In the Miscellen there occur notes by H. Giske, Zu den Chiliaden des Tzetzes; 
Th. Mommsen, Zu Ammian ; A. GemoU, Emendationen zu der Hyginischen 
Lagerbeschreibung ; F. Gustafson, ad Ciceronis Tusculanas Disputationes 
conjecturae XII ; £d. Woelfflin, Satura Critica. 

No. II. 

P. Pulch, of Strassburg, prints an interesting study, Zu Eudocia, proving 
that the famous Violarium (*Iwv«i) of the learned Byzantine Empress (flor. circa 
1070 A. D.) is really a compilation made by Constantinus Palaeopappa, a Cretan 
monk, who had been an inmate of a monastery on Mt. Athos. He and other 
Greek calligraphers and copyists of that time found generous employment at 
Paris in the time of Henry II, husband of Catherine of Medicis, and of the 
Cardinal of Lorraine, also a distinguished patron of such men. The paper is 
an interesting contribution to the history of classic philology in the XVI 
centuiy. [See A. J. P. Ill 489.] 



• 

L _ _ 



I lO AMERICAN JO URNAL OF PHILOLOG V. 

R. Foerster (Kiel): Achilles and Polyxena, two unedited declamations of 
Choricius. Choricius was a noted professor of rhetoric and literature in 
the age of Justinian,* The two pieces edited by Foerster are from a MS of 
the National Library of Madrid. Foerster edits the text with many emenda- 
tions ; these, however, are generally very palpable, as the copyist of the MS 
appears to have been very ignorant of Greek. Each declamation is preceded 
by an hypothesis and ^supla which betoken the practical teacher of rhetoric. 

Choricius appears to have been thoroughly imbued with Demosthenes, and 
he maintains very pure Atticism in construction and vocabulary ; a few phrases 
are direct reminiscences, e, g, p. 212, 1, 22 ovyKeKporfffihoc rd Toi noXifiov (from 
Dem. XXIII 3). In poetical phrases he rarely indulges, f. g, ^7j6ya Trpocayuv^ 
p. 212, 1. 25; enl yijpaog ovJ^, p. 231, 1. 30. Certain phrases are derived from 
Platonic and Aristotelian vocabulary, as y^wxaycrytlv, avXXjoyi^t(r&ai, The 
argument is wrought out with great clearness and with as much earnestness as 
such a subject-matter would admit. Choricius exhibits an excellent faculty of 
psychological analysis, although this is sometimes brought forward too didacti> 
cally, not in the proper dramatic manner. 

I. Schmidt (Halle) discusses the MSS of Serenus Sammonicus. 

Gardthausen: Ursicinus and the Inscription of Dojan. This inscription 
(edited by Mommsen, C. I. L. Ill 6159) records a victory over the Goths; 
spelling and antiquarian detail point to the IV century A. D. Gardthausen in 
the present paper endeavors to specify dates and persons, assigning the inscrip- 
tion to Constantinus II, son of Constantlne th^ Great. 

* Vahlen: Varia; Exegetical notes on passages in Cicero, Ovid, Vergil, Seneca, 
Plautus. This distinguished successor of Haupt and Lachmann shows a strong 
vein of conservatism in his dealing with texts. He hesitates to leap from the 
notation of difficulty to the utterance of condemnation. He is careful to exhaust 
the ranges of parallel literature, and he often uses the mild remedy of a change 
in punctuation. His Latin, generally limpid and dispassionate, becomes some- 
what ruffled when dealing with that band of younger Ritschelians who are 
continuing their master^s edition of Plautus. He imputes to them violent and 
subjective practices and returns some of their strictures with interest. Schoell, 
in bracketing Plaut. Trucul. I I, 60, had remarked of Vahlen ** novo igitur 
exemplo V. ostendit audaciorem esse neminem quam criticum iusto timidiorem.** 
To which V. replies '* Poteram respondere ut ilia : d^ aw^. Sed nolo cavillari 
hominem quem auguror paullatim ultro desiturum mirari si multa quae ipsi 
nunc sunt certissima, ab aliis aut falsa habebuntur, aut dubia admodum.'* On 
p. 268 V. says **admonemur, ne, si qua in veterum libris a nostra dicendi 
consuetudine abhorrent, ne veteribus quidem potuisse placere confidentius 
affirmemus." Vahlen's method while avoiding specious brilliancy would seem 
to be more apt to make thorough scholars. 

K. Lincke (Jena): Zur Xenophonkritik. This elaborate and somewhat 

rambling paper sets forth the theory that many passages in Xenophon*s Anabasis 

are not late interpolations, but additions, probably, by the first editor or some 

one near to Xenophon. Of course there are additions of grosser and quite 

I 
1 See American Joursal of Philology, I 79. 



REPORTS, III 

palpable sort, as the summaries at the beginning of books II, III, IV, V and 
VII, also the general summary of tribes and nations at the end of the Anabasis. 
In many of his bracketings, Lincke reaffirms the critical judgment of Cobet, 
Schenkel, Knieger and others. A typical addition of such early addition, 
according to L., is the note on Apollo and Marsyas, Anab. I 2, 8. Here the 
style, too, serves L. as a handle for his condemnation, there being a series of 
monotonous, poorly connected data. L. also objects to the use of ao^ia for 
musical skill. Lincke*s paper may prove ver/ handy for its presentation of a 
conspectus of doubtful or difficult passages in the Anabasis. At the same time 
one cannot help feeling that the critic cast around for more material after the 
fa9ade of his critical stricture was completed. This is the impression produced 
by his strictures on III 4, 7 sqq. (Larisa and Mespila); where slight difficulties 
are stretched considerably, and where his comments on points of detail seem 
to have been biased in advance by his general theory. In conclusion, L. 
applies his theory oi early additions to the Cjrnegeticns.^ That book, according 
to Lincke, was not indeed written by Xenophon when a young man, but edited 
after his death by a young man with a young man's additions, such as the 
heavy mythological embellishment of the preface, etc. 

The minor papers of this number are notes, by Hirzel on the Democritean 
Diotimos ; by P. Stengel on Libations of wine in connection with burnt offer- 
ings; by A. Piccolomini, De loco quodam vitae Euripidis^ and Th. Kock, A 
reply to van Herwerden on Aristoph. Ran. 548. 

£. G. SiHLBR. 



ZSITSCHRIFT DBR DEUTSCHEN MORGBNLANDISGPHEN GBSELLSCHAFT. 1882. Ill 

and IV Heft. 

z. In continuation of FlagePs account of 'Abd al-ghSnI*s third and first 
journeys (Zeitschrift, 16, 651), J. Gildemeister gives extracts from his second 
joum^, from Damascus to Jerusalem, made A. D. 1690. The interest of these 
travels lies in the geographical notices, and in the description of Moslem sacred 
places in and around Jerusalem. The author travelled in state, with a retinue 
of pupils, and was everywhere, as a great scholar, honorably received, and all 
means of obtaining information were placed at his disposal. His report gives 
a curious picture of the Moslem religious ideas of the day, which, however, seem 
not to differ greatly from what we now find in the East With the Arabic 
devotion to names *Abd al-ghini begins his work with a list of the names of 
Jerusalem, eighteen in number, most of them from the Hebrew, with various 
distortions of form, as Babush for Yabus (Jebus), though this may be a scribal 
error. Perhaps the most valuable historical statement of the book is the account 
of the (laram (described by other Moslem pilgrims also), with its mosques, 
domes, and graves of the patriarchs. The existence of two rival graves of 
Motes occasions our traveller some embarrassment, but he takes refuge in the 
reflection that bodies are sometimes removed from one grave to another. The 
poetry scattered freely through the book Gildemeister pronounces to be gene- 
rally poor. 

> See American Journal of Philology, III 199. 



112 AMERICAN JOURNAL OF PHILOLOGY. 

2. Professor Backer's article on "Abulwalid Ibn Janab and the modem 
Hebrew poetry," cites a number of poetical quotations from Ibn Janab's 
Hebrew-Arabic Dictionary, giring the names of the poets (these are of two 
classes, the liturgical or Paritanim or Piut-poets {Troiifaii)^ and the non-liturgical), 
and illustrating the new forms and meanings of words which they employ. In 
this late poetry is found not only a considerable widening in the significations 
of biblical words, but also free departure from the masoretic rules of punctua- 
tion, and Abulwalid finds occasion to go into discussions like those of which the 
Arabic grammarians are so fond; thus there is a defence of ^^1 and 3^|^ as 

Stat, const, of ')$! and 3^Pr, and of the preposition yOVt used instead of the 

fuller form ")12P2. Among words used in non-biblical senses may be men- 
tioned ]*^n (found in Old Testament only in Job xli 4, in sense of ** grace, 
comeliness"), which the Piut-poets employ in the sense *' discourse,** taking it 
from ^Jinnn, first as *' prayer," and then as speech in general. For another 
biblical hapaxlegomenon, ^K2{27 (Ps. Ixviii 18), properly *' repetition," we find 
in the Paitanim the rendering that the King James English Version has 
adopted, " angels," which was, apparently, the generally accepted signification 
among the later Jews (so Saadia and the Taxgum). The old versions all stumble 
at the word, twisting it in various wa3rs, and the Jewish interpreters seem to 
have taken the signification "angels" by a simple A;»r^/(;rr^ from the con- 
text. Bacher accompanies his citations with instructive critical remarks and 
references to Jewish authorities. 

3. In reply to Dr. Nager*s article ( see the Journal II 7) Dr. FQrst defends 
his interpretation of Azkara and Shem Hammephorash (" the distinctly pro- 
nounced tetragrammaton ") by an examination of various passages of the 
Talmud in which they occur. There is no doubt that the verb r"lD (in Pael) 
is used in the sense of " distinctly or expressly pronouncing the divine name,'* 
and that the prohibition of such pronunciation extended only to the name 
mrr*. in regard to m^TX the question is whether it is used of other divine 
names than the tetragrammaton, to which Dr. Fttrst's answer (well supported 
by citations) is that in later times when the designation Shem Hammephorash 
had been generally adopted, and the original signification of Azkara had been 
forgotten, the latter was used of the other divine names, which are included in the 
category n'**12'*D '* cognomina." 

4. Franz Praetorius expresses the opinion that the Safa alphabet contains at 
least 25 letters, instead of 23, as Hal^vy holds (Joum. As., VII series, vols. 10 
and 17), but thinks it impossible to speak with confidence on thb point till the 
inscriptions have been more certainly deciphered. 

Book Notices, i . Noldeke's highly commendatory reviews of Socin*s Texts of 
modem Aramaic dialects from Urmia to Mosul, and of W. Wrighfs edition of 
the Chronicle of Joshua the Stylite are characterized by his usual richness of 
text-critical, grammatical, historical, and geographical remark, too detailed to 
be given here ; it may be mentioned that he prefers the spelling Orhai (later 
Orhoi) to Wright's Orhdi or Crhii (Edessa). Having better material at his 
disposal, Wright has been able to produce a correcter edition of Joshua than 
Martin, and so to make accessible the contents of this valuable chronicle, 



REPORTS, 113 

whose date is given by Wright and Ndldeke as A. D. 507. 2. Kautssch gives 
a detailed statement of the contents of Stade's Zeitschrift fQr die alttestament- 
liche Wissenschaft (1881 and first number of 1882), article by article, with 
critical characterization of the positions of the various contributors. His ver- 
dict b in general favorable, but he expresses the hope that the new Journal 
will not become merely the mouthpiece of one critical school (namely, that 
represented by Reuss, Graf, Kuenen, Wellhausen, and others), but will number 
among its contributors defenders of all existing tendencies. Among the most 
noteworthy of the articles which have up to this time appeared in the Zeit- 
schrift are Stade*s on Zech. ix-xiv, which prophecy he assigns to the Greek 
period, and Giesebrecht*s examination of the language of the Hexateuch in 
order to determine whether the linguistic phenomena permit or forbid the 
assignment of the Priest-Codex or Elohistic recension to the period B. C. 620- 
4SO. His lexicographical result (herein he continues RysseFs work) is as fol- 
lows : Of the characteristic words of the PC there are found in the literature 
up to B. C. 700 at most 28, of which 12 are in Isaiah, Micah, Hosea, Amos; 
while there are 58 in Jeremiah and Lamentations, 29 in Deuteronomy, 72 in 
the exilian Isaiah, 192 in Ezekiel, over 80 in Job and Proverbs, 229 in Esther, 
Ezra, Nehemiah, Chronicles, 11 in Judges, 6 in Samuel, 31 in Kings. This 
striking result is entitled to careful attention from the opponents of the post- 
exilian date of the Elohbt. 3. Nestle has notices of Hoffmann's edition of the 
Syrian account of Julian the Apostate, and of Baethgen*s edition and transla- 
tion of the Syriac Grammar of Mar Elias of Tirhan. The latter is the only 
printed original East-Syrian grammar, and also the oldest surviving work that 
can Slake pretensions to the name of a Syriac grammar, its date being about 
A. D. 1000. 4. The Chinese grammar of Georg von der Gabelentz (Leipzig, 1881) 
is declared by Grube to be an epoch-making book. In ZDMG 32, 601 v. d. 
Gabelentz described what he thought to be the proper way of treating Chinese 
grammar, and his present work is intended to be an illustration of the prin- 
ciples there laid down. For the first time, says Grube, we here have the study 
of the Chinese language emancipated from the methods of the Latin grammars, 
and put on the basis of an examination of the facts themselves. The author 
calls in question (as Lepsius had already done in 1861) the original monosyl- 
labism of the Chinese tongue, and the reviewer adds that he himself has shown the 
impossibility of this supposed original monosyllabism by a comparison between 
the Chinese and the Tibetan and related languages, in his essay. Die sprach- 
geschichtliche Stellung des Chinesischen, Leipzig, 1881. 5. The contents of 
Ignatius Goldziher*s work on Islam (unfortunately written in Hungarian, 
Budapest, 1881) are described by Bacher as being '* rich and interesting.*' The 
six chapters treat of: The religion of the desert and of Islam; The traditions 
of Islam; Saint-worship, and the remains of older religions; Buildings, in 
connection with the Muhammedan conception of the world ; Muhammedan 
University life ; Incorrect opinions respecting Islam. Goldziher takes the field 
against Sprenger and others, and maintains that Islam is not in any sense a 
development of the Arabian national thought, but is, on the contrary, a com- 
plete denial and reversal of all the habits and tendencies of the people. This 
is doubtless an exaggeration of one side of the phenomenon, but has its rights 
over against similar exaggerations of the other side. 



114 AMERICAN JOURNAL OF PHILOLOGY. 

At the request of the editon of the Zeitschrift, Dr. H. Gathe describes his 
work on the Siloam inscription, and gives a Hebrew transliteration, German 
translation, and commentary, with a photograph of his gypsum cast. The 
transliteration is as follows (the stars represent illegible letters, those in 
parenthesis-marks are not quite certain, those in square brackets are supplied) : 

■ny3 • n2p:n • ^y\ • t<'t\ • nti • Ts::^'^:^^ • • i 
p • ©CK) • bp •••••• • Q)n':) • n?sK • "zr^ • T.y2i • ys^ • ■>« • rx • irnn « 

n • D'^ai • (?K) • • • ?s(i) • p'*^ • i>:d • HIT • n^•^ • -d on • bx • xn) « 

• w^i • 1T10) • *:5y • iTi:i • w • nipb • rx • n3>:rn • i^n • nap: * 

(X)?2i • n?2X • p)*?x(i) • D^nK?s3 • HDiDn • "JX • xxi?sn • p • c?:.-! « 

[ci'*25jnn • irxn • "^r • mr\. • nao) • n'^n • n::x • n • 

The translation of Professor E. Kautzsch, slightly modified by Guthe, is as 
follows : I. *' The cut [is finished]. And this was the manner of the cut. While 
[they were] still [swinging] 2. the picks one toward the other, and while 
there were yet three cubits to [cut through], [there was heard] the voice of one 
who cal- 3. led to another, for there was a cleft (?) in the rock on (or from) 
the south [and on (or from) the north]. And on the day of the 4. cut the 
masons struck one toward the other; pick against pick, and there came 5. the 
water from the spring to the pool 1200 cubits, and two hun- 6. dred cubits 
was the height of the rock above the heads of the masons.'* The readings of 
the photograph are in some tases nearly or quite illegible where Dr. Guthe 
expresses no doubt ; in such cases he must have got his idea of the reading 
from the inscription itself, and failed to make a complete transfer on his cast. 
Nevertheless, we are under great obligation to him for the perseverance and 
skill which he has shown in procuring and publishing the cast. The only new 
words in the inscription are H^p: and n^T, of which the former is clear (from 
ap: ** to pierce '*). No satisfactory sense has been found for rnT. Of known 
Semitic stems we could think only of *1V, which in Hebrew signifies only ** to 
boil, be proud," but in Arabic means to *' increase, be over and above, remain,*' 
whence the noun would signify ** a remaining part, or an attached part,** which, 
however, yields no clear sense. Guthe and others assume the sense '* cleft, 
fissure,'* from what they think the necessity of the connection, but there seems 
to be no etymological basis for this signification, and it must remain at best 
doubtful. The grounds for the assumption of the bracketed words will appear 
from the connection. The letters of the inscription are nearly identical in 
form with those of the Mesha-stone (9th century B. C.) except the Aleph, which 
is like the Aleph of the Eshmunazar-inscription (4th century B. C). From a 
comparison with II Chron. xxxii 30, Guthe (regarding Isa. viii 6 as not 
decisive) assigns the tunnel and the inscription to the time of Hezekiah, latter 
part of 8th century B. C. 

Other articles in this number are: On the M§nava-Gthya-Sutra, by P. y. 

Bradke, proof that the M&nava belongs to the Maitrayanl-^ikha, with remark 

that the transition from the Vedic prose to the classic metrical prose was made 

through the epic poetry; Extract from the poems of the Tatar sage Ni§ir 

Chnsran, by Prof. Dr. Hermann Eth^ ; Specimens from various Indian poets, 

by Theodor Aufrecht ; On the Ashi-Yasht of the Avesta, by Chr. Bartholomae; 

On Avestan text-criticbm, by F. Spiegel; Avestan studies, by C. de Harles; 

and several short articles. ^ „ — ^„ 

C. H. Toy. 



REPORTS. 115 

JOUltNAL ASIATIQUl. 1 882. 

No. 3. August-September (No. i contains the Annual Report, by Renan). 

I. Senart continues his studies of the inscriptions of Piyadasi (sec the Journal, 
No. 7). 

3. The Sanskrit inscriptions, collected in Gamboge by M. A3rmonier, Rep- 
resentative of the French Protectorate, and sent by him to the Asiatic Society 
at Paris, were submitted for examination to a committee consisting of Messrs. 
Barth, Bergaigne and Senart, who give a detailed report of the contents, and 
warmly commend Aymonier's work. Bergaigne points out the evidence 
famished by the inscriptions of the missionary activity of Brahmanism, which, 
indeed, prepared the way for Buddhism ; Brahmanism, says Barth elsewhere, 
failed to penetrate anterior Asia not from lack of the proselyting spirit, but by 
reason of the barriers opposed to its progress by more compact religions, 
vigorous political organizations, and national culture su|>erior to its own. 
Barth adds annotated translations of two of the inscriptions (see the Journal, 
No. II). 

3. M. Arthur Amiand gives an improved translation of the non-Semitic 
inscription of Hammurabi (Cuneiform Ins. of West. Asia, 4, 15, i) for the 
purpose of putting alongside of it an Assyrian translation which he has taken 
almost wholly from the Semitic inscriptions of the same king Hammurabi, and 
showing the striking correspondence that exists between the composition of 
the former and that of the latter. He declares that there is a similar corres- 
pondence in all the non-Semitic inscriptions of the five first plates of the W. 
A. I. He concludes that if the authors of the inscriptions wrote in two diflferent 
languages, it seems at least that they thought in only one. He says nothing 
further on this point, but apparently reserves his more explicit statement for 
another article. 

Book Notices. There are highly commendatory notices of J&schke's Tibetan- 
English Dictionary (London, 1881) by Peer, and of de Harlez's Pehlvi Manual 
(Paris, 1880) by de Dillon. Siouffi, French vice-consul at Mosul, communicates 
some very curious creation-legends of the Yezidis, together with an account of 
their present customs. 

No. 3 of the Journal, containing the index to the Seventh Series, has not yet 
appeared. * 

1883. No. I, January (beginning of the Eighth Series). 

I. Maspero gives text and translation of the love-songs of the Turin papyrus, 
and the papyrus Harris 500, and points out certain resemblances between them 
and the Old Testament Song of Songs. 

3. Client Huart, Interpreter of the French Embassy at Constantinople, 
gives a long list of Arabic words and expressions peculiar to the Damascus 
dialect, supplementing the Arabic-French dictionary of Cuche, and Dozy's 
Supplement to Arabic dictionaries. The strange forms ^arjini^ warjini^farjinly 
all meaning *' cause me to see," he derives from the ordinary stem fo'J *' to see,** 
in which the middle radical Alif has been changed into/, with prefixing of the 
conjunctions wa 9xAfa, 

C. H. Toy. 



Il6 AMERICAN JOURNAL OF PHILOLOGY. 

Mnhiosykk, Vol X, Fart III. 

The first article (pp. 225-38) of this nttmber is bf Cobet, containing criti' 
cisms OD Appian, dt teOit civiliitit. He places (int in parallel columns a pas- 
sage which occurs sabstantially in the same lenns in IV sS and III 57; "hone 
locnm ex nescio qao historico descnptum qnum semel Appianas apposaisset 
eius rei immcmor cuodem itcrum in rem saara conTertit." Cobet shows that 
errora have been introduced into the later quotation which do not occur in the 
earlier. V ga, a8 : iitrvAti Tobf hniitaf h^jiaatv tni^i/aaf if ^uaXiav tnpaioirv. 
" Debebat firi^i^Moof scribeie el iTi^ijaa Homero et Herodota relinquere, Sei 
apud sequiores nivra ft^/iara iv 6/1011 neque in quoquam eratd voiv 6 iauoaii^injv." 
V 101,4: Toic wpa^JMft Tiir vavTriryavpivuv KaUiapi vtuvo'ia octtttJc 'AtANQZ 
ffuriirruv, " ecqnid absurdins est quam oK^rif apaviit iiuTim-uv, at apnd iocosom 
poetam in rebas manifesto absurdis ludentem ; le lanturrt en tiUnee Pardts Alain 
obietBt aniuHfait safr/seniet Eiimedaas literulas et erit 'A^NQ,' qnod Appianas 
passim usurpat pro t)^llr,nAi9^ifuiA>." IV 94, 33 : post caedem Caesaris oi/rwa 
i povk^ ryy mivirv yvitiiijv iftfrpit ea^ phi hrt ml ytpa rvpantucTOVaa hpt/fl^arro 
ihtu, inurxAvro^ 6i avrob^ 'Avruviov naff inr&cpiaiv oTAPofiof, " Senatns intct- 
fectoribus Caesaris praemia decemere volebaC, scd retinuit cos Anlonins. 
Retinoil anlem *aff vwdnpiatv arapafiaf. Quid tandem esl toff inriKiHair arapa- 
fiiif? Nihil pronus. Inteipres: leJ rrooeatiat en canHlie pairti per Anionivm 
aiMt Jitnmulantem gnat exeitalunu enet Uirbas. Vides interpretem qnoqae 
Mtnte dissimulare se ea verba non inteltigere.* Lenissima emendalio claram 
Incem afleret. Rescribe loB irrdipiea' dllpajfaf. 'Airpojia est iiutitiunt qnom 
neque cum populo neqae cam patribus quidquam recte agi posset eoque prae. 
texta anus Anton ins piohibuit qno minus patres qu idqoam decemerent." He 
showihoweverthatAppian'susual word for (uiA'Awn aafrflai: iuiHHMminJuere 
= dpyio; npoypi^ai ; iiulflitan rtmiUere =: apylac avaiptiv or ffain-iinv, Cobet 
closes these notes with sotne general remarks on the style of Appian. " UtJlnr 
Appianus oratione simplici et incomta etpeispicua adintelligendum. Totus in 
rebns verba eadem de iisdem sotet ponere inops magii qnam copiosus et minime 
sectatur r^ iHi^?^UjTtaiy multorum nanfragiis infamem scopulttm . . . Non vital 
vitiosam vnlgi awffieani et horum fere incuriosus ac negligens nonnumquam in 
mirificos crrorcs le jnduil, Anxerat, ut eruditi omnes, dicendi copiam assidna 
lectione Velerum, et saepe pellucet Thucydidis diclio. saepe Demosthenis, 
laepissime omnium Herodolea. Non putide haec et pueriliter imiiatur sed 
quae longo nsa imbiberat ntitur pro suis." This statement is supported bf 
many citations which establish each point ; and he says at the end : " Complnra 
: genere hoc ex Herodoti et Appiani companitione colligi possunt, quam 
,niotum ingenio et indnstriae commendo." 
page 23g, with the heading 'AAIAKOHTA LatinaetGallica,' Cobet gives 

mitukc woald be medliied hy tbe lau ud ttn lana A*HI12.-~B. L. G. 

pouage occun in a ipc«ch of CfLUiu where bfi Darrein the eYuia that fbliovtd upoa 
I death. For my pan, T canaof teethe difficulty of ■«#' vv^fh^ir irmfimtiat laTinrof 
b'l vordt (Vll. AntOB. c. h); idnrin Pn>,iit /nuiMpiTms iri/nimmr i 'Artaruf 

lpil9^^tA mti voAtTuvraTa- A 
iiwimfiifvt bt used ai aqulvaknt 



REPORTS. "7 

M. Bfeal's interpretation * antiquissimae, ut perhibent, inscriptionis,* cited in 
this Jonmal, Vol. Ill, p. 107 ; and exclaims : " <S Zci) paatXev, rij^ Tutirr&nrro^ 
Tuv ^ftevuv. Qaam soaviter in sinu ridebit scurra Romanus, qui banc nugato- 
nam inscriptionem ioci causa de suo commentus est." For Cobet's own view 
of it see p. 246 of the same volume. 

The next article (pp. 240-46) is by J. Van Der Vliet, on passages in the 
letters of Seneca. His suggestions are often ingenious and probable ; but 
some of the changes proposed have been already adopted by the Editors; 
e. g, that on Ep. 50, §?. 

Next, K. G: P. Schwartz (pp. 247-50) gives notes * ad Platonem et Lucianum.' 
The only passage of Plato touched upon is Phaed. 62a, which he proposes to 
emend by inserting the spaced words : kolX ovSinore Tvyx&vei. r^ avOpCmt^i^ Cumtp 
Kot riAAa, lariv bre xai die pi^riov 6v d AAo; re Kal olf piXrtov rtdvdvai fj 
Ctv. This would suit the sense very well, but is unnecessary if we understand 
the preceding rcwro to mean rb fir^ Oefxirbv elvai avrbv kavrhv airoicrtwOvai^ and 
not with M. Schwartz simply rb airrbv iavrbv airoKTiwifvcu. 

We have then emendations proposed for some thirty passages of Lucian. 
Only one can be quoted. •* Bis Accusaius c. 3, roaairac i6h*vg ducag ^Xirrofirv 
. . . icaJ fi&Xuna birdaai rac iirtar^fiac Kal rk^ya^ irpbc hvBp&Kovq rivdc ^wianfaav. 
Sine dubio corrigendum rale errtar^fMug koI rixvaic, Notissima locutio est dUtf 
ftoi awioTaToi irp6g rtva,** 

Cobet next (pp. 251-57) contributes some notes on Galen. He illustrates the 
condition of his MSS by comparing quotations made by Galen from Hdt. II 
36 and Ar. Av. 471 with the texts as they appear in our editions. Galen's own rule 
as to style is quoted ; VI, p. 579 : kyot fih> oiv rolg bvbftaaiv ovtuc hxpfjo&tujv etc ol vvv 
iiSpwwi xp^fvratf piXrtnv tjyohfievoc elvai 6i6d^cu aa^Ctg ra irp&yfiaTa rov ira2nioc 
arrtKi^etv. '*Athenienses Galeni aetate utebantur t$ koiv^ tqv *EX^vuv awrj- 
Oei^ . . . sed in tali re irahtioc addi non solet, quia omnes sciebant aTTud^eiv 
non nisi de vetere lingua Attica usurpari." A passage in VII, p. 291, is quoted 
to justify the substitution of Kadfftpayiac for kqI yvfivaaiuv in Ar. Nub. 417, 
'*Absurdum est ankx^t, yvfivaaiuVf in quo nulla inest luxuriae notio. Apud Dio- 
genem Laertium, ubi hie versus laudatur, pro yvfivaaluv legitur adrf^yloQ^ quod 
nniceverum esse et Aristophani reddendum ratio demonstrat et confirmat locus 
Galeni." Several passages are quoted simply for their interest : e,g, VIII, p. 
148 : Tv66ftevoc r^ 'Apx^y^'' '"' yeypdijtdai ^ipXlov tvda diddoKei fiv^fiiK pe^^fifiivrfg 
6vdKTj^tv, evdiijc nept^Wov diraaat ftiv rag PipXioB^Kog, anavrac dk rovg /?//?^- 
ir^AoCt fliravraf 6^ oCf ijdetv larpovg ktrrovdaK&rag irepl tcL avyypdfifiara ravdpbg 
evfrop^at tov pt^Xlov irpoyprjfiivoc, VIII, p. 151 : a>f tyoye Koi vvv Oectfuu EATA 
rag tqv voooi/vruv hruyichffeig rovg larpovg iv r§ KoivoTuoiylif mnSavofiivovg aX^^hjv 
Kord, TLva X6yov rMe Pofffiijfia irpb rovde avvtP<Ajktvcav. " Optima haec erat et 
ntilissima medicorum consuetudo, sed vitium est in verbis : narh rag rCtv voO' 
ainnuv iniaid^lftig, Interpres vertit ut debuit, inter visitandtim aegros, Sed qui- 
cunque sentit haec coram aegrotantibus agi non potuisse rescribito MErd rag 
TOV voaovvTuv kirtaKitf^eig,*' *' Est operae pretium videre quam alto supercilio 
Graeci linguas barbarorum contemnerent. Paullo ante Galenus scribit (VIII, 
p. 5S5) kdv rrpoc^xv^ ^^ *•'"''' ^<**f ^uvaig ruv pappdpuv StaXiicruv elag aa^ctg rag 
fitv raig ruv avuv^ rd( di raig rov parp6x<^v fj noXoiuv i^ nopdicuv hucviag, Eiusdem- 




Il8 A ilERICAN JOURNAL OF PHILOLOGY. 

modi est qood Jallanut icribil de poetii tnuu Rbennm p«g. ysjd: tBtaaiii^ 

. . , rotfc ^ip rAv 'P^vov ^^ipmif hypia fiiiii U$» irctroniplva iropeirlbgal^ rait 
KpuyfiOit Tuv Tpaxi/ poCnnuv bf/viSiM ^Sorrof." 

la the next article (pp. 95S-SQ) Naber continues his EuripiJea. In Ion. £4, 
he propoMt BpiyKo^ioKa for jcpt^o^^o"". which be supports by comparing vr. 
315 and 414, explaining ^pf^ii^ to be 'i)«Atira(ain saeptnm Mxit ttractum qao, 
adytum templi circa mclumm erat.' In 304 he writes toi^ ft^i^MU for utf^vai, 
Mseiting that in the Tr^ic dialect the mle of ordinary Attic speech must hold 
by which filyvvTat i &v^p rj ywatid, mx fl ymi) Ty avSpi, though he is aware that 
the distinction is nokaown to Homer, Hesiod, or Pindar. la 506, otr* M 
KCpniaiv oliTc Xiyoit ^tiv dim, be feels the difficulty that has troubled othen, 
that ^iyoft expresses the time of gossip too vaguely to be properly contrasted 
with KcpKtaai. ' Quodnam praeterea tempus habent virgines, quod amicis con. 
fabulationibus dare pouunt? Dam lavandria, uti arbitror, siccant ad soleip. 
Vide modoHippoI. 125. Hinc conieci i^' M mpxiaiv obrt n-^woif.' In Helen. 
163, l^aXti^iif iic AyaX/ut is objected to as unintelligible. ' Solentnc deorum 
sigaa iiaXutB^vaiJ Cur? Qnando? Qaomodo?'' Then taking a hint from 
■ Plat. SympoB. 21 %i where Alcibiades compares Socrates to ooe of those Sileni, 
ol iix" SuiixBivTt; falimnu ivMro hyiX)iaT' ixovric Oeoy, he conjectures that 
Helen said 

tiff ifawMjT^iZtr' lif d/iiA^, oMf w6?uv 
alaX""' rWoT i>^' i"<™ laiMi 'Xi0ov. 
' Quid mimm, si Euripides incidit in eandem comparationem, in quam postea 
Plato? Quamqaam Plato melius, nam turpis Silenns inlerdnm Helenam intas 
celat, sed formosa Helena non celat SUennm.' He further argues that iyai/m 
means always a sculptared figure, not a picture. On HeL 1590 he writes: 
' Nondum cipedire potuerunt riri docti iriXiv vTiiapni Nof (av • id\nt ai. Codex 
habel a^iav, cui superscriptnm wi, unde Noffai' in Aldinam fluiit. Scrihit Her- 
manns dfiu, Faleius W virv wXluurv Matnr^fav, Badhamus avriav : sed haec des- 
perantium sunt eonamina. Eiaeuiandnm: v&^ TrXfapev St^iiv kIXtvi eb, Tao- 
talam vitium omnes laluisse! Nee potero fortasse excitare locum ubi hoc 
|egitar Se^i^v hcXiOiiv, sed qaum Aristophanes dlierit : ipBifv luXtbeic ^ ri 
fatvtrai, nihil est cur dubitemui an recte dica^r 6i(iav ittXtiieiv.' Id 
on the Bacehae he speaks of the edition of Wecklein as «ry uscFdI to 
ide praeterea in mea librorum penuria cognosceie potui quid Tiri docti 
c iTBgoediam explanandam attnlissent.' Among other books Wecklein 
oa dissertation by Middendorf, ' qui in observation jbus ad banc fabulam 
pn\» refutavit quae Boeckhias, quo eiat Ingenii acumine, persecntas 
j«r paginal triginta,' in regard to the qaestion whether the younger 
had any hand in bringing about the existing condition of this play. 
,ein Middendorfio careo ati et aliis quam plurimis, qnos minime con- 
sed qnos comparare mihi non potui, quia bibUotheca Amttelodamensis 
itinis et Graecis hand ita opipare instructa est, neqne sacculus mens 
imnia. Leidensis bibliotheca mihi quaedam commodavit: etiam 
> nonnutla debeo; sed Middendorfiui in banc urbem paloram et 

Ith* critic forgottae Kallimachn, LaTact. FatUdli.ijt Sec UaUcr, ArcbieaL. da 



REPORTS. 1 19 

paludum nondum adTolavit' Nabcr's observations on this play and on the Iph. 
T., Iph. AuL and Cyclops are always entertaining and sometimes helpful. But 
he is occasionally very perverse. On Bacch. 259 1 tl fjJj ae yvpac no^v k^ep- 
ftiero I KaBfi& hv hv pdKxcuai iic/uoc fUaat/Q, he says ' praefero optativum KoBy hv, 
Sederest non s€t&sses ; * and he argues, Bacch. 1065, that for naT^ev^ iyev, ^yev eig 
fiiXav nidou we should read nar^tv tipkfi* vpifi*. 

The next article (pp. 290-94) contains Platonic notes by Dr. Badham, chiefly 
on the Philebus ; but none of them are available for this notice. 

J. J. Comelissen (pp. 295-300) follows with notes on Petronius. One or two 
of them may be quoted. §4, *^parenUs obiurgatione digni sunt^ qui nolunt libero$ 
tuos stvera lege perjuere, Legendum est produeere^ i. e. educare ut est apud 
Plant. Asin. Ill [i, 40; Ter. Adelph. Ill 2, 16; Juvenal. Sat. 14, 2^8.' ' §80. 
*fulminatus hoc pronunHaHone^ sic ut eram^ sine gladio in ieetulum decidi^ et aUU' 
Hssem mihi damnatas [Buech. damnatus] manus^ si mm inimici victori^xe invidis^ 
sem, Absurde legitur sine gladio^ nam neque si sine gladio in Ieetulum Encolpius 
decidisset, mortis sibi conciscendae consilium iniisset et supra narravit, post- 
qnam Ascyltos gladium parricidali manu strinxerit, se idem fecisse ; legendum 
igitur, sicut eram^ stricto gladio^ 

D. L. Van Stegeren follows (pp. 301-08) with Varia Cn/lrVflr, chiefly on Plutarch. 
In Pint Cim. 13, 2?nrot> '\ijtv dpdfiov ael r^ *EA^«c^ ankx^iv doX&TrrKt Cobet 
rightly inserts ifiipac, comparing Dem. 19, 273. But Imretoc dp6fiog or Ijnrwtdv 
6i6anffm was probably a recognized distance. Cf. Sol. 23: vSfiov iyparj^ev bnov 
fiiv tar I djjfidatov ^piap kvrbg liriruciAf XPV*^^ rohn^ ' rh dh Irrirucbv diaarijfm rta- 
adpuv fpt aradujVf and this is confirmed by other quotations. Fault is found with 
Cobet, who has said ** ut enim in bello, sic in grammatica nihil contemni oportet, 
nee quisquam qui minuta haec spreverit, unquam facit in re critica operae pre- 
tinm,** that he never applies crasis to ra bwhi. If the passages in Aristophanes 
where the words are found contracted, and other analogical ones, do not estab- 
lish the rule, consider that the words * apud omnes Graecos in usu fuisse ob 
icapdyyeXfia militare. Solent enim haec iussa et irapayytk^mra omni tempore 
et apud omnes populos per ipsam rei naturam esse et brevissima et maxime 
4>erspicua. Credasne igitur, ut hoc utar, in Anabasi VII i, 22 in tumultu mili- 
tari ad Byzantium Xenophontem languida ilia napeyyv^aai^ Qia6e ra ^ttAo? 
imo iussit, puto, BiaOe Ounhi ... ad arma apud ipsum Xenophontem est elg 
SitTrXa,'* In Lysand. 2 Plutarch quotes Aristotle as rdg fuy&hig ^baetg ano^ivuv 
fttXayxoyuKdg^ og t^ ILoKpdrovc kqI TiXdrwog koI 'RpaxXiovg, 'Quis unquam 
fando audivit Herculem, qui semper bibax et edax apud Aristophanem et in 
fine Euripidis Alcestidis e.g, exhibetur . . . fuisse melancholicum . . . quis 
credat Plutarchum principem Graecorum heroum, qui Athenisadeo ut deuscole- 
batur, post Socratem et Platonem demum nominavisse ? ' He is confident that we 
should read Tleptx^^ovc, But the passage of Aristotle to which Plutarch refers 
(Probl. p. 953a 27) leaves no doubt that 'EpaxXiovg is the right reading. For 
before naming tuv icrtpov EfiireSoKX^ ical HX&ruv Kal ^uKpdTr^ kgi erepot (rvxvol 
Tuv yvopipuv, Aristotle devotes several lines to the proof that the history of 
Herakles is an illustration of the rule that irdvreg boot ireptrroi yey6v<iaiv hvdptg 
. . . ^vovrai luXayxP^'^^ bvreg} 

>A melancholy proof that van S. does not know what fAcAayxoAue^ meant. Greek fAtAayxO" 
XuiM and Dutch awaarmoedig^ drinfgtettig are not the same.^B. L. G. 



.^-l. 



3r 






' • - 




,A. ■ » r 







■ - - ■• •«» 1 - — »• • -"•» 

>, ; f.v^ji:: rnn; .tc C: r^ —r . — *,?--/ * 

', »v.,ak .. iiB >::l.*im nrrvL'Triai r*licirsi — * .* -^rwt < ^ Tm ir^4L. , r- tt 

'ii, "iov ''*.' ^/^y lO'^'r'f; ;j'tiir^— ^f^- -* ^j'^it ^ •• -rr»* c~!^— -?!tj in>t .;-? 

'1 Ik' ric*t »ft.r.ji», pp. Kf-^* contains cntiati 3crtes oa Aristop&smitSv by S. 
A. Na;><.-f, On "Jit A£Hamen::c: he proposes in Z to read Trji/u/a tec "trrupm^ 
wKn.K • aii/iot tjc e;5;na.ned MTLffactonlv. iad isincrjnststentwith the sabseqaeat 

cjiuuu idUuii. In 25 he write* ts.o\. -rputT'iv ^v/ujv for rtpi Tuui-njv ^v/jjv^ 



REPORTS, 121 

that we cannot suppose that the fifty irpvrdvei^ straggled with each other for 
the first bench, like Philocleon in Vespae 90, since all bad the irpoeSpia alike ; 
*' sed dam qaisque properabat capessere sedem, turba oriebatur intrantium irepl 
vpuTov fifAov." The ace. is found in the same sense in 692 : yipovr" airo?Jaai 
iroXtdv hvdpa nepl KXti\th6pav, He interprets 994-91 reading irpoaXa^eiv for trpoff^ 
paXeiVf of the simple employments of a country life which Dicaeopolis hopes 
he will again enjoy when peace is secured; ' vitem seret, ficum, olivam; nihil 
potest esse simplicius. Sed docti interpretes non satis sibi docti videntur» nisi 
doctam attulerint interpretationem. Sedulo quaerunt quis sit honim verborum 
dt^Ux sensus, Neqtdtia^ inquiunt, inest in allusiotu ad dpx^^t idque ne oblivis- 
camur, etiam bis inculcant, cum poeta hfintkido^ bpxov et ifupldo^ bpxov appel* 
larerit . . . ne oliva quidem innocenter seri potent . . . rursus in hoc versu 
obscoenitas htet. Roma locuta.est." 

On the Equites, 230-4, he denies that the words justify the interpretation of 
the Schol. that ohK elxtv avrov irpoouKeiov 6ia rb SedouUvai Twg OKtvonouwc ical fiif 
0iXuv fiffTt wX&TTetv fifyrt axffMTt^etv n)v bifftv tov KXiuvo^. It is merely an ironical 
jest of the poet. '* Qui histrio Cleonem agebat . . . ipso vultu totoque corporis 
habitu quam exactissime potuit referebat notam personam demagogi. Cleonem 
nno impetu omnes dum statim agnoscebant, erumpebant in cachinnos et plan- 
sum dabant Demosthenes autem lepidissime negat, larvam satis esse similem, 
qoam in ilia comica partium distortione omnes videbant tam esse similem, ut 
omm non ovo similius esse dicerent." 

In Nubes^ 50, not understanding kpUw irepiovolac, he proposes to write atp&Vf 
ipwvpyiag, and quotes Columella to show that such ccUm were apt to have a bad 
smell, " sed putidum est in tali re veterum testimoniis uti et cavendum est ne 
imitemur exemplum illius Hispani, qui docte et cum cura demonstravit, pueros 
apud Athenienses plorare solitos fuisse, quum vapularent." On 417, olvov t* anix^i 
MM yi'pvaoiuv kqI tov iXhjv avo^ruv^ he does not approve the substitution of 
aSfj^aylac for yvfivaaiuv^ though the line is quoted by Diog. L41., perhaps from 
the second recension, with that reading (see p. 1 1 7), because ** non placere poterit, 
si quis reputaverit in reliqua fabula voracitatis non fieri mentionem ; nee tamen 
Herwerdenum sequar, qui avfiKoaiuv rescripsit, nam nusquam video Socratem 
hilare convivium aversatum fuisse." Since, however, we are told in 837 and Av. 
1554 that Socrates eschewed the bath, and in 992 and 1045 ff. the ^iKotog Xbyof 
condemns the bath on tHe ground of its enervating tendency, he proposes that 
PaXaveiuv should be read in this line. But Kock made the same suggestion in 
his ed. of 1862. On 541-2 : ovdi irpeaphrtK b Xiyuv r&mj ry (iaicTifpi^ rimrti rbv 
irap6vT* a^avii^uv irovr^pd, GK^fifmra^ we have a long discussion as to the point of 
the reference to Eupolis ; and then the suggestion is made that we should read 
ippfT* fi for rimj t^ [where the ^ is hard to dispose of] and tpari^uv for a^vV^uv : 
** id tantum incommode accidit quod . . . verbum ^ril^etv nusquam in comoedia 
legitur.** On 583 : ppovr^ & kpp&yri 6t* aarpainjg^ 61* 'Apfiarog is suggested, in 
allusion to the proverbial expression brav iurrpdiffy 6C 'Ap/iaroc, In 712, among 
the other sufferings inflicted on Strepsiades by ol KoplvBioi ix rov ffKi/iwodog i^kp- 
irovrff, he mentions «oJ n)v irvx^ tKnivovaiv, Naber proposes rtfv i^Hjlkffv^ a 
snbstitution which was made by Brunck in Lysistr. 963. But he has not 
observed that in 719 Strepsiades describes the same affliction by the words 
^poifSif i^vx^t which, with the substitution of ifo^, we can hardly suppose to 



132 AMESICAN JOURNAL OF PBIU3L0CY. 

bBTe been within the compeleiice of hii tonnenhm. Id 113S, StrepMsdci ii 
indigiunt at the rigor of hii doDr-heuted crediton wbo will nol recogDiie u 
pttpm col S'mua his propouls, li iatfiitv, ri) pir n ml fn} U^Pf , rd ^ i^o- 
Pa^miioijTi S i^. Naberqaola the cxplanmtioD of G. Hennann, "qiu Kocklo 
imponit." " tM]t^ lit opoctet qui ad tun frigidam iocam lubridere pouit. 
Emendi : ri /iff ri wnd pot Xa^, Verbnin nos addMn." Tbe credilon, how- 
ever, appear to haTe andcntood the matter aa Hentunn did ; for we aie told : 
o{ faaiv iron ovru( axoi^ftaff, iAXi )MA>poval /a. 

On Vt^at 16, Naber taggeiti ^&iain arrht Btraariaam it tI)» ayapav 6pa¥ 
fttyav for /Uyov wivv, as well ai two other altetatiom of the paxia^c which 
teem leas probable. On 9S, n- dbp^ Af^wr coUv, he sbowi that while there i* 
abandant evidence that the Gteekt in*cribed the names of their favorites on 
wall* and treei, theie ii none that ther did lo on doon ; and accordiagljr he 
wovld wtittfMpf lorBbpf; and if anj one object! that nch a confirmed cit 
M Philoclcon would never go outside the walls where he conld see the trees so 
"abtucd," he endeavoim to prove that there must have been tree* in Athen* 
itself in private Kronnds or public space*. On i^tiitca'tiObt npoat&m ip^iX- 
\tt fuH T^ jt^'P' liTali^ Tim i^/ioaaiB MtEXofvimi, where i/ipiXXei occurs between 
two plural verbs, he objects to the ififfiXJumea of Hirschig, and the ifi^6XXa 
fioi rif of Meineke, and pioposcs irpoaUa- rif i/i^^t t^v xop" aTo^ tipi tri. 
Some (ixteen passages of the J'ax, also, are commented on in this article. 

H. van Herwerden follows, pp. 3B6-99, with Conjtclimu EfigrapkUae on G. 
Kaibel's Epigrammata Graeta, Two short extracts maj be made. "'Omiu yip 
inixal [at/iviif r* aya6ii( r^ i^iumni. Licet hoc poetamm genus nihil non sibi 
permitlere videatur, non tamen indicia caosa singuli damnandi sunt, ut arbitror. 
Nempe qui in hoc epigiammate retiqua piobe tcripsit, huic tribtiere nolim cum 
editore vocabulum pessimae ootae ayoBof pro ti, et potini snppleveiim beoat 
yip inxot [at^uf ff dj-Buf r^ l^iuaav." " In Hadriani scriptoris epigrammate 
repeito apud Thespias vs. i : ii Toi ro^ira KinrpiAic i^iiK Kaibelius suspicatoi 
Mnsamm sacris Theipiaa translatis Veneri Mnsaram et speciem et nominationem 
aliqnan accessisie. Quod credat qui potent. Mibi non penuatum est imper- 
alorem non dedisse yXvaliK" 

The next article, pp. 400-13, is b; Cobet, entitled Heredetea, He uses Stein's 
text, 1869-71. The apparatus furnished bj this editorfor the three chief MSS, 
the loth century, B of the tith, and R of tbe 14th. leaves nothing 10 be 
d; and no other authorities are needed: "caeteii codices omnes et quid* 
icripturamm ex ii* congestnm est citra damnum abijci possunt et vel in 
I eoniicL" It is even matter of r^rret that Stein has reported all the 
ioni he has. " In codicihni antiquissimis (id est saeculo IX, X et XI 
liE)nbique ri napayrypafi/ilvov Uira cemitor, in minus antiquis idcmubique 
itnr. Hinc farrago discrepantiam lectionnm nnllins pretii et moment! 
:ur. In A B constanter iara additur. in R eadem coustantia semper 
itur. Sexcenties annotatur ;t:p4iCu A B,3:^b>R . . . et similia passim . . . 
I has ineptias : et omnino omnes tcribamm errores, uM di vera icriflum 
eil dubiiatie. Constat inter bmnes verbi Ipxapai imperfectum in Ionia 
'la, ^j(, fiOBv, cum omnibus composili*. Nonne absurdum est tgitur III 74 
l0av afferri ex Ubris: jurov, foov, ^uaav, ^Uaav, larav, iqaiai, et dvoavl ant 



REPORTS, 123 

VI 46 pro trpoaifit scribi irpoa^iet, irpoaifti^ irpoathf, et npoaeirfe ? " Cobet then 
proceeds to discuss the merits of the three MSS and Stein*s judgment on them. 
**A et 6 duo yetustissimi tarn fideliter inter se conspirant etiam in mtnutioribus 
cum perexigua tantnm discrepantia, ut constet inter omnes duos hos libros pro 
uno et eodem habtri posse : ' tanta sunt inter se simiHtudine ut fere unius instar sint^ 
nee tamen tanta ut alter ex altero deseriptus videaiur^ ut rectissime iudicat Stein. 
Miraberis autem in libris tarn antiquis tarn paucas bonas et veras scripturas 
reperiri." The number of corrupt passages in which these MSS present the 
true reading is, by Stein's admission, quite small. But their excellence, Stein 
says, consists in the fact '* quod uni ex omnibus interpolatorum temeritate fere vacui 
manseruntf quod quale sit infra apparebit.*' As to the merits of R, the opinion 
of the earlier critics, which Stein confesses himself to have shared, was that 
they were very high. Stein has convinced himself now, however, that this MS 
has been greatly overrated : ' quidquid unus R offert id omne est aut correctum 
aut erratum.' The merits of this corrector, however^ are praised in such high 
terms by Stein that "de Valckenario aut Bentleio aut Porsono agi putes.*' 
Cobet then proceeds to sho^ how such different opinions can have been formed 
about the same codex. *' DerivatuS est enim ex libro antiquo emendatissime 
scripto et longe longeque fideliore et certiore teste quam sunt A et B. Passim 
ille liber ipsam Herodoti manum solus servavit corruptam apud reliquos omnes. 
Ex illo codice fluxerunt per complures hominum aetates apographa permulta 
alia ex aliis propagata, quorum ultima ad nostram aetatem perdurarunt. . . Hi 
omnes in mendis et erroribus et lacunis ferme inter se conspirant Optimum 
omnium et antiquius caeteris est Romanum exemplum." Cobet then gives a 
long list of manifest errors in R where A and B have the true reading. " In 
antiquo libro unde R propagatus est, versus erant litterarum 15-18. Deprehendi 
id potest III 54, ubi in R sic scriptum est r ol 6k hrtandfievoi iicTelvovrec Aaxedau 
ftcvtup pro: ol dk knurtrdfuvoi iicreivov, ei fikv wv ol nape^vrec hwudaiftovluv, 
Erat autem in vetusto codice. OIAEEHI | 2II0MEN0IEKTEIN0N \ttfjtivvw 
ol irapedvT \ TE2 AAKEAAIMONIQN, et scriba socors omisso versu tertio inepte 
coniunxit hrunrdftevoi iKretvovrec AoKedaifioviov,** To show why, notwithstanding 
these many errors, he still regards R as '* unum omnium testem optimum/* he 
says " utar in earn rem comparatione : duo antiqui libri sunt veluti duo senes, 
homines frugi et graves sed rusticani et ingenii obtusioris. Contra Romanus 
adolescentis instar est, qui nobili loco natus et divitiis affluens liberius vivit 
vino et amori dans ludum, sed idem lepidus, urbanus, elegans, venustus homo. 
Is si forte temulentus est ov6h iryik^ loquitur, sed ubi se collegit et ad se rediit 
faceti ingenii est et iucundissimi sermonis. Quem modo audivimus meras nugas 
deblaterantem, idem permagnum numerum optimarum lectionum solus servat, 
quae tantam habent Ivdpyetav, ut Stein ipse longe maximam earum partem ex 
solo R in textum receperit." Cobet then gives some instances of this, only 
one of which can be here quoted. In VI 128, Clisthenes tested the suitors for 
the hand of Agariste in various ways : ** xal t6 ye fiiyurrov iv ry aweari'g Sitirei' 
paro, Audi nunc Steinium: * Gxvtarlrfi A : awtarly Uredotnus, ryai lariifatci Z. 
Dindorf^ ipse tentabam ry (nvumfyji hretparo,^ Sardi venales, alius alio nequior. 
Quid est igitur ab Herodoto scriptum? Id quod unice verum et in Vaticano 
codice solo servatum est : hr^ owtorOl {awecroT), Dialectus lonica et vetus 
Attica habebat nomen ioT6 id est ovala, et hnear^ pro airowsla^ et ehior^ pro 



124 AMERICAN JOURNAL OP PHILOLOGY. 

ev^aiftovia et KOKtarCi pro dvarvxia et aeuorit pro aUmo^ avata^ et sic <Tvvetrr6 pro 
ovK>otx7(<2, id est <rrrova^rpo^, de hominiun inter se usa et consaetudine . . • 
Ecquis ad hanc lacem serio credere poterit omaroi, qaod nusquam alibi locoruin 
comparet, ex Graecnli correctione esse natam ? ** In R the fifth book is wanting, 
the text of which is, howerer, the most correct of alL Why u this? •* Non 
omnes habebant olim Herodotnm Mmm (nt nee Romani Liriam) sed circam- 
ferebantur codices, qnibos aut singuli libri ant pars aliqna libromm contineretur. 
Sic factum est nt libri nostri non ex nno eodemqne fonte omnes manaTerint« 
sed alii ex aliis interdnm melioribns interdum deterioribus sint descriptL** 
The writer, therefore, of the original of R was probably unable to obtain a copy 
of Book V. It is notable also that in A and B, at the end of Books V, VIII, 
IX, there is a stichometric number. ^ In caeteris libris nulla est arixofierpia. 
Cur? Quia libri V et VIII et IX ex alio libro quam reliqui desumti sunt.*' 

But little space is left for the two remaining articles. The first of them, pp. 
414-23, is by L C. Volgraff, entitled Lamx Satura. We have here conjectural 
emendations of certain passages. E. g, in Thuc I 5, 2, we read that piracy 
was no discredit, even to some of the historian's contemporaries, 01^ K6a\u>^ jcoAi^ 
TcivTo ipav. We are told to expunge icoxiJc* &s a marginal comment (cf. Cobet, 
N. L. p. 441), and the same remedy is to be applied in Isocr. IV 158. In Thuc 
1 10, 2, r:7f (Jwd/wcjf is to be omitted ; in I 134, 4, oi^rrcp rw^ aumi'pyovc is to be 
erased; and in III 82, I, for hoifiuv we are to read M'kfiuv, The last two 
have much probability. There is offered, besides an emendation of a fragment 
of Diphilus, one of Lncian's bveipo^ fj *A}Mcrpv6v, and several of Procopius de 
beUo Persico, 

The last article, pp. 424-48, is by Cobet on Julian. He commends in the 
highest terms the reoent edition of Hertlein, whose only fault is that * misellos 
libros nullius momenti aut pretii, . . . ut testes mendaces et corruptos, tamen 
producendos et audiendos esse censuit" He should have regarded the Vossi- 
anus alone, which is at Cobet*s side as he writes, and *' tam anxia cura ab 
Editore excussus est ut nikU st^enit agendum^'* In these notes there is not 
much of general interesc It is shown that Julian ^ immodicis laudibus extolltt 
in coelum Constantium principem, minimi pretii hominem,'* speaking of his 
wife as yafter^ ^aaiXka^ avSpeiov oiM^povo^ cwerov duuiiov ;(p;7(rrov Kai np^ov koX 
fuyaXtyilrhxov^ " sed veros animi sensus aperit ; <lp* ovk ovetdi^et fun tuii KaTaye?4 
T^ ftopia^^ bri rbv ^ovka iraTp6c, aSeX^ijv, avetjuov^ dirdtT^ uc hroc ein-elv r^c Kotv^ 
4fujv iariac icdi ovyyeveUiQ rdv d^fuov eif rovro tOepdwevaa ;" Several passages are 
referred to the originals from which they are copied. ** Quidquid erat in Graecia 
hominum elegantiorum ut Homeri carmina sic Demosthenis orationes tenebant 
memoriter et statim agnoscebant si quis iis lepide et scite uteretur." We have, 
of course, many illustrations of the barbarisms which are to be left **Juliano 
cum sequioribus saepius sic peccanti*,'* as dfrnprfyjii for dfiapTifaofmt, irapd for 
vit6, fdj for ov, etc.; and as to Julian*s style in general, " quam sit Julianus loquax 
et verbosus «Kc ko^ rpic rahr* etodyuv operae pretium est semel diligenter anim- 
advertere." After giving examples of his garrulity (" quae est, ut libere dicam, 
/uucpd Kol ituiki^yto^ ^7.vapla^\ and a long list of synonyms ** nulla el^^tia 
cumulatorum,*' he concludes : " possem multo plura de genere hoc congerere, 
sed, ut Juliani verbis utar : rl nkpa^ ijfuv iarat tcjv ?^}'uv et ravra ft^pro at neWet ;" 

C. D. Morris. 



CORRESPONDENCE. 

I find that my criticisms on Max MQller*s views on Septentrio have been par- 
tially anticipated by Mr. Ch. Ploix in the Memoires de la Society de la Linguis- 
tiqae, I, pp. 377 sq., a paper which I have only just seen. 

J. P. POSTGATE. 
Trinity College, Cambridge, April xz, X883. 



RECENT PUBLICATIONS. 

Thanks are due to Messrs. B. Westermann & Co., New York, for material 

famished. 

AMERICAN. 

Braune (W.) Gothic Grammar with selections for reading and a Glossary. 
Translated from the second German edition by G. H. Balg. N. Y. B, Wes' 
termann 6r» Co, 8vo. 90c. 

Condit (I. M.) English and Chinese dictionary. Am, Tract Soc. lamo. 
75c. 

Liddell and Scott. A Greek-English lexicon. 7th ed., rev. and augmented 
throughout, with the co-operation of Prof. Drisler. Harper, 4to shp. $10. 

Pliny. Selections from the letters of Pliny; ed. by G. O. Holbrooke. Allyn, 
i6mo. $1. 

Virgil. The greater poems of Virgil. V. i contains the pastoral poems and 
six books of the i£neid; ed. by J. B. Greenough. Ginn, il. i2mo. $1.55, 
without vocabulary $1.25. 

BRITISH. 

Aclfric's Lives of the Saints. Edited, from MS Julius E. VII Cotton., by 
W. W. Skeat. 8vo, sd., 256 pp. los. 

Babrius. Ed. by W. Gunion Rutherford. 8vo, 292 pp. Macmillan, 12s. 6d. 

Demosthenes against Androtion and against Timocrates. Ed. by William 
Wayte. Post 8vo, 320 pp. Cambric^ Warehouse, 7s. 6d. 

Euripides. Medea. Ed. by A.W.Verrall. i2mo, 160 pp. Macmillan, 3s. 6d. 

Guest (Edwin). Origines Celticte. 2 vols., 8vo, 960 pp. Macmillan, 32s. 

Homer's Iliad. Done into English Prose by A. Lang, W. Leaf, and E. 
Myers. Post 8vo, 490 pp. Macmillan. 12s. 6d. 

Homer's Odyssey. Books 21-4. Edited by S. G. Hamilton, i2mo, 136 pp. 
Macmillan, 3s. 6d. 

Horace. Liber 4. Ed. by T. E. Page. i2mo, T02 pp. Macmillan, 2s. 

Lenormant (F.) The Beginnings of History. From the second French ed. 
With an Introduction by Francis Brown. Post 8vo, 600 pp. Low, 12s. 

MQller (F. Max). Sacred Books of the East. Vol. 14, Sacred Laws of the 
Aryas. Translated by Georg Buhler. Part 2, Vasishtha and Baudhayana. 8vo, 
390 pp. Frowdc, I OS. 6d. 

—^ Sacred Books of the East. Vol. 18, Pahlavi Texts. Translated by E, 
W. West. Part 2, The Dadistan I Dinik and the Epistles of Manuskihar. 
8vo, 506 pp. Frowde, X2s. 6d. 






^a,*- 



X. £^ -X ^^ ..IT t ■*r"T ., .'.Til JJE 




Urnr'i.ukgg. iiIHi^ jcd. :as ttz. ^ik. 






^^irta-ltrr GcdIL Ciit^s snirris JO'i-'Tr^. .r* ir'TTi'^ «t rrmnru^ 






* 31 



^ tcm:dr=3 i^ccasnaa in Tm* Tar "Zh. FarTe Rii^Tt*. l3-r3L \i^rmier. 3 fir. 

Gr. ia-i. />7«x. 7 Sc 50. 

Inraieaeter tjmfnp^'. ftsdes T-an im i wu> . Tjme L Htnics ssr la gcftB- 
naire liL.ir/:r:r:ie 'ie let laasoc pensacL Tone II. I pBrtiir, >felxai(es Iruucas* 

FUIjoo I L. CL) Ar.at ardLeouogic^oc de ia BibLe. Ib*4 cvec 93 pluKlws. 
Dtlkommu et Bn^wet, 20 fi^ 

HiM i]. A.) La L^eade ifloee zraat Vb^He. Gx. ta<<L Zfw.r. 3 fe. 

Imbanlt'Hoart (Camillet Asecriotes. kistoriettes et boos BOts, en cbisois 
parU ; arec «ne tnuiaclioa firui^aise et des aoCcs cxplkatrres. Ia-ix. P Aing . 
Lervux. Cart., 5 fir. 

Lacume de Sainte-Palaje. Dictiofuiaire hlsto^qne de Faacicii langage inn- 
fois. Pabii« par les soins de L. Farre. Tome 10 (dernier). IB-4. CkmmipUm^ 
30 it, Le Icftac I a para en 1875. L'oarrage complet en 10 voL 300 fr. 






I 



RECENT PUBLICATIONS. ^2^ 

Landberg (Carlo). ProTerbes et dictons da peuple arabe. Volnme I, Pro- 
vince de Syrie, section de Saydi. Gr. in-8. Leyde. Afaisonneuve, 15 fr. 

Menant (J.) Empreintes de cachets assyro-chald^ns, relev^s au Mus^e bri- 
tannique sur des contrats d'int^r^t prir^. In-8, 106 fig. Maisonmuve, 3 fr. 50. 

NaTille (Ed.) Inscription historique de Pinodjem III, grand pr^tre d*Ammon 
k Thebes, traduite et comment^e par Edonard Naville. In-4 avec planche. 
Afaisonneuve, 8 fr. 

Notes grammaticales sur la langue de Lifu (Loyaltys), par le P. A. C. prStre 
mariste. In.8. Afaisonneuve, 4 fr. 

Rosny (L^n de). Essai sur le d^chiffrement de I'^criture hi^ratiqae de 
VAm^rique centrale. 4 liTraison (demi^re). In-folio. Mcdsonneuoe, 25 fr. 

GERMAN. 

Aischylos. Agamemnon, erklart t. F. W. Schneidewin. 3 Aufl., besorgt ▼. 
Otto Hense. gr. 8, xvi, 218 S. Berlin, Weidmann, 2.25. 

Anthologia lyrica. CuraTit Thdr. Bergk. Ed. Ill, ex poetamm lyricomm 
graecorum ed. IV expressa. 8, ▼!, 383 S. Leipzig, Teubner, 3. 

Aristoxenus v. Tarent, Melik u. Rh3rtbmik. d. classischen Hellenenthums. 
Cbers. u. erUutert darch Dr. R. Westphal. gr. 8, Ixxiv, 508 S. Leipzig, Abel, 
n. — 3a gcb. n. 33. 

Bartholomae (Chr.) Handbnch der altiranischen Dialekte. gr. 8, vii, 272 S. 
Leipzig, Breitkopf &• IldrUl. n. 6. 

Benndorf (Otto). Vorlaufiger Bericht Ob. zwei Osterreichische archaeologische 
Expeditionen nach Kleinasien. gr. 8, loi S. m. 5 Taf. Wien, GerohVs Sohn. n. 3. 

Bergmann (Fr.) Hohelied Salomons a. der Prediger Salomo, aus dem urtext 
fibers, u. erkUirt 8, viii, 11 1 S. Strassburg, TreuUel^ WUrtz, n. 3.50. 

Braun (W.) Der Thyestes d. Enripides. gr. 8, 3i S. Wesel, 1883, Kuhkr 
in Comm, n. — 50. 

Broschmann (M.) De y&p particulae nsu Herodoteo. Diss. gr. 8, vii, 89 S. 
Leipzig, 1882, Fock, baar n.n. 2.5a 

Baschmann (H.) Bilder aus dem alten Rom. gr. 8, iv, 283 S. Leipzig, 
Teubner, n. 3.6a 

Brugsch (H.) Thesauras inscriptionum aegyptiacamm. i Abth. A. u. d. T. 
Astronomische u. astrolog. Inschriften altigypt Denkmiler. hoch 4, vii, 194 
S. Leipzig, HinricKs Verl, baar n.n. 50. 

Caneilo (U. A.) La vita e le opere del troTatore Amaldo Daniello. gr. 8, 
Ti, 282 S. Halle, Niemeyer, n. 9. 

Commentaria in Aristotelem graeca. Vol. XIII, partes I, II. gr. 8. Berlin, 
G, RHmer, n. 9. Inhalt : I. Sophoniae in libros Aristotelis de anima para- 
phrasis. Ed. Mich. Hayduck. viii, 175 S. II. Anonym! in Aristotelis cate- 
gorias paraphrasis. Ed. Mich. Hayduck. iv, 86 S. 

Cruel (R.) Die Sprachen u. Volker Europas vor der arischen Einwanderung. 
Streifzttge auf turan. Sprachgebiete. gr. 8; rii, 174 S. TitXmoX^^ Meyer in Comm, 
baar n.n. 2.50. 

Culmann (F. W.) Etymologische Aufs&tze u. Grunds&tze. VI. gr. 8. 
Strassburg. Schmidt, n. 1.20. (I-VI, n. 9.90.) Inhalt: Umschau auf dem 
Gebiete der historischen * Zeitformen u. ihrer Augmente. Zweiter Beitrag sur 
Anfkliirg. grammat Geheimnisse. 48 S. 



S Ic — wr T?fftg. 



-- ^— aBi . ■ 7 J '9C vaS^C I^ZBCSUL .^ 



^C C ffi^ 



=. JT^ 



'~"*" *•" "^"^ '?''' ■~i'ii.»ll*'^£ i^ •fk^fc i;* 9^ JV, ^9 



(i) 



■■T " 



"?T^iTIir'""T-U 



-=T-r. Fi. £ xriii. «3 S. 



^rr:C * -;— r^irrrfT I Iti. ^ £. 









L^rril r. 




fv-T.irr; 5 TL Epsrr^ gr. 8, xii, 
ST. Ill 5l i^r-^r:^ ^.-tr il 5. 



Jsrdxjs H. S7ati«cih.&f ^ kjscaruaK Ttla^auaa italicimK. gr. 4, 27 S. 

Klc:>fr ^L.» <^ii Tx-r.tiia^ ia diiI:Kro Trjarbcs scriptoKbcs debcat. Diss. 

gr. 5, 9C> St Hili5 Sxx.TEsa. Berlin, AVttf^ ^ -VjLir: b*ar m. 1.2a 
Klussmaan vKcd.) Cxrac iifricuue. 4, 15 S. Genu JuuniM" 5*rt 1.5a 
KoU'.n^ vE^.) Die Dordiscbc ». die cn^asdic Versioa der Tristan-Sage. 

a ThL Sir Tiistrem. gr. 8, xciii, 19a S. Heiibrwm, Bamimfftr, (Jt) o. 12. 



RECENT PUBLICATIONS. 1 29 

Kramm (Frz.) Cb. Konrads v. Heimesfurt Sprache [Laut- u. Formenlehre] 
u. Vcrskunst. gr. 8, 80 S. Strassburg, TrUbner. n. 1.80. 

Kuhfeldt (O.) De Capitoliis imperii Romani. gr. 8, 83 S. Berlin, Weid- 
fnafiH. n. 3. 

Kurschat (Frdr.) Littauisch-deutsches WOrterbuch. gr. 8, xii, 530 S. Halle, 
Buchh. d. Waisenhauses. n. 12. (cplt. n. 39.) 

Larfeld (Guil.) Sylloge inscriptionum boeoticarum. Praemittitur de dialect i 
boeoticae mutationibus dissertatio. gr. 8, xxxvi, 233 S. Berlin, G, Reimer, n. 10. 

Livi (T.) Ab urbe condita libri. Scholarum in usum ed. Ant. Zingerle. 
Pars IV. Lib. XXVI-XXX. 8, xxiv, 233 S. Prag, Tempsky, Leipzig. Frey- 
tag, n. 1.20. 

Livi (Titi). Ab urbe condita liber XXVIII. Fttr den Schulgebrauch erkldrt 
▼. F. Friedersdorff. gr. 8, vii, 127 S. Leipzig, Teubner, 1.20. 

Livius, ansgew&hlte StUcke aus der 3 Dekade. Mit Anmerkgn. v. W. Jordan. 
3 Aufl. 8, xiv, 187 S. Stuttgart, Neff, n. 1.50. 

Ludwig (Thdr.) De enuntiatorum interrogatiyorum apud Aristophanem usu. 
Diss. gr. 8,69 s. KOnigsberg, 1882, ^<7^r. baarn. 1.20. 

Marx (Frdr.) Studia Luciliana. Diss. gr. 8, 100 S. Bonn, 1882, Behrendt. n. 2. 

' Merguet (H.) Lexikon zu den Red en d. Cicero m. Angabe siimintlicher 

Stellen. 4 Bd. 1-4 Lfg. Lex.-8, S. 1-144. Jena, Fischer, (i) n. 2. (I-IV, 

4. n. 137.) 

MGller (Herm.) Das altenglische Volksepos in der ursprOnglichen strophi- 
schen Form. 2 Tie. in i Bd. I. Abhandlungen ; II. Texte. gr. 8, 160 u. 78 

5. Kiel, Lipsius b* Tischer. n. 5. 

MQnzel (Rob.) De Apollodori Trepi i9e6v libris. Diss. gr. 8, 39 S. Bonn, 
Bthrendt, n. I. 

Neisser (Walther). Zur vedischen Verballehre. Diss. gr. 8, 35 S. 
GOttingen, 1882, Vandenhoeck 6r» Rupreckt, baar n. — 80. 

Nissen (P.) Der Nominativ der verbundenen Personalpronomia in den 
dltesten franzdsischen Sprachdenkm&lem. Diss. gr. 8, 83 S. Greifswald, 
1882. Kiel, Lipsius &• Tischer. n. 1.60. 

Ovidii Nasonis (P.) Carmina selecta. Scholarum in usum ed. Henr. Steph. 
Sedlmayer. 8, xviii, 139 S. Prag, Tempsky. Leipzig, Freytag, n. — 80. 

Werke. Deutsch im Versmasze der Urschrift. 4 Lfg. 8. Stuttgart, 

Werther, (i) n. — 35. Inhalt : Metamorphosen. Ubers. u. erlilutert v. Rhart. 
Suchier. 4 Lfg. 4 Aufl. S. 129-68. 

Petri Hispani de lingua arabica libri duo, Pauli de Lagarde studio et sump- 
tibus repetiti. Lex.-8, viii, 440 S. G()ttingen, Dieterich*s Sort, in Comm, baar 
n.n. 20. 

Plauti, F. Macci, comoediae. Rec, inslrumento critico et prolegomenis auxit 
Frdr. Ritschelius, sociis operae adsumptis Gust. Loewe, Geo. Goetz, Frdr. 
Schoell. Tomi II, fasc. 3. gr. 8. Leipzig, Teubntr, n. 3.60. Inhalt : Mer- 
cator, rec. Frdr. Ritschelius. Ed. II, a Geo. Goetz recognita. xiii, 124 S. 

Plutarch*s ausgew&hlte Bic^raphieen. Far den Schulgebrauch erklart v. 
Otto Siefert u. Frdr. Blass. 3 Bdchn. Themistokles u. Perikles. Von Dr. 
Frdr. Blass. 2 verb. Aufl. gr. 8, 136 S. Leipzig, Teubntr, 1.50. 

Roscher (Dr. W. H.) Nektar u. Ambrosia, gr. 8, viii, 116 S. Leipzig, 
Teubner, n. 3.60. 







V.iti3 V.;^! Die mpcla^Lentf Btijgffat k,C— yirnai JgaedJopilrfiii. 
f .1.1) ' H.:iuir. t'afffrtg»_ hzmoid. Srhafrr ^i ■iifii~] fE- 9. 5o S. Bobb. 
I»3. iltf-MD. m. IJXt. 

SopfancLu Aou^Mie, idwiwxs » o^ c^ JxAz^ Stfcnhw t- S. xii, 4! S. 
Pta^ Tomftkf. Lerpaif, frrytKf. L — ^Ol 

Sophnfc:««' Oeii.p«» aiC Kouiooi. Ft J«» S " " " 
Pr.&wMnaL p.>,:ii,665. i^^*". r'^. ^ ■■ 

Tuagl (T.) Bocthii c 
DiM. p. a. IT, ic^ S Goci:^ iSa*. A A. PvtJka. m. ».4Dl 

ftCBiJies. tm^achfc. Oi^a t. CBfc'- pliiJ: j>e*' ■>"* ■■III iHi I !■ hi d. cb^ 
■Btemcluci uf br'-tterea idnlea. Hng. ■. PioC Dr. E^. KAIbiaB- ^ Bd. 
~, I Hfl. 163 S. Heilbrou, Hem^npr. ■. 15. 

— fruiAiisdc. Hng. t. C KMti^ a. E. EoKhviu. 4 Bd. i a. 3 Hit. 
>, EM. U| B. MO- Inkili : >- Nntilc -le Ix Ckswcc's Lelia ■. Woke. 
J-ibt. UtboC ii. 67 S. >. Die Qtuniiui da becoam Toole ia Ncn- 
P>j«I»e». Vow JaL J^er. ir. 63 S. 

>e>enb>i2 <H.) De PfTif"— -■'-■''■v-" T"*™*"*« ** r-n.^.— ci Ad- 

I eit UlvU fcofiapiika (li:h.f. fi. 4, ao S, Lcipiif. HimritU Sirt. bur 

ao. 

»dTi. Conclii. ajtuJci. Ftr den Schalsebrwdl erkUn t. Dr. W. Pfiuner. 

Icha. Bach I a. II. gr. a, n. ijoS. Gothz, /: ^. Avtjbr. 1.9a. 

— ih eiccMa diri AogaMi libri. Rec W. PfitiseT. Putic L er. B, 71 
EM. —60. 

heoHalfi cptfcopi Aareliaoenm de iadicibiu venos ab Hena. Higea leco- 

i. p. 4, xiii, 31 S. Bern. iSSs, Aij«. n. i.n. 

bcophaoi* chionogtaphia. fee Caiolni de Boor. V'oL I, leztam giaecnm 

inent. p. 8, viii, joj S. Leipdf, Ttuintr. n. >a 

bacy'lidii de bello pelopoaoeiuco libri VIIL EipL En*l Frdi. Poppa. 

II, loani anxit ci emeodani Joa. Haitb. StaU. Vol. FT. Sect. II. ^. 

r> !i. LeipilE. Tni6tier. 3.70. 

flicr (F.) Qoaeitionet aliquot Sallutiaaae gnutmuiicae et crititae. DU*. 

>. 5]!i. BeroliDi, 1SS3. GOUiagpi, Van^tMiMcJl ^ Xt^ncht, baariul.SOL 

'Irkh (J,) RhUoTomanitche Cfarettomatbie. Teite, AnncHcgn., Glottar. 



RECENT PUBLICATIONS. 131 

I Thl. A. a. d. T. Oberlslndische Chrestomathie. gr. 8, viii, 275 S. Halle, 
Niemeyer, n. 6. (i u. 2, n. ii^.) 

Vergils Aeneide, f. SchQler bearb. v. W. Gebhardi. 3 Tl. Der Aeneide 5 
a. 6 Buch. gr. 8, xii, 183 S. Paderborn, F, Schdmngh, n. 1.60. (1-3, n. 4.40.) 

Wiedemann (Alfr.) Die Sltesten Beziebangen zwiscben Aegypten u. Griecb- 
enland. gr. 8, 22 S. Leipzig, Barth, — 60. 

WQstenfeld (F.) Per Tod d. Hasein ben 'Ali u. die Racbe. Nacb Hand- 
schriften ttbers. [Aus Abbandlgn. d. k. Gesellscb. d. Wiss. zu GOttingen.] 
gr. 4, ix, 213 S. GOttingen, DietericKs Verl, n. 9. 

ITALIAN. 

Avolio (C.) Introduzione alio stndio del dialetto siciliano. Torino, 1882. 
8iro» Tiii, 480 pp. 4m. 

Campello della Spina (P.) Demagogbi e conservatori al tempo di Cesare, 
ed altri scritti. Firenze, 1883. i6mo, vi, 259 pp. 3m. 

Canini (M. A.) £tttdes ^tymologiques. Torino, 1882. 8vo, xvi, 2gi pp. lom. 

Decea (C. A.) Saggio d'arcbeologia ed araldica biblica. Piacenza, 1882. 

8to, 112 pp. 2m. 40. 

Fumi (F. G.) I. Note latine e neo*latine. Palermo, 1882. 8vo, xx, 150 pp. 
6m. 

SPANISH. 

Mdida Q. R.) Sobre los vasot griegos, etniscos i italo-griegos del Museo 
arqaeol6gico nacional. Madrid, 1882. 4to, 48 pp. 2m. 50. 



BOOKS RECEIVED. 



Braune (W.) Gotbic Grammar. Transl. from the second German edition 
by G. H. Balg. New York. Westermann 6* G;., 1883. 

Campbell (L.) The Theaetetas of Plato, with a revised text and English 
notes. 2d ed. Oxford, At the Clarendon Press ^ 1883. 

Demosthenes. The First Philippic. Ed. after Rehdantz by T. Gwatkin. 
London and New York, MacmiUan &* C^., 1883. 60 cts. 

Deutsche Litteraturdenkmale der 18 Jabrbunderts in Neudrucken heraasg. 
V. Bemhard Seuffert. No. 11, Der Messias, Erster, zweiter u. dritter Gesang. 
Von F. G. Klopstock. Heilbronn, Geh', Henninger, 1883. 90 pf. 

Englische Sprach- u. Litteraturdenkmale der 16, 17, u. 18 Jahrh. herausg. v. 
Karl Vollm6ller. (i) Gorboduc, or Ferrex and Porrex. A tragedy by Thomas 
Norton and Thomas Sackville, A. D. 1561. Ed. by L. Toulmin Smith. Heil- 
bronn, Gebr, Henmnger^ 1883. 2m. 

Euripides. Medea. Ed. by A. W. Verrall. London and New York, MacmiUan 
<y Co,, 1882. 90 cts. 

Fasnacht (G. E.) Synthetic French Grammar. London and New York, 
MacmiUan &* Co,, 1882. $i. 

Goethe. Goetz von Berlichingen. Ed. by H. A. Bull. London and New York, 
MacmiUan 6r» Co,, 1883. 55 cts. 

Homer, The Iliad of. Done into English prose. By Andrew Lang, Walter 
Leaf and Ernest Myers. London and New York, MacmiUan &* Co,, 1883. 






* • 



' ,,- 



-zr z m- 



--J-- c- 



rams: 









£■". » L- TLT.rif r_ 



i „. XT 3 t 



r 1.5^5 






t¥' -' / 



' *■ ^ ' 



f / . <- ,,'. /v " ' '' '•i''/%»i A -!ti* vv.-'r.tr'ajj^ Wirt if -wiici 
-' ,-•-*/■'■ '-'^ ?"/-*/« ^;i -^^ -■ y\T:kA V -».:gv mrr XL. U-?iil\* o^i 

;,' ' *f h ;,"/>• '/f ■,> ^ .v^'r*/ '^ '''^*/ ^c^ futre j.TOi is. £acs;^Ue, ^^-g 
n ' . ^ f^ ./, TrA f w*,' f, >-*' ■>^^^ itfy. ^'*'\t cire Ha* beea toJten tn the sclcctioa, 
/,*>,<.♦ f^A f*tf* • (/f /*" ir ^*fy^ '•//fAj/f^/' jIr**-»rat.oo of the derelopment of this 
^f'/ * fHj^f'ff'itti 'itt'i hfitt''tif thntnnt-r. The repatation of the editors is a 
•nOf \> tit pnif^tiif ffit ♦!•* f/ ' Mf.^/ y w»fh which the work has been done. 

AMf/M tN **|,*,(.#M, f'/f ¥r)tom fill* fK^e^i to I^atin manascripU is attended 

Mil -h pthfii tWih' mIM'», »»t»t/hl Iff wflttnne with especial interest the pnblica- 

^\ iH M^ j-Mt li /Mi'ltHlU-, h»m1 AiiifiihnM HhmricK ought to contribute liberally to 

III. rti|)ijiHH m( mm* h Mhil*>f lMl*lh^^ 



AMERICAN 

JOURNAL OF PHILOLOGY 



Vol. IV, 2. Whole No. 14. 

L— STICHOMETRY. 

Introduction. 

The following investigations have been undertaken in the Hope 
of obtaining some critical conclusions with regard to the extent of 
early documents, chiefly Biblical, from the apparently insignificant, 
yet highly important data furnished by certain numbers appended 
by ancient scribes to the books which they copied. It is only lately 
that I have come to regard, with any other feeling than complacent 
pity, the labors of those Masoretic editors of the Hebrew Bible who 
so carefully inform us as to the number of verses and the points of 
bisection of the separate books ; the natural impulse of one's mind 
being towards the conclusion that such work might perhaps be 
agreeable at some period of involuntary incarceration accompanied 
by a most plentiful lack of books. The Masoretes themselves, how- 
ever, seem to have been sensible of the importance as well as of the 
arduous nature of the work of book-measuring, since they preface 
their annotations with the word plH, which is generally under- 
stood to be an encouragement (Jortis esto) either to themselves 
or their readers. How much more strongly would they have ex- 
pressed themselves if their task had been, like ours, the inverse 
problem of restoring the ancient books from their accredited 
measurements! Doubtless their sympathy would have flowed 
(after the approved Rabbinic fashion, which I remember to have 
noted somewhere), in votive offerings of midnight oil for the labors 
of the devoted calculator. 



*'•! 















• • 












-- • . -» 



- _i n-^ 



I — — 



^^e :r rbe 5th 

- — I — 
: =1 J 



.1 t^.ZZ .t 






r r. i^c oodd 







38 



^- >4.i^ 



■^, ■£ # ^ *" x*^'* ^T* « * 



X - v*.».:v'xir*-n:i? its xa-jtu ^'c ib; Tr*k~^ r jr the Pa py r u s 
^ ..v^vjL.K"i:i r,J:> rr-.-.-of i:? wzi xru_^djj^ instances of 
,, V .. .1^ :o ;>.vrr:>. xrc r^c^*ic::Ly ibe Cii;r rVr:= of Greek 



STICHOMETRY, 135 

numeration presents itself, as, for instance, n. 1027 (ed. Oxon.) has 
the subscription 

KAPNEI2K0Y ♦IA12TA B. APie. XXXHHAAAnill, 

which implies that a certain portion of the writings of Kamiskus 
contains 3238 verses. What these verses represent in prose writings 
is a problem presently to be considered. 

Other instances of the preservation of the more ancient Greek 
numeration may be seen in the MSS of Herodotu^, Cod. Lauren- 
tianus LXX 3, and Cod. Angelicanus C i, 6, and in several im- 
portant MSS of Demosthenes. 

Stichometry earlier than the Alexandrian Library, 

It is sufficiently evident that the custom of measuring literary 
works by <mx<M is coeval with literature itself, and instances may 
be given which establish the continuance of such measurements, 
both for prose and verse, down to the twelfth century, if not later. 
It is possible, however, that these more modern subscriptions are 
to a great extent traditional measurements from an earlier time. 
Ritschl,* in his important researches on the subject of stichometry, 
came to the conclusion that Callimachus, of the Alexandrian Li- 
brary, was the inventor of the stichometric method ; the chief 
authority for such a statement is found in the following extracts 
from Athenaeus : 

Tov Xaip€<p&vros jcal avyypofifJM apaypa<j>€i KaKkiftaxos cV r^p rwv nayrofknr&v 
vtMuei ypaicfttiv ovt»s * Aciirya ocra typa^tuf * Xaip€<l>&p Kvpfjfiimn ' uff t^rjs 
Ttpf dpX'l'^ \miO/iK€V * *Eircid^ fu>i voKXclkh iirianiKai * arlx^v roc. Athen. 
VI, p. 244 A. 

'AWypo^ dc avT^p (v6fAOP riya ovcatTiKhv) KaKXifMaxos iv r^ rpiT^ mvam 
Tm¥ p6iMnfj KoX apxflp airrov rrivdt irapiOtro * *0dc 6 tf6fiog tcror €ypa<f)ri jcal 
o/«oiO£* mix^P TptoKoaitiW ctico(ri rpi&v* Athen. XIII, p. 585 B. 

It will be evident, however, that these quotations really imply 
nothing more than a general statement that Callimachus entered 
books under certain catalogues, in which were found, with the 
name of the author and the title of the book, the first line of its 
contents, and the number of lines. And M. Graux' has pointed 
out that we have evidence anterior to Callimachus of the exis- 
tence of prose works measured by their author in ftn;, which is 

* Opusc. Philolog. I, p. 84, ' Revue de Philologie, April, 1878, p. 97. 



'j^ ?*ZI^^Z*x~I 



r^aercaccis 
e5 of 




inc^BT » ~ 






X :"s.--- 



^ . . u -' '-r _'*!*-! »-::r t /t-^^ a nnt^- w, ir-s "ve isk bow 

. , -^ ». • '*^- T -: ^^yr ^in-r^*' rmsinr iientsr ir scasdArd of 



->^ >w * --' '^ r' rr*r .X n r:^ rr-xises. Tr*K.5e ri terms of 
>v . . J5«^ ->^ It- ;:i.nvrir uoi n»± iirrr nc rrz=«cer. and the 



. .V. V*: .\ :rt: -'\m:r 3- it w r.^: prssr^'e 1^:^ s thrown 

.V ," ' -i.. ..* - v*v: :•?:?, cr n^rci:!!:^. re a ro:«:ot the Iliad, 

v V ^x- -'*-^ -"* -.^^-ii::?. w-i-i^ v^ jr:^ r:i!'i bv that writer 

> . , . V :^ K^v ^.fc^ >.\ ,xv r^ ■ «* i^i 2a- *; ^r .fior^ss to the MS 

Xn- . .v^ - v\*^i>c:x r^'^'-^^-r;. t' ixirr^i^e whether the word 

. V .*vx ^t'Xvtt.'^z 3VirT C5 :s:r:/i*i j.tc rzoecnite meaning of 

.,N^ .'.vO \x\^ 5;.>?rar Trtrj^r^-vr wh^ch ::raT >d«:tiiV it as a 

V c v:*^^. .>. >u.rxSt; XT r:::KS ^ ^rc!i the urif jmiiry of printed 

\ V vs^vi. cs 'k^^tmI =>-jLn:r:§ 35 s£n:rCy row. lioe, or verse. 
.. ..V 'o** 01 stoc^ in the breastplate of the high 

. . W > ^rv\«w * •« v^Ex. XXVIII 17), whidi the Vul- 



STICffOMETRY, 137 

gate renders by in primo versu erit lapis sardius, etc. In a military 
sense the <rrixo^ is used of either a rank or file of soldiers, but more 
properly belongs to the latter. Thus we find in Montfaucon. Bibl. 
Coislin-, cod. 347, some fi"agments of a little work De Tacticis, and 

here arixos Koi d€Kauia Koi \6)(09 t6 avrS iarip * paSot cWi ^xiXayyof t6 fitp 
rh fitramov Snap xal 6 a7r6 Xo^oyov tni ovpayhv (rri\o^ Koi ^dBos Xfycrcu, and 

the definition o{\6xos contains the following interesting statement, 
showing that the fondness for particular numerical arrangements 
was gratified on every opportunity : mw /Wv tA av<rn)fia t6 i^ dvdpci>y 

oxro> oc de r A c^ dv^pSiv do»d€Ka, ol dc t6 (k dcicac^ av^fMV ifkriOos h fcai Ttk(i6v 

<t>aai Koi (rvfjLfirrpov, We see that a preference is shown in arranging 
the men for the numbers 8, 10, 12, and 16. 

Precisely similar statements are found in iElian, Tactic. IV, from 
which we may take the following : 

O dc \6xos c'otIv apidfi6s ay^pS>v om6 rivot riyovfiivov koi rStv ficr ax/rov 
IhriaSiv rtroiUimv fifXP^ ^^^ T€\€VTaiov * top dc dpiBfxiv tov Xd^pv ol fup oicra> 
dpdpStp €*iroii;(ray, ol Ac da>dc«ea, ol dc dcKOt^ * ^orcii) dc pvp c«citaidc«ea dpdp&p 6 
\6xot * avfifi€Tp<af yap ?;(€i np6s re t6 fuJKos r^s <f>dKayyof ' 6 \6xos dc SKos 
ffoXcirac vrlxo^i ^popa^erai dc kclI dtKcufiaj xmo dc rtpap cyoi/iorta. 

2Tixos a measure of syllables rather than words. 

We shall then not be surprised if we find that the scribes, in 
arranging or in measuring their lines, show a preference for 
particular numbers; and any such plan of fixing the length 
of the line must evidently be by the enumeration, either of the 
letters, syllables, or words which the line contains. The last of 
these suppositions may be rejected almost at once ; the continuous 
writing of early times pays little regard to words, which are broken 
up by the line-endings with the greatest freedom. On the other 
hand, the very greatest respect is paid to the division of syllables ; 
it is true that this is somewhat obscured by the feet that the ancient 
division of syllables is different from the modem English method ; 
but if we observe that the ancient syllable, in Greek manuscripts, 
ends with a vowel or weak letter, we can easily trace in most of 
the early MSS a complete system of syllable-section ; and this 
respect paid to the syllable is a transcriptional phenomenon of great 
importance.* 

In fact, in many cases where we should speak of words, the 

' Cf. Kuhner. Grammar I 273, and Wcstcott and Hort, Introd. to N. T. 315. 



'X - 



A. 



Z=S « S. ^« I 



xr . 



■ M il II ^ f 



in 



• ■ -*^ - 






'*^* ' :- — 2t 



« C 









»b- ■*• *^ \ ""*- '"' 






- JT ^ r:t^rz:Jii 



T •■ 



t* M' 



crtbe 






lull I ^1 



ict: 5iE^; aad 
-:c**rLX be able 



. ^' ^ V siV- u "^-irrt*" 1 n* "5>.it KT "zr:^ 



-^ '•^ *r«i. -■«M«k-*Vk. fi^ -■»»■ ^»' .mtm t Sk, '-•, *w m Aft r» r" 

,. v*. v« '•* mt^^A ""^r rmi rr—y ?-— Tt .*« "^ i-iCtt-jmw "Ts V^ X"W €wMmw 



<^ 



tX^Bot 



• m * * * 



^v 



I-rtr;^xs5. e£. E^w* P- 69. 



STICHOMETRY. 139 

^Tt'^oy identified with hexameter of i6 syllables. 

According to Galen then, 39 syllables of prose writing are equiv- 
alent to 2 J hexameters ; 83 syllables represent 5 hexameters ; the 
two quotations together, 122 syllables, do not amount to more than 
eight hexameters. From which it is obvious that the prose hex- 
ameter of Galen is 16 syllables ; and we observe further that this 
line-unit is dignified with the alternative titles of ftrw c^d/Acrpov, ciror, 
and fFTixoi rjfHaiK6i, The peculiarity in the use of these words seems 
to consist in the extension of the meaning of ftror which is implied 
in the use of an adjective, from its normal meaning of a heroic or 
hexameter line to the more general application which includes any 
written line whatever ; while, on the other hand, the term orixoy, 
which normally represents any written line whatever, undergoes a 
contraction of meaning until we frequently find it used synony- 
mously with hexameter, even to the exclusion of lines of other 
lengths. A curious instance of this may be seen in a tenth-century 
MS, written on Mount Athos, and described in Montfaucon, Bibl. 
Coislin., p. 597. Here we find frrixps used of hexameter verses, in 
distinction fi'om iambics. 

ocot dt^ arixetv Koi Idfipiop ttfipaaav* 

''Ofiripos trrlxovs, *Airok\o»vtos arixovft 0€6Kpvros 6fwi<as, 

"ApQTos ofioims, NiKapdpos o/ioioDff, Mtyay^pof Idfi^vs kt4. 

So far, then, everything tends to the assumption that the <rrlxof is 
equivalent to the average hexameter, a conclusion which will be 
abundantly verified by an actual reference to texts and documents. 
It also seems that there is a preference shown for measuring the 
average hexameter by syllables, probably sixteen in number. The 
number of syllables in a hexameter is an instance of variation 
between fixed limits (cf. the definition quoted from Hephaes- 
tion) ; but the number sixteen invites especial attention, as being 
that suggested by the first line of Homer, and also on account 
of its symmetrical or square character, which, as we have already 
seen, gave it a preference in the determination of the conven- 
tional number of ranks in a phalanx of soldiers, and which was 
always an important feature in the eyes of thbse who saw special 
Pythagorean virtues in numbers. 



MO AMERICAN JOURXAL OF PHILOLOGY. 

AlUmatioe of a UtUr-Hne, 

On the other hand we must enquire whether there is any 
ground for asserting the existence of a letter-line in preference to 
a syllable-line ; for it may be assumed, I think, with safety that the 
art of transcription undergoes a double development : first, it 
changes from letler-by-letter writii^ to a writing syllable-by- 
syllable, and from this, for greater ease in reading, to a transcrip- 
tion word-byword ; so that the lines for successive periods of time 
would end, in the first case with the geometrical limit of the line, 
in the second and third cases with the most convenient syllable or 
word. And this change is evidently in the direction from a very 
regular line, such as those found in many eariy inscriptions, to one 
not quite so regular, such as occurs in early vellum MSS, and so to 
the somewhat irregular later writing. We should expect then to 
find some traces of the measurement <^ the actual number of letters 
in a line. The following are the only instances with which I am 
acquainted. 

On the back of an astronomical work of Eudoxus,' dating 
from the second century before Christ, are twelve verses form- 
ing the acrostic eyaosoY teXNH ; these are arranged so that 
each of the letters is a day, each of the lines represents a month, 
■■ poem a year of 365 days : according to the verse 
r (im, ypoftfia If wupa.' Another instance is given by 
ippus Alexandrinus (11 17,4; H 23) in which the 

'ApTtiudot lAtiTt tpirot tfoji^ ifpta Kovpat 

'. 38 letters (rrfi of» ypiitfiara ivnr Xq rov aTi;i[ov^. 
hcse instances bears very exacdy upon our enquiry ; 
>wever, traces of a method of measurement which 
en common in early times, when the letter, rather 
ible, was the basis of metre ai>d prose alike. It is 
evident principle that a MS written on the basis of 
be reckoned by the number of its letters, and a MS 
reference to the syllable will be numbered by its 



Gt. Palneognpbic. p. 7. * Birt. Bachvresen, p. 161. 

■Bin. p. t6a 



STICHOMETRY, 141 

Actual Calculation of the Length of Lines for Various Authors, 

We shall now confirm these results by the examination of 
actual data supplied by MSS and authors, following closely the 
results of M. Graux, with such changes as may be necessary in the 
arrangement of the matter, and some additions and corrections. 
Where the results deduced for the value of the artxoff are given in 
letters, we have only to remember that the average hexameter, 
taken by M. Graux from 50 lines of the Iliad opened at random, is 
37.7 letters ; and where the result is given in syllables, the average 
is 15.6 syllables, as deduced by Diels* fi-om the first fifty lines of 
the Iliad. In every case we must divide the estimated letters or 
syllables of a book by the number of traditional lines. We begin 
with Herodotus; stichometric notes are found in Laurentianus 
LXX 3, and Angelicanus C i, 6, to books IV, V, VIII, IX. 

M. Graux gives as follows : 

Lines, Letters to line. 

Book IV XXXHHPIII 3253 37.6 

V XX HH 2200 37.5 

VIII XXHHHAAII 2322 37.6 

IX XXHHni 2206 37 

Diels measures the syllables, giving : 

Total lines. Total syllables. Syllables to line. 

Book IV 3253 48940 15.08 

IX 2206 32640 14.8 

For Thucydides we have the following fi-om Dionysius of Hali- 
camassus (Judic. de Thucyd. c. 10) : 

Book I, c. 1-87 dtcrxiXioi 2000 35 

I, C. 1—23 irtwaK^aioi 500 35 

Diels estimates the syllables for the second passage to be 7740 
and deduces a normal line of 15.5 syllables. There are several 
other stichometric noteb in Dionysius to other passages of Thucy- 
dides, for which M. Graux did not quote the results, because they 
seemed to diverge from the preceding. The difficulty in such cases 
is that the numbers are approximate and the passages not clearly 
defined. They will be found, according to Birt (p. 198), to give 
results agreeing closely with a line of 35 letters. 

» Diels, Hermes, XVII Bd.. 3 Heft. 



142 



AMERICAN JOURNAL OF PHILOLOGY. 



In Isocrates we have a single subscription from Codex Urbinas, 
together with some other marks to be discussed later on. This 
gives us : 

Busiris HHHPAAAA 



Lines, 



LeiUr litu, 
374 



Diels gives 6070 syllables and deduces 15.5 syllables to the line, 
which is sufficiently near, though his estimate is in reality in excess 
by 30 syllables. Fuhr repeated M. Graux*s calculation and made 
37.66 letters to the erlxos.^ With the same datum corrected to 395, 
as suggested by Fuhr, we have a line of 15.2 syllables. 

For Demosthenes we have a valuable collection of data from 
Graux and W. Christ,* which may be exhibited in one table, with 
the corresponding MS authority and the deduced value of the 
oTixos. The notation of the MSS is based on that of Vomel, and no 
account is taken of documents inserted in the text. Obvious 
errors are corrected. 

Corrected Letur 
MSS. Data. Hnes. Hne. 



I Olynth. 


2 BF. 


HHPAH 265 


34-8 


2 Olynth. 


BFA.j 


HHAAAARf ^^^ 


35-3 


3 Olynth. 


2 BF. 


HHHAAR 325 


36.6 


I Philipp. 


2 BF. 


HHHHPA 455 


364 


Peace 


2 BF. 


HHni 206 


35-7 


2 Philipp. 


2 BF. 


HHPAAAA 290 


35 


Halonnesus 


2 BF. 


HHHAAAAn 345 


36.7 


Chersonesus 


2 BFA.. 


FPAAAA 590 


37-3 


3 Philipp. 


A.) 
2 BF.| 


PPAAAAl ,80 
PPAAA) ^^ 


36.3 


4 Philipp. 


. ^■1 


PHAAAH 

PHAAA 

PHAAAII 


^ 634 


35-8 


Letter of Philip 


2 BFA.. 


HP AAA AH 


1 196 


351 


Ilfpl (Tvvrd^Ciur 


2 BF. 


HHHAAZ 


^ 330 


35-8 


HfpX rS>v 2vfifu>pi«ov 


2 BF. 


HHHPAAAZ 


^ 390 


34 


Liberty of Rhodians, 

• 


2 BF. •) 
Corr. 




HHHAAAAn 
HHHAAA 1 


} 334 


34-5 


Megalopolitans 


2 BF.1 
Corr. J 


HHPAAAA 1 
HHPAAAH 


\ 288 


33-9 


Treaty with Alexander, Vat. "1 

Corr. j 


HHHAAAH 
HHPAAH 


} 277 


34-6 



* Rhein. Mus. Bd. 37, Heft 3, p. 468. * Die Atticusausgabc dcs Demosthenes. 





STICHOMETRY. 






Correchd 




MSS. 


Data. linei. 


Crown 


."^■1 


fxI^BB^iRIII}-- 




De eds. leg. 


I BF. 


XXXHHPAAA 32SO 


Leptines 


£ BF. 


XPHnill i6o8 


Midias 


I BF.l 
Corr.j 


xxiin,,., 
xxHi;""' 


Andration 




PHHPAAA 7So 


Aphobos 1 


s BF.) 
Coir.f 


"iift^ii} '- 




Aphobos II 




HHAAAA 240 


Adv. Onet. 


s BF. 


HAAAA 140 


Lacritos 


2 BF 


HHHHAAA 430 


Nausimachoa 


5 F. 


HHPAA 270 


Boeotos I 




HHHPAAA 3S0 


Boeotos 11 




PPAA 570 


Macartatos 




PHPAA 670 


Leochares 




PHAAAA 640 


Stephanos I 




PHHPAAAAIM 793 


Corona Trierarch 


Corr. } 


HHnH\ ,,, 




Callippos 




HHHAAIM 323 


Nicostratos ■ 


Corr. } 


HHnn ,^ 
HHHni; 3<* 




Conon 




HHHHPA 460 


Eubulides 


;co„.} 


PPAAAAl 6„ 

phpaaaa; *'° 




Neaera 


^ Corr. [ 


XHHHPn,,,, 
XHHPi;"=' 




Epitaph. 


'corr.} 


HHHHMin „, 
HHHPnil) 357 




Eroticos 


Corr.; 


^%^i\ 3- 




Prooemia 




XHHHPAA 1370 


Epistle I 


^o„.} 


NAAARl ,,, 
HAAAn; "3= 




Episde II 




HHAnil 217 


Episde III 


;co„.} 


PHHHPAA) „. 
HHHPAAJ »° 




Epistle IV 




HI loi 


Episde V 




AAAA 40 



144 



AMERICAN JOURNAL OF PHILOLOGY, 



The majority of the corrections in the previous table (due to 
Blass, Sauppe, and Graux) are sufficientiy obvious. The results 
exhibit a remarkable constancy, though they are slightly in defect 
of the full average hexameter. 

On the application of these data to the study of the genealogy 
of the MSS of Demosthenes, we must refer to W, Christ's valu- 
able paper, previously alluded to. 

Reserving the question of Biblical and Euthalian stichometry 
for later consideration, we have the following further references 
from M. Graux, 

For Eusebius : Praeparatio Evangelica ; from the MS Paris 451 : 









Lints. 


Letter line. 






Lib, I 
Lib. II. 


A«Hr=A 


♦Nr=i553 
= 1483 


37.2 

37-2 






AYnr 






Lib. IIL 


AONH 


= 1858 


36.1 




For Gregory of Nazianzus ; from the MS Laur 


.VII 8: 




Homily I 


PH 


36 


HomUy XXIII 


TMB 


35-4 


II 


Aa9c 


354 


XXIV 


y9e 


36.2 


III 


PMB 


37 


XXV 


*3fe 


36 


IV 


BYNH 


36.6 


XXVI 


♦Kr 


36.4 


V 


AMB 


36.7 


XXVII 


CI=^CO 


36.7 


VI 


XKE 


36 


XXVIII 


*Ne=^Ne36 


VII 


♦IH 


35 


XXIX 


♦9 


37 


VIII 


«3e 


35-9 


XXX 


¥=x 


36.5 


IX 


PMe 


37-4 


XXXI 


^I'OE 


36.5 


X 


P 


35-9 


XXXII 


OIA 


36.9 


XI 


ce 


36 


XXXIII 


YM 


36.6 


XII 


PN 


37-7 


XXXIV 


9e=c9e 


36 


XIV 


AIZ =APZ 


36.2 


XXXVI 


TAr 


35.5 


XV 


YA 


36.8 


XXXVIII 


YNE 


36.6 


XVI 


XKC 


35-3 


XXXIX 


*N=*H 


36.6 


XVII 


TAE 


36.8 


XL 


AYie 


36 


XVIII 


ACAH 


35-8 


XLI 


YDA 


37-2 


XIX 


YIZ 


365 


XLII 


*AB 


36 


XX 


TA 


366 


XLIII 


BY 


35 


XXI 


APSA — i^BA 35.6 


XLIV 


c9e 


35-6 


XXII 


YAH 


368 


XLV 


onr 


35-9 



STICHOMETRY, 145 

And for the letters of Gregory from the same source : 

Ep. CI TM 36 

Ep. CI I P=:p» 36.9 

It must be sufficiently patent from the foregoing researches of 
M. Graux that every speculation as to the equality between the 
cmxor and the average hexameter is abundantly confirmed. The 
only thing that does not appear from the results is whether the 
lines are measured by their letters or their syllables ; but this has 
been already discussed, and we have arrived at a high probability 
in favor of syllabic measurements, at least in the case of later 
authors. 

Alternative arixot of twelve syllables. 

The next question that arises is whether there are traces of any 
other normal lines ; and in the first place, are there any instances of 
lines measured by iambic trimeters, or lines whose normal extent 
is 12 syllables? Now if we measure this line by letters, we at once 
find from 25 lines of the Medea of Euripides 29.96 letters ; and this 
number 29.96 is extremely suggestive when we examine the 
following passage from Josephus, at the close of the Jewish Anti- 
quities : cV( ^xnw.'s re tLtrraatadfrvi ri\v dpx<uoKoyituf, /3i/3Xoir fup tiKoo'i it€piti- 
Xflfiftispipff ^ dc fAvpiaai iTrl\(»p, 

If we take the assertion of Josephus literally, remarks M. Graux, 
we should find for the value of the arlxoi the inadmissible quantity 
28 or 2g letters. The statement is then explained to be a rough 
expansion of the assertion that each of the 20 books of the Anti- 
quities contained 2000 or 3000 <rrixoi. And Birt (Buchwesen, p. 
204) suggests the alternative reading c for cf by which the Josephus 
line will be 34.2 letters. Obviously the lines are really iambic lines : 
and this is confirmed in several ways by other considerations 
which I have adduced elsewhere.* It will also be more apparent 
as we proceed with our subject. 

The importance of the result is mainly thb, that it establishes 
the habit of writing iambic lines, at least so far as regards the 
first century and the locality of Syria, a conclusion which may 
affect our views as to the character of the originals of the New 
Testament 

*Amer. Journ. Phil. 12, Suppl. 



146 AMERICAN JOURNAL OF PHILOLOGY. 

AlUmoHve of a longer line. 

Dieb' believes that he has also found traces of a line even 
longer than the hexameter. He bases this belief on quotations 
which Galen makes from Hippocrates. From these we have:" 

SyUabUs Utters 
HippocraUs {ed, KAhn), Sr/j^w. Syllabus, to lime, to line. 

I. 348 — 360, 18 240 4360 18 40.8 

1.348 — 371, + 616 — 625,9 less than 600 11420 19 42.7 
I. 624, 17 — 625, 9 about 10 212 21.2 49 

Moreover, in another place, Galen (V 716. Kahn) reckons 86 
syllables of Plato, Tim. p. 70 D, as 4 arixoi' 

The difficulty of admitting these results is considerable ; for we 
have already shown that Galen employs a sixteen-syllabled line for 
measuring cn-ixoc, and it is difficult to see how he should have 
varied his standard for another so nearly coincident with it as 18 
syllables would be. Moreover, Diels has shown, with high proba- 
bility, by very appropriate quotations, that not only did Galen use 
a line of 16 syllables as his unit, but that the early copies of his 
works were written in an exemplar of that very length. This he 
establishes by the following quotations and measurements : 

Oribasius III 662, 3 (ed. Daremberg et Bussemaker) yivcrm hi 
irvr€ KT€. From Galen, Meth. Med. XIV (X 1009, 4 sqq. Kiihn), 

on which the Scholiast remarks (p. 689, 12) cM rov id r^ Bepmrtv 

riic^r o>r jrp6 ay (rrixovs (1. <rri;(<ttv) rov rAovr, K€(f}, ircpl ipfmjTos, Three 

similar quotations are given from the same source,' and finally we 
have: 







Normal 


Galtn (fd, Kuhn). 


'Lrixoi, 


Syllables, Letters. 


X. 1007, 4 — 102 1, 19 


ca. 250 


16.6 39.6 


X. 445i 7 455. 12 


ca. 200 


15.9 41.2 


X. 448, 4 455, 12 


ca. 140 


16.5 42.9 


VII. 705, I— 7i7> I 


ca. 200 


16 41 



Galen, therefore, measures and perhaps even writes i6-syllabled 
lines ; and the only contlusion we can come to is that his copy of 
Hippocrates must have been slightly in excess of the ordinary 
pattern, rather than that it was written on a new pattern. 

> Hermes, Bd. XVII. Heft 3. « Galen, ed. Kuhn. XV 9, 10. 

•Oribasius, IV 179. 4; IV 181. 2; III 598, 11. 



STIC HOME TR Y, 1 47 

Subdivision of lines in MSS, 

The existence of the normal hexameter and iambic lines is, how- 
ever, so little obvious from surviving MSS themselves, that an objec- 
tion arises a^inst the previous investigations on the ground of want 
of actual paleographic evidence. Perhaps the deficiency on this point 
is due to two causes. First of all, the cataloguing of an exactly 
written library edition, such as would be found in the library 
at Alexandria, rendered the preservation of the stichometric 
form unnecessary and prepared the way for the breaking up of 
that form ; and in the next place, the breadthof the columns of the 
papyrus-rolls did not generally admit that the lines should be 
written in full, and they were consequently subdivided into two, 
three, or more narrow lines. Conspicuous instances are furnished 
by the celebrated Vatican and Sinaitic codices of the Bible ; of 
which the lines represent respectively a somewhat curtate half- 
hexameter and a similarly divided iambic trimeter. This I 
have shown to be the case in the two MSS in question by the 
actual examination of the text for the accidental hexameter in 
James I 17 and for a quoted iambic verse in i Cor. XV 34.' The 
supposition is confirmed by Baehrens,' in some good remarks on 
the Ancient Book-Form of Roman poets. And Baehrens points 
out that these subdivided lines may actually be seen in a papyrus 
roll represented on a Pompeian painting, where four lines are 
found divided into sixteen. This, however, may be nothing more 
than artistic license. In Montfaucon, Bibl. Coislin., for example, 
the Gospel of John is pictorially represented as being written by its 
author in lines of about a syllable each. The most likely place to 
find these subdivided lines is in epistles, which seem to have 
been written on shorter models. 

Partial Stichonutry. 

A fiirther development of the simple stichometric subscription 
is found in those MSS which inform us, by means of marginal 
notes from point to point, as to the number of trrixoi contained in 

* American Journal of Philology, 12, Supp). p. 18. 

•Ncne JahrbQchcr fUr Philologie, ElftesHeft, 1882. p. 785: "aber dafUr gab 
es nar cine mOglichkeit, nemlich indem man die seiten schmaler machte ; und 
dies fahrte wiederum notwendig dazu dass man grOssere verse (hexameter u.s.w.) 
auf zwei oder mehr xeilen yertheilte/* 



5:^:.En=r- x* ioi iat tlxse 



zer TO Older 
si: both. 




X -i-. 7: Tuv; ~.« zre j'.^r: n zk :o:it b«rre 3 rrpreseots 
cs rn; ;c^=* .-n i:e TSirj-ii :e:=i£ rie rjod'^SKes «' the 
:^-:r-ri= :i it^.ii^.-ir^n^ ~~:t^ —e nxk s ^sosured in 
^ ■_- ■^; I— -m. T^ i^-f ra; TiiLTx 3 a :r>; laxb 'iae and 
^— n -iT^ i. 1 T»;— =■ zxe 'Izies ics i llrzle sh.:n so as to 
:5.^ «■■ I_:l>=i. ■-; ti-'; 3 a: i^e sxf^ je a=d r at the 
;■;, vi.:ii s tstt '-ri.~ riii rs; r.ti^ "rori is now 395 
t :.ca si7ors Fi.iri — t— i-.t-i-r- "H^-ese marks are 
■ -M >--s -t't i^i;::Tim;:z7- sii-uii; 3;c --orG^-r-c : as tbeyare 

■.^->- Or^« r^^i-i.— tr -C s r:;'" r," jssrr.« that tbej' are 
•.:■;. A~d ■»•? joaI -r.cJ-.-T *>k rs«iOG to conclude that 
.■x':-:ir.-icry j*. i= .ts r.s;:ri:ul 6fvt:r=eE^ always later 

e. ei:;;.\'.'>-l:ij: ■->; I^ttjrs ot" :b; ^rr. ib« :-m^ the successive 
bi^t aot gT'.-^--ii-i L:r«>3 Cit i^d^ju sraem as in the 

.li-i ^^eri! o:her mirks on the :r.irg:a of this MS which 
vc been explained. At Busiris 10 s:ai]ds the figure t 
iK- wv.^iis anXoytmr cv^yirMrAu. This represents the 82d 
lie jOJiie length as the measured verse), and if we allow a 
liv sp^"? at the b^inoing of the document for its title, it 

■ Kbetn. Hbs. 37 Bd. 3 Heft. tSSa, p. 468. 



STICHOMETRY. 149 

may very well be the dose of the sixth page of the exemplar copied, 
each page being 14 hexameters. 

The mark .)iC' also occurs, three times, once with the previous 
mark, once at the 345th arixo^y and once at the 368th. These are 
probably the marks of the ^opQwri^ or MS corrector, and may 
refer to simple pauses in the work of revision, or perhaps to pages 
either of the MS copied or of that used in the process of revision. 
In the actual case in question, the first pause was at the sixth page 
of the MS copied ; while the proportion of the numbers 345 and 
368, which are 15 X 23 and 16 X 23, shows that the other two 
marks may be the conclusions of the 15th and i6th pages respec- 
tively of the revising MS. 

The Urbinas MS has also other annotations of various kinds, 
the most prominent being the paragraph mark, a horizontal stroke 
against the beginning of the line where the pause is to be made. 
All these marks may be found quoted in Fuhr*s article already 
referred to. 

For Plato. 

Schanz* has discussed a precisely similar question for the Plato 
manuscripts. He remarks that the Bodleian Plato (Clarkianus) 
has partial stichometry in the Cratylus and Symposion, the letters 
running continuously to ^. Counting the lines of Clarkianus 
between the successive marks, we have 68, 69, 70, 71, 72, 73, 74, 
75; 71 being the most frequent interval. Now this gives us a 
arxx^ of 35.56 letters for the Cratylus, and 34.32 for the Symposion, 
which are sufficiendy in accord with M. Graux*s results. Similar 
stichometric marks are found in another MS of Plato, Venetus 185 
(n of Bekker, D of Schanz.) Here again they are confined to 
Cratylus and Symposion. Between two following letters lie on the 
average 68 lines; and the same sections are marked off by the 
letters as in Clarkianus. An interesting application is made by 
Schanz to determine the authenticity of a passage in Cratylus 437 a^, 
where certain words are wanting in MSS B and T. We can at 
once verify that these words were wanting in the exemplar that 
supplied the stichometry. 

W. Christ has studied in a similar manner the partial stichometry 
of Demosthenes* (Codex Bavaricus), and applied the results to 

• Hermes, XVI 309, 1881. 

* Die Atticasausgabe des Demosthenes. MQnchen, 1883. 



148 



AMERICAN JOUFNAL 0! 



the preceding portion of the book. A- 
are found in many early writers, who 
<m';{ai precisely as we quote page ar ' 
Schanz has given the name of i'arti ' 

Precisely as in the case of total 
MS notes have no special conn'" 
the documents in which they o 
copies, or to fixed and uniform ir 



<s of Demosthenes. 

i^d in the preface to 

15 to have been igno- 

He gave the references. 

r.ef statement that the 

aiportant than the con- 

- -ii the determination of 

- arom which the numbecs 



For / 



For example, we ha\'e alr^ 
Isocrates, Busiris, in Cotlcx 
ginal references, which h;i\ i 
fol. 22. io(S25\ before t<i'.t-\ 
tra^-r the letter r; bettt. :. 
which evidently repres-iu 
we ought also to find th..; ■ 



^rses marked by hundred 
•.:nbrosian Pentateuch ;' and 
.jsurement of the Acts and 
- ;^iitio^ of Euthalius. Some 
:t im'j^ are found in Diogenes 
ire mostly in round numbers 
1 Tou» ;(iXi(iirt tmjcovc), 
ases the exemplars 
srmediate measure- 
ices more at length 
of the quotations 
1 the similar quota- 
re is no reason to 
tations made being 
circa Uriium, and 
' end instead of the 
.er, sufficient actual 
liod of citation by 
d, it was the only 
i quotation. 



elopment of the art 
>f letters to one of 



s of word- 
i.34N.F,p.38.i879. 






.V / 1 C no ME TRY, 151 

.' s corresponding to the well-established syl- 

\ will have made their appearance first in those 

Mit where the distinct enunciation of a sentence is 

t. with the object of removing the causes which 

•li.i and vocal effect. That is, it is evident that in 

ii J re publicly recited, an effort will be made to render 

t^isk of reading orally a continuous text. This is 

\. th the works of the great orators, as well as witli the 

1 .. . nns ; and we may expect to find in such works a ten- 

\ ii. the direction of sense-Hnes rather than space-lines. In 

1 /l instance this tendency will only be manifested by the 

. . lion of the paragraph mark, as it is found in the Hyperides 

.:, the MS of Isocrates, and the early Bible texts. But this 

,:.iph mark, perhaps accompanied by a rude interpunction, is 

t found by the rhetoricians to be a sufficientiy obvious and 

. jljatic division of the text. Sense-lines are therefore introduced. 

V le change seems to be made in the first case with a reservation 

tluit the text when broken up shall still represent the same number 

• 1 lines, or sensibly as many, as the archaic copies. And the 

natural effect of such a change is that the arix^ undergoes a new 

deflection in the direction of sentence^ the sentence being not very 

different from a hexameter. 

The evidence for these statements may be arranged as follows : 
St. Jerome, at the commencement of his preface to Isaiah, informs 
his readers as to the nature of the book that he is translating. 

'* Nemo cum Prophetas versibus viderit esse descriptos metro 
eos aestimet apud Hebraeos ligari, et aliquid simile habere de Psal- 
mis vel operibus Salomonis : sed quod in Demosthene et Tullio 
solet fieri^ tU per cola scribaniur et commaia, qui utique prosa et 
non versibus conscripserunt, nos quoque uiiliiaii iegeniiutn provi- 
dentes, interpretationem novam novo scribendi genere distinximus."* 
St. Jerome introduces for the convenience of readers a new kind 
of transcription similar to that which was in vogue for Cicero and 
Demosthenes ; this division of the text is by cola and commaia. 
From Suidas' we find that when the crnxoff forms a complete clause 

it is known as a colon I koXov ow h awrjpTKTfuvriP fwoiap tfxo^^ OTixos. 

From Joann. Sicul. in Hermog. i, 63 (Vol. VI, p. 127, Walz), we 
find that writing by cola and commaia is the invention of rheto- 

• Migne, PatroL Lat. XXVIII, col. 771. 's. ▼. KuiKov, 



152 AMERICA// JOURNAL OF PHILOLOGY. 

ricians in imitation of poetry : itr™ /trtiS^ Ttoofths oi p^ropn fufioiWai 

KoKar \iyova\ ri J)r4 tvvia <7vXXiii3wi> Sr fMJItpi na tnraiuulliKa ' ri B( wXiof 

KotMir olriH mXouffir Antrrfc <I ^mw (Siraprifbifi' JntHOf. In this pas- 
sage it is interesting to observe that the standard of measurement 
is still the syllable, but, as we should expect, there is no longer a 
fixed number of syllables to a line, but we have three rough divi- 
sions 1 viz. if the clause be less than eight syllables it is called rAii/w, 
if between eight and seventeen it is called caXov, and if greater than 
this, (rxotrormc pT a long-drawn-out sentence. Such a long line is 
actually termed a verse in a quotation given by VOmel' from Aquila 
Romanus de Figuris c. 40 : " Ponam . , . Demosthenicum versum ; 
Et non dixi quidem haec . . . persuasi quidem." The passage 
(De Corona, §179) contains 20 words. We may actually see in 
operation the process of dividing the text of Demosthenes into tSXa. 
In a passage of the rhetorician Castor,* of the fifth century, we 
find the following; 

BliiTOfHu rir SKoy AqpHrAnicir \6yor rir ivtypalfiivTa Upit rqv ivumArpi 
tiXiinrai.' ' roOror yip (m'fo/Mi', irvv 6tif <f)drBi, mr^ KaXo* ■onur^mu'rct (it 
TJjP JTOiriiTifra tSc xiikav nari ric aptOiiht rir iyKfi)M90V Jy toIe apxi^ou 
^(^Xlalt, ■« iiHrpif^tr abrit i Aijpxr4<r^t r^ iSior \6yor. 

Castor proposes, that is, to punctuate a passage of Demosthenes 
so that the numeration of the broken-up text may agree with the 
number of verses found in the old copies. Whether he supposes 
Demosthenes himself to have divided the text in this way, or 
whether he implies by the word (fuV^njovf a regular and uniform 
measure, is not very apparent at first sight ; but a little considera- 
tion will show that it is not important to decide such a point, for it 
is sufficiently demonstrated that the stichometry of the M5S of 
Demosthenes is hexameter stichometry ; and it must be the number 
of such verses that Castor wishes to preserve. Dionysius Halic. De 
Comp. Verb. XVIII gives explanations of the methods employed 
in breaking up the text of Demosthenes into cola and periods. 
For instance, in De Corona the first period is to consist of three 
cola, as follows : 

ry JTjpl TDu trrttftarou X jyy, rpta iiir ionr i rj* rpvnpi wtploSor 

1 KaAa ' oi Si ical raOra naraiaTptivmt 0% ttaiy al puOfmi. 

• phi, & 'aipts AA]vouh, rait d<iHC tSxopai rraai icat iraaatt . , . 

' Rhein. Mai. N. F. II 453. 'W«Ii. Rh. Gr. Ill 711. 



STICHOMETR K 1 53 

Tov df d€vr%pov KoaXov rovdc. 

Oarjv c^yotay ?x^^ ^V^ diorcXw 1^ re ir<$Xf i koi iracriy vfuy . . . 
Tow dc Tpirov kvXov, 
Tov TotravuiP vndp^i fioi vap vp&v th rovroyl t6p <ry«0ya. 

It is evident that this custom of colon-writing introduces a meas- 
ure of confusion into the subject ; the more so because colon-writing 
is sometimes accompanied by colometry, of which occasional traces 
may be found, as in Dionysius Hal.* who makes the proem to 
Thucydides up to ov xaXcTr^r airaWoroyro to be 30 cola, and the 
beginning of the Aristocratea to be 9 cola. Misled by this peculiar 
dissection of the text at the hands of the rhetoricians, F. Blass' 
maintained strongly that the ancient (rrixos was not a space-line but 
a sense-line. And with remarkable skill, which M. Graux honored 
with the term habiUU de main^ he proceeded to divide various 
passages, principally in Demosthenes, into a number of cola, suffi- 
ciendy nearly in accord with the traditional number of verses. 

Besides this, he reasoned that if the arlx^ were a fixed quantity 
there ought to be a sensibly uniform ratio between the number of 
verses and the number of lines occupied in the printed text. This 
he maintained not to be the case. 

In this, however, he seems to have ^liled almost completely, if 
we allow for the small margin of variation necessary in the measure- 
ment of the lines, and the small variations in the sizes of the Teubner 
pages to which he referred. A single instance will suffice. Taking 
the data for Herodotus, Blass gives : 





'Lrixoi. 


Teubner Lines, 


Ratio, 


Lib. IV 


3253 


2764 


.849 


Lib.V 


2200 


1866 


.845 


Lib. VIII 


2322 


1952 


.840 


Lib. IX 


2206 


1849 


.842 



If this does not demonstrate the use of a uniform verse-measure 
for Herodotus, it would be difficult to prove anything. 

The merit of Blass* work consists, however, in the light it throws 
on the early rhetorical studies, and not at all in its bearing on 
stichometry. Blass himself, after making his colon division, came 
to the conclusion that the colon could not be very different from 

* Dion. Hal. de Comp. pp. 169, 199. 

♦Zur Frage Uber die Stichomctrie. Rhcin. Mus. N. F. XXIV, 1869, p. 524. 



154 AMERICAN JOURNAL OF PHILOLOGY. 

the hexameter. " Die Zeilen smd mitunter lang, aber selien Idnger 
als ein Hexameter." ' "Das rhetorische Colon enispricht dem 
poetischen Vers."* This is precisely what we should expect to find. 
for we have indicated that the colon was introduced as an alterna- 
tive for the hexameter, and was made as far as possible equivalent 
to it. Another instance of this tendency, besides those which have 
been already quoted, is found in Cicero, OraL 222 ; " E quattuor 
igitur (sc. menibris) quasi hexametrorum instar versuum quod sit, 
constat fere plena comprehensio. His igitur singulis versibus quasi 
nodi apparent continuations, quos in ambitu coniungimus." 

Herodes Atticus' is said to have had a clepsydra made which 
was the time-equivalent of 100 hexameters, aviifuittrpiuUvrir is iKorii' 
fffij, by means of which his enunciation was regulated. 

Scrivener's pay and price of books. 

We now turn to the question of the employment of stichometric 
measurements in determining the pay of scribes and regulating 
the price of books. For investigations on this point the best 
researches are those of Graux and Birt. 

It is established by means of the celebrated edict of Diocletian 
(A. D. 301), which was a tariff of maximum prices for the Roman 
empire, that the pay of scribes was by the hundred lines ; and M. 
Graux very justly remarked that this assumed the fixity of the 
line, and would be altogether illusory upon any other hypothesis. 
I have discussed elsewhere the sUtenients of this edict and their 
stichometric value.' It is only necessary, therefore, to give a brief 
recapitulation of the points thereby established. The edict from 
which the data are supplied is found in greater or less complete- 
ness in many localities, but the most important form is presented 
in an inscription from Stratonice ; the figures being edited in the 
Corpus Inscriptionum from another inscription found in Phrygia. 
We have then ; 

Membranario in [qua]t[r]endone pedali pergamena, [XL denani] 

Scriptori in scriptura optima versus No. centum. [XXV] 

Se[quentis] acripturae versuum No. centum. [XX] 
ni in scriptura libelli vel tabular[um] in 

us No. centum. [Xj 

' P- 530- 
.ua Sophist. 11 10, p. 185, quoted by Wachsmulh, Rh»n. Mas. 34, 
* American Journal of Philology, 13, Suppl. p. iz sqq. 



STICHOME TRY, 1 55 

It is clear from the inscription that there are at least two principal 
types of writing, if not a third ; and in every case the measurement 
is by verses, no distinction being made or imagined between prose 
and poetry. 

It is inconceivable that the difference in price should be due to 
a difference in the quality of the writing (as Birt suggests), for it 
would be somewhat difficult to graduate such uncertain things as 
the hands of scribes, to say nothing of dividing them exactly into 
good and bad ; it must, therefore, be of different lengths of line 
that the edict speaks, optimus and sequens being the common 
terms all through the edict for first size and second size. 

If the prices are correctly edited in the Corpus, the ratio 5 : 4 
(=35 : 28) is very nearly that of the normal hexameter to the normal 
iambic line, 36 : 28, which confirms our previous speculations as 
to the existence of the iambic lines. The difficulty in all such cases 
is to reduce the brass denarius of Diocletian's time into an equiva- 
lent of modern money. If we may take the values given by Birt' 
from Hultsch,' the payment is sufficiently small ; 100 denarii being 
worth no more than 2.4 marks. The denarius is then .6 cent ; the 
scribe's pay being 15 cents for a hundred hexameters and 12 cents 
for a hundred iambics. On this basis I have calculated the cost 
of production of the complete volume of which the Codex Sinaiticus 
forms a part; the result being approximately 180 dollars, the cost 
of the vellum being included. 

It is not uncommon to find in early codices notes of the prices 
for which they were sold ; Montfaucon (Bibl. Coislin. p. 57) observes 
that the price on the first leaf of a Psalter is y^xJo-a d' = grosa sive 
drachmae quatuor ; and at p. 83 he notes that codex 29 was bought 
for 24 aspra^ the book itself being a commentary by Chrysos- 
tom on S. Paul's epistles. 

A cursive MS of the Gospels (No. 444) sold in A. D. 1537 for 500 
aspra ; upon which Scrivener* notes that ** the asper or asprum 
was a mediaeval Greek silver coin (derived from aairpos = albus) ; 
we may infer its value ft^om a passage cited by Ducange from 
Vincentius Bellovacus, XXX 75, * quindecim drachmae seu 
asperos.* " Since the 4 Gospels are not more than twice as long 
again as the Psalms, it is difficult to see why the Psalter should 

> Birt, p. 209. « Hultsch. Neue JahrbUchcr fQr Philologie, 1880, Heft I. 
'Scrivener, Introduction to N. T., p. 208. 



156 AMEHICAX JOCMXAL OF PHILOLOGY. 

seQ ibr 4 drachmae and the Gospels for 5CXX And it is possible that 
Mont£aucon*s price is incorrect. 

M. Graux' grves us the further important information with regard 
to the pay of scribes, that the custom of regulating, if not the tariff, 
at least the measure of lines written, continued right into the middle 
ages, especially at Bologna and other universitv' towns in Italy. 
He quotes Sa\*igny,' Geschichte dcs Romischen Rechts im Mittel- 
alter, to establish this point 

The unit of measure b the pecia^ which consists of 16 colimins, 
each containing 62 lines, and the number of letters in each line 
being 32. *• Secundum taxationem studii bononiensis firmamus 
quod petia constituetur ex sedecim columnis quarum quaelibet 
contineat sexaginta duas lineas et quaelibet linea litteras XXXI I/* 
The numbers here are peculiar, and it is extremely difficult to 
believe that as many as 62 lines were normally written on the p^^. 
It is interesting, however, to observe the survival of ancient custom 
in the columnar writing, and the measurement of lines by letters. 
The statute is, therefore, in all probability the relic and modification 
of previous laws. 

Whether the line of 32 letters has any reference to the Italian 
poetry, as Birt suggests, b extremely doubtfiiL It b more likely 
to have been suggested as a multiple of the feivorite number 16. 
We have no reason to suppose that such a statute as that mentioned 
required that MSS should actually be copied in columns or lines 
of the pattern indicated ; all that was necessary was the adoption 
of thb unit as the standard, and the record by the scribe of the 
number of pedae. M. Graux remarks that these notes of the scribe 
as to the progress of hb work, " finb pede I," are sometimes found 
in the body of the pages or the text 

Upon the whole, I am inclined to believe that the text of the 
statute b incorrect in reading sixty-two lines, a most improbable 
number. If we read 72 for 62, the peda b almost exactiy 1000 
hexameters of 36 letters each ; stricdy speaking it b 1024. And 
thb is an extremely likely unit of work to have been handed down 
by tradition from the early scribes. 

An interesting survival of thb early manner of determining the 
pay of a scribe b found in the modem custom among Indian 
copybts. Here the basis b the gloka, an iambic metre of 32 syl- 

» Rcvuc dc Philologic. p. 139. « T. Ill, c. XXV, §579. 



STIC/iOATETI/y. 

lables, which is applied as a unit of measurement t 
kinds.' 

We shall now turn our attention to the bearing w 
have upon the restoration of the early book-form, 
upon the texts of the New Testament. Thus far 
almost entirely any reference to the stichometric c 
Biblical M5S, because they constitute so impor 
textual criticism that they deserve a separate dis 
more complete than has hitherto been accorded 
same reason we have reserved any allusions to E 
edition of the New Testament. 

J. Ren 

I Note by Dr. Blooinfield in Amer. Joom. Phil, i j, SuppI 
by Gaidlhausen from NOtdekc. in Griech. Palaeogr. p. 132. 



IL— STUDIES IN PINDARIC SYNTAX. 

III. — AoRisT AND Imperfect. 

In older grammatical study there was much teleology. I do 
not mean merely teleological expression, for language was made 
by teleologists, who could not have understood any result without 
conscious agency somewhere. The final is always earlier than the 
consecutive ; and even when the consecutive comes in, the final 
element may reappear at any moment So the evolutionists are 
not to be assailed because they use the only vocabulary that the 
dynamic thinkers of the earliest days have left them ; and the 
grammarians of the future will use to some extent the consecrated 
expressions of the past. But the attitude is changed, and though 
our grammars speak of certain forms, of certain phrases, as if the 
demiurges of language had gone to work deliberately and framed 
forms and phrases to a clear end, we no longer wonder at 
the marvellous mechanism of speech. The most varied, most 
pliable, most subtle language on earth is only a congeries of sur- 
vivals. The harmonies of speech are the result of the indolence of 
the human organs of utterance. The close texture of composition 
and inflexion is due to slurring impatience. At first, it is true, the 
scientific study of language heightened the admiration with which 
the faculty of speech was once regarded. The human mind, 
unscientific as well as scientific, delights in the variations produced 
by the combination of a few elements. Given a short list of radicals, 
a handful of terminations, and the language with all its arborescent 
growths is there. So the identification of the personal endings of 
the verb with the pronominal stems for /, f^ou^ that seemed to our 
fathers a revelation. This revelation we look upon coldly now. 
We go a little way in Greek, for instance, and the scheme seems 
plausible. A little further, and we are perplexed beyond measure. 
The terminations seem to have wandered off from the bodies to 
which they originally belonged and to have grafted themselves on 
alien trunks. Primary endings attach themselves to stems which 
ought to have secondary endings — as -/u of the optative — and 
secondary endings are equally capricious, as is shown by the second 



STUDIES IN PINDARIC SYNTAX. 1 59 

person singular of the present indicative active. And so after many 
disillusionments we come down to the sober view that language 
serves its purpose only after a rude fashion. Physiologists have 
declared the eye as an optical instrument to be a wretched failure, 
and our students of linguistics smile at the enthusiasm which once 
clothed the subject of language with the purple light of rhetoric. 

This changed attitude of the grammatical mind toward language 
may be illustrated by the treatment of function. In old times the 
grammarian cudgelled his brain to find the meaning common to 
all the functions of the ablative case. The principle that each form 
must have its function was a logical necessity. That any language 
which had once developed a form should lose it, should care- 
lessly merge three or four forms into one, or use a dominant form 
for different functions, seemed impossible proceedings on the part 
of the personification called * language.' Modem research has no 
scruples on that score, and goes so far, in fact, as to ignore utterly 
what remains of consciousness are preserved, not by the personifi- 
cation called language, but by the actual users of language. On the 
other hand, many distinctions which seemed to be rooted in the 
nature of things prove on examination to be mere afterthoughts. 
The early speakers put forth a variety of forms with no sharp 
distinction, and those that came after made a regular differenti- 
ation, sometimes on symbolic principles, sometimes on no dis- 
cernible principle. 

So while we have not a weltering chaos, we have no beautiful 
k6<tho£. What we have may be something practically better than the 
fancied Kdafios, as it certainly is practically better than any Kdafio^ 
that human wit could devise. A compromise is often better than 
a thorough measure, and while our study has not the charm of 
logical symmetry, which the average individual of our race prefers, 
it has the charm of conscientiousness. 

The distinction between imperfect and aorist is one of the old 
landmarks that have suffered from the closer study of language, 
and in exploring the borderland in which grammar and literary 
art meet, I have examined recentiy whether the Pindaric use of 
aorist and imperfect gives any reason to suspect any indifference 
on that score. Of course, in the ordinary school-grammars there 
b hardly a whisper of doubt as to the universality of the difference, 
and this may doubdess be considered sound practice. But if the 
student is to grapple with Homer — and he is generally introduced 
to Homer as soon as he can make out Xenophon's Anabasis without 



l6o AMERICAN JOURNAL OF PHILOLOGY, 

frightful exertion — he will be met by the remarkable statement of 
commentary and text-book that it often does not matter which is 
used. So La Roche on B 43 copies from KrUger Di. §53, 2, i, a 
long list of indifferences. KrQger says that the choice is often 
arbitrary, often dictated by metrical considerations. So we find 
without * any considerable difference ' /Saivoy and /35, A 437 and 439 ; 
/SoXXcTo and /SoXcro, B 43 and 45 ; Br^K^v and t W^«, ♦ 653 and 656, c 265. 
267 ; d^Kc and didov, H 303 and 305 ; XiVc and Xcwrc. B 106 and 107 ; 
still more striking is /uorvXAoy side by side with Smn^iray ir€fH<f>pMm, 
A 465. All these are given in the same order by Professor Good- 
win in his Moods and Tenses ; only Kriiger's ' erheblicher Unter- 
schied' becomes 'perceptible difference/ and the indifference is 
referred to the meaning of the verb, an explanation which lacks 
clearness. It is a little remarkable that Delbriick, in his Grund- 
lagen der griechischen Syntax, takes /So/vciy and firjpai and /SdXXciy 
and PaKtw as striking examples of the Homeric differentiation of 
durative and aoristic tenses ; did6vai and dovpai, are sharply distin- 
guished elsewhere. 

In prose the MSS are sometimes to blame for the confusion of 
cXcifToy and IXtn-oi/, but there is a translatable difference everywhere, 
and it is hard to admit, without better evidence, that Homer, so 
exact in the use of the tenses, should have admitted the imperfect 
mein causa, though metri causa is coming to honor again. At 
the same time it must be acknowledged that the attempts to con- 
stitute a difference of conception are often lamentable in the extreme, 
and it would be better simply to note the difficulty as a problem 
than to hazard such breakneck mental positions as commentators 
sometimes indulge in. After all the preaching that has been done 
on the subject of the tenses, grammarians, the sermon over, are 
apt straightway to forget that the imperfect has nothing to do with 
the absolute length of the action, it has only to do with the vision 
of the narrator. So Nagelsbach's notion that the imperfect might 
refer to the abiding character of the result, though almost demon- 
strably false, has been echoed by so good a scholar as Classen. 
So rooted is the tendency in beginners to consider imperfect * pro- 
longed ' and aorist * momentary * that a course of cv^ur with the 
imperfect and of high numbers with the aorist is necessary to get 
them into right habits of thought ; but certainly veteran schc^ars 
ought not to be tangled with such formulae. 

What I wish to bring out in these remarks is the substantial 
justification of the difference between imperfect and aorist, from an 



STUDIES IN PINDARIC SYNTAX, l6l 

aesthetic point of view. Let it be conceded that the imperfect is 
nothing more than an aorist which has a present indicative, whereas 
the second aorist has no such present indicative. The present 
indicative is associated in the mind of the Greek with the idea of 
duration. He has no aoristic present, as a matter of fact, in the 
crystallized language. If he wishes to express the notion, he must 
use the aorist indicative as an approximation. Otherwise he must 
let the aoristic idea come out as best it may from the environment. 
It is useless to inquire into an earlier type. Contrasted groups 
such as <^€v>c» and cc^uyoy, XeiVo) and tkmov preserved clear samples 
of durative and complexive, and that is enough from the point of 
view of the users of language. ''Etp«I>€ was as durative as ?</>fvye. 
First aorist and second aorist, though formed on different principles, 
unite in the complexive notion — the first aorist keeping for itself 
the special notion of ingress. First aorist and second aorist, then, 
we may regard as one, so far as the contrast to the imperfect is 
concerned, and they are so regarded in the statistics I am about to 
submit — statistics which seem to show how intimate is the associa- 
tion of aesthetics and grammar. 

Of course, in separating aorist from imperfect, difficulties and 
doubts arise. Some of the old preterites, ordinarily classed among 
the imperfects, may fairly be claimed as aorists or indifferents. So* 
?r may be considered indifferent, and ll<f>fjv is decidedly aoristic. 
Then there are variations in reading, coincidences of form, ufiwtv 
and wrpwwvf and the balance produced by durative and complexive 
forms of the verb, outside of the indicative, has also to be considered, 
so that the investigation is laborious — ^laborious out of all proportion 
to the possible result. Still, having begun with a few characteristic 
odes, I was encouraged to continue ; and now that the work is done, 

^fac is aoristic. Clearly so II. 9, 35; 14, 126; not so clearly II. 3, 44. In 
Attic all ambiguity is remored by the bifurcation into ^okuv and ^a^. 
f&fuvo^ has no offsetting middle of ^doKu^ and, while It^ro is clearly aoristic, 
fdfuvoc may be durative (cf. Hdt. i, 176; 2, 22. 28. 148. 174; 3, 31. 68. 69. 74. 
75 Hfy etc.). In Pindar ^fiev(f> (I 5, 47) is complexive, while TrafjifKifiha (N 5, 
31) would naturally be considered durative : no^Xci yap viv navri dvpift nap^fUva 
?uT6vevev (Schol. eXtrdveve koI frapaireiOetv knextipti awtTSelv airnj. The whole 
subject of these early verbs is full of difficulties. So a long chapter might be 
written on ja and its forms. In Homer the group is aoristic. In Attid the 
absence of an imperfect to ipxofiai^ as well as the presence of elpt, forced on 
the group an imperfect sense which commentators (e. g, Classen on Thuk. i, 2, 
4 ; 3, 33, a) have not failed to develop. 



1 62 AMERICAN JOURNAL OF PHILOLOGY, 

it may be worth while to register the facts, without putting sledge- 
hammer emphasis on the conclusions. 

The logaoedic and the dactylo-epitrite odes of Pindar differ 
from each other in many points, one might say in every point in 
which the law of the epinikion and the individuality of the poet are 
not involved. It is not necessary to expand on this familiar theme. 
The organic difference which expresses itself in the rhythm, 
expresses itself in the build of the poem, the development of the 
story, the order of the words. Why not in the use of the tenses ? 
The quicker measures of the logaoedics would seem to call for 
more short forms and consequently for more second aorists. Pindar 
himself, when he is professedly quickening his pace in the narrative, 
multiplies aorists,* and almost tells us that he considers the aorist 
complexive. Should we riot expect, then, to find in the myths of 
the dactylo-epitrites a larger proportion of imperfects, in the loga- 
oedics a larger propK)rtion of aorists ? Of course the aorist will 
preponderate in both classes. That is the rule of the language. 
And regard must be had also to special emergencies. The law of 
the rhythm is crossed by the necessities of the story. There may 
be dactylo-epitrite stories that demand sharp, rapid handling. 
The poet may pause for leisurely contemplation in the logaoedic. 
But taking a sufficient basis for induction, we find that on the whole 
the proportion of imperfects to aorists in the dactylo-epitrites rises 
in a marked degree — a proportion that can not be seriously affected 
by recount, by elimination of errors. 

I cannot pause to justify the selection I have made out of the 
forty-odd Pindaric odes. Pindaric scholars will understand most 
of the omissions. 1 have hmited myself to the narratives, and 
have counted in them, not only imperfects and aorists, but also 
duratives and complexives outside of the indicative. After a 
rough preliminary count, which led me to think that a more exact 
examination would be remunerative, 1 requested Mr. C. W. E. 
Miller, Scholar of the Johns Hopkins University, to go over the 
ground carefully, and compare his lists with the result of my 
second examination. In this way, it is hoped, most of the errors 
have been eliminated. The result is the following table — ^worked 
out by Mr. Miller. The references are to Christ's edition 
(Telibner). 

' Sec P 4, 247 : piaKpa fiot vel<r9ai kot* hfia^irov ' upa yap cwaTrrei Koi Ttva | 
vt^ni' Ictifit /^/*rt,v» »'. From this point to the end of the story. P. has eight aorists 
and but a single imperfect. Contrast the statistics of P 4 given below. 



STUDIES IN PINDARIC SYNTAX. 



163 









I 


NDICATIN'K. 




Modal. 




LOGAOBOIC. 


a 


u 



< 


Ratio. 


• 

5 


• 

E 



U 


Ratio. 


Oi 


25-92 


Story of Pclops 


^6 


23 I: 


3.833 


II 


16 I 


: 1.454 


09 


42-79 


Pyrrha and Deukalion 


3 


12 I : 


4 


5 


II I 


:2.2 


10(11; 


) 24-77 


First 01. games 


6 


18 i: 


3 


9 


10 I 


: I. Ill 


13 


63-92 


Bellerophon 


2 


12 i: 


6 


II 


8 I 


: .727 


P2 


21-48 


Ixion 


3 


15 i: 


5 


9 


3 I 


: .333 


P5 


55-95 


Coming of Battos 


2 


II I : 


5.5 


6 


5 I 


: .833 


P6 


28-42 


Antilochos 


2 


6 i: 


3 


3 


4 I 


: '.333 


P8 


39-56 


Alkmaion 




4 0: 


4 


3 


4 1 


: 1.333 


Pio 


31-48 


Perseus 


I 


4 i: 


4 


4 


4 I 


:i 


Pii 


17-37 


Orestes 


I 


9 1 = 


9 


4 


4 I 


:i 


N3 


33-64 


Aiakidai 


9 


7 I 


: .777 


10 


5 I 


: .5 


N4 


54-68 


Peleus 


2 


5 I 


: 2.5 


I 


3 I 


:3 


N6 


53-61 


Aiakidai 




4 


: 4 


I 


2 I 


:2 


N7 


27-48 


Aiakidai 


3 


10 I : 


3.333 


6 


2 I 


: .333 


17 


17-60 


Achilles 


8 


19 i: 


2.375 


14 


10 I 


: .714 








48 


159 i: 


3.3125 


97 


91 I 


: .938 




DACTYLO-EWTmT«. 














03 


13-38 


Finding of Olive 
Story of lamos 


5 


9 I 


: 1.8 


7 


5 I 


: .714 


06 


29-70 


6 


18 I 


: 3 


12 


14 I 


: 1.166 


07 


27-76 


Founding of Rhodes 


6 


21 I. 


3.5 


9 


20 I 


: 2.222 


08 


31-53 


Story of Aiakos 


4 


5 i: 


1.25 


7 


5 I 


: .714 


P3 


7-58 


Story of Koronis 


4 


24 i: 


6 


13 


18 I 


: 1.384 


P4 


70-262 Argonauts 


37 


66 I 


1.783 


57 


61 I 


:i.07 


P9 


5-70 


Kyrcne 


6 


13 i: 


2.166 


14 


16 I 


:i.i43 


P12 


6-23 


Origin of Flute 


2 


7 I 


' 3.5 


4 


5 I 


:i.25 


Ni 


35-72 


Infant Herakles 


4 


14 I 


: 35 


6 


12 I 


:2 


N5 


9-39 


Aiakidai 


9 


8 I 


: .888 


5 


9 I 


:i.8 


N9 


11-27 


Adrastos 


5 


5 I 


: I 


5 


8 I 


:i.6 


N 10 


4-18 


Glory of Argos 





9 


: 9 


3 


4 I 


: 1.333 


Ii 


17-31 


Kastor and lolaos 


3 


4 I 


: 1.333 


4 


I I 


: .25 


I3 


15-37 


Kleonvmidai 
Aiakidai 


2 


7 I 


: 3.5 


5 


I I 


: .20 


I4 


19-42 





6 


: 6 


5 


I I 


: .20 


I5 


24-56 


Aias and his sire 


I 


13 I 


:I3 


4 


10 I 


:2.5 






Total Dactylo-Epitrite 


94 229 I 


: 2.436 


160 


190 I 


:l.i875 






Total Logaoedic 


48 


159 I 


: 3.3125 


97 


91 I 


: .938 








142 388 I 


: 2.732 


257 


2S1 I 


: 1.093 


It will be seen at a glance that the aorist 


prepor 


iderates i 


n both 


classes. 


This 


is the rule everywhere, mus 


it be the rule i 


Q lyric 



poetry. The lyric poet is unresting. He does not linger, he only 
touches on the K^fftaKaia X&yuiv (P 4, 1 16). It is of the essence of his 
profession that he is not to weary his hearers ; he must speed from 
theme to theme as a bee from flower to flower, as a ship from shore 
to shore (P 10, 51), as an eagle from quarry to quarry. We cannot 
expect the lingering imperfect. Even the slowest measures will 
hardly tolerate a leisurely unfolding. Only when the long voyage 
of the Argonauts slackens the flight of the poet, or when he pauses 
to watch the rearing of Achilles, favorite theme with all the Aiakidai, 
does the imperfect mount, does the imperfect surpass. The state- 
lier measures, of course, favor the imperfect, but the stateliness of 



164 AMERICAN JOURNAL OF PHILOLOGY, 

the measure is often tempered by the brevity of the myth. Most 
of the Isthmians are dactylo-epitrite, but the story of Aias and his 
father is a long one for the compass of the ode (I 5, 24-56), and 
there is a note of impatience at the end (v. 56) : tftoi di fxaKphv wdaas 
avayri<Tatrff dprrds. We must be Content with approximate results. 
Absolute uniformity would be fetal to vitality. If every dactylo- 
epitrite were full of imperfects, if every logaoedic abounded in 
aorists, that might gratify a certain sense of the fitness of things. 
It is enough that it is the rule. The rhythm is not all. We have 
to consider the bulk of the poem, the proportion of the narrative 
to bulk. Sometimes the poet says to himself rpi'a tlnta duip«<r«, and 
that must suffice us even in a long poem. 

Pick and choose and you can prove anything. So it might be 
said that a logaoedic poem will show the highest run of imperfects, 
a dactylo-epitrite the highest run of aorists. In I 5 (dactylo- 
epitrite) there is but one imperfect to thirteen aorists. In N 3 
(logaoedic) the imperfects outnumber the aorists more than they 
do in N 5 (dactylo-epitrite). But take all the figures and see how 
persistently the logaoedics continue on the high ranges. Leaving 
out Mr. Miller's third decimal place as too cumbrous, we find : 

Logaoedic, Dactyto-Epitrite, 

Impf. Aor. Impf. Aor. 



0:4 P 8, N 6 





:9 
:6 


Pii 
13 




: 5.50 

5 

4 
^3.83 


P5 
P 2 

O9 P 10 

I 




3-33 
= 3 


N7 

10(11), P6 




: 2.50 
2.37 


N4 
I7 



0:9 N 10 

0:6 I 4 

I • 13 I 5 

1:6 P3 



I •• 350 O 7, P 12, N I, I 3 
I : .^ O 6 



1 w 7 N 3 
The clustering speaks for itself. 



I : 2.16 


P9 


I : 1.80 


O3 


I : 1.78 


P4 


I • 1.33 


I I 


I : 1.25 


08 


I : I 


N9 


I : .88 


N5 



STUDIES m PINDARIC SYNTAX. 

As has been intimated, a careful calculation will r 
the account the effect of the durative and complexive 
of the indicative. An aorist infinitive in oratio ob]i< 
as much as an indicative aorist, and is selected on the i 
and it may be maintained that as the difference betu 
t^tvY>' '^ ^ mere matter of kind of time and not of ! 
so aor. subj. ^v^ and pres. subj. ^rvyai give the s 
one as f^vyw, the other as l^tvyav. If students 
distinguish between aor. and imperf. indie, they mi 
distinguish with equal sharpness between the prese 
in£ But after all the indicative gives the main line; 
moods only the shading ; and this comes out very 
treat the fibres already gained in a different way. 

If we lump imperfects and duratives as durativ< 
complexives as complexives, the difference between 
of poems will be reduced. In the logaoedic poems 
145 duratives against 250 complexives (about i 
dactylo-epitrites, 254 duratives against 419 comp 
I : 1.65). Thisisawamingagainst the'straining of 
ever just that theory may be. 

B. L. GiLi 



III.— WORDS FOR COLOR IN THE RIG VEDA." 

A grave responsibility rests upon philologians. The so-called color 
theory, which assumes that the human eye was incapable three or 
four thousand years ago of perceiving certain colors of the spec- 
trum (chiefly green and blue) that are now clearly seen and 
distinguished, has been abandoned by Magnus, who was in this 
respect the representative of physical science. Philology, there- 
fore, remains alone in its support of this theory ; and it is mainly 
through the writings of two philologians, Gladstone (on the color- 
sense in Homer) and Geiger (on color in Rig Veda, Zend Avesta, 
etc.) that this responsibility has been incurred. The latter author 
has been received as authority and upheld by Weise (Farben- 
bezeichnungen, Bezzenb. Beitrage II, 273 flg.), and the essay in 
question (Geiger : Ursprung und entwickelung der menschlichen 
Sprache und Vernunft, Bd. II, 3tes Buch) is quoted by the non- 
philological world generally as proving that the disputed colors 
were actually not mentioned in the Rig Veda, and were in all 
probability unknown at that time ; in Geiger's own words, ' the 
men of that time did not and could not call anything blue.' 

Of all the literatures which have been dealt with in this connec- 
tion, those of Greece and India are paramount in importance, for 
the others (Koran, Hebrew Bible, Zend Avesta, etc.) are either not 
so early a growth or are less extensive or less distinctively national. 

It is the object of this paper to inquire into the correctness of 
Geiger's deductions, and in doing this the writer may state at the 
outset that from an independent investigation of the color words in 
the Rig Veda he was led not only to question the facts adduced 
by Geiger, but also to doubt whether his application of these facts 
(even if proved) be admissible. 

In Part I is subjoined the occurrence and application of all color 
words found in the Rig Veda (Geiger treats in detail only the dis- 
puted colors), and in Part II some remarks have been added in 
regard to the results deducible from the use of color words in the 
Rik, the methods of Geiger, and the inferences to be drawn from 
this special study in regard to the color theory in general. 

» An Essay read before the American Oriental Society in New York, Oct. 1882. 



WORDS FOR COLOR IN THE RIG VEDA, 1 67 

Part I. 

§1. On examining the words in the Rig Veda which indicate 
color, we notice as in other languages that there are a great number 
of epithets which, strictly speaking, are not designations of color at 
all, but simply imply or suggest it, and, furthermore, that the optical 
effect thus suggested is always white, while the simple idea of 
glancing, shining, dazzling is generally fundamental to this attempt 
to reproduce by verbal signs the most striking effect which light 
produces on the retina. Foremost among such words in the Rik 
are the derivatives formed from the roots arc and bha. The first 
of these gives us ard, arcis, that which glances, flame (cf arka), 
and thence arc, ardn, ardmant, arcivant, employed as epithet of 
the A9vins and Maruts and of the " flaming " stars ; the adj. ardn 
in ardna pada seems to indicate the rapid movement of the feet, 
but in area mdsa it is the shining of the moon that is prominent, and 
arcaddhama illustrates the adjective force of the participle " (fires) 
of which the smoke glances." It is possible that we have also in the 
word rksa, bear, J/Mcror, which Kuhn regards as derived from the 
same root, an allusion to its gleaming (reddish) color. It is at any 
rate noteworthy that in the epic poetry we find the same word 
used to indicate the color of horses distinct from white (Mbha VII 
132, 30) rksavarnah hayah karkair migrah (bear-colored steeds 
mingled with white ones). If this be the case we have here an 
example of what will be noticed below, where words have 
evolved from a root that contained in itself no idea of color a 
a distinct color sense, through applying derivatives of such roots to 
objects that have color. The second root, bha, indicates the gen- 
eral idea of glance. The adjectives and compounds derived from 
it (bhamin of Agni and clouds ; the substantives bhas, bhama ; 
bhasas of Agni ; vibha of Usas ; vibhanu of Agni ; vibhavan of 
Agni and Usas with the feminine form vibhavarl of Usas and as 
substantive in the later language " the bright," /. e, night ; vibha- 
vasu of Agni) contain no idea of color more complex than that of 
simple glance or brilliancy.* We may say the same of the deriv- 
atives of div (dyu), dyumat (of Agni, Soma and other gods, the 
chariot, etc, as well as in other appUcations where it is difficult to 
say if clearness of glance and tone or beauty in general is meant), 

'Although Tibhavail occurs in the meaning *' yellow ginger,** yet this use is 
common to all words meaning night, and the idea of yellow in this applica- 
tion is later and foreign to the Rik. 



l68 AMERICAN JOURNAL OF PHILOLOGY, 

dyumna (glance) and its compounds, together with didyut and 
vidyut (lightning), cf. sudyut and possibly sudina (of the day and 
the morning) unless the derivation from dl, shine, be rejected and 
sudina be only dina in composition as purudina and madhyamdina. 
None of these has a developed meaning, nor is there in the Rik 
any instance of their use where the simple idea root is not suffi- 
cient explanation of the sense. Another root which may perhaps 
be included under arc (= arj) shows two distinct inclinations, if we 
may so speak, first, that expressed in arjuna, which preserves 
throughout the root-meaning of shining, and is applied to the day 
(ahar), the thunderbolt (vajra), the dawn (Usas), etc., in opposition 
not only to black (krsna), but also to the yellow gold color (pi9- 
ahga). In the same way rajata, another derivative from the same 
root, occurs once in the Rik in the meaning glancing (white), and 
in the later literature acquires as substantive the meaning glandng- 
white-metal (silver, compare the unconscious pleonasm of Horace, 
argenti splendor). The other inclination of which we spoke 
is that shown in rjra which stands in relation to aij as rksa to 
arc. The idea of glancing passes into that of glancing-red, 
and even the darker red (distinguished from rajata and arusa) 
is so indicated, the application of the word being chiefly to 
horses (cf. rjra9va), the celestial , steeds, and to fire (compare 
rjriya of agniyoni, and rjiti of offering, beams, etc., and rj»ka 
variegated). Of like origin are the words rajas the sky, and 
rajani, night, in reference to the shining stars (cf. prabhatayam rajan- 
yam). Among the verbs of shining, glancing we must include the 
most frequent of them, vas, with its many derivatives (usr, us, usas, 
the aurora); the adj. formed from the same stem, vivasvan, vivasvat, 
and the possibly connected usra, usriga, ustr, ustra, all of which 
unite the idea of glancing to the indistinct suggestion of the red 
glow of morning, and so substantively the dawn; or to that of the 
red gleam of ox and steer, etc., and so substantively ox, buffalo. 
But in all these cases simple glance is the fundamental meaning. 

Bhraja from bhraj (fulgeo) used of the sun (cf. agnibhrSjas = 
ignis fulgor^ and bhrgu) ; rajin, contained in a-rajin of mountains ; 
pajasvat from pajas (sahasra") gleam (cf. prthu") are not used often 
or clearly enough to see if any idea further than that of glancing 
has been reached by them. Two more examples complete this 
list where the color idea is marked enough to be noted by the 
addition of that word (varna). The first is candra from cand, to 
shine, especially called the candravarna, and this developed color 



WORDS FOR COLOR IN THE RIG VEDA. 169 

is not reddish but yellowish, so that it is the moon-color, as, indeed, 
candramas is the moon and candram is gold. It is used with 
hiranya (golden) of divinities, garments, fire, etc. (cf. hari9candra, 
9candra being the older form of the word). It is found again 
compounded with puru (puru9candra) of deities, sky and chariot 
Like the candravarna is the ru9anvarna, from rue to glance, which 
characterizes the (yellow) glancing soma. From the root-mean- 
ing shine which appears in rue and its derivatives ruksa, glancing 
(of Agni) roka, light, virukmat, viroka (once inviroke usasas) and 
virokin, the glancing (of divinities) we find an apparent approach 
to distinct color in rukma, the shining (-yellow, gold) and in the 
ptc. form ru9at (Xcvicdr) which is, as color, opposed to 9yava 
(brown) and krsna (black) and is particularly termed a color in the 
compound with varna given above, with which we may again com- 
pare the compounds ru9atpa9U (usas) " who has glancing herds,'' 
referrii^ to the red or yellow clouds (cf. ru9adgo, ru9at of dh^si, 
milk, and rugadvatsa) and ru9adtirmi (the fire) " which has glan- 
cing yellow or red waves (flames)." Other forms in the Rik are 
vasuruc, vasurocis, which (whether from Vasu (B. R.) or vasu as 
simple adj.) exhibit the two roots vas and rue compounded together. 
One root is found, the primary signification of which is doubtful: 
fviiy if it mean simply to shine, belongs here. B. R. give the 
meaning " be white," while Weise, in the essay referred to above, 
insists on the fundamental meaning "to bum." The adjective 
formed from this, 9veta (opposed to krsna, cf. Zend spaeta, Gothic 
hveita, Eng. white) means simply white, and is so used as epithet 
of the gods, the white steeds of Agni, the kala9a (elsewhere 
hiranyasya kala9ah), while in later literature applied to parvatah, 
the mountains, 9veta describes the appearance of the snow. So 
too 9vitra is later the white leprosy and 9vitrin in Manu is leper. 
Compare with these 9vitlci, 9vilna, 9vitnya, 9vityahc, 9vetya, and 
also the substantive 9vetand, the lighting up of the dawn (like 
sQrya9vit). In all these words whiteness is the color expressed, 
if indeed any real color is expressed (sQrya9vit contains a 
second idea of glance in so far as sQrya itself is ultimately only 
the glancing, svar-sar). On the other hand the form 9yeta seems 
to mean reddish white, as the feminine (yen! is the aurora, and the 
adj. itself is applied to the colors that appear at the rising and the 
setting of the sun, and to the fire. In 9yena, eagle, we have pos- 
sibly the same word. According to Indian authorities 9yeta itself 
is regarded as equivalent to 9veta, white. Whether 9yama, black, 



170 AMERICAN JOURNAL OF PHILOLOGY, 

is from this root is doubtful, the word does not occur in the Rik» 
and the apparent contradiction of white and black could be explained 
only by granting Weise's proposed rendering of 9vlt, etc, as 
denoting the color of burning (glancing and thus white) and burned 
(so black) as in Germ, blank against Eng. black. 

There are, however, other methods of forming adj. of color than 
by developing the idea of red from roots meaning glance. Weisc 
has sought to prove that the original and most primitive color 
words arose from roots which signify to bum (namely, 9vit and 
ghar), but in the Rig Veda we cannot resolve the roots meaning 
" shine " back into a more primitive " bum," except in isolated 
cases, and outside of these we fii)d other roots which have at first 
neither of these ideas, but which form similar color words. Among 
these are the derivatives from ci, ki, and cit, where the original 
meaning is that of perception. From the idea * this object is per- 
ceptible,' we have as the next step * this object is conspicuous, bright,' 
and so we arrive at the idea of brightness, glance, gleam, and this 
in turn develops into the more distinct notion of color. In such 
examples we have, to be sure, the idea of glance preceding that of 
color, but not as the ultimate meaning. So citra, from dt, is used 
to characterize gold, garments, chariots, as well as divinities and 
the sky (cf ex gr, I 115, i, citram devanam anikam), while in the 
post-Vedic language it means variegated. Other Rik formsirom this 
root are cetana, cetas, vicetas, where the meaning is uniformly that o^ 
bright, sparkling. Similar is the growth of those words of which ketu, 
glancing (banner), is a specimen, from ci (ki) to perceive ; cf. ketu- 
mat, sahasraketu (of the chariot of deities). It may be questioned 
whether real color enter into the meaning of these derivatives at 
all, as they stand in relation to the developed root meaning shine 
much as candidus in Latin does to candeo. 

The same process whereby * perceptible * develops into * glan- 
cing ' and possibly thence into an idea of color, is gone through 
with by roots meaning * be pure.' Thus 9undhyu, bright, clear, 
used as epithet of the sun's horses, is derived from 9udh (9undh) 
purify (ica$-ap6i), Thus also from pQ (Lat. pums) to be pure, clean, 
we have pavaka, used of Agni, Vamna, Aurora, etc., not only to 
denote ** purifier," but also to express the gleaming color-effect 
which is more expressly indicated in pava, found, as it is, only in con- 
nection with hiranya (golden), to describe the soma (hiranyapiva). 
Like these words which denote brightness by an idea of purity 
are the negatives adhvasvan, aripra, arepas, used of physical 



WORDS FOR COLOR IN THE RtG VEDA, 171 

objects and meaning literally^ spotless. With all of these we may 
place the root 9uc and its derivatives often employed in union with 
pflvaka. The noun 9uci and the adjectives 9ukra (9ukla) denote 
glance, brightness (cf. ^ucivarna, ^ukra-varna and substantive yocis, 
9oka, etc) and express the idea of brightly adorned, as does 9U- 
bhra, a similar word meaning either clear-shining or well-adorned, 
and, in the later language, white, while in 9ona (especially of fire 
and the steeds of the deities) we have the idea of red predomi- 
nating, which later gives the neuter substantive 9onitam the mean- 
ing blood. It is, however, doubtful if ^ona be connected with 
9UC at all. 

Again, in 9ukrapi9 we have another train of thought in develop- 
ing the color idea. This pi9, which in pi9ahga and pivahgarapa 
has developed into the meaning gold-color (cf. hiranyape9a) or 
yellow-brown (contrasted with arjuna) and characterizes divinities, 
rays of the sun, etc., has as ultimate meaning the idea '^ adorn,'' 
prepare (by cutting), decorate, as in pe9a and supe9a, and this 
decoration is applied to color in pi9ahga (and possibly in pi9a, the 
(red) stag. Possibly the same idea is to be found in su9ilpa, once 
applied to day and night, from 9ilpa, adornment, in the sense ' well- 
adorned,' and so * many-colored ' (so B. R.) 

Beside these we find in one, perhaps two forms, that the idea of 
sharpness may be capable of a similar development. The first 
of these is tejisjha, properly a superlative adj. from tij, be sharp ; 
from the meaning of very sharp we have first the idea of hot and 
then of sparkling. The primitive idea is seen in its application to 
the rays of the sun which are sharp or hot, while in limiting 
water (soma-drop) we have the transformation into glancing; the 
gleam is piercing bright, cf. dhsr^ agneh the glare (sharpness) of 
fire. Possibly too, tapu may belong here. 

Under 9vit we have purposely omitted 9iti in its Rig Veda com- 
pounds 9itipad and 9itiprstha (the word is not found in the Rik un- 
compounded). It is doubtful if this word is derived from 9i, bum 
(9U9vit) or from 9a, to sharpen (so Grassmann). Not only the deri- 
vation but even the meaning of this word is matter of dispute, as it 
may mean either white or black (v. above under 9vit). (^itiprstha 
is applied to the steeds of Indra and figuratively to milk, 9itipad 
is epithet of chariot and the brown steeds (9yava) of Savitar. The 
assumed double meaning of 9iti may have been the result of the 
later 9itikantha used with nllakantha as epithet of Rudra. In the 
later form 9ita B. R. assume a confusion of 9 with s, sita being a 



172 AMERICAN JOURNAL OF PHILOLOGY. 

post-Vedic word for white. Another development of the glancing, 
shining idea (if not of positive color) is to be seen in tvislmant, 
tve^ and the substantive tvi§i, where the simple root denotes only 
rapid movement, and thence glance. These adjectives are applied 
to fire, the countenance of the Manits, the rays of the sun, etc, 
while the substantive expresses an attribute (glance) of fire 
and sun. 

We leave this division with the remark that rudra, which in 
Grassmann is said to mean glancing, must, to effect this, be derived 
fi'om a suppositious root, rud to shine, and phalgva (phalgu) said 
by Grassmann to be * reddish ' is defined by B. R. in accord with 
its derivation from phalgu, weak, worthless. 

§2. We pass in this divbion to those words where the first color 
of the spectrum, red, begins to be indicated by direct comparison 
with red objects. First we mention agnirapa, fire-color, like agni- 
9rr (of the glory of fire), ^rl itself (cf. hari^rl) meaning glory, beauty. 
Both of these terms are applied to the storm -gods, like varcas 
(glory) and its compounds. So indhanvan, having the brilliancy 
of lighted fuel, and indhana (c£ dlBwi) of the clouds (dhenu). The 
root indh means to bum, to light a fire. Similar in sense b ang^Ura, 
a coal, so called from its fiery appearance (aflj, adorn, glow, shine, 
Grassmann) and in the later language one of the names of the 
** red planet Mars." Perhaps the best of the few examples in this 
list is rohita (later lohita), from the root *rudh, to be red, which 
underlies rudhira (A. V.) blood. The shorter form rohit appears 
as feminine substantive in the meaning * red mare,* and the femi- 
nine rohinl shows a like usage. The short form occurs again in 
the compound rohidayva, and as adjective rohita is generally 
applied to cattle and horses (hari, prasti, vdjin); while substan- 
tively rohita means a horse, just as lodha (from the form with 1 
like lohita) means a red animal. Kalmalikin, said to mean flam- 
ing, as epithet of Rudra, is derived from kalmali, gleam {yielleicht 
glanz B. R.) which occurs in A. V. 

Together with these we may place the word su-kim^uka, derived 
from the kim^uka, a tree with red blossoms. The adj. thus formed 
by the prefix su (cv) is said by Grassmann and B. R. to mean ' beau- 
tifully ornamented with blossoms of the kim^uka tree.* The adjec- 
tive occurs, however, only as epithet of sorya (figuratively as 9al- 
mali with the epithets vigvarapa and hiranyavarna) and in the 
Nirukta (quoted by B. R.) the adjective is explained by sukagana, 
fair to see. It seems to me probable that we have here a direct 



WORDS FOR COLOR IN THE RIG VEDA, 173 

comparison with the main characteristic of this tree — its red 
appearance, as in the following with the characteristic of gold 
(hiranya). This use we may illustrate from the later language, since 
the form is so isolated in the Rig Veda, and by comparing the 
epithet with Mbha. IX 58, 34, we can understand how the red 
blossoms of this tree provoke a direct comparison of color. The 
epic passage is as follows : 

. . . rodhireni 'bhisamplutau 
Dadr^ite himavati puspitav iva kim9ukau. 

(The two warriors) dripping with blood looked like two kiih^uka 
trees blooming on the mount of snow ; or again, XIII 30,43: 
" down fell the warriors, wet with blood their limbs, like two kini- 
9uka trees cut down." Like this is also XII 166, 62-3 : "And the 
foul earth grew all filled with bodies dripping with blood like 
hills with kiniguka trees." So in the Rig Veda passage it seems 
to me that the tertium comparationis is rather the fiery red 
common to both the sun's chariot and the tree's blossoms. We 
have in another form saryatvac (cC hiranyatvac) an attempt to 
express the color of the Maruts by referring their appearance to 
that of the sun, cf. sQro varnah (IV 5, 13) and the two forms sQr- 
yara9mi, surya^vit ; and this word expresses also the glance of a 
chariot or the radiance of a goddess. 

§3. Words for red and yellow, or reddish-yellow, which are not 
the result of comparison with physical objects and are not derived 
from roots meaning glance, etc. 

Foremost among these is aruna (cf. arunapsu) red, bright-brown, 
golden-yellow. Any distinct standard whereby we may estimate the 
exact worth of this adjective is wanting. It is applied to the dawn, 
the sun, the soma (plant), the color of the wolf and catde, and 
means as substantive the* red-yellow cow (dawn). As compound 
adj. it appears in aruni^va, of the horses (c£ aruni) of the Maruts, 
and as varna (I 73, 7, naktflca cakrur usasa virope krsnanica 
varnam arunahica samdhuh) is the glow of dawn in antithesis to 
the blackness of night Similar to this in form and meaning is 
aru^a, which is the fiery red of the thunderbolt, the fire, the steeds 
of Agni (c£ of Agni aru^astdpa, with red flame), the sun, the dawn, 
and is found like aruna in the fem. as substantive, arusT, meaning 
red cow (dawn). An]9a (in'aru^ahan) is perhaps (B. R.) only 
another form of the same word. With both of these compare the 
compounds tryaruna, tryarusa, the first being a proper name, the 
second used as epithet of cattle, ' red in three places.' With these, 



174 AMERICAN JOURNAL OF PHILOLOGY. 

and in sense closely related, stands bradhna, light red, and yellow 
which is used as epithet of fire, soma, and of the sun's horses. It 
occurs but infrequently and with no great variety of application. 
In an interesting compound of this word (9atabradhna) bradhna 
seems to denote the shining metal end of Indra's weapon (isu). 
Six colors, we are told (X 20, 9), accompany Agni the fire god : 
krsnah 9veto Vuso yamo asya (agneh) bradhnah rjra uta 9ona 
yagasv^. This list includes many of the words discussed above, 
and it is safe to say that there is not (with the exception of krsna, 
black) one color word here mentioned to which wc can give a 
meaning that will answer all cases ; at times applied to yellowish 
objects, at times to reddish, again to objects partaking of both 
colors, these words are as vague in meaning and as wide in appli- 
cation as the English use of purple in purple blood or purple grape. 

Gaura, which is applied as adj. to milk, is used as substantive to 
denote a kind of buffalo, and in the frequent expression " gauro na 
trsitah piba *' seems to be a general designation of cattle. Latin 
galbus has been compared with this word. 

Like the use of the kim^uka tree in denoting color (v. above) is 
that of the udumbara (fig-tree) in the once found form udumbala, 
to which B. R. assign the meaning copper-colored, although its 
use as epithet of Yama*s messenger might lead us to compare it 
with 9yava, used of Yama's steed. It is possible that the compar- 
ison is not with the leaves but with the yellow fruit A darker red 
than any epithet hitherto noted is that in kapila, the color of the 
ape (kapi, cf. vrsakapi) used once of garbha. The development of a 
general color-term like kapila from the name of an animal is rather 
surprising, and Geiger maintains that both gaura and kapila are 
indicative of color first and later of names of animals. The inverted 
process is common enough as we have seen, and many animals are 
named from their color, just as they are fi*om the sound they pro- 
duce (kaka, cakravaka, krostr); nevertheless we may compare 
pidaku-s&nu, maytira roman (v. next paragraph) to show that the 
characteristics of animals are used in determining color. DaHc 
yellow or reddish brown is the meaning assigned kadru (in kadro) 
which appears only as substantive, the dark brown or dark yellow 
soma vessel (cf trikadruka). Babhru is a more genuine color word 
than any of these. It is really the mingled color of dark red brown, 
but apparently darker than pingala, with which it is in Sk. fi-e- 
quently connected (pingala does not occur in Rik, but pinga, bow- 
string, is referred to pinga, yellow, apparently the same as pingala). 



WORDS FOR COLOR IN THE RIG VEDA, 1 75 

Its use is varied, being applied to Agni, Rudra, soma, horse, cow, 
and the (brown) dice (nuts). It is distinguished from aruna 
bradhna and hari, and is used substantively to denote certain 
plants. 

As the last word in this category we have the root ghar, to bum 
(cf. gharma, heat) which develops in the adj. forms into a color 
meaning. The idea of bum or burned is the real basis of the color 
idea, for this idea does not come through the meaning glance and 
sparkle, but is the fire color itself. The word is, however, peculiar 
in its application, especially in later Sanskrit, having there the 
meaning green as well as reddish-yellow. The first group of deriva- 
tives comprises hari, harita, harina, the second (with stem -vowel 
weakened) hiri and hiranya. The prevailing meaning in all these 
words is yellow, fallow, and, in general, any change from this mean- 
ing seems to incline rather to the lighter than to the darker end of 
the spectrum, or, to speak more correctly, the underlying notion 
of burning, fier>% includes at first red and yellow, and, though use 
has almost confined the application to the latter, yet the former 
meaning is also occasionally brought into prominence. The most 
frequent use of the short form hari is in the dual as substantive 
to denote the two steeds of Indra (cf. harya9va of Indra), and these 
receive in turn epithets which show that these steeds are regarded 
as yellow, red, or whitish (arusa, rjra, and the once used rohita hirl, 
together with 9itiprstha, v. above). Hari is applied again to soma, 
which as plant or moon is yellow (though elsewhere soma is called 
aruna, babhru, rjra) ; cf. IX 97, 9 (somah) diva harir dadr^e nak- 
tam f jrah — the moon looks pale-yellow in the daytime but reddish 
at night. To the soma itself hari is applied as limiting the ani9u 
(plant), indu (drop of the plant), and to the press stones (yellow 
with soma juice). Beside this frequent dual form we find singular 
and plural used to characterize the steeds of Agni, the sun, Savitar, 
the wind (vayu), the A9vins, and soma (moon), or the gods are 
conceived as steeds and receive the same epithets, or agni, not as 
god, but as fire, and vajra, the thunderbolt, are so called. The 
form harit (fcm.) is used (like arusl and rohit) to denote the red or 
yellow mares of the sun, Agni, Indra and Soma, or the fingers 
spoken of metaphorically as steeds. Haritvat, gold-yellow, is 
epithet of the glor>', brightness of the sun (haritvata varcasa) where, 
if varcas itself means brightness, we may translate *'by the glaring 
gleam." 

The longer form harita, fem. harini (c£ harina, gazelle, Mow- 



levbat wider applka- 
tD the divinities and 
[ in the same way as 
, thunderbolt, the ape 
bnns <^ the gods, the 
d, referriDg to soma, 
tvo boms, and ^ipri 
itive to vanaspati, the 
pithet hiranyaya, the 
be assumed for hariti 
ndra. Once, too, we 
I wood as opposed to 
a meaning approach- 
ipplied apparently to 
lyQpa in hariyopiya, 

ika) where we would 
han by yellow, Omit- 
: is spoken of without 
in antithesis between 
hat is haiita, and also 
imanam nama)," but 
Ij. virupa, variegated, 
J is so probably by 
refore, as &r as this 
yf yellow or its later 

rivarpas, which signi- 
ish greenish appear - 
idhlyasam, prthivim 
[e Ludwig translates 
sends (us) down the 
covering." And we 
appellation which we 
r "green," and such 
the rest of the hymn 
aning in this word to 
ar compounds which 
nn (harivarpas, in X 

lis (yellow) jaundice 



WORDS FOR COLOR IN THE RIG VEDA. l^^ 

(hariman) shall pass upon the parrots, Geiger seems to draw the 
conclusion that the (green) parrots are regarded as yellow. But 
the color is not spoken of as passing over to the parrots, all parrots 
are not green, and finally the same disease is desired for the h^ri- 
drava, an unknown yellow bird (from the same root with lengthened 
stem vowel, as in h^yojana * the harnessing of the yellow steeds ')• 
The compounds of these words show much the same usage as the 
application of the simple adjectives, harike9a and hari9ma9^Uii, 
yellow-haired (of Indra, Agni, and the sun) ; harijata, bom in yel- 
low red glance (of Indra). We have, too, some compounds where 
hari as first member means Agni, as harivrata, or Soma, as harivat, 
haripS, or the press-stones, as haridru and harisac, or the thunder- 
bolt, if harimanyusslyaka be correctly interpreted, or the steeds of 
the gods, as hariyoga and haristha. Like the use of harit is that 
of hari in hari9ipra, and in hari^candra (cf. hari^rl, mentioned in i) 
we have the two roots, one of fire, one of whiteness, united 
(c£ hiranyacandra). From the other form, hiri, we have similar 
compounds, hirigipra, 9ma9ru (of Agni and Indra), and hirlmant 
(of Indra) gleaming, to which we have the corresponding form 
harivant The form hiri does not occur except in composition, and 
that but seldom. Hiranya (Zend Zaranya) with its many deriva- 
tives is employed in the universal sense of the yellow-gleaming 
metal (gold), and since this meaning is universal we note only 
those compounds which may illustrate the compounds of hari. 
Thus this word is also compounded with ke9a (hair) to describe 
Agni, and hiranyaya is an adjective applied to cattle, thunderbolt, etc 
So hiranya-rfipa or hiranya-varna, gold-color, which underlies the 
meaning of hirany&ksa, golden-eyed, of Savitar, is like the use of hari 
in Rig Veda when applied to the persons of the divinities (har- 
yaksa, however, is post-Vedic and in epic, Mbha X i, 38, means 
yellow-green-eyed: [so*pa9yat] ulukam haryaksam babhrupih- 
galam). 

§4. We come now to the second color that our forefathers are 
said to have " had no name for, because they could not distinguish 
it fi'om black, grey and brown" (Geiger. bd. II, s. 356). There is 
but one word under discussion, nlla, which means in classical San- 
skrt dark blue. Geiger asserts that in the Rig Veda this word 
means only black or dark-brown, and even in the later epic he 
would prefer to understand grey as the meaning of the word (s 
307). The Petersburg lexicon defines nila as dark-colored, parti- 
cularly dark-blue, blue-black. Before entering on the discussion 



ITS AJiEMICAN JOURNAL OF PHILOLOGY. 

xsk regard to the R% Veda we would like to point out its exact 
megnrng in the later Cterature, as the cases where the word occurs 
in the Rig Veda are few. On glandi^ over the various compounds 
of nila found in the classical literature we see at once that the pre- 
vailing idea b dark-blue, and that the meaning "blue" is so 
inherent in the word that the accessory notion of dark easily van- 
ishesy whereas the meaning bliu remains, so that we may oftoi 
render the word simply as blue, but never simply as black. Espe- 
cially where nlla is introduced into languages that have no 
affinity with Sanskrt thb prevalence of the blue b conspicuous, as, 
for instance, in Tamil, where to-day nila b synonymous with sky, 
L ۥ the blue. But even in pure Sanskrit the nllavarna b blue, and in 
the substantive form nila b synonymous with the blue-water-lily 
or the blue sapphire. Generally, however, the word denotes a 
darker blue, as in expressing the color of the snake's back, the 
waters of the sea, the neck of the peacock (cf. nilakantha, nlla- 
griva of ^iva), or the indigo (cf. Rood, Modern Chromatics, p. 21). 
The form nllaka has the same meanings, though used perhaps 
more decidedly in the sense blue than dark-blue. The derivation 
of the word b unknown, for its assumed connection with Lat niger 
(Geiger, s. 306) is at best but a guess. If we turn from the classi- 
cal to the Vedic use we find first that nila in the Atharva Veda b 
distinguished from lohita and from pi9ahga (red and reddish-brown) 
and occurs in a few compounds^ of which the exact meaning b 
doubtful, although they seem to denote more the dark than the 
blue color, but as both these forms (nllayikanda and nilanakha) are 
mere names of demons, the latter being dbtingubhed from the 
green or yellow demons, called harita (19, 22, 4 and 5 ; nilanake- 
bhyah svahah, haritebhyah svahah) we can assign no exact mean- 
ing to the adjective (nilayikanda is like nilalohita an epithet of ^iva). 
In the same way nilam udaram (15, i, 17) b opposed to the lohi- 
tam pretham of the anthropomorphized deity, as in (^t Br. 9ukla, 
the gleaming white b dbtinguished from nila, where, however, 
black b not necessarily the meaning of nila. 

Geiger and Weise assume that thb word nila, which may occa- 
sionally mean dark without reference to blue (?), but which in the 
majority of cases in the later literature has a distinctly blue tone, 
ia, in the Rig Veda, entirely devoid of the meaning blue. Ifi now, 
the derivation b unknown and the word everywhere except in the 
Rig Veda appears to mean dark blue, and even blue alone, it will 
be necessary, in ordet to support Geiger's statement, to prove that 



/ 



WORDS FOR COLOR IN THE RIG VEDA, 1 79 

nda in the Rig Veda is applied in such connection that it is here 
impossible to attribute to the word any meaning of blue at alL 

We examine then the use of nila in the Rik. The word never 
occurs alone, but always in composition. First as adjective with 
the suffix -vat, nllavant, the word is used once as epithet of drapsa. 
If we isolate the word from its connection we could here learn 
nothing of the meaning, for drapsa (literally drof) is elsewhere called 
white (9veta), black (krsna), and red (aruna). The use is here, 
however, metaphorical, and drapsa means the spark of the fire, the 
verse in which it is contained being part of a hymn to Agni (VIII 
19, 31) : " Thy spark, O Agni, is nilavan, kindled in good time," 
Now we connect no especial idea of darkness with a spark of fire, 
and if we compare this passage with VII 87, 6, we find drapsa is 
used metaphorically of the moon and receives the epithet 9veta, 
the white, the shining spark in the sky. Since in the first passage 
the time of lighting the fire is meant, I have thought it possible that 
nilavan here might betoken the blue color of the first small flames 
that arise when wood is kindled. Not insbting on this, however, 
it remains to be proved that the spark of a newly kindled fire is black 
or dark — yet such must be the meaning if Geiger is correct Nila- 
vant occurs in one other passage as epithet of sadhastham, place, 
home, refuge. It is, however, questionable whether the word which 
belongs in this passage can have any influence on the discussion at 
all, for if we read with Mailer's large edition, nilavant, we have an 
entirely different word meaning nest-like (so understood by Sayana). 
The former reading nilavant is, however, introduced into Muller*s 
small edition, and is endorsed by the Petersburg lexicon. Regard- 
ii^ this reading as correct, we find that in the passage in which it 
occurs the poet is speaking of Bf'haspati, who is said (IV 50, 4) 
to be arisen out of light, and (II 23, 3) to be the enemy of dark- 
ness. We are therefore inclined at the outset to be sceptical, when 
we are asked to believe that in this passage (VII 97, 6) the same 
god is represented as one whose strength is a dark place, nor are 
we more inclined to believe this when we examine the whole pas- 
sage : lam arus^o a^va brhaspatini . . • vahanti saha9cid yasya 
nilavat sadhastham nabhona rupam arusam vasanah, t. e. "red 
steeds carry this B^haspati of whom the power is a place nilavat, 
being clothed as with a cloud in red color." In the following verse 
we are told of his golden weapon, and the whole picture of the 
god is one of bright color and glance. Why then should his 
strength be a place of darkness ? It is not here the picture of a 



l8o AMERICAN JOURffAL OF PHILOLOGY. 

god of light bom from the darkness that precedes the dawn. 
With this idea we are familiar, but here we have sadhastham, 
either a place of refuge for those who invoke the god, or his place, 
his home (as in A. V. II 3, i, divi te sadastham^ is itself a source 
of strength. In the latter case we might translate : " Brhaspati 
whose strength is the dark blue place," i. e. the sky. But however 
we take sadhastham, what sense do we make from the pass^^ 
by translating nitavat dark-colored ? 

Beside these cases niU is used twice (III 7, 3, V 43, 12) in com- 
position with pi>tha (back) as epithet of Agni. Nilaprstha has 
been translated " he whose back is black," i. e. the fire. We are 
accustomed to the idea of black in connection with fire, for we 
have it, for instance, in the passage quoted in §3, where the path 
of the fire is black (kr^na), and such compounds as krsn&-yiuDa, 
kr^na-vyathis, k^'sna-vartani, kr^na-slta, krsnfidhvan, give the same 
idea of the black path. But in all these expressions the adj. is 

■ ... _._■_.__. __^ jjj^ substantive invariably means the Irack 

K> in kr^na-pavi and in the doubtful passage 
irayah*) while in ki^nagarbha and kr^na-yoni 
which the fire comes is meant. We have, 
dmony to the effect that the "back" of Agni 
u-k, but as light : I 58, 3, prstham prusitasya 
the back of the fire shines like the back (vault) 
could prstha refer to the track of the fire 
hat nila is here dark, for the path of Agni is 
ocii^, cf. citra-yfima, of Agni (III 3, 13). This 
Tiployed to characterize the hamsa (pi.), a bird 
as goose, crane, swan, flamingo, etc In VII 
scribed as nilapr^h&h. Whether nlla here 
black remains an open question, but that the 
ressarily mean dark without any idea of blue 
The hamsa as steed of the A^vins recdves the 
lered (hiranya-parna), but this is figurative, 
ime for the Indian crane (from this same nda 
1 is said to be bluish in color ; and nil&k^ is 
)ose fix)m its (not dark but) blue eye (ak^a). 
as the case is, the similar usage of the later 
us rather to incline to the meanii^ blue than 

cases in the Rig Veda in which nlla occms, 
ige (X 85, 28) where it is fiwnd compouoded 



WORDS FOX COLOR IN THE RIG VEDA. l8l 

with lohita (late form for rohita) ; and, as Geiger says» the passage is 
probably taken from the Atharva Veda, as it is identical with A. 
V. 14, I, 26. Even if genuine, and if lohita were not itself a sus- 
piciously late form, we should learn nothing from the passage of 
the color of nila, for reference is made to bewitchery, krtya, and 
the verse states that the color of this power is red and nila, which 
m^ht be black, blue or any other color. It is from the fact that 
the sky is not called blue in the Rig Veda that Geiger doubts if 
bhie was known as a color, and after examining the use of nila he 
draws from the above few, and, if I am not mistaken, in part con- 
tradictory data the conclusion ''blue was not mentioned and 
was not known in the time of the Rig Veda, but develops out of 
die idea black or brown, the only meaning that nila can there 
have." The facts of the case are, however, that the word for dark- 
blue occurs in the Rig Veda, and Geiger has not proved that this 
customary meaning is impossible in the Vedic application of the 
word. Why the sky is not called blue will be discussed in Part II. 

§5. Without any notion of lighter color is jy&va (dark-brown), 
employed as epithet of horse and chariot, and substantively as 
(brown) horse, nomen proprium, and (in the feminine form 9yavi) 
Ae brown mare, the night. It is contrasted with red, rohita, and 
white, 9iti(-pad). The later post- Vedic word 9y&ma (dark green 
or dark blue) appears to be a related word. 

Blackness, or darkness (andhas, tamas, cf. andha with tamas, 
and tamisra, night) is often implied by negation of light, as in aruc, 
ar§jin, acitra (cf. aketu), while black itself is denoted by the adjec- 
tives asita (sita, white, does not occur in the Rig Veda) and krsna 
(varna). The latter occurs in many compounds, and the feminines 
kr^na, krsni are used as substantives to denote the blackness of 
night or night itself. Krsna as proper name is equivalent to Rama 
(the dark), and its related ratri, night.* 

§6. It remains to say a few words in regard to some designa- 
tions which come under none of the above heads : palita, grey, 
and palasti, said by Sayana to be the same as palita (in palastija- 
madagni) means the greyness of years, as substantive (pi.) grey 
hairs. We have several words and expressions meaning varie- 

* Night, as we have seen from many examples, may be tenned either the dark 
or the bright, according as the dominant idea is that of gloom on earth or of 
light among the heavenly luminaries. In dosi and pradosam (evening, dark- 
ness, opposed to usas, dawn) Grassmann sees a derivative from dus, to spot, 
darken. 



l82 AMERICAN JOURNAL OF PHILOLOGY. 

gated, as rjlka, vi-rOpa, vi9varflpa, puni-rtipa (c£ X 169, 2 [givah] 
yah sar(ip§h virQpil ekarupilh) used especially of the cows, but 
also of the cloud, milk, sun, frog, snake, etc. So prsant (prfati) 
spotted (sprinkled), of the cows, the clouds, Maruts* steeds, etc 
^bala is used in the same way of Yama's dogs (cf. Udumbala) 
.and 9arvari are the variegated steeds of the Maruts (cf. B. R. s. v.) ; 
.'SO, too, kilasi, property leprous. In mayura-roman, said of the 
steeds of Indra, we have an attempt at color-description that 
reminds us of kapila (v. above), literally the word means " pea- 
cock-haired," that is, many-colored. Like the compounds of rtipa 
is sprhayadvarna (II 10, 5) literally " striving after color," 1. e. 
changing color, of Agni ; but varna does not occur compounded 
with vi, puru, vljva (and is therefore a later word for color than 
rupa ?). Both r&pa and varna denote color, as in sa-r&pa sa -varna 
(of like color), su-rupa (vi-su-rupa) su-varna, (of good color, cf. 
sudr9rkarupa). Vyeta (f. vyenl) is strengthened from eta (£ eni 
and eta) with the same idea of many-colored, used of Usas. 
Strength of color b denoted by vi in composition, or by the accom- 
panying adverb brhad, strong, as in brhadbhilnu, or by rabhasa, as 
in rabhasana. 

The word madhu (sweet, honey) is compounded with varna 
(color), and characterizes the ghrta, melted butter, A9vins, and 
chariot This madhuvarna may mean, as Sayana says, " having 
an agreeable color," or we may take it more literally as " honey- 
colored." 

Part 1 1. 

We find from this investigation that the use of color words in 
the Rig Veda is not unlike that in other poetic literatures. The 
light colors predominate in frequency of occurrence and breadth 
of application. All that glances, glares, sparkles, is more frequently 
described than that which is dark and gloomy. Light and dark 
are the broad general antitheses, real color is less often mentioned, 
and to fix any exact standard for these colors is impossible, as this 
is forbidden by the general meaning of the root, or by the uncer- 
tainty we are in in regard to the real color of the objects described. 
It is, finally, impossible to mark off distinct meanings for the 
majority of color words used in the Rig Veda, as no one color 
term is precise enough to answer to any one spectrum color ; an 
indefiniteness that lies, however, in the language alone, since we 
have no proof that such indefiniteness was the result of physical 



WORDS FOR COLOR IN THE RIG VEDA. 183 

inability to distinguish between the various colors or shades of 
color which in the literature are grouped under one universal term. 
In regard to the disputed colors, we have two words, hari (harita), 
and nlla, which in the later literatvure may mean, on the one hand, 
yellow and green, on the other hand dark and (dark-) blue. A 
few cases render it possible that the later meaning of these adjec- 
tives may exist in the Rig Veda ; that such later meaning cannot 
exist in the Rig Veda is neither proved nor disproved. Unless nlla 
include the idea of violet we have no term for this color in the 
Vedic literature. 

We pass now to a discussion of Geiger's theory in regard to the 
explanation of the infrequency or, as he would say, the absence of 
those terms which denote green and blue. Before criticizing this 
theory from the standpoint of the Rig Veda, I would call the 
reader's attention to a few statistics on this very point drawn from 
much later literature. 

In Milton's Paradise Lost we have all the light gleaming colors 
used in abundance, chiefly red and yellow, especially in union with 
hirid, dazzling, etc., whereas green is mentioned but fifteen times, 
only nine places show purple (which is indefinite and may mean 
red, of blood and roses, or blue-black, of grapes), and the test 
adjective, blue, is found but once (XI 206), although the lapis 
lazuli is occasionally made to do duty as blue in the term '' azure." 
In the rainbow Milton sees "three listed colors" (XI 866, 897). 
Going back to the thirteenth century we find in the use of color 
words in the Nibelungenlied a still better proof of how fair an 
index of ocular development is given us by the employment or 
non-employment of certain colors in the literature. For here 
red is, as proved by the overwhelming majority of cases, the 
£ivorite color word. Yellow occiu^ less often, and a pure color 
word for yellow occurs but once, the idea being generally suggested 
by comparison with gold. Green is mentioned only four times, 
but never except in the fixed expressions " green as grass," " greener 
than grass " (graz or kl^), applied to jewels, marble and silk, but 
never to fields, trees or mountains. Finally, in the 9516 verses of 
this poem we have not a single mention of blue, although every 
opportunity is presented the poet for describing sea and sky by 
this color.' 

' The precise use of color words in the Nibelungenlied is as follows (the 
immbers refer to the stanzas in Bartsch*s edition) : The chief verbs denoting 
£Unce and shine are liuhten and schlnen ; glancing is denoted by lieht (adj.) 



184 AMERICAN JOURNAL OF PHILOLOGY. 

Prof. March (in his essay before the Am. Phil. Ass. in Cambridge, 
July, 1882) has told us that in Beowulf (circa 7th century) no word 
occurs for blue or green. It may, if I am not mistaken, be 
added that in the Heliand, about the same time, there is no word 
for blue. 

I now return to the Rig Veda. Weise says, quoting from Geiger 
the statement diat fields and fruits are often ^>oken of while we 
never hear of green fields : *A11, at first, was vague in color, but 
gradually a difference was perceived, and men were compelled to 
find some term to express this*newly observed appearance '; and 
again : ' green was for a long time regarded as yellow.' Geiger 
himself (Bd. II, s. 305) says of the period represented in the Rig 

and (adv.) liehte schlnen (2006) ; as noun lieht is candle (1005). The adj. (to 
which, in 80, the word gevar (color) is added, in waete lieht gevar) is applied 
to cloth, garments (kleit. wit, gewant, 572, 80, 363, 586), pfeellc (570, 833), porte, 
gewtthrte (429. 573), precious stones (586, 1783), bant (von golde, 1654) ; weapons 
and armor (helm, 1783, 1744, 20s, helmvaz, 2279), cf. helmenglanz, 1841, and 
goldvaz, 1328; schild (212, 2170, etc., cf. der liehten schilde schin, 1602), ringe 
(214, 2218), swerte (233, strengthened by vil, 1972), brUnne (66, 406), wicgewant 
(2317, in 1597 wlcg. "in herlicher varwe"); furs, riuhe (9S4): again, used of 
flesh (hende, 587, wange, 618, ougen 84, 1286, with vil 1069, 1349), the red color 
of the face (varwe, 240) ; of gold (183, 255, 570) which is red, or of the moon 
(283, 8x7, 1620 der liehte mane), of the sun (1624, liehtez schlnen), the dawn 
(1360, den liehten morgen schlnen sach). Lieht as adj. corresponds to the verb 
and noun (schlnen, schIn) of glance, or is united with rdt, red, to intensify the 
redness (2068, liehtez golt vil rdt, or 400, schellen von liehtem golde rdt, cf. 
bouc . . . lieht unde sch6ne was er von golde rdt). 

White, wiz: is used of flesh (hant, 274, loii, 1358, 1701, cf. vil wfze hant, 
661 ; arme, 451), of dress (compared to snow, mit sn^wlzen geren, 555, 392, 
compared to saben, in saben wizer hemede, 632, 976), of sails (wTzer dan der 
sn^, 508), of silk (sider wtz also der sn$, 362), of armor (wize brOnne, 79, 188, 
halsperge wize, 171 7). White, blanc: is used of cloth (hemede blanc, 670), of 
horse and garment both (von snS blanken varwe, 399), of sweat (1882). 

Loss of color, paleness, is denoted by bleich, generally opposed to red : varwe 
(155. 1665, 1734. etc.), just as the " liehte varwe," in 987, is *' des t6des reichen." 

Red, rdt (occurs most often of all colors) : is used chiefly to describe gold, 
golt vil r6t, golt daz r6te\(268, 854, 951. 999. 569, 687. 1784. 4795, 1427, 1554, 
2130, 68, 92, 71, I79S)> AnA goldesrdt characterizes garments, saddles, horse- 
trappings, or goblets (von golde r6t, 606). Red alone is the characteristic of 
the bougen (1322. 1634, 2204), or intensified by vil (1550), and with all the rdt 
is epithet of gold (von alrdtem golde, 435, in 1595 ein zeichen r6t); or, again, 
rot betokens the blood color (1624, 1006, bluotes rdt), and in 2309 (brUnne rdt, 
with gold or with blood, 428, 2309), more particularly var nach bluote (21 «) is 
blood-colored, the same as bluotes varwe (218), so rdte bare (239 reddened with 
blood), von bluote rdt (loii, 1932); armor is bloody red, an sfme rdtem (t. r. 



WOJiDS FOR COLOR IN THE RIG VEDA, 185 

Veda : " Not only was the sky not called blue, but nothing was 
called bhie, and it was impossible to call anythmg blue ... No 
word for such an idea could have existed, because the idea of blue 
was early and late bound up with another idea.'' We pass over 
the fact that Geiger, in mentioning VIII 19, 31 (where the sparks 
according to his theory are called black), fails to mention that the 
adj. is ho-e applied to fire and contents himself with the remark 
" in this place nlla cannot mean bluish, but blackish " (s. 306), and 
concern ourselves with the theory alone. We have seen that in 
other Uteratures no great reliance can be placed on the occurrence 
of color words as indicating development or lack of development 
of color sense. But particularly is this true of the Rig Veda. 

bloody) helme (191); so (2279) von pluote r6t nude naz, and (ao88) die blut 
▼arwen helde. In 921 the blood is bluomen r6t, and r6t is the color of velvet 
(samft rftt, 705). Fire is thus characterized (firver rdten vanken, 186, 2053, 
fiwer r6ter wind, from the clash of swords, 3062, 2275). Fire-red is blood in 
S072 (brOnne fiwer r6t, from blood). Rose-red is the blush (rosen rot, 241) or 
the color of the face in general (282); the blush of joy, freuden r6t (1497), vor 
▼renden rdt (448), vroQden iti (770), or of shame (614), or of an^er (465 in zome 
r6t) is thus described ; as, too, the redness of the lips (r6senvarwer munt, 591). 
As mentioned above it is the antithesis of bleich, pale, as in 285, er wart . . . 
Til dieke bleich unde rdt ; the alternation of each produces a mixed color (1666, 
gemischet wart ir varwe. bleich unde rdt). In 281 we have the very Vedic 
description: No gie die roinnecllche, also der morgenrftt tuot dz den trUeben 
Wolken. 

Yellow: it is occasionally dubious whether red or yellow be meant when 
gold (golden) is introduced to express color. With rdt expressed, as is generally 
the case, gold is represented as red, otherwise *• goldvarwe ** and ** guldin,** or 
gold alone may betoken yellow, as in 434, 954 (though here goldes zein may be 
made of gold), or in 712, die golt varwen zoume; 376, waz golt varwen g^ren ; 
gohvarwen schilde, 376; or gnldin in 570, die guldlnen scaemel; 679, ein 
guldin vingerhn ; so the sword-hilt, gehilze, 1784, and 956,guldInetQlle (though 
in the three last guldin may mean made of gold). Once only have we a genuine 
color word for yellow, the same word that occurs so often in Beowulf: Die sach 
nan valevahse, f. /. (women) with yellow hair (vahs). The passage (1783 and) 
1784 (quoted above) gives us in short compass almost all the color adj. of the 
poem which betoken gleaming, red, yellow, and green : ein vil liehtez wifen, 
Az des knopfe schein. Ein vil liehter jaspes, grQener danne ein gras. Sin 
gehilze daz was guldtn, die scheide ein porte rOt. 

Green is mentioned very rarely. Once, in the passage just quoted, grUen 
designates the jasper, greener than grass, and the same comparison is found 
(404): von edelm marmelsteine grUene alsam ein gras; so, too, the jewels 
glance with grass-green light against (wider) the (yellow) gold (436), cf. 577, 
Ifihte gegen der wAt, and 799, where the color of the face surpasses the glance 
of gold. Instead of grass we find clover once employed to illustrate the color 



l86 AMERICAN JOURNAL OF PHILOLOGY, 

It is in bxX misleading when one tells us diat green things are 
mentioned so often and yet not called green. The natural impli- 
cation is that we have the green color left out in a mass of litera- 
ture filled with descriptions of earth's products. So, for instance, 
Montagu Lubbock understands Geiger (v. in his recent article in the 
Fortnightly Review). But, we ask, what are the trees and fields 
called ? How often in the Rig Veda does the poet allude to these 
objects of nature and in what connection do they appear ? Is the 
character of the work such that the omission of green and blue 
has really the weight that Geiger lays upon it ? It will be seen, 
I think, if we ask how often the Rig Veda writers took the trouble 

of the green silk " von Zazamanc," as opposed to that of Arabia which was 
"wiz also der sne,*' whereas the other is *'grtten alsam der kl^." In all these 
cases green is used merely in this one set expression. 

The adjective blue [bl4o] does not occur although the heavens are frequently 
mentioned, journeys are taken on the water, etc., etc. In 894 the verb zer- 
bliuwen which occurs in ptc. form zerblouen, has no reference to color, even if 
we admit the identity of " blue " and " blow." 

Neither violet (purple) nor brown are mentioned. Black occurs a few times : 
pfelle (cf. 952) swarz alsam ein kol (365) ; clothes are said to be von rabenswar- 
zer varwe (402), as is sabel (zobel, 1826). Harnaschvar is the black color of 
grime. The often mentioned moere (75, 570, etc.) are said to mean originally 
black horses (cf. Bartsch, note to 75) but in Nib. no trace of color is associated 
with the word. Mixed in color is grey, gri, ** ze kleidern gra unde bant " (59) ; 
so grfs (1734) especially of grey hair, mit einer grisen varwe (gemischet war sin 
h&r), and grls means simply grey with age, den alt gifsen man (497) ; so grise 
(old men). is opposed to "die tumben (young) in 1798 (cf. 768 die tumben ant 
die wisen). Variegated is expressed by bunt (59 and oft) or by gemilet (1294). 
Color itself in Nib. is regarded as increasing, merte sich ir varwe (561) ; lighting 
up, erzunde sich (292); paling away, erblUete ir liehte varwe (240) ; or as simply 
glancing, ir varwe . . . diu lilhte ir dz dem golde (1351). It may be bad.missevar 
(= pale, of men, 1590, and of ringe, 2218). In 1702 color seems to be con- 
sidered as a covering (varwe, cf. Sk. varna, var) : ein hulft von liehtem pfelle 
ob slner varwe lac , the covering lay over the color (cf. 439). " Falsified color " 
is rouge (gevelschet frouwen varwe, i6s4). Both schlnen and liuhten are used 
abjiolutely, or with gegcn (wider) to express contrast of color. (282) Idhte vil 
edel stein, (1761) garments, (cf. 1663, 570, and erlCihte, 806); as adverb Idter- 
lich is used (283). Schlnen and liuhte are united in 434, Kihte mit schlne (cf. 
647, licliten schildes schln, and 399, 2348, etc.). We find also the substantive 
mvaniuK of thexe verbs expressed by schln or glanz (passim). 

IVi iilirtr v*^!)** of gUnce aie blicken (of the sword-stroke, 2077), stieben (the 
\\\i\\\% (nil K«> fMpitUy on the hero daz er stieben began, 2277 cf. 2278), so loogen 
(4S7. A'^'^td, (»it\, cf. 1 61 a). In 1620 we have prehen used with sdilnen of the 
miutit, 4i)ut in 1 841 loh^n of breastplate (with glanz). As an active verb appears 
beliahiisn In I7t\a (<Wr tac , . « l>el\lhte den schilt). 



WORDS FOR COLOR IN THE RIG VEDA. 1 87 

to speak of nature's beauties in grass and wood and meadow (t. e. 
where green would be used), that these objects are rarely referred 
to except in a most unpoetically practical manner, where not only 
green but almost every other epithet that enhances the idea is gen- 
erally omitted. I consider this view as worthy of notice, for Geiger's 
remark in regard to the " green fields " is frequently quoted, and 
his general essay is I believe all that has been written on the sub- 
ject from a Veda standpoint A few statistics may, however, ser\'e 
to show how comparatively litde ornate mention is made of these 
natural products anyway, and, when alluded to, how scanty a 
description of any kind is added. 

In the first place grass (trna) is mentioned only five times in the 
Rig Veda, and in not one of these cases is any epithet at all applied 
to it Not only are the poets silent in regard to its color,' but in 
the same way they are silent in regard to its growth, general 
appearance, luxuriance, etc. It is always in such pictureless 
expressions as " eat grass,*' '* bum grass," " provide grass," " lie 
on the grass," " bring grass and water," the bare grass without 
epithet So the plants and vegetables (virudh and osadhl) are, 
although occasionally provided with epithets, almost always regarded 
from a practical, utilitarian standpoint They are full of sap (pay- 
asvat), sweet (madhumat), strengthem'ng (dijasvat), various in kind 
(vijvaropa, this may mean many-colored), they have a hundred 
different appearances (9atavicaksana). Only once is the aesthetic 
side touched upon, for they are supippala, provided with pretty (su) 
fruits, and even this is explained by B. R. as having good fruits, i. e, 
useful. Babhru (see above) is used substantively to denote certain 
plants, ^ada (a word of doubtful meaning, B. R. " grass," Grass- 
mann ** somagefass ") occurs only once and without epithet 
Finally, yava, com, grain (without descriptive epithet) gives us the 
ward yavasa (n.), the field of grain or cora-meadow, and may be 
translated simply " the meadow." This word is used upwards of 
twenty times, but always alone by itself, and confined to use in 
such expressions as '' rejoice as cows in the meadow/' " grow fat 
in the meadow," "a wild beast in the meadow," "feed in the 
meadow," " retum from the meadow," " like rain on the meadow." 
This is all, we are not told that the meadows are beautifiil, or sunny, 
or shady, or pleasant, or soft, or wide, or sloping — surely then if 
we are not told that they are green it need not surprise us. Another 
word for field in general is k^tra (the later ked&ra is not found 
in Rig Veda), and here we find certain epithets attached : V 62, 7, 



l88 AMERICAN JOURNAL OF PHILOLOGY. 

a field is wished for " fruitful and good ** (tilvila, bhadra), again it 
is called pleasant (ranva) X 33, 6: or distant farana) VI 61, 14: 
but undesirable is the fruidess field (agavyuti) VI 47, 20. (II 31, 
15, mahi XiscXx^m purugcandramy wide glancing, has reference not 
to earthly but to heavenly fields.) In all these we see that use 
and not beauty is in the singer's thought The earth itself (bhami, 
ksam, gma prthivl, etc) is called the immeasurable, the extended, 
the strength-giver, die wide, the great, the firm, etc, the four- 
cornered (caturbhrsfi), and once as we have seen (cf. §3, prthi- 
vlm harivarpasam) a doubtful color word is given it. For foliage 
we have parna (n.), literally feathers (later pattra and dala do not 
occur) used once X 68, 10 "as trees robbed of their foliage " (once 
also as special tree). Often as the trees, vana, vanaspati, vanin, 
vrksa, (tarn and druma do not occur) are alluded to, we find no 
variety of poetical description, though vrksake^a occurs as epithet 
of mountains (girayo vfksake^fih), but once as soma plant vrksa is 
termed red (aruna), and supal§9a, well covered with foliage, also 
occurs. The boughs (vay& and 9akha) are spoken of as branching 
(dividing) and ripe, but not otherwise. They are noticed but sixteen 
times in the thousand hymns (and three of these allusions are in 
metaphor.) 

We must then, I think, admit that there is another cause than 
that given by C^eiger and Weise for the lack of certain color words. 
It is because those objects in nature such as woods and fields are 
not brought in to be embellished — are not meant to be described. 
It is merely that they may embellish an idea that they are intro- 
duced at all ; they are subordinate objects to the greater purpose 
of the poet. For we are not reading rhapsodies over nature when 
we peruse the Rig Veda. The aim of the work is different — to 
praise the gods and increase in worldly goods, or rather, to praise 
the gods in order thereby to increase in wealth is the one aim that 
inspires the overwhelming majority of the hymns. Earthly objects 
are therefore chiefly introduced to point a song, to specify what 
good the singer demands in return for his hymn. Thus it happens 
that the objects of earth, com, trees, fields, are, so to speak, merely 
hurried into the song and then drop out of it, while the poet pro- 
ceeds to glorify the deity by extolling his might and beauty. 
And thus we come to the explanation of the second fact — the blue 
fidls amid the descriptions of heaven. But it is not quite exact to 
say the description of heaven. The vault of heaven (divos) nakah-, 
n§kasya pr^tham, is spoken of only about once in every forty 



WORDS FOR COLOR IN THE RIG VEDA, 1 89 

hymns, and then accompanied by no color word at all. Heaven's 
vault is not called blue, but it is also not called yellow or red, the 
only epithet given it that suggests color is in citra90cis, clear- 
shining, and agrbhrta90cis, of inconceivable brig htness ; beside these 
the only other epithets of any sort applied to it are wide (X 1 13, 4), 
lofty, mighty (VII 86, i, and VII 99, 2). This firmament is not 
addressed as a deity, it is simply a locality, and is therefore merely 
introduced as an incident ; he mounted in the vault of heaven, he 
stands therein, he upholds it, he adorned it with stars (I 68, 5) (cf. 
X 68 11), etc. The cosmology of the Hindus placed between this 
vault and the earth the real often-named three-fold sky — the div, 
dyauh. This heaven is, however, in itself color, " the glancing "; 
it is light But the firmament (n&ka) is not dark though it lies 
beyond the region of light. According to the development theory 
the blue vault should be dark, but in I 19, 6 we find nakasya rocane. 
We cannot, however, regard this as a mere confusion of firmament 
and realm of light, for these are carefully distinguished (cf. Zimmer 
Altindisch. Lebens. 358). One has to dimb above dyauh to reach 
the divo naka (A. V. 4, 14, 3) (cf. R. V. VI 8, 2, I 34, 8) which is 
supported like a pillar (IV 13, 5) and has a (rounded) back, nak- 
asya pretham (I 125, 5, cf. Ill 2, 12) or nakasya sanu (VIII 103, 2). 
Dyauh, svar, vyoman, rajas (antariksam), give the theatre for the 
color-display of the clouds and storms. If we understand the 
meteorological notions of the Veda we shall not be able to say that 
the blue heaven is often described but never called blue. The 
blue heaven, 1. e, firmament, is not " often described'' it is scarcely 
described at all, and stands above the realm of which the Vedic 
poets give us their glowing accounts, outside the stage on which 
the wonders of heaven are enacted. That which is described is 
the lower heaven, not the blue firmament — where, had we the same 
views in regard to the heavens, we also should find no occasion to 
speak of blue. The mass of color description, as of all other des- 
cription, falls where the whole interest of the poets lies, upon the 
active powers of the atmosphere and the lower sky. These were 
the subject of their hymns, and to describe these powers was to 
glorify and to worship them. It was not the blue firmament with 
which the Vedic singers were impressed, their deities do not live 
there, and therefore, as the whole object of the hymns is bound up 
in the gods, it is easy to see why this firmament was so seldom 
alluded to as compared to the innumerable descriptions of the 
(lower, not blue but shining) atmospheric sky. Here is the real 



igo AMERICAN JOURNAL OF PHILOLOGY. 

abode of their divinities. It is, therefore, not strange that, being 
so seldom alluded to, naka has no real epithet of color given it, 
while we find the colors which appertain to the lower heaven, the 
home of the gods, are often employed by the poets. This lower 
heaven is not only the home of the gods, it is a god itself, dyos 
pita, whereas the naka, firmament which is blue, is not conceived 
as divinity. The gods whose praises the poet sings are red and 
yellow, white or dark, the colors of sunrise and lightning, of the 
morning clouds and the gleam of the sun. The Vedic singer 
occupies himself with the foreground, he was too busy with the 
prominent features and characters of the scene to care much for 
the background, of his picture. Beautiful things per se he did not 
celebrate, for beauty, was an incident of his song, not the theme. 
So, too, on earth, beauty of color did not often appeal to the 
unaesthetic mind of the Vedic poet And then, for instance, 
we find when the lotus is mentioned, it is more as something worthy 
of notice for its beauty of smell than for its beautiful color, and 
amid all the luxuriant vegetation of India we have only some half 
dozen plants mentioned by name (v. Zimmer, Altind. Leb. s. 71). 
The singer gave them no adjectives, for only to mention them was, 
so to speak, an adjective to his train of thought, and hence all his 
adjectives of description are meagre, except where they apply 
to the persons of the gods, or douds, or the particular object 
longed for by the poet* 

There is a second point that is involved in the quotations given 
above from Geiger and Weise. It is in regard to the lack of clear- 
ness, the inexactness of many Vedic color terms. From this, how- 
ever, we cannot draw the conclusion that the sense of color was 
inexact, for we could scarcely affirm this if it were proved to exist 
in the language used. There is, however, no literature where color 
words are applied to objects which are so constandy indefinite in 
point of color. The gods, the natural forces of the atmosphere, 
the clouds, the steeds of the gods, horses, wolves, cows — these are 
the objects of which color is chiefly predicated. The application 
of color words is made in most cases where we ourselves would be 

' Even as late as the second century B. C. we find an anal3r8is of light that 
gives only three colors and excludes green : Anugita, Mbh. XIII 50, 46 (cf. XII 
184, 35) jyotisagca guno rupam, rupam ca bahudhi smrtam (the quality of light 
is color which is manifold), and these colors are given as (black, white) red, 
yellow and blue. Yet in the Rig Veda (X 55. 3) light is declared to consist of 
thirty-four different parts: catu9trim9ita purudha vicaste sarupena jyotisa 
vivratena. 



WORDS FOR COLOR IN THE RIG VEDA, 19I 

at a loss to say what exact term might best describe the object. If 
we found that a word which means yellow alone b applied to some- 
thing which is and can be only red or green, then we might admit 
a confusion of epithets and of idea. But such cases do not occur. 
It b, again, impossible to say at what period the idea of the root 
that underlies the color word has passed into abeyance and left a 
fixed color term. Green and yellow may both at the same time 
be denoted by the bum-color hari with the subsidiary notion of 
fiery or shining. The conception is not indefinite because the 
appearance is described by a general term. " Blue," to our minds, 
embraces many colors, but the one name covers all shades. We 
cannot say that yellow preceded green in the Teutonic languages 
though gelu and gruoni come together from this same root ghar. 
Nor can we say at what time the " glare " (ghar) in gelu ceased to 
be prominent 

We conclude, then, by affirming : ist. Non-mention of the colors 
green and blue is not proved for the Rig Veda literature ; 2d, That 
the sky is not called blue nor the fields green rests on reasons which 
have nothing to do with the development of the retina ; 3d, We 
cannot admit that either color words or color perception of those 
who composed the Rig Veda were inexact or imperfect, for the 
cause of the apparently inexact employment of words lies in the 
variable and uncertain color of the objects to which the color terms 
are applied. 

The theory of the development of the color sense rests, from a 
literary point of view, in great part on negative data. From the 
standpoint of physiology it has no support. Lubbock has shown 
that savages have perfect sense of color, Wallace has affirmed that 
non-mention of color is in general no proof that it was not appre- 
ciated. We have endeavored to show that this is true at all times 
and to explain the reasons for this fact in the Rig Veda. 

If die Vedic literature fail to support the theory, one of the 
strongest of these negative proofs is withdrawn, and even the 
absence of certain colors in Homer may be deemed perhaps of less 
significance than has been claimed when we consider that the 
Nibelungenlied exhibits, twenty centuries later, the same absence 
of corresponding colors, and a like ratio in the greater use of terms 
denoting red and yellow. 

Edward W. Hopkins. 



IV.— THE HARBORS OF ANCIENT ATHENS. 

I. 

It is not necessary to enumerate here the various positions which 
have been assigned, during the last fifty years, to the different 
harbors of ancient Athens. It is sufficient to say that, in the early 
part of the century, the easternmost natural haven (Phanari) of the 
Peiraic peninsula was identified as the port of Phaleron. Later, 
the investigations of several distinguished German scholars, espe- 
cially of Ulrichs and Curtius, led them to place Phaleron at the 
eastern extremity of the bay of the same name, near the spot now 
known as Haghios Georgios. This theory is now generally 
accepted ; and the port of Phanari is known as Mounychia, and 
the southeastern harbor of the peninsula (Pasha- Limani)* as Zea. 
There has been no' dispute about the identity of the main harbor 
of the Peiraieus, which has now resumed its classic name ; but the 
subdivision of this harbor, attempted in accordance with ancient 
texts, is a matter of much uncertainty. 

II. — Phaleron. 

'* In the maritime towns of antiquity, the seaport was fi*equendy 
separate fi'om the city proper, and at some distance from it In 
early times there were very few artificial harbors, surrounded by 
quays, divided into basins, and protected by jetties, breakwaters, 
and fortifications, as in many modem seaports^ • . • The ancients 
chose as a rule, for their ports, a small natural gulf or inlet, sheltered 
fi'om the fury of the open sea, and provided with a gently inclined 
beach, upon which their vessels could be drawn up.'' * An exam- 
ination of the conformation of the Athenian coast renders it doubtfiil 
whether these conditions are fiilfilled in the site ascribed to Phaleron 
at Haghios Georgios. This site is described as follows by M. 
£mile Burnouf, ex-Director of the French School at Athens : " It 
would be impossible to establish a harbor near TpeU Uvpyoi, except 

* Stratiotiki— (Leake). 

' Charles Lentheric — La Provence Maritime Ancienne et Modeme. Paris, 
1880, p. 209. 



THE HARBORS OF ANCIENT ATHENS. 193 

by the construction of breakwaters of great extent ; and even such 
breakwaters would afibrd incomplete protection against winds from 
the west and south. There remains no vestige of a breakwater, 
or of engineering works of any kind ; while the cape at the extremity 
of the bay would afford but scant shelter to a single fishing boat"* 
In the harbor of Phanari, on the other hand, at the western end 
of the Phaleric bay, we have a beautiful litde natural basin, almost 
circular, and about one-fifth of a mile in diameter. This basin has 
a single narrow entrance, contracted still further by ancient Hellenic 
breakwaters, which remain almost perfect The harbor is sheltered 
on three sides from the wind, and it possesses the sandy beach 
which was sought by the ancients for their ports. At the water*s 
edge are remains of numerous shipways and houses, both cut in 
the rock and constructed of blocks of hewn stone. Even taking 
into consideration that, before the Persian war, the naval power of 
Athens was comparatively inconsiderable, and that the ships were 
small and drawn easily up on the shore, it would seem reasonable 
that so excellent a natural harbor should be chosen in preference 
to the open coast near Haghios Georgios, exposed to storms and 
difficult to defend against a hostile surprise. It must be conceded 
that Haghios Georgios is considerably nearer Athens than Phanari ; 
but we shall see below that the distance of the latter place from 
the dty accords better than that of the former with the length of 
the Phaleric Long Wall as given by Thucydides.* 

III. 

I will not repeat the arguments of Ulrichs and Curtius in favor 
of the identification of Haghios Georgios with the ancient Phaleron. 
These arguments are reviewed and summed up very clearly in 
Curt von Wachsmuth*s Die Stadt Athen im Alterthum,* a work 
of much erudition, in which is brought together a mass of ancient 
information with reference to each question discussed. I will now 
consider some points in the scanty ancient testimony that remains 
to us regarding the topography of the Athenian seaports, which 
seem to throw doubt upon the solution generally accepted. 

The Long Walls to Phaleron and the Peiraieus were begun in 459 
B. C* If Phaleron was at Haghios Georgios, nearly two miles of 

» La Ville et TAcropole d'Athines. Paris, 1877. p. 136. 

• Book II, chapt 13, * Leipzig, 1874, p. 306 et seq, 

*Thucydidcs, 1 107. 



194 AMERICAN JOURNAL OF PHILOLOGY. 

shore along a plain, in some places indeed marshy, but in general 
smooth and accessible, lay exposed to a hostile attack from the sea 
between the Long Walls/ of which the usefulness was thus seriously 
impaired. It is true that Athens had, at the time of the construc- 
tion of the Long Walls, almost reached the zenith of her power both 
by sea and by land. However, it would not be safe to assume 
that she could consider herself secure against even a raid from the 
sea. An old rival and bitter enemy — Aigina, the ** eyesore of the 
Peiraieus" — ^lay only a few miles distant across the Saronic gulf, 
her temple of Athena in plain sight from Athens three miles inland. 
Aigina was indeed much crippled, but she still retained some 
semblance of independence.* It was not until 455,' four years after 
the Long Walls were begun, that she was forced to surrender her 
last ships to Athens. Many of the allies of Athens had considerable 
naval power until long after thb. It was thirty years later that 
Lesbos revolted and was crushed ; and the presiding city of the 
confederacy had before her the example of the revolt of Thasos,* 
to warn her against over-confidence in the fidelity of her allies. 
Yet the Outer and the Phaleric Long Walls were begun some eigh- 
teen years after the Peiraieus had become her principal seaport ; 
and there was therefore no urgent necessity for seeking to assure 
the connection between the metropolis and Phaleron ; while the 
attempt to do so in the way that Thucydides tells us it was done, 
always granting that Phaleron was at Haghios Georgios, would 
have introduced an obvious element of weakness into the whole 
system of fortification. 

The following is the main passage of Thucydides which bears 
upon the defences of Athens and her ports at the beginning of the 
Peloponnesian war : ** The length of the Phaleric Long Wall was 
thirty-five stadia, to the fortifications of the city. The circuit of 
that portion of the fortifications of the city which was kept under 
guard was forty-three stadia, in addition to the portion left 
unguarded, between the [outer] Long Wall and the Phaleric WalL 
The length of the Long Walls to the Peiraieus was forty stadia ; and 

> Cf. Wachsmuth— Die Stmdt Athen im Altertham, p. 55S. 

' Cf. G. von Alien, in the Erl&uternder Text of Curtias and Kaupert*s Karten 
von Attika, Berlin, 1881. Heft I, p. 10, ** Die N&he des feindlichen Aegina, 
von welchem man jeder Stunde eines Uebcrfalls gewirtig sein konnte, allein 
machte einc seiche Sicherang [the fortification of the seaports] n6thig.** 

'Geoige W. Cox — The Athenian Empire (Epoch series). London, 1876, 
p. 31. •465-463 B, C. 



THE HARBORS OF ANCIENT ATHENS, 195 

of these the outer one was guarded. The whole circuit of the 
Peiraieus, including Mounychia, was sixty stadia, of which the half 
was guarded." ' 

Even if we allow that the entire land side of the Peiraic peninsula, 
including the circuit of the promontory of Eetioneia, was held 
under guard without reference to the Long Walls to Athens — a 
condition which is highly improbable — we must fill out from the 
sea-walls of the peninsula a large part of Thucydides' thirty stadia. 
The inference is easy, that at the beginning of the war, although 
an attack from the sea may not have been much dreaded, still it 
was thought necessary to take proper precautions.' Yet, according 
to the accepted theory concerning the harbors, we must believe 
that a long stretch of sandy beach was left unprotected between 
the Peiraic peninsula and Phaleron. We know that the middle 
Long Wall was not guarded, and that there was a portion of the 
dty wall, " between the Long Wall and the Phaleric Wall,** which 
was not occupied by the garrison. We must imagine, therefore, 
about three square miles of land, in great part fertile, of which the 
value to Athens would have been inestimable, during the Pelopon- 
nesian invasions, exposed to a bold nocturnal raid at the hands of 
such enemies as the Lacedaemonians. Worse than this, the middle 
Long Wall might have been seized, or even an entrance to the 
city have been gained by surprise over the undefended section of 
the fortifications. 

An argument perhaps still more forcible against the existence of 
this great intervening space between Phaleron and the Peiraieus 
is found in Thucydides* description of the crowded state of the 
city at (he time of the first Peloponnesian invasion. Thucydides' 
words are as follows : *' When the country people arrived in Athens, 
some few of them found lodgings in the houses of friends or 
relatives; but the great majority established themselves in the 
open spaces of the city, and in all the sacred enclosures of gods 
and heroes, except the Akropolis and the Eleusinion, and some 
other places which were kept resolutely closed.* Even the spot 
beneath the Akropolis, called the Pelasgikon, was thus occupied, 
in spite of curses which had been proclaimed against its setdement, 

> Thucydides, II 13. 7- 

'Later, the Athenians became more careless in their watch toward the sea, 
as we know by the amusing incident of the planned Spartan attack upon the 
Peiraieus, described by Thucydides, Book II, 93. 

* ILai el n AAAo pe$aluc kkyorht ^v. 



•196 AMERICAN JO UR^r A L OF PHILOLOGY. 

and of the words of the Pythic oracle, ' It is better that the Pelas- 
gikon should remain fallow.' I think, for my part, that this oracle 
meant the opposite of its popular interpretation, and that it was 
not on account of the impiety of inhabiting the Pela^gikon that 
disasters befell the city, but on account of the war that it became 
necessary to occupy the Pelasgikon. The oracle was doubdess 
rendered with knowledge that this place would never be given 
over to dwellings in time of prosperity, although it does not state 
this plainly. Many of the newcomers constructed quarters for 
themselves in the towers of the city walb, and wherever else any one 
was able to find accommodation ; for there was not room enough 
in the city for so large a number as were crowded into it Finally, 
they took possession of [the space between] the Long Walls, and 
of the greater part of the Peiraieus." * 

If three square miles of ground had been available, between the 
Long Walls and the Phaleric Wall, it would hardly have been 
possible for the want of room to be so pressing. That this space 
could not have been left unoccupied for fear of attack is shown by 
the &ct already often alluded to, that the middle Long Wall and 
a certain portion of the city wall were left unguarded. The Phaleric 
deme, as Strabo tells us, began at the boundary of the Peiraic, and 
extended along the adjacent shore.' Yet no mention is found of 
the occupation by the refugees of the territory of this deme, which 
would have been, in great part, within the walls. 

Another argument against the identification of H^hios Georgios 
with Phaleron is furnished by the very nearness of this point to 
Athens. The intervening distance is only about thirty Attic stadia ;' 
while that to the city firom the little promontory on the northern 
side of Phanari agrees much more closely with the length of thirty- 
five stadia assigned by Thucydides to the Phaleric WalL To 
explain away this and other difficulties in the measurements given 
by Thucydides, Curtius supposes that the historian used a stadion 
measure smaller than the usual Attic ; and other scholars suppose 
inexactitude on the part of Thucydides, or excessive windings of 
the walls. The latter supposition is very unlikely in the case of 
fortifications of the nature of the Long Walls, upon such ground 

> Thuc)^iiae$, II 17; cf. II 51, 

'^timbo» 39S. 21 : Mcrti 6i rdr W^tipeuk ^a>Jip(i^ 6^/iO( h r^ efef^f wapaXl^ . . . 
^ Wachsmuth, p. 330. 



THE HARBORS OF ANCIENT ATHENS. 197 

as this part of the Attic plain, and the first two seem disproved by 
independent evidence.' 

IV. — MOUNYCHIA. 

Mounychia, the Akropolis' of the Peiraieus, is identified by the 
German scholars with the steep hill above the harbor of Phanari. 
There seem to be weighty reasons for doubting the correctness of 
this assumption. The smaller peninsula, which forms the southern 
extremity of the Peiraic peninsula, is, there can be no doubt, the 
'Aicn7 of the ancients ;* famed for its quarries of building stone, 
abundant remains of which still exist. In Herodotos, VIII 77, we 
have preserved the following words of an oracle :* " When they 
shall make a bridge with their ships between the sacred headland 
(oicnyv) of Artemis of the golden sword and sea-girt Kynosoura, 
etc.** But Pausanias tells us that " the Athenians have still another 
harbor, that at Mounychia [where there is] a temple of the Mouny- 
chian Artemis.*** As no other Artemis is mentioned in connection 
with the seaports, except the Thracian Bendis, whose sanctuary 
was in the neighborhood of that of Artemis Mounychia,* this is 
enough to establish a presumption that 'Axt^ and Mounychia were 
merely different names for the same locality. This presumption 
is strengthened by Herodotos* account of the disposition of the 
Persian fleet before the battle of Salamis: "Those who were 
stationed near Keos and Kynosoura brought up their ships and 

' See Wachsmuth. pp. 330 and 334, etc., for this evidence. Milchhoefer says, 
in the explanatory text of the Karten von Attika» 1881, Heft I, p. 34, §6, that 
the Phaleric bay extended probably, in ancient times, much further inland 
towards the city; and that even now it is impossible to walk dryshod in a 
straight line from Athens to the site at Tpe?f Wipyot (Haghios Georgios). The 
sea at the eastern side of the bay is shallow and even obstructed by reefs, so as 
to be ill-fitted for navigation. Towards the western side of the bay, remains 
of ancient houses exist ; these must have been in the deme of Phaleron. 
Milchhoefer {Joe, cit,) seems inclined to the opinion that the port of Phaleron 
occupied a position now wholly inland, upon the supposed ancient inland 
extremity of the Phaleric bay, and not far distant from the southern Long 
Wall. It is probable, however, that this inlet was already, in the earliest 
historic times, extremely shallow. 

•Wachsmuth, p. 307. 

•See Wachsmuth, p. 316 et seq,y for proof of this. 

^Wachsmuth, p. 317, and note 6. ' Pausanias, I i, 4. 

*Xenophon — Hellenica, II 4, ii. Cf. Plato^IIoA/rcta, a', I. 



198 AMERICAN JOURNAL OF PHILOLOGY. 

blockaded the whole strait as far as Mounychia. This movement 
was made in order to cut off the retreat of the Hellenes, . . ." ' 

A glance at the map shows that it is unlikely that the blockading 
line was extended further than the extremity of the Peiraic penin- 
sula. If the line of ships had been carried beyond 'Axr^ to the 
Phanari harbor, a large number of shij^ would have been in such 
a position as to be unable to render any service — the whole Paraic 
headland being necessarily between these ships and the scene of 
batde. 

Under the word Mowv^'W we find in Phodos the followii^ 

explanation : 'H/wi^ td>oi noBitpimiarTQS aiirifv (Mmarvxlof 'Apn'fuSa) iwi 

Tf Tov Ilci/xuar oKpur^pif.' Wachsmuth quotes this sentence as 
evidence that Mounychia was the Akropolls of the Paraleus. The 
word axparnipiif descHbes excellently the peninsula of 'a«t4, which, 
too, was peculiarly fitted to be the Akropolis ; not only by nature, 
since it is connected with the main peninsula merely by a narrow 
isthmus, and since it commands completely the entrances both to 
the main Peiraieus harbor and to the harbor of Pasha-Limani, but 
also by art :' for considerable remains of its ancient fortificadons 
survive. The hill above Phanari, called Mounychia by the Germans, 
is higher and steeper ; but before the invention of gunpowder, 'Act^ 
wa& plainly a more advantageous site for the Akropolis. Strabo's 
description of Mounychia runs as follows : Atf^ ff Jirriy 7 Mmiwvxia, 
X^pponiinaitMi koI KtHkot ecu umtw/ur* iroXv lUpot lf>v<ru n ml tirirtfdti iar' 
alK^rtit iixtaOm, trro/uf At pitp^* ■np' titroiw Ixtn- ' vicaKarrovtn S ovr^ 
Xi^'mc Tpttt. Ti fMV otr ffoAai^ iTKnlxurra ul iruvf uirro ^ Mourv^ia vapa- 
trXi^mW imrtp ^ rvv Podi'w' K6Ktt, trpomiXij^uui rf ir(pi|3iJXai nSr n tltipiua 
tal Toiir Xi^'tnc v\^ptit vtupiur, tr iXi nai 11 6n\o$iiai, tAttrtn Ipyon ' S^tir 
Tt ir vavtrraBiMii rate rtrpaiaiviaii rotioiv, tnr oliK AoTTOin ?OT«XXof 'aAjkiun. 

' Heiodotos, VIII 76. • Wachsmiiih, p. 307, note 6. 

""' *'■- use of the word aKpuri/P'ov with reference to this very 'Ajct^, or to a 
a Pin taruh. Them istokles, fr^. i, of Mailer: Frg. Hi»t. Grace. 11, p. 
ihsniDth, p. 330. note 4.) 

:h»muUi, p. 315, note 4. Diodoros. XX 45, nnd XIV 33. 
rominent tcholars consider that the expression uniJj/c mi vtrdvofiot 
b peculiar aptnesi to the hill nearest the inainland, on account of 
able passage hewn from the rock in very ancient times, and COD- 
ight of steps which descends to a greal depth in the sonthweat 
e hill. This explanation seems, however, rather far-fetched. This 
id passage has not yet been Ihoronghly investigated. 
fi luxpii applies vei; well (o the narrow peninsula by which 'Aiit^ ii 
le rest of the Peiraic peninsula. 



THE HARBORS OF ANCIENT ATHENS. I99 

T^ ^ Tfi'xci rovr^ avvrfirrt rk Ka0(iXicva'fi€tHi €k tov Aareos o-kAi; * ravra d* 

Ijv /uucpa T€ixr}, rrrrcipdKOvra aradioiv t6 fJ^KOs, frvvarrrovra rA 5cm; r^ IIci- 

1 
paui ... 

The word x^PP^^i^^C^^ — forming a peninsula — seems to adapt 
itself admirably to 'AjcnJ, while it cannot without a stretch of meaning 
be applied to the hiU above Phanari. 'a^tiJ, again, and the neck of 
land by which it is connected with the rest of the peninsula, are 
much better " adapted for dwellings," and for the wide streets and 
symmetrical plan of Hippodamos — resembling those of Rhodes in 
beauty ' — than the steep, rough slopes of the Phanari hill. The 
rest of the description appears to suit equally well either site. 

Wachsmuth mentions* the remains of a Doric temple found upon 
the shore of the Pasha-Limani by Colonel Leake, and says that 
** Leake attributed these ruins incorrectly to the temple of Artemis 
Mounychia." He gives, however, no reason why they should not 
belong to the temple in question as well as to any other. Again, 
Wachsmuth thinks* that only one theatre can have existed, in 
ancient times, in the seaport city. As considerable remains of a 
theatre survive upon the northwestern slope of the Phanari hill, 
and as Thucydides mentions t6 Trp6s rj Movwxla Aiowa-uiKhy BiarpoVf* 
he argues that the Phanari hill must be Mounychia. This argument 
is upset by the discovery in 1880 of another theatre at the north- 
eastern extremity of 'Aim},* close to the bay of Zea (Pasha-Limani), 
which it overlooks. If, therefore, Pasha-Limani is the ancient 
haven of Mounychia, we have in this new theatre t6 vp6s rg Movwxl^ 

Bicerpov* 

' Strabo. IX 395, 15, Ed. Didot, 1853. ?• 339- 'Wachsmuth, p. 319. 

•Wachsmuth, p. 328. •Wachsmuth, p. 320, note 3. 

•Wachsmuth, p. 320. note 2. Thucydides, VIII 93, i. Cf. Lysias, XIII 33 
and 35. 

•See Karten von Attika, mit erUuterndem Text, herausgegeben von E. 
Cnrtius und J. A. Kaupert. Berlin, 1881. Heft I, Bl. II. 

'Mr. Dragatses, in his article on Td diarpa roit Heipcuoc Kal 6 Ko>^f Tuftfyv^ 
published in the TlapvaaaSc for 1882, p. 257 et seq.^ gives satisfactory evidence 
that both theatres existed before the Peloponnesian war. He proceeds with 
an attempt to show from a study of Xenophon's account of the campaign of 
Pausanias against Thrasyboulos, that the K(*>^c Af//^ was not, as is usually 
accepted, either the inlet west of Eetioneia or the marshy bay, now in great 
part filled up, at the northern extremity of the Peiraic harbor ; but that it was 
the first of the subdivisions of the main harbor near its entrance. Even in 
connection with the usual theory of Peiraic topography, this part of M. Dragat- 
ses* essay can hardly be considered successful ; while if Thrasyboulos* head- 
quarters were on 'A«ri^, the Spartan commander's scouting expedition towards 
E^ioneia would explain itself. 



200 AMEBIC AN^ JOURNAL OF PHILOLOGY, 

V. — The Peiraieus. 

We read in Pausanias that "before Themistokljes came into 
office . . . the Peiraieus was not the port of Athens, but Phaleron, 
where the sea is [comparatively] very near the city . . . But 
when Themistokles became prominent in the government, seeing 
that the Peiraieus was better adapted to the needs of navigation 
than Phaleron, and that it had three havens while Phaleron had 
but one, he took the necessary steps to create this seaport for the 
Athenians. And down to my own time, ship-houses have existed 
there ; and the tomb of Themistokles is situated near the largest 
haven . . . 

" The Athenians have still another harbor — that at Mounychia, 
where is the temple of Artemis Mounychia — besides the port of 
Phaleron, which I have mentioned already. Near the Phaleric 
harbor stand the temples of Demeter and of Athena of Skiras, 
beyond which is that of Zeus. Here, too, are the altars of the so- 
called Unknown Gods, etc. 

"... Twenty stadia distant [from Phaleron] is the promontory 
of Kolias,* upon which the current cast up the wreckage after the 
destruction of the fleet of the Medes [at Salamis] . . . "' 

*E;(€t dc 6 Ileftpatcvr Xt/icvar ^P^^r, nayras icXctorovr ' €& ficv tariv 6 Kay- 
Bapov Xififfv KaXovfuvotf cV f t6, vtSpia i^rfKOpraf cfra [t6] *A<f>podia'iov, eira 
icvkX^ tov \ifitpof drool ircvre.' 

Zca • • • ciff r&y iv Utipaul Xifitpap,* 

Graser is of opinion* that by " the three harbors of the Peiraieus " 
are meant the three divisions of the main harbor formed by two 
projections of its shore-line. He thinks that these three havens 
were described as K\€i<rrovSf because the fortifications at the entrance 
defended at once ^ the inner subdivisions of the harbor. This 
opinion is shared by Colonel Leake and by M. Burnouf, among 
other scholars of high standing. The adjective icXeiorour could refer 
equally well to the fact that these inner harbors were protected — 
" closed " —from the violence of the sea. 

• 

* Pausanias, I I, 5. This distance corresponds very closely with that from 
Phanari to the promontory at the eastern extremity of the Phaleric bay. 

* Pausanias, I I, 2, 4, 5. 

•Frg. 4 in Mtiller*s Frg. Hist. Graec. IV, p. 450. (Wachsmuth, p. 310.) 
*Hesychios, at the word Zia. (Wachsmuth, p. 307, note 5.) For other 
authorities mentioning the three harbors of the Peiraieus, see Wachsmuth, Part 
II. pp. 306-28 /kissim. 8 Wachsmuth. p. 311. 



THE HARBORS OP ANCIENT ATHENS. 20I 

The most important point in the passage from Pausanias is that, 
after speaking of Phaleron and the Peiraieus, with its three harbors, 
he mentions Mounychia as another harbor, implying that it was 
not one of the three havens of the Peiraieus proper. This militates 
against the modern theory that the three havens in question are 
the Peiraieus, Pasha-Limani, and Phanari, and that the two last 
are the old Zea and Mounychia. We know that Mounychia was 
on the Peiraic peninsula ; if, then, its harbor was not one of the 
Xi/imr r^li avTo<l>v€W the three havens in question must have been 
subdivisions of the main harbor. 

VI. 

From all that has preceded I venture to infer that the topogra- 
phical arrangement of the chief harbors of Athens set forth last by 
M. Bumouf,* but not defended in detail by him, and agreeing in 
the main with that of Colonel Leake, is not only a possible, but 
even the probable arrangement According to this theory the 
small peninsula at the extremity of the Peiraic peninsula is Mouny- 
chia or 'Ajct;7 ; and the port beneath it to the northeast is 6 iifi Mow- 
wxi? Xi/A^v. Phanari is the ancient Phaleron, and the hill above it 
is the Akropolis of Phaleron. 

It still remains to settle the relative positions of the three bays 
of the main Peiraieus harbor — Zea, Aphrodision, and Kantharos. 
Different students have proposed in turn every arrangement of the 
names rendered possible by the existing number of bays ; but no 
one of these arrangements seems based upon conclusive evidence. 
The chief naval establishment was on the harbor of Zea; we have 
therefore some reason to identify 'as Zea the largest of the three 
interior bays — the first on the right hand sid^upon entering the 
harbor. This position, commanding the narrow entrance and 
protected itself by the Akropolis of 'AjcnJ, would have been especially 
fiivorable for the naval station ; and the opinion that it was here is 
supported by the discovery near the modem Custom House, which 
stands on the point between this bay and the nop^/xcm or commercial 
port, of the important naval inscriptions first published by Boeckh. 
In these inscriptions reference is fi*equently made to " the Arsenal " • 

> Thucydides, I 93, 3. (Wachsmuth, p. 307, note 2.) 

• £mile Burnouf— La Ville et I'Acropole d*Athines. Paris, 1877. Plate XI, 
and p. 1 36 ^ seq. 

•Sec A. N. Meletoponlos — 'Av^KcIorof 'En-fypo^. h *A&/vatc^ 1882, p. 6, for 
quotations from the inscriptions. 



AMBMICAS JOURXAL OF PHILOLOGY, 

in flKh terms as to Icaye Ihde doubt that tbe arsenal in question 
was the ^mocs Arsenal of Philon, which, as appears from the loi^ 
insoipdon relating to it found last' year near the Pasha-Limani 
(Moimychia)— as 1 belie\'e, at some distance from its original posi- 
tioo — stood ** in Zea«" I think it therefore probable that this first 
boy is tbe andent Zea, and that the great arsenal stood near it, 
pcxbaps, as suggested by Milchhoefer/ on the ridge between it and 
Moonychia, which was the second in importance of the old Athe- 
nian naval stations. It may be remarked that these positions for 
tbe naval stations and for the Akropolis would have been especially 
convenient for the transportation to the Akropolis and to the arsenal 
of the spars and rigging, etc., of which there is frequent mention 
in the naval inscriptions. 

VII.— The Long Walls.* 

I have touched already upon the question of the length of the 
Long Walls to Phaleron and to the Peiraieus. I will give one 
other passage which, with that quoted already from Thucydides,' 
establishes clearly that there werfe three Long Walls — two from 
Athens to the Peiraieus, and one to Phaleron : "Antiphon [says] 
to Nikokles that there were three walls in Atdka, as Aristophanes 
tells us in the Tpi^ijs — the Northern, the Southern, and the Pha- 
leric Walls. The wall which ran between the other two was called 
the Southern Wall ; it is mentioned by Plato, also, in his GorfriasJ^* 

In connection with this r6 deck fUvw rtixof, we meet with a difficulty. 
The Scholiast on Plato's Gorgias tells us that : dc^ lUtrav ruxoi Xcyct 

A possible explanation suggests itself fit>m the topography of 
the ground. $\ e. that the middle Long Wall was carried along the 

1 Cf. Karten von Attika« lS8i« BUtt II«, mod explanatory text, p. 48. Drums 
of Peiraic limestone and a IXkric capita) of Pentelic marble hare been foond 
upon this site. The diroensiv^ns of the drams corres^xvnd Tery well with those 
given in the new in*cnjn»on for ihvvseof the ArMrnal ; the cajMial is a few inches 
higher than it shouKl be» but jxxs^mMv the speciti cations of the contract as to 
measurements were n\^ *dherf>l to ri^KilT* — See American Joamal of Philology, 
No. II, October iSJ*. |v 517 M ^y* 

• See Wachsmuth. »\rt U, p^v 5*^%^ » Thvcrdides, 11 13, 7. 
•Harpokrates (SukIj^jV at the ww\U /.i *.*:*•• '^'r-"*r. ^Wachsmnth, p. 328, 

note a.) 

• Scholiast upon PUlv^V Gv>i^'Jkv >v >\4* llcnH. \W*achsactK p. JtS^ note 2.) 



THE HARBORS OF ANCIENT ATHENS, 203 

crest of the steep hill above Phaleron, and down to the little point 
at the northern side of the entrance to the port of Mounychia, 
forming thus a dividing wall between Phaleron and the rest of the 
Peiraic peninsula.* This course would give to the Middle Wall 
approximately the length of forty stadia assigned to it by Thucy- 
dides, while other courses suggested heretofore make it fall short 
of this measurement To arrive at definite results, and to settle, 
perhaps, this whole question of the harbors and of the Long Walls, 
it will be necessary to institute a thorough investigation on the 
spot. 

The construction of the Middle Long Wall by Perikles, although 
the Phaleric and the Peiraic Long Walls existed already so near 
together, can be accounted for as a measure of extra precaution, 
to ensure communication with the sea if one or the other of the 
ports should chance to fedl into the hands of an enemy, or one of 
the exterior walls to be carried by storm. 

The German scholars do not claim to have found any vestige of 
a Long Wall between the shore near Haghios Georgios and Athens. 
They mention only some scanty ancient remains dose to the sea.* 
These may well mark the site of an ancient settlement ; my con- 
tention is merely that, for the reasons enumerated, this settlement, 
if it ever existed, cannot have been the port of Phaleron — the earli- 
est seaport of Athens of which we have historic record. Bumouf, 
on the other hand, says : ** The line given by the German scholars 
for a Long Wall from the cape near Treis Pyrgoi to Athens is 
entirely imaginary. In the whole intervening space there exists 

no vestige or trace of such a wall." ■ 

Thomas W. Ludlow. 

' An ancient boundary monument of a public space before a gate was found 

in its original position on November 27, 1882, on the southern side of the hill 

in question, just within the exterior fortifications. I have no map sufficiently 

detailed to show its exact position ; but from the description, the monument 

may very possibly refer to a fortification wall between the Peiraieus and the 

eastern haven. The inscription, which is prior to the IVth century, is as 

follows : 

{F)POrTA (7r)po7rifX 

OAEMO£(:) 01; 6rffioa{l) 

0H0P02 ov bfio^. 

Tlapvcujadc^ Nov. 30-12, 1882, p. 862.) 

• Wachsmuth, p. 330. ' Work cited, p. 137. 



v.— THE DYING ALEXANDER OF THE UFFIZI 
GALLERY AND THE GIGANTOMACHIA 

OF PERGAMUM. 

No. 318, of the sculpturQ3 in the Uffizi collection, has long been 
known as the Dying Alexander. This name has been retained for 
want of a better, archaeologists having come to no agreement 
concerning it further than a general recognition of the truth of 
Otfried Mueller's remark (Ancient Art and its Remains, § 129, 
Note 4) : " The head of the Dying Alexander at Florence is an 
archaeological enigma," The Work represents the head of a young 
man whose beardless face is turned to the right and upward. The 
agonized tension of the eyebrow muscles and the open lips con^ 
spire with this turn of the countenance to express deep physical or 
mental suffering ; akin to th^t of the Laocoon, it is represented by 
the same means. Long and wavy hair, rising from the forehead 
and falling, manelike, down either side of the face, serves as a frame 
to this picture of pain. The work has undergone considerable 
injuries and has been subjected to much * restoration ; according to 
H. Meyer, a large portion of the hair on the back of the head 
and of the curls about the face is new, also most of the nose, 
and the breast and shoulders with part of the neck. 

As the pedestal has engraved upon it the name ALESSANDRO, 
the identification with Alexander the Great is probably as old as 
the restoration. It is not without interest to trace it to the com- 
plex misapprehension on which it rests. Plutarch, Alex. M. 4, 

writes as follows : T^v luv oZv Iblap rod aoifioTOf ol AvaiinrtUH fAakurm 
rmv drdptdpray ffAffnutHJvatPf vif> oZ ia6vov koI avrht rj^iov irXarrcfr^oi. Koi 
yap & fioXurra iroXXol r&v duM^o^v v(rr€pov kqI t&v (^uXoy oyrc/u/iovmro, r^y r 
avaraoty tov avxivoi th fl&Wfiov rfovx^j iccxXt/icvov, Koi r^p vyp&njra r»w 

dfifiarmvt diorcn^pi^Kcv dicpi^s 6 rcxWn/r. Another passage that may 
very probably have had to do with the naming of the bust is 
Plutarch de Alex. M. virtute aut fortuna II, 2 : Avdmrov dc r6 vpinw 

*AXff(aydpov irXdaayros &v<a ffkhfovra r^p vpoirmrip Trp6s t6v ovpav6vf &tnr€p 

Tii ovK arriBavws * 

avdaaovvri d* Hoikw 6 xaXxtos cZr Ata Xtwrciau * 
ycof vn ipjoi rlBtpaif Z€V av ^ "OXvpfirov ?;(€. 



206 AMERICAN JOURNAL OF PHILOLOGY. 

has entwined itself about his limbs and rendered him helpless ; it 
was probably also biting him. His doom is sealed, as is that of 
his brethren ; Ge, rising, near by, out of the groimd, is powerless 
to save her child, for already Nike brings Athena the crown of 
victory. It has been noticed that the care taken by the artist to 
leave the magnificent torso unhidden by the folds of the serpent 
suggests the Laocoon, where the same caution is conspicuous. The 
head, thrown back in despair, reminds us of the Alexander. The 
pendant to this group was that represented on Plate III ; here also 
a god has overcome a young giant of human figure. To quote 
Conze: ''Mightiest of all the gods, Zeus, in wide wind-blown 
mantle, his body imcovered, strides in batde; his head, unhappily, 
is lost, his right hand wielded a thunderbolt, with his left he 
advances the a^^, his shield and weapon. On either side a van- 
quished giant falls ; the one to the left with the shield, his thigh 
bored through and through by the three-pronged, flaming light- 
ning-bolt, raises his right hand in supplication ; he to the right, in 
fi'ont of the god below the aegis, rests on his knee and with his 
left hand seizes his right shoulder — as if struck there, was my 
notion. But Herr Bode recognizes in this motion, in the knotted 
muscles of the right arm, in the contracted sides, a being actually 
writhing in a fit before the god's aegis. As I hear, he has the 
approval of physicians, and his explanation is one not at variance 
with the spirit of these reliefe.** The face of this &llen giant is 
broken off, but it is highly probable that in the " Dying Alexander " 
we have a copy of the head that once occupied this place. The 
giant whose skin-covered arm is outstretched above this one's 
head has been found imitated on a Roman sarcophagus, so that 
there is nothing remarkable in the supposition advanced ; for its 
substantiation it must depend on the coincidence of the required 
and given features. As this is a question to be decided by the 
eyes rather than by the understanding, I have prepared a drawing 
of the giant with the head restored, that is to say, copied in firom a 
photograph of the Alexander taken before the finds at Pergamum 
were made. It is noteworthy that I did not have to alter the angle 
of vision, inasmuch as this shows that the point of view most 
advantageous for the Alexander naturally presents the head in the 
position which is the only possible one in the relief. I have taken 
no liberties other than making a few changes in the restored por- 
tions of the hair and slightly lengthening one side of the neck 
above the giant's left shoulder. This last was necessary in order 



THE DYING ALEXANDER, 207 

to direct the giant's gaze to the aegis that, whatever we may think 
of Herr Bode's remarkable suggestion, so strongly affects him. 

As the appropriateness of the expression of the giant's recovered 
&ce is self-evident, the perfect correspondence with that of the 
match figure in the pendant group, the giant subdued by Athena, 
is the only thing that remains to be pointed out. A subde Greek 
sense of proportion would require the adversary of the greater 
god to be cast in a larger mould, and this holds good of the pair. 
Perhaps the rather too large proportions I have given to the head 
exaggerate this impression. 

It is a curious corroboration of the theory advanced that Over- 
beck (Kimstarch. Vorles., p. 137, quoted by Wieseler in Miiller's 
DAK.), seeing in the Florentine bust the expression of " a sudden, 
surprising pain," suggested an altogether analogous subject : " Cap- 
aneus, at the moment when Zeus' thunderbolt strikes him in the 
neck and is about to hurl him from the scaling-ladder." 

Alfred Emerson. 



NOTES. 

Propertius III (IV) 7, 47-50. 

While discussing Mr. Postgate*s edition of Propertius* with a 
friend who had found it helpful in his classes, the verses cited at 
the head of this note came up for consideration, and on the spur of 
the moment I suggested an interpretation, which I felt to be ven- 
turesome, but, as my view excited lively opposition, I b<^;an to 
take a deeper interest in the passage, and a few hours afterwards 
lighted on a confirmation of my theory, which, if I mistake not, has 
never been advanced before. 

Propertius III (IV) 7 is an elegy on the loss at sea of Paetus, a 
young man about town who had undertaken to mend his fortunes 

* Mr. Postgate's excellent edition of Select Elegies of Propertius interested 
me so much when it first appeared (in 1881) that I called the attention of 
some of my Latinist friends to the book, in the hope that some special student 
of Propertius might give the readers of the Journal a just appreciation of the 
labor and thought and ingenuity that Mr. Postgate has expended on his author. 
But among the many troubles of the editor of a philological journal in America, 
not the least is the difficulty of procuring reviews by those best qualified to make 
them, and as my own knowledge of Propertius did not and does not warrant 
me to sit in judgment on Mr. Postgate's special work, I have not thought it 
worth while to write a notice which should contain little more than a string of 
points in which I differ with Mr. Postgate on general principles. Such a review 
would have produced an unfavorable effect on the reader, while in point of fact 
I hold the book in high esteem and have studied it with great pleasure. Slips 
there are, such as a curious mistranslation of so familiar a passage as Xen. Anab. 
2, 3, 25 : ovx ^«fv ^J<ri>* oi 'EA?,^ef k<^p6vTi(^(}v, • he did not come ; so (we may 
conclude) the Greeks were wise*; and in the grammatical notes Mr. Postgate sees 
too much, and sometimes beclouds a very simple matter by a mass of verbiage. 
So where he explains an everyday construction like ' si patiare, levest ' (II 5, 16) 
in this way : **An * allied fact * (a * general truth ' Ifvest) is here substituted for 
the proper hjrpothetical apodosis (the particular statement 'you will be relieved*). 
See Roby, 1574 (i). * You will see the truth of the general statement that the 
woe is light, supposing you bear it.' All this on * omne in amore malum, si 
patiare, levest,' where poHare is the ideal second person, not Cynthia, but any 
loving soul. My admiration of Mr. Roby*s syntax has its limits, but I hardly 



NOTES. 209 

by engaging in mercantile pursuits. His ship went down on the 
voyage to Alexandreia. 

Tu (Pecunia) Paetum ad Pharios tendentem lintea portus 
obniis insano terque quaterque mari. 

Then we have the usual homily on land and water and the more 
or less familiar mythic parallels. The close of the poem dwells 
especially on the hardship that so young and tender a lad should 
have perished by so cruel a death. And the characteristic of the 
youth begins v. 47 : 

Non tulit hie Paetus stridorem audire procellae 

et duro teneras laedere fune manus, 
sed thyio in thalamo aut Oricia terebintho 

efTultum pluma versicolore caput. 

This is the text as Mr. Postgate would have it, though he prints 
^hunc Paetus * both in text and notes. Baehrens, v. 47, has hoc in 
anticipation of audire which is more simple, but ^hic Paetus * brings 
before us the style of the man, *this Paetus of ours,* whatever 
another Paetus might do, and if Propertius had been gifted with 
prophetic foresight he would have known that there would be a 
Paetus of a very different stamp. Notice the iteration in what 

thought it possible that he could have stated so common a case so badly, and 
on turning to his grammar I found that he provides for this class in 1546, 
although he has not been careful to separate it from the other and less common 
class of sentences of which the type is : 'si verum excutias, facies, non uxor 
amatur,* in which the real apodosis is the ascertainment of the predicate 
(reperias faciem, non uxorem amari). Then, as Mr. Postgate has taken a 
dislike to Ovid, who had genius enough for half a dozen small poets, he is not 
sjitis6ed with calling him an * inferior Cicero in verse,* but hounds him down 
as a purloiner of Propertian tidbits, and that on the slenderest grounds ; * post 
cineres/ for instance, is cited as a theft, a phrase which that *■ conscious pedant 
Persius * also twists into * cinere ulterior,* and even Minucius Felix cribs in his 
Octavius 1 1 : post mortem et cineres et favillas. Ovid cannot even use t nunc 
and wnbra in peace, though Propertius himself, as Mr. Postgate tells us, has in 
his possession a phrase which coincides remarkably with a passage quoted from 
C. Gracchus by Cic. De Orat. 2, 67, 269. That phrase is : Quid tibi vis, insane? 
But I am going to be more generous than Mr. Postgate. Macaulay, in his Ballad 
of Virginia, says *And now mine own dear little girl, there is no way but this.* 
* There is no way but this * occurs in Shakespeare totidem verbis (Twelfth Night, 
Act III, Sc. 2),' but I do not accuse Macaulay of plagiarism despite his pro- 
digious memory. However, I am determined not to lapse into -a faultfinding 
criticism of a book which is not only far superior to the run of editions, not 
only useful for classes, but is full of genuine learning and manifold suggestive- 
ness. 



2IO AMERICAN JOURNAL OF PHTLOLOGY. 

follows V. 51 kuic, V. 53 hunc with the troXurranw so characteristic 
of artificial poetry. For effuUum Baehrens retains et fuUum. 
Into the criticism of the rest I do not enter. Non iulU is <Ak frX^ 
= mm isfuit qui ferret, from which we get for the contrast sed is 
fuii qui mallet. ' This Paetus was not the man to bear the sound of 
the piping storm, but he was the man(to have)his head propped on 
feather pillow of shot colors in a chamber of thyine wood or (oO 
Orician terebinth.' This chamber the commentators have sought on 
land and sought in a real chamber. But we know that Paetus was in 
narrow circumstances (J>aufier, v. 48) and had no such luxurious 
chamber or bed as Mr. Postgate would render it. Propertiua 
simply tells us what Paetus would have preferred. But the tktda- 
mus is not a chamber on land nor yet a bed. It is a stateroom, the 
stateroom of such a ship as the Romans must have known as well 
as we know Cleopatra's barge in Shakespeare, the ship of Hieron, 
built under the direction of Archimedes and fully described by 
Athenaios, 5, p. 206. Of this ship we read 0aXa/iovc JH rpAt rfx« 

rpiicXiiwr (p. 207 C), and further : ai^pokimov Kartatiijairra TpiAim . . . 
rovt Toixovr f tJx' ""' ^'' o/x^V •K'irapirrov rdt 8* Bipat tXii^arTot ical Sue v. 

This was the kind of seagoing environment that our Paetus was fit 
for, not the rough work of the deck that the mannish Roman Udy 
jvenal delighted in (duros gaudet tractare ruderttis). . 

B. L.G. 



CONIECTURAE BaBRIANAE. 

i6, 17, Rutherford: 

W iM flfHHrif*! voTor ivmixot <7ti/3ij, 
Kol icaC/ia ffoAtru, irairnt luil mmuyain ; 

haps KBi Kavpa AiXirti navrax^ KOTOKmia, 

V 8 t Tat d Idjoi diP^Kt iiaitpa XiftAmw. 

: seems possible that tdim is a mistake for ^/Jpas, the tame goats, 
has just before mentioned the other atyat Ktpoixpv aypiat wvKi 
pvc Or abrhc (yr. 

C 1 2 : ui Jb- ffKiiroiTo riv irAai ri ySouXnut. 

Lutherford reads after Gitlbauer : 

wc if {SXnrat ri njv irAar W jSovXtuot 



NOTES, 211 

against the Babrian rules of rhythm. It would be better to retain 
/SXriroiro as a passive, and reading rov fr€kas, make the genitive depend 
on the substantival notion contained in W /3ovXevoc, ' that so might be 
seen in one's neighbor, what he was purposing ' = * one's neighbor's 
intention.' • 

LXIII 9 • KtuMP dc irdvmp Srt (rvvtanv apBpwroi^ 

donijpts ^fi€h. 

Perhaps irr Uttrnv, At any rate Srt seems impossible. 

Rutherford seems right in supplying a negative to iytwfiOriv ; but I 
would then recast the verse as follows : 

/yi> V(pva'ip6s ; cV eros ovk tyfvu^Bfjv. 

I do not believe Babrius could have admitted so faulty a rhythm 

as OVK rir cror tytviniBfjv, 

XCV 75 • ***^ ^*^ €K€i¥os irKfiov fj aif BvfMvrai. 

BvfAoiyti is an obvious correction. 

XCIX 2, 3* X*^ XcW rt Koj>Kv€i *, 

irp6t avT6y elirtv, dXX' iv^xypov dfucecr 
TCiKirtrnpa) <rov ftri luBUvai mariv. 

Rutherford writes oXX* ^n iv^x^pt^ ^awrcw and adds in his note that 
he considers this conjecture certain. I should much prefer, taking 
a feather from his own wing, to write 

aXX iv^xypov ov dttcrcif 
raKVfrripvt aov fi^ fiiOUvtu ttIotiv ; 

* But won't you give your two quill-feathers as a pledge of your 

fidelity?'* 

R. Ellis. 

* Professor Ellis has published a review of Rutherford's Babrius in the Philo- 
logische Rundschau of May 19, 18S3. Among the certain emendations he 
classes f/fiet for olfwi 34,7 (which is, indeed, perfect) ; a?^rp£i'0)v for XarpEiorv 129, 
5 (also very good); O/jpayprural^ for (pihiypevrai^ 107, 10— a large percentage of 
successes. B. L. G. 



CES. 

; and E. Koschwitz. 

m the litat ediLion in 
;aage which has often 
:o orthogiaph]'. Their 
aphy that are contrary 

language long since 
:es should corresponii 
D course without any ' 
; old language, as has 
ajice of which is thus 

that time. Of course. 
arejt lo Ihe primitive 
ilopment and historic 
itions noted, both for 
iraied periods of the 
full of interest lo the 
scholars, Gaston Paris, 

But while the details 
the middle ages down 

IS been almost wholly 
onp of languages. It 
the first deGnile and 
celebrated Girart d« 
in, nach alien erreich- 

r step forward is beset 
cDces of conscientious 
:r of the paper before 
rit of having given ns 
tempt, here and there, 
have hitherto been 
«ls himself unable to 

. the reader will find 
I dM Oiforder Textes 
a, f, », H are treated, 
ring oasal and, with 
iced by a reinvestiga- 



REVIEWS AND BOOK NOTICES. 213 

tion of the much-yexed MS-problem, in the solution of which, with reference 
to their language and relations, he sides with Profs. F6rster (Roman -Stud. V 
95 seq.) and Stengel (Jahrbuch XII 119 seq.), who maintain that parts of the 
Oxford Codex are interpolations by a later hand, in opposition to Paul Meyer, 
who draws precisely upon these parts to prove the identity of origin of two of 
the most important MSS of this epic. 

The general results obtained by the writer from his investigation of the 
assonance may be summed .up about as follows : This epic was originally 
composed in neither pure Proven9al nor pure French, but in a mixed language, 
which must be supposed for the first cast of this Roman, just as for the Croisade 
contre les Albigeois, for Daurel, Beton and Aigar. The essential characteris- 
tics of such a mixed dialect have been determined for the eastern branch of it, 
by Ascoli in his Arch. Glott. Ill, Schizzi Francoprovenzali, which extended in 
the beginning along the whole boundary line between the Langue d'oc and 
Langue d*oil territories, and is preserved to us in a number of scattered linguis- 
tic remnants outside of the above-mentioned epic productions. The disappear- 
ance of this species from the main body of the literature of that time is attributed 
to the overwhelming influence of the Limousin court language, which had 
become so popular with the Troubadors as to be characterized the drtg Limosi^ 
the development of which was wholly similar to what took place two hundred 
years later for the Isle de France. To this influence must also be ascribed the 
difference of language which exists between the MSS of these fragmentary 
epics on the one hand and those of the rhymed compositions on the other. As 
to the question whether these epics preserved to us contain one and the same 
dialect variety, or whether different dialects show themselves in this mixed 
Franco-proven^al speech, M. holds to the latter view, and supports it with 
abundant proof throughout the course of his work. Daurel and the Croisade, 
for example, belong to the western group, while Aigar and Girart are assigned 
to the eastern division of the linguistic medley. The difference between Aigar 
and Girart consists principally in the characteristic treatment of e (/, i) and o 
in the latter, which separates it sharply from the other three texts. 

Everything pertaining to this celebrated Girart Roman has been clouded in 
such mystery that, notwithstanding the large number of works published on 
the subject, little has been done to clear up the uncertainty which clings about 
its origin both in point of history and language. The present contribution has 
thrown decided light on the latter of these difficult problems, and has inciden- 
tally given us many interesting details concerning the more exact relations of 
the Provencal dialects, and especially those varieties that make up the belt of 
mixed speech lying along the border line of the Langue d*oc and Langue d*ort 
species. 

6 Heft. 

It is now about ten years since the modem school of young grammarians, 
amone whom stand Paul, Brugman and Osthoff as chief representatives to-day, 
began to make iu influence specially felt in Germany in opposition to the 
analytical processes and dissecting mania of the old advocates of descriptive 
grammar. Their fundamental doctrine, that all phonetic change takes place 
according to absolute and inviolable laws, was boldly stated and developed 



REVIEIVS AND BOOK NOTICES, 21 5 

For the strong perfects the writer sets apart a special division in his work, 
since they occupy an exceptional position, and have as their basis in Latin an 
already modified present stem. A few interesting results obtained for this 
class are worthy of note, viz. the second person singular of such modern prae- 
terita as vins (venir), tins (tenir) — in the old language ven^is^ Un^is^ respec- 
tively—have passed through the intermediate forms veins^ teins^ before reach- 
ing the present contracted stage. So, too, with the corresponding stem-vowel 
forms for the first and second persons plural. The limits of this transformation 
period, in which veniSy through veins ^ passed into vins^ are put down from 
I 450-1 5 50. 

Again, for perfects of which oi (habui) is a representative (for example, poi 
pavi, ploi placui, poi potui, soi sapui, toi tacui), four distinct conjugational types 
are traced in the dialects that throw much light on the mode of growth of the 
Modem French so-called irregular forms. For the singular, ^we have (i) oi ous 
out^ (2) oi eus ot^ (3) euch eus eut^ (4) au awis aut — the plurals of which are oumes 
omstes ourent^ eumts eusUs orent^ eumes eustes eurent^ awimes awistes aureni. No. 
I represents the Norman t3rpe ; No. 2 comes up in most non-Norman documents ; 
while Nos. 3 and 4 are dialect productions of the north and northeast. In and 
round about Amiens particularly is to be found the original home of the eu' 
diphthong species, which afterwards spread and became very abundant through- 
out the north. A forcible illustration of the passage, by analogy, from one 
grammar category to another presents itself in the sigma-perfects of such verbs 
as prendre (pris), mettre (mis), dire (dis), etc. The theory generally held with 
reference to this class is that the second person singular of the modem language 
is an analogical formation on the third singular, that is. Old French i^tx^, pres-is, 
prist = modem pris,/m, prit, by the simple syncopation of medial s in presis 
and the contraction oi preis to /m. The sibilant never falls out, however, in 
such cases in French, and hence the {otTQ.% preis, preiims^preistes would become 
nnexplainable by this hypothesis, but acccording to B.*s investigation these si- 
perfects have passed over into the class of t-perfects, that give the regular 
model vi (vidi) ve-is vit, and by analogy to it we have our present second person 
singular, and first and second persons plural. 

In the same way a number of perfects that originally belong to the sigma 
division have passed over within the literary period into the tft-class, or settled 
down altogether in the weak verb conjugation. 

The author adds two very full alphabetical registers to his work, one for all 
the verbs, the other for all nouns and adjectives treated in it. 

One only has to glance through such a contribution as this to recognize the 
great difficulty of writing to-day a general historical French grammar that shall 
in any degree represent the present status of the science. Diez troubled him- 
self very little about dialect influence in the production of grammar forms, 
being satisfied to set down the resemblances to or deviations from the Latin, 
but the grammarian of the present is expected to trace the tangled threads of 
each dialect variety that helps to form the complex texture of the modern lin- 
guistic fabric. To this end the paper before us is a most valuable auxiliary, 
and will do much, without doubt, to stimulate further research in the rich field 
of dialect effects upon the composite body of our modem grammar. 



2l6 AMERICAN JOURNAL OF PHILOLOGY, 

7 Heft. 

Old French syntax has been the subject of investigation in a long series of 
monographs that present us with the laws of word-position for a given, isolated 
period of the language and for a certain author, without in any way establish- 
ing his relation to his times, and often without even mentioning the model 
types of expression of the mother-tongue which were his constant companions. 
That these special studies were the necessary forerunners to a general system 
of syntax is evident, but that they only give us a partial and, for the most part, 
a very imperfect idea of the complex phenomena out of which they naturally 
grew up is manifest to any one who, through them, attempts to get at the philo- 
sophy of thought-expression for any particular phase of the language which 
they claim to represent. The tendencies of language are so diverse and 
depend so much upon the varying products of intelligence, developed out of 
what is gathered* from those about us, that the characteristic coloring of an 
author's phrase can only be appreciated in many cases by a study of his inner 
life as the result of all the changing influences of his epoch. It is for this 
reason that the separate treatises just mentioned often seem to give us contra- 
dictory results, in only tracing the rich and varied growth of the Old French 
sentence for the individual author ; but these discrepancies either disappear 
altogether, or are reduced to a few peculiarities of special style, when each 
literary monument is considered as a simple factor, a single link in the chain 
that binds the historic traditions of the mother-tongue to the set and rigid for- 
mula of the modem idiom. The differentiating tendencies of the Old French 
4n the earliest stages of its structural development are so numerous, the modes 
of expression often so naYve and original, that the collection of them into a 
well-rounded, systematic whole, so that they may be viewed from the stand- 
point of a more general word-relation, cannot fail. I think, to be of interest to 
many scholars who are not versed in the details of this particular branch of 
syntax. For the student of Latin, especially, does this early period of modem 
phrase-building offer a rich source for fruitful research, in that he may frequently 
6nd here the more logical, natural expression of thought — the so-called excep- 
tions to rule — highly developed, but of which he has only the meagre traces 
in the artificial constructions of the classic writers. It is, therefore, with 
pleasure that we greet any attempt to give us a general survey of word -relation, 
for however limited a period it may be, of this transition stage between the 
latest Low Latin usages and the more settled forms of the twelfth and thirteenth 
centuries. Such is the little work of fifty-six octavo pages published as the 
last number of Vol. Ill of the Franz&sische Studien — Die Wortstellung in den 
&ltesten franz5sischen Sprachdenkmalern, von Bernard V&llker — in which 
the author covers the field for all the most important monuments of the ninth, 
tenth and eleventh centuries, that is, Les Serments de Strasbourg, Cantilene de 
Sainte Eulalie, Fragment de Valenciennes, Passion du Christ, Vie de Saint 
Leger, Vie de Saint Alexis, Gormond et Isembard, and the Chanson de Roland. 

For all cases where these texts do not agree in their word arrangement, each 
yjue is treated separately, and note carefully taken as to whether a given pecu- 
lUiity iH due to assonance or some other probable cause, and whether it has 
v;<k|jktiuucd to live as a typical or exceptional form in the later language. For 
VU<^ «u^thod of investigation we find here many points taken from Morfs Wort* 



REVIEWS AND BOOK NOTICES, 2\^ 

stellung im Rolandsliede, but for the real division and distribution of his mate- 
rial the author depends on the system adopted by Prof. K5rting of MUnster in 
his Franzds. Grammatik, Leipzig, 1872. An interesting feature throughout the 
work is the precise way in which, when we have two varying constructions 
belonging to the same grammar category, the percentage of each and their 
changing relations for each individual document are stated. Thus in the 
Latin, as we know, the personal pronoun subject was indicated, as a rule, by 
the terminations of the verbal predicate, while in modem French the constant 
nse of the pronominal subject is required. Here the Old French holds a 
middle position between these two, and in its oldest period shows a strong prefer- 
ence for the Latin usage, especially in dependent clauses. For principal 
clauses the constructions nearly balance, as is seen by the results of V.'s inves- 
tigation, viz : 






Expressed, 


• Omitted, 


Principal Clauses. — Eulalie 


50 per 


cent. 


50 per cent 


Passion 


54 




46 


Alexis 


49 




51 


Gormond 


54 




46 


Subordinate Clauses. — Passion 


62 




38 


Alexis 


60 




40 


Gormond 


67 




33 



In these literary monuments of the Old French, however, the language can- 
not be considered as identical with the folkspeech, since their authors in 
many cases not only read and wrote Latin, but also thought in it, and hence, 
however exact the results may be for the texts that have come down to us, we 
may presume that the differences of construction were much less marked and 
the speech more uniform among the people. What Schlickum discovered with 
reference to the inversion of the subject for the thirteenth century French of 
Aucassin and Nicolete is here confirmed by the tenth and eleventh century 
language, that is, that no such thing exists as arbitrary inversion, but that it is 
dependent upon fixed rules, which, though not so absolute and inviolable as in 
A. and N., yet hold their sway with rigid force for certain well-defined posi- 
tions. As the narrative flows on in this early stage of the language by princi- 
pal clauses of limited compass the inverted order is much more frequently 
developed here than in the subordinate clauses, which number only about half 
as many examples as the former. No trace is yet found of that species of 
inversion so common to Modern French, where in an interrogative sentence 
the substantival subject is expressed at the beginning and then taken up again 
after the verb by a pronoun : e,g, Le doge . . . n^a-t-il plus rien & dire ? 

Marx, in his Wortstellung bei Joinville (XIII century), fails to notice a mode 
of treatment of the relative pronoun which characterizes in a forcible manner 
the Old as contradistinguished from the Modern French according to V.*s 
investigation. Where the subordinate clause is introduced by the relative, 
which in the later language has a decided predilection for post-position with 
reference to the verb, in the earlier idiom a very strong tendency is shown 
against this freedom. In fact, we only find here pre-position, except in cases 
where, for metrical reasons, another distribution of the phrase elements is con- 
ditioned. 



2l8 AMERICAN JOURNAL OF PHILOLOGY, 

Both in its mode of dealing with predicative attributes and in the relation 
of object to verb the French language up to the end of the eleventh century 
(epoch of composition of the Clermont documents, Passion and St. Leger) 
shows a most decided tendency to make them always precede the verb. Here 
we recognize Latin influence as the basis of construction, from which the lan- 
guage in the course of time cut loose, and Bnally the object receives its char- 
acteristic position after the verb — an arrangement of the phrase elements 
necessitated by the complete disappearance of flexional endings. For the first 
five texts examined the average relation stands (object -f- verb) 97: 56 (verb-f- 
object). In the Passion we find it reduces to 56 : 30, while a little later in the St. 
Leger the difference becomes still less, 30 : 20, and finally, half a century further on 
— middle of the eleventh century — the order is inverted and the Alexis gives us 
43: 87, wherein we see that the language has acquired more stability, a more 
constant, definite shape, and is rapidly nearing the Modem French type. 

For the construction of the adjective we find a confirmation of Diez* and' 
Kriiger^s assertion, in opposition to Morf, that the tendency of the Old French 
is towards pre-position, which takes place almost universally in the oldest 
poetic compositions. Traces, however, of the Modem French post-position 
mle appear in the later compositions of this period, and particularly in the 
Chanson de Roland (end of the XI century), while in Joinville (XIII century) 
the present usage has become fully established. 

For the adverb a striking contrast to the modem rule presents itself here in 
that it always shows a certain mobility in reference to position, but notwith- 
standing this changeable nature it always keeps up a close relation to the 
primitive word to which it refers, and in a great majority of cases precedes it. 
It is to the first half of the twelfth century that we have to assign the change 
of position for the adverbial attributes, as in the Chanson de Roland pre-posi- 
tion is predominant, while in Crestien de Troyes (XIII century) the Modem 
French post-position has become richly developed. 

The writer of this paper promises us another soon, which shall continue the 
work on down through the twelfth and thirteenth centuries and thus give us a 
complete system of Old French syntax. The principal merit of his contribu- 
tion is its general character. In it he has acted rather the part of compiler, 
throwing together and succinctly stating results of research and observations 
scattered through a large number of special treatises. The sharp contrast to 
or agreement with the Modem French syntax is noted with care, but the 
student who is not familiar with the older stage of the language will find the 
lack of examples a very great disadvantage for comparative study. References 
are given in abundance, and yet but few citations, which will make any practi- 
cal use of it clumsy and unsatisfactory. The Latin construction, too, is fre- 
quently called up by way of illustration, but here again all examples are wanting, 
and the force of the illustration is much weakened or lost altogether in the effort 
to seek out the cognate word arrangement. No mention whatever is made of 
similar forms of sentence in the other Romance languages. In spite of all 
these minor imperfections, however, both the Romance scholar in particular 
and the general student of syntax will welcome this little work as a valuable 
help towards filling in one more important gap in our knowledge of the char- 
acteristic modes of expressioi^ belonging to a special domain of human thought. 

A. M. Elliott. 



REVIEWS AND BOOK NOTICES. 219 

L'Egitto al Tempo dei Greci e dei Romani. Di Giacomo Lumbroso. Roma, 
Fratelli Bocca, 1882. 8vo, 204 pp. 

Time was when Italians were the great scholars of Europe, and, though in 
the last two hundred years they have in some respects been outstripped by 
the Germans, the Dutch, and even at times by the English and French, there 
has been no period since the revival of letters when Italy did not contain a few 
men devoted to learning. Unfortunately, however, amid the fanaticism for 
things German that has lately become epidemic in the learned world, Italian 
scholarship, like everything else Italian of any value, has been wellnigh lost 
sight of. The result has been that while the rawest and flimsiest productions 
of new-fledged German doctors have been reviewed and pufled and circulated 
everywhere, the well-weighed works of ripe Italian scholars lie unread and 
unheeded on the shelves of Italian libraries. This is all the more to be deplored 
that ripe Italian scholarship is of a very high order, indeed of a type perhaps 
superior to any other. It is as exact and painstaking as the German without 
being unwieldy or chimerical; elegant as the French without being superficial ; 
as solid as the English without being prosaic, and as comprehensive as the 
Dutch without losing itself in minutiae. In a word, Italian scholars combine 
exhaustive knowledge of facts and texts with correct judgment, well regulated 
imagination, orderly statement, and a clear, manly style of expression. 

Of these characteristics the work before us is a favorable example. Egypt 
in tht Time of the Greeks and Romans reads almost like one of Ebers' novels, 
and yet there is hardly a statement in it that is not solidly based upon authori- 
ties, ancient or modern, carefully weighed and often very shrewdly commented 
upon and corrected. The variety of subjects embraced in it may best be seen 
from the titles of the twenty-five chapters into which it is divided. These are 
I. Worship of the Nile. II. Representations of things from the Nile: the 
Palestrina mosaic. III. The Nile from a practical and positive point of view. 
IV. Deserts bounding the Nile valley : Ancient roads. V. Manners and cus- 
toms of travellers in the desert. VI. Southern confines ; Pescennius Niger in 
the Thebaid. VII. Pelusium. VIII. Cyrene. IX. The Egyptians under 
Greek and Roman rule. X. Greek citizenship. XI. The military class. XII. 
Alexandria. XIII. Character of the Alexandrines. XIV. Games and spectacles. 
XV. Worship of Dionysos. XVI. Worship of Serapis. Sacred medicine. 

XVII. Alexander in the Greek romance, The Acts of Alexander the Great. 

XVIII. Worship and priest of Alexander the Great. XIX. Temple and Hymn 
to Augustus. XX. View of Alexandria at the beginning of the Empire. XXI. 
The architect and inscription of the Pharos. XXII. The Necropolis of Alex- 
andria and the mummies of the Middle Ages. XXIII. Hills of broken pottery 
at Alexandria (Monti Testacci). XXIV. Pompey's pillar. XXV. Epilogue. 

The whole chapter upon the worship of the Nile and the ceremonies con- 
nected with it sheds a curious light upon the growth of religious ideas. Towards 
the end of it a tempting suggestion is thrown out as to the meaning of the 
6rst line of Euripides* Helena : ** NdAov yilv ai6t KaXXnrdp^evot /x>ai" The 
description of the Palestrina mosaic is very graphic, and well worth consid- 
ering the suggestion (due to E. Q. Visconti) that it was made in imitation of 
the Egyptian carpets, so much prized by the Romans under the Empire. 
Indeed Prof. Lumbroso shows that the influence of Egypt upon the Romans 



220 AMERICAN JOURNAL OF PHILOLOGY, 

was very much more extensive than is generally supposed, affecting almost 
every sphere of life, action and thought. 

The account of the canals of the Nile, their construction, locality, and the 
officers appointed to take charge of them, is full of curious facts which help us 
to form a picture of life in Egypt in the centuries immediately before and after 
Christ. Who would suspect that the modern Egyptian term Djerme or Gtrme 
is only the Greek didpjffm}^ In chapter IV is collected all the information 
obtainable with respect to the little-visited desert lying between the Nile valley 
and the Red Sea. It contains long extracts from the MS journal of Matt. Bert 
and Raffenau-Delile, who visited this desert in 1800. This journal was long 
supposed to have been lost, but was discovered by Prof. Lumbrosoin the King's 
library at Turin. According to this, it appears that there still exist consider- 
able remains of the old Roman roads that crossed the desert, with their vSpeiftara 
or castle-like watering stations. 

Passing over several chapters we come to the ninth, which describes the con- 
dition of the native Egyptians under the Greeks and Romans. And what a 
condition ! As we read the details of it, we ask ourselves : Is it possible for a 
conquering people to be just ? And to think that this condition has lasted for 
considerably over two thousand years ! Whether ruled by Ptolemy. Caesar, 
Khalif, Sultan or Khedive, the poor Egyptians have always been the same 
down-trodden, suffering people ; and their future, alas ! looks no brighter than 
their past. Va^ victis ! 

The Greek settlers in Egypt, as described in Chapter X, present the same 
phenomena that Greek settlers in all regions did — personal selfishness, social 
corruption, political injustice and tyranny. It is sad to think that the most 
gifted people that the world ever saw should, when they lost the instinct of 
liberty, have sunk into depths of moral debasement which it is almost impos- 
sible to fathom. As we read Prof. Lumbroso's account of social life in Alexan- 
dria, we seem to be reading an account of Naples under the Bourbons. And 
even at this hour Naples suffers the awful consequences of being a Greek 
colony. If the Greek settlers were bad, the foreign military class described in 
Chapter XI were certainly no better. 

The i^emaining chapters treat mainly of Alexandria, its buildings, public 
works, people, amusements, library, light-house, necropolis, etc., etc. Alexan- 
dria was in many respects the Paris of the ancient world and the Alexandrines 
were its Parisians. Prof. Lumbroso's account .of both is most vivid and telling, 
but by no means flattering to the latter. In Chapter XIII he gives a very 
shrewd interpretation of a passage from Lampridius, which hitherto has baffled 
scholars. It is this : ** Volebat (Alexander Severus) videri origincm de Romano- 
runt gente irahe^qma eum ptuUbat Syrum dici^ maxime quod quodam tempore festo 
ut iolent Antiochensks, i^^GYPTll, Alexandrini kuessiverant eum convUiolis^^ 
After quoting the opinions, suggestions and Teutonic guesses of the various 
editors, he proposes simply to omit the comma after Antiochenses, which omis- 

>As Siipi}fia is not in Liddel! and Scott, some of the readers of the Journal may like to know 
that this strange word occurs in Procop. de Aedif. 6« i, p. 109 A . , . it Kip^wx 6c r^ 
AiyvirTiov vlrov . . . fACTa/3i/3do'arref , ovvircp ffoAeiK iiapijiJLara Kcvo^iiKOtfty cW. Professor 
Sophocles in his lexicon s. r. evidently identifies the modem Egyptian with the Greek word, 
for he adds : ' The modem Egyptian j e r m has usually two large laUtH-^ails.' — B< L. G- 



REVIEWS AND BOOK' NOTICES, 221 

sion makes the last clause mean, ** On a certain festal occasion (carnival) those 
Egyptian Antiochenes, the Alexandrines had jibed him." This gives just the 
sense required, and Prof. Lumbroso shows that the Antiochenes were as 
famous in ancient times for their rude jibes as the Sachsenhauser are at the 
present day, and that the Alexandrines were not far, if anything, behind them. 
Very apt is the quotation from Ausonius respecting Antioch and Alexandria : 

*' Ambarum locus unus . . . Turbida vulgo 
Utraque, et amentis populi male sana tumultui'' 

The amusements of the Alexandrines, described in Chap. XIV, bespeak a 
people of brutal and depraved tastes. It seems they had great skill in training 
animals to all sorts of human-like accomplishments. They had elephants who 
crould speak Greek and wf ite (hieroglyphics ?), and monkeys who could dance 
the Pyrrhic, drive tandem, and read! Prof. Lumbroso makes it evident that 
many of the most barbarous amusements to which the Romans of the empire 
gave themselves up were direct importations from Alexandria. 

In no way superior to their amusements were the religious ceremonies of the 
Alexandrines. Their chief divinities seem to have been the god of drunkenness 
(Dionysos), the god of quack medicine (Serapis), and their own dissolute kings. 
The Roman practice of divinizing emperors was plainly borrowed from the 
Alexandrines. Very curious is the story told of how the Christians, when they 
got the upper hand, transferred the healing sanctuary of Serapis to the Saints 
Cyrus and John and continued the quackery on their own behalf, to the great 
disgust of the regular physicians. It seems the saints were homoeopaths, while 
the Asklepiads were allopaths— which perhaps explains the mutual jealousy.^ 

In Chapter XVII Prof. Lumbroso shows that the Greek romance The Acts of 
Akxander tht Create though untrustworthy is regard to the facts of the hero's 
life, was written by some one well acquainted with Alexandria, and is, there- 
fore, of great value in connection with the topography of that city, whose ex- 
tent appears to have been at one time almost equal to that of London. The 
account of the worship rendered to Alexander and Augustus is repulsively in- 
teresting. Chapter XX, entitled View of Alexandria at the beginning of the 
empire^ contains many curious pieces of information, among them this, that the 
two obelisks, the one of which has recently been removed to London and the 
other to New York, were conected with the Kaiadpeiov^ or temple of Caesar 
Epibaterios (i. e, Augustus). 

In Chapter XXI the author throws cold water upon the story told by Lou- 
kian respecting the inscription on the Pharos at Alexandria. He is possibly 
right ; but his explanation of the origin of the story seems to us in the very 
highest degree fanciful and improbable. 

It is curious to learn (Chap. XXI) that the term Nmp<$.to>uc, now so common, 
was not used in ancient times except to designate the great cemetery of Alex- 
andria, and that the old name for mummy, namely gdbbara, was originally the 
Arabic name for the same spot Mummy itself, It seems, comes from the 
Arabic mum meaning wax. Prof. Lumbroso quotes authorities to show that 
the tombs of Alexandria were rifled of their mummies in the middle ages. 

>The writer was present some years ago at the great festival of the miraculously healing 
Panagia in the bland of Tenos, and can testify that priestly quackery is as much alire now in 
the Greek Church as it was in the early centuries. 



222 AMERICAN JOURNAL OF PHILOLOGY, 

The author thinks a good deal of light might be thrown upon the commerce 
of Alexandria if the numti testtuci which still exist near the city were opened 
and their contents examined, and those who know M. Dumont's work upon the 
inscribed handles of earthenware jars found in the neighborhood of Athens 
will agree with him. After relating all that is known concerning ** Pompey*s 
pillar" he finds it, of course, ** a misnomer*'; but cannot determine ** quale sia 
stato propriamente il posto, lo scopo e Tintegro aspetto del monumento nella 
citt4 antica." The book closes with a quotation from Chateaubriand. 

Pending the general ignorance of Italian on the part of English and Ameri- 
can scholars, it would, we think, be well worth while to translate this enter- 
taining book into English. 

Thomas Davidson. 



Reale Accademia dei Lincei (Anno CCLXXIX, 1 881-2). L'Omelia di Giacomo 
di Sarug sul Battesimo di Costantino Imperatore pubblicata, tradotta ed 
annotata da Arthur L. Frothingham, Jun. Roma, coi tipi Salvincci, 
1882. 

Jacob, or James, of Sarug (A. D. 452-521), bishop of Batna in Mesopotamia, is 
one of those voluminous Syriac writers whose works have little present intrinsic 
value, but great incidental importance from a linguistic, historical, or critical 
point of view. Ecclesiastically (a quality never to be lost sight of in a Syriac 
author) he was orthodox ; or, as we should put it, he adopted all the superstitions 
and corruptions that were catholic in his time. All the writings fathered upon 
him number 763 ; but the genuine are only 231. Of these, two noted ones were 
a liturgy and one of the (seven) rituals of baptism in use among the Syrian 
churches. The rest were epistles and homilies ; the Syriac homily being regu- 
larly a sort of sermon in verse, like Young's Night Thoughts, or Pollock's Course 
of Time. The Homily on the Baptism of Constantine, now first published by 
Mr. Frothingham, is extant in two MSS.one in the Vatican (lOth cent.), the other 
in the Brit. Mus. (gth cent.?), besides a fragment in the Bodleian. These MSS are 
well described by Mr. Frothingham, who takes the Vatican MS as the basis of 
his printed text, emending it once or twice, and very slightly, from the Brit, 
Mus. MS. It is the most complete; it is dated probably A. D. 919; and its 
written character is the Seria^ or that used by the Jacobites and Maronites, 

The editing and printing of the Syriac text appear to be very well done. 
Even the misprints are rare. If the editor has noted all the difficult or appa- 
rently erroneous spots of the MS, the original script must be exceedingly plain 
and correct. Once in the printing the first nun is changed to ^yudva the name 
Constantine ; and scarcely anything worse appears in the text. The criti« 
cal annotations, giving the variant readings of the Brit. Mus. MS and the 
Bodl. fragment, show great care, and are full of important matter. The varia- 
tions which they present leave no doubt that the original composition (as in 
the case of modem songs and hymns) was unscrupulously altered to suit the 
taste of each editor or copyist. The alterations, however, do little harm ; and 
in one or two instances they clear up a difficulty in the Vatican text. In print- 
ing the variant readings, the diacritic points have not been treated with the 
same care as those in the text — ^unless, indeed, the MSS themselves were some* 
times deficient. 



REVIEWS AND BOOK NOTICES, 223 

The translation is executed with understanding, and on a scholarly basis. It 
professes to keep '* strettamente al testo/' and generally does so ; but still the 
translation is not as literal in all respects as an English scholar would demand. 
Thus the phrase *' exalt the horn of* is reduced to the simple " exalt," although 
the Italian Bible retains the full expression. ** Neglect not, O Lord, thy flock " 
(p. 33) is rendered **Abbi cura, o Signore, del tuo gregge." Apart from these 
matters of taste, the translation exhibits a number of oversights, most of them 
semi-clerical, which mar the beauty of the work, though they are of a compara- 
tirely unimportant character. Thus, in the former part of the homily, two 
words are used to characterize the leprosy of Constantine ; one meaning *' stink- 
ing," the other ** hateful." They differ but by one letter, and the translation 
sometimes confounds them. Words not in the text, but supplied by the trans- 
lator, are put in parenthesis, but in several instances the parenthesis is wrongly 
placed, and in others wrongly omitted. Sometimes an essential word is omitted 
in the translation, and here and there an inaccurate translation seems to occur. 
Most of these spots, however, seem to be oversights only, and not errors of 
understanding. 

A few promiscuous examples will show the character of these oversights or 
preferences. P. 33, ** una narrazione del tutto maravigliosa '' is, more closely, 
'*a narrative which is all of it marvellous." " In ogni bella guisa" is rather 
•* with all good fruits " (or, produce); but the sense is retained. P. 34, line 2, the 
word for ** righteous," as an epithet of Noah, is omitted in the translation. P. 
37, " Error thy mother" is translated " TErrore, tuo padre," and so repeatedly ; 
although Error, mother of Satan, seems to play a female part on one side, 
that offsets the part taken on the other side by Helena, mother of Constantine. 
P. 40, *' (I'animo) " is wrongly put in a parenthesis that probably belongs about 
•• dicendo," eight lines below. P. 44, parenthesis is put about the 6rst ** Maria" 
instead of the second. The word for " baptize," though technical baptism only 
can be meant, is generally translated by ** immergere," but sometimes by " batte- 
zare." P. 46, *' perch^ secondo quel che ho udito Cristo Signore per questo 
venne," though perhaps justifiable in one view, is rather '* because I have heard 
that for this the Lord Messiah came." P. 47, the words " Allora al comando 
. . . arme spirituali" are not in the text, but supplied from the substance of i\it 
annotations, and properly belong in the footnotes. Same page, ** dalle acque 
battesimali " is, literally, " from the midst of the depths of the baptism." Of 
such spots as these there is an average of rather more than one to a page. 

The introductory part of the work consists chiefly of a historico-cntical inves- 
tigation of the accretions of fable about the actual nucleus of the conversion 
and baptism of Constantine, with a brief notice of the place the fables have 
occupied in art. All this is ably and carefully done. The place of the com- 
ponent parts of this homily in the progression-series of increasing fables is 
pretty well shown ; and the whole investigation is interesting and valuable. At 
the same time the editor fails to point out (if he recognizes for himself) the 
fact that Syriac homilies of the sort were written and understood as poetical 
expansions, not as sober fact. Ephrem's *' Repentance of Nineveh " or Milton*s 
** Paradise Lost " marks likewise the existence or the invention of fabulous 
embellishment To treat this homily of James as a fable believed by either 
himself or his auditors is erroneous. A Syriac homily is not a Syriac chronicle. 



224 AMERICAN JOURNAL OF PHILOLOGY. 

The editor has crowded so much valuable matter into this work, and, further- 
more, has approached it in such an original and fruitful direction, that he may 
well be excused for leaving to others the matter of its Biblical words and phrases. 
With one or two exceptions he has avoided in his translation the apparent 
allusions, even though they might help interpret the homily. The work is to 
be heartily welcomed. It is much nearer perfection than is to be usually 

expected of an editio princeps, 

I. H. H. 



Sammlung kurzer Grammatiken German ischer Dialecte. Angels&chsische 
Grammatik von Eduard Sievers. Halle, Max Niemeyer, 1882. 

A treatise on Old English grammar, scientifically conceived, independently 
wrought out, abreast of the highest contemporaneous scholarship, discriminating 
between the various stages in the development of the language, as well as 
between the coexistent dialects, and paying due regard to it both as a separate 
entity and as a member of the Germanic family, has hitherto been a desuUratum. 

How inadequate have been the conceptions which living scholars, as well 
as those recently deceased, have entertained of phonology, for instance, may 
be illustrated by reference to one or two manuals lately published. K5rner, 
in his Einleitung in das Studium des Angelsftchsischen (Heilbronn, 1878), thus 
disposes of the Lautlehre in a note to p. 2 : '* Eine wissenschaftliche Darstellung 
der angelsftchsischen Lautverh&ltnisse ist von Grein, Koch und Holtzmann in 
ihren Grammatiken versucht worden. Sie erfordert Kenntniss der verwanten 
Dialekte, ist aber, wie sich schon aus dem Folgenden ergeben wird, fdr das 
AngelsSchsische von besonderer Schwierigkeit und geringem Nutzen; daher 
ist hier auf sie verzichtet.** 

Grein, in his Kurzgefasste Angels3chsische Grammatik (edited by Professor 
WUlcker, of Leipzig, in 1879), devotes 7 pages to an introduction, and nearly 
15 to a sketch of the Old Anglo-Saxon and Northumbrian Literature, but only 
9 to the Anglo-Saxon vowels, and less than 12 to the consonants. 

Holtzmann*s Altdeutsche Grammatik (1870) contains valuable paragraphs 
treating of Anglo-Saxon phonology, but the premature death of the author 
prevented him from finishing even the volume devoted to the phonology of the 
Germanic dialects. Since that year a number of monographs have been pub- 
lished, chiefly in the form of contributions to periodicals and the proceedings 
of learned societies, and it was from these scattered pages, not always to be 
collected without great difficulty, that the student was obliged to glean the facts 
and theories which would enable him to construct the outlines of Anglo-Saxon 
grammar. From this labor he is at once relieved by the appearance of Sievers* 
manual, of which it is scarcely too much to say that it fulfils the various condi- 
tions specified in the opening paragraph of this review. 

The series of brief grammars of which this forms the third volume, has thus 
far issued from the hands of the so-called * Junggrammatiker,' Braune contribu- 
ting the Gothic Grammar, a model of accuracy and elegant simplicity, and Paul 
thi> Middle High German Grammar. We say the so-called * Junggrammatiker/ 
Hu thl» in an appellation bestowed upon them in derision by their opponents, 
AUv\ uover accepted by this little band of earnest and high-hearted scholars ; 



RE VIE WS AND BOOK NO TICES. 225 

but by whatever name they are known, it can no longer be disputed that they 
are accomplishing a revolution, at once beneficial and inevitable, in the methods 
of comparative philology. 

An epoch of riotous and over-fanciful speculation is to be succeeded by one 
of sober induction ; abstractions are no longer to usurp the place of concrete 
existences, and serve, in the hands of philological jugglers, to mystify at once 
an uninitiated laity and the hierophants themselves. In short, the scientific 
temper is to prevail in matters linguistic, an event which is sure to be regarded 
with dismay both by super-subtlety and by dilettanteism. But whatever be 
the outcome of the movement, its leaders have contented themselves with very 
modest professions. Their aim, so far as it may be gathered from their authori- 
tative statements, is but to rescue one small province, that of phonology, from 
the dominion of caprice and to bring it under the reign of law. 

One or two quotations from Paul, whose fondness for philosophical discus- 
sion has made him the most prominent expositor of the new doctrines, will 
serve to characterize the points at issue, and to show how they are regarded by 
the ' Jnnggrammatiker.' In the Beitr&ge zur Geschichte der Deutschen Sprache 
und Literatur, Bd. IV, Paul says : 

** So lange man es mit den Lautgesetzen nicht sehr streng nahm, so lange 
fand man nicht sehr viele erhebliche schwierigkeiten bei der vergleichung der 
germanischen dialecte oder der indogermanischen sprachfamilien untereinan- 
der in bezug auf ihre declination und conjugation. Es geniigte eine unge- 
f&hre flhnlichkeit der formen.die allgemeine mOglichkeit oder wahrscheinlich- 
keit der bei der vergleichung postulierten lautiiberg&nge '* . . . pp. 317-8. 

'* Die voraussetzung, von welcher dabei ausgegangen wird, ist die, dass jedes 
lautgesets mit absoluter notwendigkeit wirkt, dass es ebenso wenig eine aus- 
nahme gestattet, wie ein chemisches oder physikalisches gesetz. Mit dieser 
voraussetzung steht und fflUt die von mir befolgte methode. Wer sich 
entschliesst die erstere zu verwerfen, der braucht auch die letztere nicht anzu- 
erkennen. £r verzichtet aber damit iiberhaupt auf die mdglichkeit, die gram* 
matik zu dem range einer wissenschaft zu erheben.** Beitr&ge VI i. 

** Eben das vertrauen zu der absoluten gesetzm&ssigkeit der lautbewegung 
ist es, wodurch die sprachwissenschaft der naturwissenschaftlichen evidens 
nahe kommt, und wodurch sie in bezug auf sicherheit ihrer resultate alien 
anderen historischen wissenschaften so sehr iiberlegen ist. Dieses vertrauen 
dient ihr wie jeder naturwissenschaft als fundament, auf welcher sie aufge- 
baut wird. Es wird ihr dadurch das ziel gesteckt, alle lautlichen ver&nder- 
ungen unter gesetze unterzubringen, die mit absoluter consequenz wirken. 
Dieses ziel dient aber zugleich als prtifstein fUr die richtigkeit der aufgestellten 
gesetze und liefert die probleme, welche durch die forschung zu losen sind. 
Nirgends darf man sich bei einer vielfSlltigkeit oder inconsequenz der behand- 
lung eines und desselben lautes unter denselben bedingungen beruhigen. 
Kann nicht durch andere .fassung der lautgesetze abgeholfen werden, so ist 
die unabweisbare consequenz, dass von den verschiedenartigen verHnderungen 
unter gleichen verhftltnissen immer nur die eine auf physiologischen wege 
entstanden sein kann, w&hrend die andere oder die anderen sich auf psycholo- 
gischen wege, durch formenassociation eingedr&ngt haben miissen." Beitr&ge 

VI 3. 



226 AMERICAN JOURNAL OF PHILOLOGY. 

But to return from the general principles advocated by this school to the 
author of the particular Tolume which we have undertaken to notice. Sievers 
has made his mark upK>n the science of comparative philology by his labors in 
the two allied branches of grammar and phonetics. Before he had undertaken 
any serious original work he had shown himself a competent translator from 
modem Scandinavian ; his version of Wimmer's Oldnordisk Formlsre, the 
standard grammar of Old Norse, was made in 1 871, and that of Thomsen*s Den 
gotiske sprogklasses indflydelse pa den finske was completed about the same 
time. His conscientiousness and laboriousness as an editor have been well 
illustrated in his Tatian (1872), Murbacher Hymnen (1874), and Heliand (1877). 

In addition to remarkable industry, his contributions to Paul und Braune's 
BeitHLge display acumen of a rare order ; in them he attacks no difficulty which 
he does not in some measure clear up; and even those who repudiate the 
doctrines advanced by the * young grammarians* are obliged to concede that 
papers so rich in lucidly arranged material, and bearing in every part the 
impress of a master^s hand, are indispensable alike to the student of Common 
Germanic and of the particular dialects treated. But it is as the leader of German 
phonologists, and the peer of Ellis and Sweet, that Sievers has won the widest 
and most indefeasible reputation. A keen perception of all shades and varieties 
of speech-sound, and a rare command of his own vocal organs, enabling him to 
reproduce any sound articulated in his hearing, and instantly to catch what is 
popularly denominated the * accent ' of the most difficult and unfamiliar tongue, 
are the special qualifications which have given him his present standing in this 
department. His GrundzUge der Lautphysiologie was published in 1876, and 
his GrundzUge der Phonetik in 1881. 

It was not as a tyro, therefore, that he approached the difficult task of writing 
an Anglo-Saxon grammar. Indeed, the only risk to which he exposed himself 
was that of disappointing extravagant anticipations. As to the manner in which 
he has answered these expectations, it is enough to say that his grammar, though 
we should not dare to affirm that it is the final word on Old English phonology 
and. inflection, does in truth mark a distinct and notable advance upon any 
similar work which has preceded it, and practically supersedes them all. Since 
the publication of Grein*s ' Sprachschatz,* no book so indispensable to the non- 
professional student of Anglo-Saxon has appeared. Nor will it be hardly less 
welcome to the English philologist, whatever his attainments, since he here 
finds collected, under one point of view, what must else be sought through many 
volumes, and is nowhere to be found in equal fullness and clearness. All that is 
important in the utterances of Old English scholarship for the last ten years is 
summed up in its pages, and the compilation is enriched by a great number of par- 
ticulars supplied by the author's own observation. Yet these particulars by no 
means represent Sievers* full share in the materials of the volume, many of its 
most noteworthy paragraphs being mere abridgments of his own articles in Paul 
und Braune*s Beitrftge. In illustration of this fact it will be sufficient to com- 
pare the treatment of the ^-, jA-^ and t-stems, §§252, 256 and 262, with PB 
1 486-504 ; the syncope o( medial vowels, §§143-8, with PB V 70-82; the succinct 
note on cuman, §390, Anm. 2, with PB VIII 80-9 ; and the statement regarding 
the instrumental, §237, Anm. 2, with PB VIII 324-33. 

Not only is there a notable accumulation of facts, but they have been 



REVIEIVS AND BOOK NOTICES. 227 

Arranged in an orderly and perspicuous manner which leaves little to desire. 
The exceptions which will be noted further on scarcely detract from the 
pleasure with which the student greets this lucid exposition of a most difficult 
subject. In particular is this true of the chapters on the vowels, hitherto the 
most hopelessly perplexed of all the intricate webs which the student of Old 
English grammar was called upon to disentangle. 

The system of cross-references adopted, while at first blush it seems unne- 
cessarily minute, proves in the end to be a convenient guide through the mazes 
of phenomena presented by the vowels. No less helpful are the pages (13-22) 
at the beginning of chap. III. Here the Anglo-Saxon vowels are considered 
in their relation to those of the Germanic and West Germanic, the treatment 
being ranged under two heads : (i) The Vowel Systems of Germanic and \yest 
Germanic, and (2) The Correspondences of the West Germanic Vowels in West 
Saxon. 

The quantity of the Anglo-Saxon vowels has here been rigorously observed, 
the authority of the manuscripts being accepted as paramount. Besides the 
introductory remarks in §8, the fluctuations of quantity are examined in §§120-5. 
Of these the most unaccountable are the prolongations of vowels followed by 
single consonants in monosyllabic words, though these are supported by the 
same evidence as similar prolongations before nasal 4~ consonant or liquid 4~ 
consonant. 

The prominence given to the unstable y and ^, §§31-3, together with the 
remarks on the t-umlaut of ea^ ed^ eo and ed^ §§97-ioo, are reassuring at the 
outset, since they seem to contain the explanation of a puzzling phenomenon ; 
closer inspection shows, however, that the difficulty is only shifted to other 
ground, remaining at last as inexplicable as before. Perhaps Sweet's suggestion. 
Pastoral Care, p. xxvii (cf. pp.xxix and xxx), is the most satisfactory yet advanced ; 
cf. also ten Brink in Anglia, I 51S-19. 

Intimately connected with the last is the paragraph on palatal umlaut, {:oi. 
Its effect, according to Sievers, consists chiefly in the transmutation of the eo^ 
f^, produced by breaking before h 4- consonant, into iV, which u ultimately 
suffers change into i and y. The discovery was made by Paul, who (Beitr&ge, 
VI 46-7) first called attention to the phenomenon, and provided the explanation, 
though he probably owed something to Moller, Die Palatalreihe der Indoger- 
manischen Grundsprache im Germanischen, pp. 56-7. 

The f-umlaut of short 0^ {93, deserves a passing notice. Sievers shows that 
the true umlaut of o is e. His theory is based upon Paul, Beitr&ge, VI 242, 
with which may be compared Cosijn, Kurzgefasste AltwestsSlchsische Gramma- 
tik, p. 36, D; for the older view regarding ek see Sweet, P. C. p. 491, though 
this explanation is modified in Anglia, III 157. 

The plan of this book is essentially that of Braune*s Gothic Grammar, t. e, 
the two exhibit the kind and amount of similarity which would naturally be 
expected between two grammars emanating from the same school. Braune 
had, however, the advantage of dealing with a much less complicated subject, 
and hence is not obliged to resort to long and frequent digressions. From the 
very nature of the case it is impossible for Sievers to attain equal symmetry in 
the disposition of his materials. The vowel system of Old English being 
extremely complicated, and the plan of the book including some account of 



228 AMERICAN JOURNAL OF PHILOLOGY. 

ttinlectic variations id th« kniiriiage, nun; or the ineqiiAUties to be noticed 
would seem lo be unavoidable. Others mi; be explained from the fact that 
the grammar has grown by accretion from a teC of University lectures. Sievers 
himself says in the preface : " Der knne abriss, den ich jetit der Offentlichkeit 
Obergebe, machl demgemlH nicht den anspruch, mehr lu sein, als eine sokhe 
nberatbeitung, obwohl Uber der umschrift und durcfaslcht manches ergSniend 
hinzugetreten ist, was der DTSprQnglichen fassung fremd war." Accordingly, 
the booit is neither i mere outline of West Saxon grammar, containing only 
the essentials, nor an exhaustive treatise, discusiing the various dialects in fulL 
II is rather a compromise between the two, with a distinct leaning toward the 
original plan of the series, which is (hat of compendious mannals. 

This conception might have been more strictly carried out, without material 
detriment to the value of the booli, and with a decided gain as regards symmetry 
of execution and unity of impression. Such details, for eiample, as are intro- 
duced in §37t, Anm. 3, gaSj, Anm. 3, and §374, Anm. are rather lexical thui 
grammatical, and will be more welcome to specialists than to the great body of 
the students for whom the boolc, or at least the series, is designed. Who 
constitute the latter class may be learned from the preface lo Braune's Gothic 
Grammar, which closes with these words: " Diese grammatilten sollen gedrlngte, 
jedoch nicht lu dilrftige darslellungen bieten und besonders anflngern sur 
elnfilhrung In das philologische sludium der betreffenden spriehsiufe dienen." 

A numt>er of misprints and minor errors have been noted, but they are hardly 
serious enough lo occasion the student much difficulty, and will no doubt be 
corrected in a second edition. 

Albert S. Cook. 



Vrchia. Texle Latin pnbli^ d'apr^ les travani 
louvelle collation duGemblaccnsis.un commen- 
ne introduction etun index par £milb Thomas. 
■ 63 pp. 

poet Archias Is not, as Tacitus says, one of the 
tat, and although in point of argument it is far 
ised a great charm for scholars and book-lovers, 
nigh literature the world lives. The present 
tractive form, with a good introduction staling 
; life of Archias, reviewing briefly the attacks 
Duineness of the speech, witharefulalion of the 
lurces of the text, and an analysis of the oration. 
and the editor shows a much greater familiarity 
nan scholarship than was formerly common in 
egelsbach, SeyETert, Me[^uet and Mueller are 
Editor has preserved an independence of jndg* 
^ keeps inrrpstrunt of the MSS against Baiter 
i, and explains thus, making ihe idea of time 
d'^trangersontpdn£lr^ dans les villesltaliennes, 
droits franduleusement acquis, repoussera-t-on 



RE VIE WS AND BOOK NO TICES, 229 

Archias?*' In the same section he makes the mistake of referring civitaU to 
the Roman franchise ; the franchise of Heraclia is meant, as the context shows. 
In §19 the note on suum^ which is omitted by some editors, is not full enough ; 
the preceding dicunt suum conditions the use of smtm here, for which Cic. De 
Rep. I 27 uses /rd? suis vindicare and elsewhere siH vindicare. The orthography 
of the edition is not as good as one might reasonably expect. Not to speak of 
the genitives in -iV which are kept throughout, a point where there is room for 
hesitation, one is pained to find inJui<n^{\mmt6\Alt\yht^QTtconfiUor)f conditioner 
solatium^ dampnationem^ contempnenda^ and there is a certain inconsistency in 
reading §24 innumerabi/t> copias and in §31 apud omn^^i. It would be well, 
however, if all editions were as free from mistakes. 

M. W. 



REPORTS. 

Revue de Philologie. Vol. VI. 
No. 4. Nov. 

1. Pp. 193-203. Critical remarks on Liv. XXIII-XXV, by O. Riemann. 
The author is preparing an edition of these books of Livy, and has made some 
emendations, 65 in all, which he here presents and discusses. The most of 
them, as he remarks, are not of great importance. I cite a few as samples. 
XXIII 5, 15 : Nee Hannibal se vicisse sentietnec Romani victos esse. Read : 
victos [se] esse. XXIV 6, 7 : Himera amnis, qui ferme dividit insulam, etc 
Read : ferme [mfdiam] dividit. XXV 28, 6: inopiam quaeque (P quequae)\^ 
inter se fremere occulte soliti erant, conquesti. Read : inopia[m alia]c^t quae. 

2. Pp. 203-4. L. Havet discusses Quintil. I, I, 24. For etiam brevia the 
Ambrosianus has H *imomna^ which \\ for ft and via with something like 
meliore between, and no doubt represents the true reading. 

3. P. 204. O. Riemann suppresses bv before *AKapvaviav in Thuc. II 80 as 
having grown out of a dittography of uk, 

4. P. 204. L. Havet expresses the opinion that Satumian distichs were 
common at a certain period, and that the word elogium^ i. e. iXeyeicv (distich)^ 
thus originated. 

5. Pp. 205-8. Book notices, chiefly by O. R. 

6. Revues des Revues^ pp. 225-337 (end). France (completed). Great Britain, 
Greece, Italy, Holland, Russia, Sweden, Switzerland. 

Vol. VII. No. I. Feb. 

1. Pp. 1-5. On the exclamation malum ^ by Constant Martha. The author 
published an article on this subject in Vol. Ill, pp. 19-25, maintaining that 
malum as an exclamation always denoted impatience at some sort of folly. 
(See this Journal, No. i, p. 84). This View was questioned by E. P. Morris in 
this Journal, No. 10, pp. 208 ff. M. Martha writes the present article for the 
purpose of making known to the readers of the Revue the new examples found 
by Mr. Morris, and of defending the views expressed in his first article. 

2. Pp. 5-6. Ten passages of Pomponius Mela emended by L. Havet 

3. Pp. 7-12. On certain omissions in the text of Demosthenes, by Henri 
Weil. Seven lacunae pointed out and filled. This article merits attention 
from students of Demosthenes. 

4. Pp. 13-22. On a Latin grammar in a MS of the eighth (?) century belong- 
ing to the library of Nancy, by A. Collignon. This grammar, or Glosa de par^ 
tibus orationis^ is composed of a series of extracts from ancient grammarians 
without much system. Sometimes the author is given, but in many cases this 
is neglected. The chief value of the work consists in the fact that it contains 



REPORTS. 231 

a considerable number of hitherto unknown passages of the grammarian Vir- 
gilius Maro. [In one of these the question whether the nominative can properly 
be called a case or not is discussed and decided affirmatively. The subject is 
viewed, however, from a standpoint very different from that of modem gram- 
marians.] 

5. P. 22. L. Havet emends a verse of Naevius (Nonius, ed. Quicherat, p. 
159 M., 6) : put cum before sis, 

6. P. 22. In Xen. Resp. Lac. 2, 6, O. Riemann proposes koI tl^ [rd e<c] 
fiijnoi \hv\ av^dveaOai Kxi, 

7. Pp. 23-32. On the Paris MSS of the Distichs of Cato, by Max Bonnet. 
The only edition of Cato (F. Hauthal, Berlin, 1870) that pretends to make any 
nse of the Paris MSS is exceedingly unreliable. H. J. Muller called attention 
to this fact in 1876, and in the same year the Revue Critique (II, p. 187) 
exposed the carelessness of Hauthal. But these warnings were fruitless, as is 
shown by the most recent edition (Poetae Latini minores, ed. Aem. Baehrens, 
vol. Ill, Lipsiae, 1881). Bonnet examines and classifies the MSS in question, 
comparing them with others, and discusses a considerable number of passages, 
taking occasion to investigate some metrical and grammatical points. 

8. Pp. 33-60. Criticism of Greek texts at the £cole des Hautes Etudes 
(Continuation. See this Journal, No. 12, p. 491.) II. Demosthenes. Y. dis- 
cusses KoX y6p Toi and submits 43 emendations. I. V. shows that Kal yap rot is 
not a quasi-synonym of Kal ydp, but rather of roiydproi (itaqtU), One difficulty 
(?• 358, ch. 56) he proposes to remove by emendation (koI yap [uv\ toi^ or some- 
thing of the sort). 2. The emendations, though rarely convincing, all merit 
attention. 

9. P. 6a Y. calls attention to the fact that in Dobree's Adversaria on Soph. 
Trachin. 574 we find ** aenigma SophocU dignum," for which Y. wishes to read 
* spkinge dignum.* 

xa Pp. 61-64. Seven passages of L. Annaei Senecae didlogorum, Lib. I, 
discussed and emended by J. van der Vliet. 

11. P. 64. For excoltdsse in Martial, Epigr. VI 52, 4, Henri Le Foyer pro- 
poses expoHisse^ which he finds on the margin of a variorum edition. 

12. P. 64. In Prudentius. Cathemerinon 2, 12, E. C. restores palUscet for 
palUscit from a good MS. 

13. Pp. 65-77. Emile Chatelain publishes a work entitled Exempla diver- 
sorum auctorum, contained in a Vatican MS (Reginensis 215) of the ninth cen- 
tury, and, in abridged form, in a MS (4883 A) of the tenth century in the 
national librfiry of Paris. It contains a list of 250 verses selected from different 
poets for the purpose of illustrating quantity in various words. In Reginensis 
215 the words illustrated are written opposite the verses in which they occur, 
and are provided with quantity marks. In most instances the author is given, 
but this is so often erroneously done that no confidence can be placed in the 
assignment of verses not otherwise known to us. The examples are evidently 
taken, not from the texts of the authors cited, but from grammarians and from 
anthologies now lost. A supposed ** interim,*' for instance, is illustrated thus : 
*' lav.: Interim veteres laudat lasciva patronos," which should be: *' Martial. 



232 AMERICAN JOURNAL OF PHILOLOGY. 

(V 34, 7) : inter tam vetcres ludat 1. p.** Chatelain has succeeded in assigning 
all the verses except about twenty to their authors. Some of the twenty are 
barbarous. 

14. Pp. 78-81. Remarks on certain passages of the Libellus pro synodo of 
Ennodius, by L. Duchesne. This author uses the inflated, affected style which 
we And in the Opus Paschale of Sedulius (see this Journal, No. 9, p. 115). 
Duchesne emends a few passages. 

15. P. 81. Michel Breal expresses the opinion that in Liv. Ill 33, 8 (with 
Dttring*s emendation, est for esset) pHvatus is equivalent to rtus, 

16. P. 82-94. Unpublished text of Domninus of Larissa on arithmetic, with 
translation and commentary, by Ch. Em. Ruelle and J. Dumontier. This 
work is devoted to the solution of a single problem — ^how to divide one frac- 
tion or ** ratio" by another. The author, being unacquainted with the method 
of inverting the divisor and multiplying, resolves the dividend into two factors, 
one of which shall be equal to the divisor. This he does in four ways, one 
of which I give : To divide ^ by J. Let Jj* = J = { X }• But 3 : 12 : : 4 : x 
= 16. Hence V = V X }. and J^-7- } = J^ = }. In the original, of course, 
no equations are written. Diagrams, however, are used; but they seem to 
have been added by some one else, for the text does not mention them. 

17. Pp. 94-96. Note on a MS of Bourges containing Cicero's Letters, by E. 
Chatelain. This note demonstrates the worthlessness of the MS in question, 
and is published for the purpose of saving others a useless journey to Bourges. 

18. Pp. 97-101. Note on two MSS of the Historia Apollonii regis Tyri, by 
O. Riemann. Among the many MSS of this romance, two of the thirteenth 
century, now in Rome, have a common text which is totally different from that 
of the other MSS. They present a developed text which in places is a literary 
improvement. I give a sample. Ordinary text (ch. 32): "Et cum* puella 
Deum dcprecaretur, subito piratae apparuerunt." The two MSS (MR): 
** Puella levavit manus et ait : Pater Apolloni, si vivis, vale, et sancta desideria, 
valete. Et dum hec diutius loquitur, supervenerant pirate." Riemann collates 
a few chapters. 

19. P. 102. L. Havet gives a new metrical division of a verse of Naevius 
(Nonius. 486, 27), and emends Statius, Achil. i, 73 {imum for unum), 

20. Pp. 103-12. Book notices, chiefly by the editors. 

M. W. HUMPHRSYS. 



NBUB jAHRBtCHBR Ft^K PHILOLOGIE UMD PIDAGOGIK. FlKCKBISSN U. MASIUS. 
, I88I. 

I YIIMX. 

j ^$* PP* 5^3"33' Review by W. Clemm of Meyer's Griechische Gram- 

! nxatik, Leipiig, 1880. A connected, scientific presentation of the subjects per- 

taining to Greek grammar was lacking. Ktthner*s work was a dozen years old, 
«ud when it appeared was not up with the times. Only one class of the forms 
Kad be«n treated^the verb, by Curtius. The other systems of inflections as 



REPORTS, . . 233 

well as phonology needed a new discussion which would use all the material 
which had been gathered. So we were all ready to welcome the work of 
Meyer. He takes the newest hypotheses as his starting-point. In this he is a 
little intemperate. Sometimes he is not ready to retain the old view even 
where the new view offers no more plausible explanations. As long as no 
strong root-forms are shown for OTrd, hv6L^ hv&o^^ d^^, etc.. the reviewer does not 
share the author's prejudice against 2 where it is not a weak form of a. He 
considers 6-^k-vw to be quite unsupported, and thinks the root Jox as possible as 
ay and apx* He believes Meyer to be unjust toward sporadic changes. Kidvarat 
and OKidvarai cannot be explained as dialectic variations. Meyer's view of 
final a in adverbs like icaXof (that it is brought by analogy from if, ficrdf, etc.) 
is not so probable as that of Curtius (that -<i)C=-arj. A chapter on the use and 
force of the accent is needed. The work, then, gives us a skilfully arranged 
view of Greek grammar according to the latest theories and based on exten- 
sive material. It does not, however, give the reasons for the views which are 
expressed, nor a complete and reliable repertorium of the facts of the language. 

(76.) pp. 534-36. G. H, Mttller in Soph. Trach. 651 reads XP^^'^ na?Mi6v 
for X' ire/idywv, comparing Soph. O. T. 561, Ajax 600, Phil. 493. In Trach. 958 
for fiowov he reads fiuXw (* weak,' * exhausted/ cf. Nicand. Ther. 32) to give the 
desired contrast to iXxifiov, In Ant. 351 for e^erai he reads e^i^erai, construed 
with two accusatives on the analogy of naiSevu^ etc. 

In Soph. £L 1394, J. Golisch reads olfut for aifta^ comparing Hom. 11 752. 

86. pp. 537-42. H. StadtmQller oflfers conjectural emendations to the 
Homeric hymn to Hermes, fiiuv for fir^ id6v^ 92 ; ap6ev^eioa^ for d^p6a^ ohaac, 
106, to introduce a reference to the watering as well as the feeding of the cattle ; 
similarly dpSfwi^ for aypoig, 399, cf. 2 521 '; Pi^a^ ^oal nvpirdk&firiaev for Pt,?^^ 
icoaX KapiraXifiotatv, 225, cf. 357 ; Kivrpuv for icipTOfwv^ 336 (cf. Ar. Clouds 444 
and the scholion Ktvrpuv Xtyerai xal 6 k?ihrTrfi) ; ovpavSv for Kpaivuv 425 ; 
CTTVpr&il^ijv ioT airovdy idv^ 305, cf. Photius airvp&i^etv at^d^etv^ etc.; vXaxruv 
for oKuwjVf 280 ; avk^^rjva^ for fUfiijhigf 437 ; im* bfiokX^c for vKd noXX^g^ 373 ; 
ianero for iTrXero^ 117, cf. 426,440 ; kvmijv^ for iavrSv, 239, cf. A 402, k 448, e 446, 
etc. 

(64.) p. 542. R. Dressier in Stobaeus An th. CXX 27 reads ^avdrtf) yap 
yiyvtrai didXvaig KafidvTog adparog tov {nrdfiov ittj pu&ivTog ktX 

87- PP- 543-52. K. von Jan defends his view (Jahrb. 1879, pp. 577 fg., 
see Am. Jour. Phil. I 373, II 531) that the Greek flutists regularly played 
two flutes (oboes) at the same time, one flute giving the melody, the other 
accompanying it on a higher note which was probably like that in the modem 
Greek churches, merely the keynote, dominant, or sub-dominant. The use of 
the two flutes is shown by the Greek vases which bear scenes of banquets, etc. 
In the museum at Munich a double flute is represented on 20 vases with black 
figures, and on 37 with red figures. On one side of another vase is a satyr with 
a flute in each hand ; on the other side is a satyr with a single flute in the left 
hand, but the position of the right hand indicates 'that he has thrown the other 
flute into the air. If this one example be granted, then the Munich vase collec- 
tion contains 59 examples of the double flute and not one of the single flute. 



234 AMERICAN JOURNAL OF PHILOLOGY. 

If the two flates were cnstomary at banquets, doubtless they were at conceits, 
where the music would be mor^ elaborate. 

That the two flutes sounded different notes is indicated most clearly by the 
words of Aristoxenus in Plutarch de mus, 36 irroKpivete yap &p rtc ixobvw atX^nm 
Tr&rep6v irore ffvfi^uvovctv d aihtl 1j ot, in connection with the definition of 
Pseudo-Euclid demns, 8 aiffi^uva fitv obv [duuniffMTa] tan 6ta rtfjoapufw, dta whrt, 
ktX, and iart di ovfi^uvia fihf Kpaotc Si)o ^6yyuv b^kpow koI papvripov, 

88. pp. 553-61. J. S. Kroschel on the IS eftXKvarui6i¥ in the oldest MSS of 
Plato. Greek grammarians assert that Attic prose-writers used this r in the 
dat. plur in -ai, the 3 pers. sing, in -e, and verbs ending in •at, no matter whether 
consonant or vowel followed, and that to this statement fiofrrvpti waaa pifi^. 
We must then believe that this was the custom in the MSS used by these 
grammarians ; but that the Athenians in the time of their earliest prose-writ- 
ings did not regard the v as a fixed element of the endings is shown by the 
usage of Aristophanes, who has always been considered a model of Attic style. 
He prefers the v e^, to elision, but avoids it before consonants as far as he can, 
except where he introduces old-fashioned forms like rowrtv, fia^rrraloiv^ etc. He 
uses it before consonants only about once in 50 verses, most frequently in 
karlv. But the copyist changed even the words of poets by attaching the ». 
So in Eur. Andr. 275 iJX^' was changed to iJA^ev, Hel. 316 kl^? to IXe^tv, 
Probably the final vowel was written in the copy and fj^'Qe 6 rdno^ seemed 
obviously wrong. In poetry we are generally able to detect such insertions by 
the metre. 

In Plato the wearisome repetition of the forms with v is due mainly to the 
codex Clarkianus. This codex is compared by Kroschel with the codex Vene- 
tus which was written in the XII century. The independence of the latter is 
shown and that its readings have not been corrupted by conjectures. It is 
thought that the sources of these two codices were divided before the V cen- 
tury. Since the Ven. gives in many points the older tradition, it deserves more 
weight than the Clarkianus. This would remove the v from at least half the 
cases. It is noticeable that in the dialogues of the first tetralogy, a '* second 
hand" (conjectured by Kroschel to be Arethas himself) has erased a large pro- 
portion of the unnecessary cases of v. It is possible that the MS from which 
the Clark, was copied had many abbreviations and used the same mark for 
i<rri and koriv. This conjecture is confirmed by the fact that lor* no longer 
appears in the MS and io^'* only rarely. Moreover, in the Platonic MSS this 
V is wrongly inserted in the quotations from Pindar, Simonides. and Euripides. 
In this connection Kroschel conjectures in the ode of Simonides quoted Prot. 
346 d, that the original was inel ohriv\ etc., (for Itcei-d* vfuv elp6v where Bergk 
now reads iiri r' ifi/uv), comparing the gnome ascribed to Simon. Amorg. 
irdpwav d* ifiufw( ohric ov6* didfptoc, 

89. pp. 561-64. K. J. Liebhold in Plato's Parmenides 135 d for e^xwrav di 
oavrbv Kal yvfivaoai reads iKkvaw xrX. ; in 160 d he inserts 0i6v re between 
yiyv6(JKe(r9at and flrav. In 164 a for oire ravrd, oi^* irepd iartv avrtft he reads 
ohrt rahrh avrift oid^ irepd iertv avrov (sc. rov ps^ 6vrof). In Gorgias 500 c he omits 
cv before ri &v ua^^ov and iirl before rdvde rdv piov. In 501 a read ^ d* hipa 
r^ ^dovvf irp6i fjv 1} 'Bepantla avrff iariv anaaat oVtc ri n^ ^iv OKt^^tafiivq ohre 



::;resi 



REPORTS, 235 

■•'t vi'wf in-' avrffv ipxfrai aA<Jywf ktX, In 503 c, tovto 6i rkxyri 

..' for nvat. In 513 b, to fonn the desired contrast with rtf 

•'i\u) T<^ Adyif) is suggested for "Xxyofikvuv tuv "kSyuv, In 513 

ff(^ovTjv ofiiltiv is illustrated by 521 a, Phaedrus 272 d, Laws 

' ; . 317 d. In 513 e it is suggested that aKonei may have been 

' copyist after b ^epawebofiev, 

;f5-68. Critical notes of H. Marquardt on Galen irepii yfwxvi 

?''K»-5f)2. A notice by F. Hultsch of Vorlesungen Qber die 

<!er Mnthematik von M. Cantor. I. Leipzig, 1880. The book is 

>t and instruction. The treatment of the decimal and sexages- 

\ms is commended and amplified. The Hbrews like the Egyptians 

> fn! lowed the decimal system; cf. the tithes of the Mosaic legislation, 

Us ions of the tabernacle, and the report of the temple of Solomon, in 

1 k of Chronicles. In this last, however, as in the (post-exilian ?) descrip- 

f Noah's ark and in the vision of Ezekiel, we see also the influence of the 

'onian system. But the 300, 60, 30 ells t>ecame decimal at once when 

ferred to their value in Hebrew roods. So the Babylonian measure of 360 

becomes the Greek stade of 600 feet or 100 bpyviai. So the superficial 

^-'n• was loo ft. square. The Greeks took from Babylon their geometry 

V as it pertained to astronomy. 

'i2. pp. 593-637. An elaborate discussion of the templa of the augurs 
;. P. Regell, intended to serve as a sort of supplement to Nissen's work 
n the templum. Much confusion has prevailed because the different Umpla 
uere not distinguished. A comparison of the various statements of the an- 
cients shows that while the augur turned to the south for the observation of 
the lightning, he generally turned towards the east to observe the flight of birds. 
In Livy I 18 (the inauguration of Numa) the clause dextras ad meridiem partes^ 
laevas ad sepUntrionem esse dixit [originally dicit'\ is thought to be an interpola- 
tion from an old marginal note. Livy would have used the technical sinistraSy 
not laevas^ and the templum in question is the whole heaven, thus for the obser- 
vation of lightning and not of birds, as Romulus was declared king by light- 
ning on the left. At the close of the article the Umbrian templum is discussed 
in opposition to the views both of Br^al and Kirchhoff'. The discussion, which 
is accompanied by figures and full quotations from the monuments, cannot be 
condensed to a brief statement. 

93. pp. 637-40. K. P. Schulze criticises Die Elegien des Tibullus erklilrt 
von Fabricius. Berlin, 1 881. The edition evidently is intended for dilettanti 
of whom the editor is one. He has followed Baehrens blindly in spite of all 
warnings and has allowed important works to escape him. Schulze in the main 
commends F. Leo^s treatment of some elegies of Tibullus in Kiessling and Wila- 
mowitz-MoUendorfi^s Philolog. Untersuchungen II. Leo endeavors to follow 
in the path opened by Vahlen (Berlin Acad. 1878). He is successful in his 
characterisation of the poet and in tracing the development of thought in the 
first six elegies of the first book. 

94. pp. 641-50. R. Klussmann criticises Engelmann*s Bibliotheca Script. 
Classicomm 8th ed. I, 1880. Philology had a right to demand more careful 



236 AMERICAN JOURNAL OF PHILOLOGY. 

levEsion of the work. The editor needs a more intimate than ntual acqaaintance 
with (hehistoiy ofGreek and Roman liteTature,<:on<bincd with perfect familiarity 
with the ^/utoTuiii' of philoli^y and long years of training in the book trade. 
Thus Prenss, the editor of this revision, catalogues hooks which do not exist, 
does not distinguish properly between homonymous writers, and does not give 
with sufficient care the particulars, if an article has been republished. 

95- PP- 650-53. C. Fricic holds that h x<-^ypa^k fiwif, Sirabo II j, 17. 
does not refer, as has been thought of late, to the map of the world which was 
prepared by Augustus. The article is generic, used to denote the whole class 
of maps. The adject itc differs from ytuypa^adt only as inclading the details 
(irauii^/uiTa) of the situation of cities, nations, etc 

(57-) PP' 653~55- ^- Daring on Horace, Car. T 6. Kiessling, In Philolog. 
Unlersnch. II, i8Si,endeaTon to save the generally condemned fourth strophe 
(guit Mortem tunica Uettait aJamantitta, lU., cf. aiw^iihioi (uAtHri x'^^'^if) hy 
assigning to these scenes from the Iliad a symbolical relation with the rest of 
the ode. Doring goes much farther. The strophe assumes Agrippa's per- 
fect familiarity with the Iliafl. He remembers who were put to flight by 
Diomed ^ PaSadu, Horn. E 330 fg., 850 fg. Ares and Aphrodite correspond to 
Antony and Cleopatra; thus a Homeric allusion is introduced which has a 
highly characteristic reference to the achievements of Octavian and Agrippa. 
In Menones, Horace bad in view Horn. N 398-305, where Idomeneus and 
Meriones are compared with Ares and his son ^^of. With these now Octavian 
and Agrippa are compared. 

In the last line of the ode, nan pratlirieliltim lives, the negative belongs only 
iofrtuler soUltan as a liletii, a strengthened mere lelile. llie poet represents 
himself ai a true Anacre«ntist. 

X. 

96, pp. <i57-7a Fr. Sasemiht on the dale of the composition of the Phaedrus 
of Plato. In the JahrbQcher, iSSo, pp. 707 fg. (see Am. Joum. Phil. II Jjl) he 
had examined Usener's argnments in support of the view that the Phaedrus 
was written 403 or 4<» B- C. Since then Wilamowiti-M Ollendorff had entered 
the Geld with a series of new arguments for this view. Wilamowili says that 
the Phaedrus is the program of the Platonic dialogue, a new branch of litera- 
ture, and thus it is hardly conceivable that other diali^ues should have pre- 
ceded it. But. says Susemihl, how can a composition be a program of the 
ae in which there is no express mention of the dialectic form for the 
tatioD of thought, and which is not itself a pure dialogue, but in which 
.peeches txt introduced? To regard this as a program of Plato's work 
lutbot would be as ooe-sided as to regard it as a program of hii work at 
r. It is at least by no means a pn^rsm of the Sfcraiic dialogue. In it 
es is far from being a barren critic and intellectual midwife for other 
ideas. Krische held that the Phaedrus was composed during the life- 
f Socrates, and he strove to show that it presented a particularly faithfal 
t of the historic Socrates. Now (he attempt is made to persuade as that 
dest deviation from truth in its representation of Plato's muter is the 
rgnmenl for the early composition of the dialogue, even dnring his life- 
Bat is it conceivable that yean before Socratea's death, Plato, at the age 



REPORTS. 237 

of twenty-four, from the standpoint of his own theory of ideas, should put on 
the mask of Socrates to announce to the world what and how he will teach, 
both in his writing and orally? The Phaedrus is indeed a program, but a pro- 
gram of Plato^s metaphysics and logic, of his dialectics and theory of ideas. 

97. pp. 670-72. W. H. Roscher recognizes the hero Adristas (Pans. VIII 
4, i) as named from the art of weaving and preparing wool which Areas learned 
from him. (For such names cf. ^airuv, VL6,ttuv^ Kepduv^ cooks and butlers in 
Lacedaeroon.) The name is derived from irpiov (Attic ^rpiov), which Curtius 
derives from the root va, ' weave.* *ATpiar^ then would be * weaver.* For the 
d for r before p cf. *Arpia and *ASpia^ *ATpapvTTiov and ^ASpafiimov. 

98. pp. 673-91. M. Mailer offers linguistic and critical notes to Livy, 
Books XXIV-XXVI, a companion to his text edition in the Bibliotheca 
Teubneriana. 

99. pp. 692-96. H. ROnsch on cophinus and faenum, Juvenal III 14, VI 
542. In III 14, quorum cophinus faenwnque supeUex introduces the reader to 
the domestic life of the Jews who dwelt in the once sacred grove of Egeria. 
In VI the ludctea tremens is represented as she leaves this home for her busi- 
ness. {Jnterpres Ugum Solymarum^ as the teacher of the Mosaic law to the 
Roman women who were proselytes ; magna sacerdos arboris^ mockingly, in con- 
trast with the magnus sacerdos templi of Jerusalem, since she dwelt in the grove ; 
summa fida inUmuntia caeli^ probably as held in higher esteem as prophets 
than the augurs.) The basket and hay were characteristic of the Jews as Jews, 
and the correct explanation is preserved in an old scholion to VI 542, according 
to which the basket was for the keeping warm of cooked food and hot water 
for the Sabbath. The hay served to protect it and keep in the heat. The 
rabbinical strictness in the observation of the Sabbath and the rules for the 
keeping of food warm are well known. 

loa pp. 697-706. J. Beloch discusses the arrangement of the history of 
Timaeus. The work was divided into books by the author, and the number of 
books can be determined approximately. The history of Agathocles, covering 
23 years, was narrated in five books, but this was the story of his own time, in 
which he naturally would be most interested and of which he would write 
with the most details. The century from the Athenian expedition cannot then 
have filled more than 15 or 20 books. It is known that he described the Attic 
war in his thirteenth book, so twelve books remain for the earliest history of 
the West. Thus to the death of Agathocles we get a general estimate of 
35 to 40 books which agrees well with our citations, which go to book 38. The 
fragments are discussed and the conclusion reached that we can form a pretty 
satisfactory conception of the plan of the first half of the work, but from the 
seventeenth book to the end we have only six citations with the number of the 
book added. The following scheme is proposed for the work : I, till the cap- 
ture of Troy 1334; II to Ol. I, 776; III-VI, Greek colonization of the West, 
to about 01. L, 580; VII-IX, history of the West, to the time of Gelo, about 
01. L-LXX, 580-500; X, to the battle of Himera, Ol. LXX-LXXIV, 500-480; 
XI, to the overthrow of the Deinomenids, 01. LXXV-LXXVII, 480-468 ; XII, 
the democracy to the Attic war, Ol. LXXVIII-LXXXVII, 468-428; XIII, 
Attic war, 01. LXXXVIII-XCI, 428-412; XIV, first war with Carthage, Ol. 



238 AMERICAN JOURNAL OF PHILOLOGY. 

XCII, 412-408; XV, siege of Acragas. Ol. XCIII 1-2, 408-406; XVI-XXIV. 
Dionysius I and II, Ol. XCIII 3-CV. 405-356; XXV-XXVI I, anarchy, Ol. 
CVI-CVIII. 356-344; XXVIII-XXX. Timoleon, 01. CIX-CX, 344-336; 
XXXI-XXXIII, Oligarchy, 01. CXI-CXIV, 336-320; XXXIV-XXXVIII, 
Agathoclcf, 01. CXV-CXXII. 320-289; Appendix, Ol. CXXIII-CXXVIII, 
288-264. 

lOi. pp. 707-20. H. Peter commends warmly Ribbeck*s life of Ritschl. 

XI. 

102. pp. 721-31. M. Wohlrab defends himself against the charge brought 
by Schanz uf critical incompetence in the treatment of the text of Plato. 

103. pp. 732-39. Emendations to the text of Plato's Laws by K. J. Lieb- 
hold. 

104. pp. 739-40. F. Blass queries whether we have a work by Simmias of 
Thebes in an anonymous treatise published by Stephanus in an appendix to 
his Diogenes Laertius (by Mullach, Frag. Phil. I 544) under the title ^Avtwiffiov 
Tivbi SiaXt^tt^ AopiKy diaAi/crv). A reference to the disaster of Aigospotaroi as 
recent proves the date of the composition. From one passage the name of 
the author has been thought to be ^Ifuig, The treatise is identified by Blass 
with Nos. 5-1 1 of the works of Simmias as enumerated by Suidas. 

105. pp. 741-48. £. Rohde, *'Leucippus and Democritus once more.*' He 
inquires how it happened that Leucippus, who, according to Aristotle and 
Theophrastus, had developed the complete atomic theory, is not mentioned by 
Lucretius or Sextus Empiricus, or indeed by any one outside the circle of 
Aristotle and Theophrastus, while all the praise is heaped on the head of Demo- 
critus. It is unexampled in the history of Greek philosophy that the inventor 
of an original theory and founder of a school based thereon should have left 
no traces in the memory of his own school, and that Epicurus could deny that 
he had ever existed. Rohde thinks that certain works which were ascribed to 
Leucippus in the time of Aristotle were afterwards assigned to Democritus. 

106. pp. 748-52. Critical notes by O. Schmidt on Xenophon's Hiero, 
II and Vn. 

107. p. 752. H. R6hl. in Kaibel, Epigr. Graeca 706 (Welcker, Syll. p. 91), 
would read, not KiSuvo^ irAjT^tov, but Khduvnf (= Kvduvlov fi^Tuov) irX^iov, 

108. pp. 753-63. H. Schweixer-Sidler praises Havet, de Satumio Latin- 
orum versu, Paris, iS8a *' The book as a whole is so important, shows so much 
learning, thoroughness and acumen, is so instructive for the old Latin language 
and poetry, that slight criticisms upon it are not becoming.** 

(42.) pp. 763-65. Conjectural emendations by M. Hertz to Seneca rhetor. 
Apuleius, Fl. Vopiscus, Ammianus Marcellinus. 

(57.) pp. 766-68. E. Hoffmann on Horace, Car. II I, 4 fg. and 21, ntwAx 
crucribus he considers unlatin (we should expect uncta cntore) and would read 
functa cruoribus. In v. 21 he would read amUre . . . tndto for cmdire . . . videor, 

109. pp. 769-83. J. Woltjer, de ankiiypo qwfdam codice LMcreHant. After 
an examination of the six books he comes to the conclusion ** fhisse olim codi* 
cem, cuius paginae temos denos Tersus continerent, faunc codicem parum acca> 
rate transcriptum fuisse, multis locts schedanim supremas et infimas partes 



REPORTS. 239 

mutilatas et corruptas fuisse, vix ut leg^ possent, singulos autem versus dimidi- 
atos fiiisse, in binis lineis ut script! essent, haud inepte contend! posse videtur, 
huic exemplari lector quidam et Lucretii arte et Epicuri philosophia satis 
imbutus suas adnotationes inseruit.versibusquibusdam coroparationis causa alios 
adscripsit, aut ut poetam sibi ipsum contra dicere ostenderet.'atque singulorum 
locorum argumenta paucis verbis, interdum ex Epicuro sumptis, expressa in 
margine adnotavit, qui codex cum deinde transcriberetur ab homine nidi et 
imperito, fieri non potuit quin multi versus e margine in contextum irreperent 
et saepius in fine paginae adderentur." 

110. pp. 783-84. K. Dziatzko, vs. 648, 649 of the Hecyra of Terence 
belong after 654. 

111. pp. 785-801. E. Zamcke, on the so-called vocabula Graf canica in the 
titles of the odes of Horace ; a supplement to his dissertation on the same sub- 
ject (Strasburg, 1880). in which he had discussed these headings as found in the 
MSS. Here he treats of them as found in the old editions and finds his former 
conclusion confirmed, that they had been thrown overboard rightly, as useless 
pedantry of the schools of the rhetoricians. 

1 1 a. pp. 802-04. J. H. Schmalz "notices" Theilmann tlber Sprache und 
Kritik des lateinischen Apolloniusromanes, 1881. 

113. pp. 805-07. K. Welzhofer holds that the MS of Pliny which was 
bought by Cosimo de' Medici was not of the younger Pliny, as conjectured by 
Voigt, but was of the elder Pliny's works, probably cod. L. 

(83.) pp. 807-08. Miscellaneous conjectures to Latin authors by K. E. Georges. 

114. p. 808. W. H. Roscher reads nacfus for f actus. Veil. Paterculus 
1117,3. 

XII. 

115- pp* 809-15. H. Stadtmiiller. Conjectural emendations to the Homeric 
Hymns. 

116. pp. 815-16. C. Cron on Plato's Gorgias corrects a view which Hertz had 
expressed and shows that in 521 e the expression hnztp irpb^ JloXov iXeyov is 
exact, since 463 fg. Socrates is continually aiming at Polus although in conver- 
sation with Gorgias. 

117. pp. 817-23. H. Flach denies the Indogermanic origin of the Prome- 
theus m3rth, asserts that it sprang up on Greek soil, and that the original mean- 
ing of the myth is made certain by the et3rmology of the Titan's name, which 
cannot be separated from npofi^eta, irpofiijSff^, etc., and must be identified with 
1lp6vooc, whom the scholion on Thuc. I 3 names as the son of Deucalion. But 
in spite of the indications of Aesch. Prom. 85, 381, Pindar Ol. VII 44, Arist. 
Birds 1 51 1, we are expected to swallow Kuhn's theory whole. Perhaps in this 
myth two originally distinct versions, one from Peloponnesus, the other from 
Lemnos, have become mingled. The connection of the Heracles myth with 
that of Prometheus is of subordinate importance. It is conjectured that at the 
bottom of the Prometheus story was the worship of a Pelasgian divinity whom 
the Greeks identified with various local heroes, as ^paveic (perhaps the fire* 
thief), Ilpdvoof, llpofufiei}c. The Aeschylean Prometheus as son 'of Taia is 
avT6xduv,}}ist as the Argive Phoroneus according to Acusilauswas the first man. 



240 AMERICAN JOURNAL OP PHILOLOGY. 

tnd lite L«(niiuui Promelbent (hrougb faU murUge wiih Hesione f Asi^) tu 
conceived t* the prog;eaitoi' or ibe humsn race jost u Pboranciu at Aisoi. 

IiB. pp. S>3-34. Ad inphora in the Berlin Mnseum it declared by K. 
Wieieler to repreienl Heracleito the decks as aScytho-Germanic god. Hel- 
aclci, according to hii view, wa« not Phoenician, bat a German war aad nin 
god ; he went to Greece and there was changed into a hero. The name Hera- 
ctei and the words on the Taie are caplained by the writer from the German. 

119. pp. 825-31. F. Kern, Critical and exegettcal note* to Soph. AnL 392, 
601,1061 Tg. 

I30. pp. 831-33. A. Lowinifci. CoDJectaril cmendalioai to Aescfa. Sept. 
to-l3. to read {fdf il XP^ "''*'• ""l ^^ iUttmnrrd ri | ^^k ot/iaiat, tal riv l^sPmr 
Zpivif \ 0Xi<miii' It' iTtAiivmn-a o^/urot iroU, | Apov r' ijto*^ uaarm, iiaircp 1A9 
nplKti ktK 

lai. pp. S33-3S. H. Rampf finds a remarkable example of Ihe digamma 
on an inicrlption from the fint ccntary B. C, which was fonnd near Sebaslopol 
in the Chcrtoneie Heracleia, and lirst pnbliihed in iSSa to hvtya it found in 
the sense of rd Ivo'rovia, vttu liiatione iruKtla. Over the standi which cannot 
be taken as a correction for «, since no (race of u for ot nor of on for « is found in 
the inscription. Thit vwasused sometimes as a representative of the digamma 
is shown by Curlius, Gzee. $64 fg., and it is to be remembered that peculiarities 
of this kind would be retained longeT in the fonnulae of sacrifices than else- 
where. 

laa. pp. S39-4a W. H. Roscher in Caesar, Bell. civ. Ill 109, S. would 
read gmmim oiler tuctpto vtditert terpori (or rigort^ eccupatta per not pre xcitt 
ItMotui, alter inlerfeelus til. 

113. pp, S41-49. O. Wichnunn justifies his view of Schwan's work on the 
Demonax of Lncian, which was criticised in this volnme of the JahrbQchet pp. 
317 fg- by Ziegler, who disputed Ihe disordered state of ibc Dcmonai. Wicb- 
mann endeairon to prove interpolations and other evidences of revision. 

t>4- PP- Sso-s6. J. G. Cuno. Etrascan Studies : Tarqninius Priscns, 
Tullioi, Tanaquil. The original story probably brought Tarquinios 
cction with Corylhns, a country, not the city Cortona as is often snp- 
^oiytbos is the coantiy which was promised to Aeneas. See Verg. 
[O, X 719. Frotn Corythns to Corinthns was bat a step, and thence arose 
ibatTaTqaio'i fatherwas a Corinthian exile, ymnu must be the Latin 
lO of Ihe Etruscan Lucumo (cf. Livy I 34) which meant rtx. He is not 
in contrast with L. Tarquinius Superbus; be would be maiar in that 
In Livy I 33, the pritci LaUm are the ciliieos of Latiam with fall 
irrespondiug to the p^tdm Rfmanta Qmr iti tm t . So also Servins 
tias nothing todo with Mnnu.bnt is a liaosUtion of the Etmscan Mas- 
uidification of the Latin magUler with the soffix -w>. which is very 
in Etruscan. From Fabretti's Corpus Insciipt Consen citei matitrr, 
with magiiUr. So Tanaquil is said 10 have received in Rocne anothei 
«cilia. Bui this is mily a translation of her other itame. It is con- 
lat Ihe myths pertaining 10 these three persons were bioogbl to Roaiaa 
tnscan conquerors and colonists, and were adopted there as 10 mnch 
le Etrascan civiliiatioa was adopted. 



HEPORTS. 241 

(18.) pp. 857-67. F. Hankel, the ancient Roman camp according to Poly- 
bias, a reply to Nissen*s article, pp. 129 fg. of this volume of the JahrbQcher. 

125. pp. 868-70. Ph. Thielmann commends Heerdegen's Untersuchungen 
zur lateinischen Semasiologie, speaking in detail of the use of orare, 

T. D. Seymour. 



Weisks on thb Articular Infinitive. 

The pedagogical section of the JahrbQcher, which is edited by Professor 
Masius, does not fall within the scheme of these reports, although the articles 
are often of considerable interest, both theoretical and practical. In the volume 
for 1882 (pp. 494-504, 529-42) Dr. G. A. Weiske, of Halle, has undertaken to 
collect and arrange according to the categories of Koch*s grammar, the articular 
infinitives occurring in Plato, Thnkydides, Xenophon and the Attic orators. 
He has not gone into the matter of proportion as I have done, so that his paper 
does not serve to correct the results reached by my pupils and myself. See 
Transactions of the American Philological Association, 1878, and Am. Joum. 
of Phil. Ill, 193-202. Nor has he noticed the occurrence of such rarities 
as the fut. articular inf. and the articular inf. with dv. The nominative 
infinitive he considers to have little grammatical interest, a point in which 
many will not agree with him, as the transition of the inf. from dat. to ace. and 
from ace. to nominative deserves the most careful study, and would have been 
furthered by a complete list It is doubtless true, as Dr. Weiske says, that the 
development of the articular infinitive checked the development of the abstract 
substantive in Greek, but we must remember that abstract substantive and 
infinitive do not cover each other, and when we look over the complete list 
that Dr. Weiske gives of the articular accusative infinitives, we find that a very 
small percentage of the verbs cited have not a corresponding abstract substan- 
tive. 

Of the advantages which the infinitive has over the abstract subst. I have 
written already, and I will not repeat here what is tolerably evident to any one 
who thinks on the subject. Dr. Weiske speaks of the * leichtigkeit des satz- 
baues * attained by the use of the articular infinitive. If he means by ' leich- 
tigkeit ' ' compactness ' I should agree with him, but I have shown sufficiently 
that the articular infinitive is really a norm of artificiality and that an excessive 
use of it toughens the style.* On the chapter of the use of the abstract sub* 
stantive and the articular infinitive in combination. Dr. Weiske has touched 
but lightly. He says the articular inf. is often combined by means of Kai 
with a preceding substantive in the teme case, the infin. giving the narrower, 
the subst. the wider sense. This, he says, is especially common in Dem- 
osthenes. So 8, 13: n^ liiv Ix^P^^ '^ ^^ PobX^a^at Ku!khttv\ 18,296: n^ 
d* iXtv&eplav ital fitfdiva i;t^iv deairdr/pf airruv ; 20, 45 : li^ npo^v/ilav Koi t6 avrbv 
ivayyeiXdfievov iroulv; 57, 2 : rbv itaipbv 6k naX rb irapo^bv^cu t^ ndXtv npb^ t^ 
amnfrrf^iaet^. He might have added that this is in exact conformity with 
Demosthenes* way of working out a problem before his audience, who are 

1 DIonyiiot makes a similar remark about the excessive use of the gen. absol. in Isaloe (De 
Isae. lud. p. 598). 



rSR^AZ OF PHILOLOGY, 

ni OS he fieels his way to a just expression. The 

. . .vatmt any regard to period or department or author, 

., J^4^^J. need to be woriied over again to get an historical 

.'t i^Kc [he school grammar should enter under the accosa- 

. , .s .* ..Q -^yuiiv^ 'Xvyetv^fvXdrrefrdaif ^peia^ai^ KaXetv, vofd^eof, 

. . .^lerenctf ' as to* ; under the genitive the complementary 

. .3 ...a ..•I'vu/AiO, dJeuZy dneipia^ dd^a^ divafiiq, t&tofi6c, e^ovoia^ 

■ ■.it*titf; /itt, TpofrwTtf, ffiffuiav, racfi^ptoVf <^^ and xP^^'i 

■> . <»(.iuiy combined with the gen. of the art. inf. (uTiaa^atf — 

— t^- titty — km&v/jteip •" aTToarepeiVf airixeiv, offaAAdrrciv, 

,,^ .le aajettives oZriof, Ki;f.toc^ i^toc. He is right when he 

. ^ .... ."UipArison occurs frequently in the articular infinitive, 

... . > . :e iccommendation of the gen. absol. which is a compar- 

.-. uc.iuQ, lor reasons given, Am, Jour. Phil. Ill 198. The 

. . . ,i.c 1 hiik. 3, 2, 3; Lys. 12,13; Isokr. 3, 6 ; 6. 3 ; Xen. Mem. 

- .>uein. 265 E, Gorg. 509 E; Polit. 310 E; Kriton 49 D; 

.'cui. 5. 2; 20,25; 23. 13; 25,17; 61,28. Against these we 

^v. .luuiocr oi ace. participles with the simple infinitive, which 

. . .i^.iucuoQ. The dative artic. inf. is far less common than the 

c c>v oi reasons (see Am. Joum. of Phil. Ill 201). The inf. is 

! '^t^ iuo>t important use is the dative of cause. Dr. Weiske 

>. .X ic combinatioa with cvavrlog, dun^ipetv, XPV^^<^^ 

^ .^. .11 ;'ie<H>!ii lions with the articular infinitive is not without its 

'I V.V .a^c ihe plastic uses of the prepositions are excluded. There 

. v. 1 .N xvaice enottgh, in prose, as it is), no xard with the gen., no 

..v» .i.le j^cope in prose), no imip with ace, no ^rrpi with dat. (which 

... .V. .iL any rate), no irapd with gen. or dat.; napd with ace. is rare. 

u%c 'earned to exclude practically from model Attic prose, is 

>y W . 1 rom Dem. 8, 65 : fi^ avv ev irenotr^oTuv ruv iTokXMv *OAw- 

.^ v..^ T4uat><u, but avv ei Ktjrovddruv is a rude quasi-compound 

. .. aud the dative depends on the totality. 

B. L. G. 



» V 



CORRESPONDENCE. 

Sir : I ask a little space for reply to some of the criticisms in a review of my 
translation of " Beowulf " contained in No. 13 of your Journal. For the gen- 
erally favorable opinion expressed and for some of the bibliographical references 
I am much obliged, but, without going into details, I may say that I do not 
think it necessary to include in the bibliography of a work every book that in 
any way makes reference to it. As to the difficulties of Anglo-Saxon inversion 
for the general reader, they are sufficiently commented on in my preface, and 
as to the Unwdrter charged, none are specified, and I do not think that any 
words used would be unintelligible to the general reader except the dozen or 
so for which a glossary is given. 

The reviewer would seem to require that a line-for-line translation, which 
tries to preserve two accents to the half-line, should follow the principle of the 
Revisers of 161 1 rather than that of the Revisers of 18S1, as he objects to ** the 
perpetual recurrence of such words as ' victorious,' * jewel,' * treasure,* *' and 
thinks *' a subtler insight would have perceived picturesque shades of meaning." 
When it is recollected that of sige and sigor = ** victory," there are over a dozen 
compounds, of sine = "jewel" or " treasure," at least ^vn^oimd^um = "jewel** 
no less than fourteen, and of hard = ** treasure,** about twelve, besides the 
*• perpetual recurrence " of these words by themselves, I think the " subtler 
insight ** would have been overburdened, and it were better to decline the 
attempt to find " picturesque shades of meaning." 

I cannot notice each one of the thirty-Jive references in which, for one reason 
or another, exception is taken, and in some of which I concede that the correction 
is more exact than the word or phrase used ; but many of them touch very small 
points, often silently passed over in Heyne*s translation (which, being in ten- 
syllable iambic measure, is much freer), some are due to the failure to use 
Grein*s text, the one translated, in the comparison instead of Heyne*s, and 
others are inadmissible. . 

Some examples of each of these will be given : (i) Exception is taken to the 
omission of the particle hAru in 182 and 369. This important word = saltem, 
quidem, certe, yk^ and is omitted entirely by Heyne in both passages. I find 
that I have translated it ** now,** '* indeed,** in seven other passages, but while 
my translation professes to be line-for-line, it does not profess to be word-for- 
word, and I should not have considered it much of a blemish if I had omitted 
it in all of them. In one of these, 862, where I have " now,*' the reviewer 
corrects to ** nevertheless,** Heyne's glossary giving doeh^jedoch for this refer- 
ence. Let us substitute it and read the line : 



i< 



They did not nevertheless at all their dear lord blame " (!) 



Surely some license in particles may be allowed to even a line-for-line trans- 
lator. 



24* AMERtCAN JOURNAL OF PHtLOLOGY. 

This ii Ihc only posiage ia which I follow Heme's explaDitioD io text and 
relegBte Gcein's to tlie cotn. Heyne has in gloisftij " inep lA/t, mit d«ta 
koilbaicD Schwerte? odei mit wtichtigem Schwerle?" and in hia tmnslation 
"mit dcm wachtigen Stkhl," hence "weighty " wonld have been better thai) 
"mighty." Grein follow* Thorpe in taking itugc as a proper name =i log, 
King of the Danes, but then we must read Ifgtli] with Thoipe, who nys: 
" My interpretation is quite conjectural, the word inegt being unknown to me," 
and he translates " with Inge's Tetic." Amold, too, follows Thorpe, translating 
" with the Dane's (?) bequest," but %a.y% : " No one has suggested an explanation 
for inegt." The reviewer snggests "with the eiig'e of the sword," to which 
translation (here would be no objection i/ there was any authority for the 
reading, but the scribe uses ecg in the same line and might easily have written 
etg< here, if that were the reading, and Holder gives plainly iW gt laft (p. sg, 
1870, 1. It), BO that ecgt mutt be rejected, and inegi still awaits a satisfactory 
expianalioD. T.-B. has not yet reached the word. Heyne suggests a connec- 
tion with Ugt gold, 1107, :^"Schat^old, reiches Gold?" but that word is 
equally as nnknown. Wackerbarth translates "And with his mighty Relic 
Brand." Etlmuller follows Thorpe and Grein.' 

To shorten this reply, I notice lastly only 3S30, dim ^ " doom," which trans- 
lation might be "ambiguous" if the context did not show plainly what wa* 
meant, but it is certainly literal. It means here, of course, " heavenly glory," 
" Hertichkeit " (Grein and Heyne), "iibMnof the jnst" (Thoq>e). "idwmof the 
•oolhfast" (Arnold), bat it does «0f mean "realm," nor will any support for this 
rendering be found in T,-B.,q, v., s.t. Ill, p.307, where "numerous references" 
for ddm ^ "glory" may be found. 

I am obliged for some of the references, as they wilt enable me to supply 
further notes, but I am glad that the philological miscroscope, even when of 
ilrong magnifying power, has been able to detect so few " inaccnracies." 

James M. Gaknttt. 

■ If ibe llsu an v&W opeu, t would auueit iiitc-pI4/i (cf. m^^um-muiTd, icnj), allller- 
biIdi with '<••, which alIl»rHIioB,a]th(ni(n ran, li>l[[lidmlu[ble: 1 at IxcinDing of Ihe MS 
UncntghihiTebecndroppeduroidily u>rterf( hI [bio the Hoc. Tbc fac>imllcaflh« Beo- 
vulfMS.JuiIpubliihcd by the Early Eniliih Tm Society.readi inc^^/i.Eppuently uooa 
word. Thorkelln prinled/K /i/n/i.bul RlUlnlciprElsl the puuge; Onindtvii, iKfi-^<, 
butlUfgctti Ingvina U/t I KEmble ilyl In glmuT. >, t. Idf, " incft-Za/, aaii. I CUUWI 
eiptaLa the linl ward, snd believe it 10 be a conuptioD at icfft-fd//' and hegivci "hgtf vef- 

." with 1107 ai the aole nfanuc*. The con)eclure baa, at lesutithe 

r inielliglblc. 



RECENT PUBLICATIONS. 

Thanks are due to Messrs. B. Westermann & Co., New York, for material 
furnished. 

AMERICAN. 

Horace, Horati Flacci Opera, Parch. L. AppUton. $1.25. 
Xenophontis libri Socratici. Harper, 50c. ; pap. 32c. 



BRITISH. 

iEschylus. Suppliant Maidens, translated into English verse by E. D. A. 
Morshead. Post 8vo, pp. 60. Paul^ Trench 6* Co. 3s. 6d. 

Bentley (Richard). Dissertations on the Epistles of Phalaris, etc. Edited, 
with an Introduction and Notes, by the late W. Wagner. (Bohn*s Classical 
Library.) i2mo, pp. 620. Bell &* Sons, 5s. 

Demosthenes. First Philippic. With an Introduction and Notes. Ed. after 
C. Rehdantz, by Rev. T. Gwatkin. i2mo, pp. 104. Macmillan, 2s. 6d. 

Evans (George). An Essay on Assyriology. Published for the Hibbert 
Trustees. 8vo, with 4to tables of Assyrian inscriptions. Willams &* Nor- 
gaU, 5s. 

Frankfurter (Dr. O.) Handbook of Pali : being an Elementary Grammar, a 
Chrestomathy, and a Glossary. 8vo. Williams 6r* Co. i6s. 

Gardner (P.) The Types of Greek Coins : an Archaeological Essay. Folio, 
pp. 350. Cambridge Warehouse, 31s. 6d. 

Geddet (W. D.) Flosculi Gneci Boreales ; sive Anthologia Gneca Aberdo- 
nensis. 8vo, pp. 264. Macmillan. 6s. 

Geldart (E. M.) Simplified Grammar of Modem Greek. Post 8vo, pp. 66. 
TrUbner. as. 6d. 

— ^ A Guide to Modem Greek. Post 8vo, pp. 386. Trabner, 7s. 6d.; 
Key, 2s. 6d. 

Livy. Books 2i>25. The Second Punic War. Translated into English, with 
Notes, by Alfred John Church and William Jackson Brodnbb. With maps. 
Post 8vo, pp. 370. Macmillan, 7s. 6d. 

Persian Wit and Humour: being the 6th Book of the Baharistan of JimL 
Trans., for the first time, from the original Persian into English prose and 
verse, with Notes, by C. E. Wilson. Crown 8vo, pp. vii, 40. Chatto dr* Win- 
dus, 

Plato. Theaetetus. With a Revised Text and English Notes by Lewis 
Campbell. 2d ed. 8vo, pp. 296. Frowde, los. 6d. 

Rockhill (W. W.) Udanavarga ; a Collection of Verses from the Buddhist 
Canon, compiled by DharmatrOta, being the Northern Buddhist version of 
Dhammapada^ Translated from the Tibetan of the Bkah-hypur, with Notes 



248 AMERICAN JOURNAL OF PHILOLOGY. 

and Bxtncts from the Cominentaij of PndiDaraniun. (TrObDci'i OiicnUl 
Seriei). Sva, pp. zja. TnAjttr. 9*. 

Sophocles. Seven Playi In English »eise bj Lewis CampbeU. Post 8»o, 
pp. 433. Favl, Trtnth &• Co. ^*. 6d. 

Vinaya Pitakam, one of the principal Bnddhiit Hoi; Scriptures in tbe Pali 
langnage. Edited by Dr. H. Otdenberf. Vol. 5, The Parivara, and Indices. 
8»o, Orient. lAbrary. ais.; Complete in J toI*., ^5 %%. 

DANISH. 

Soronsen (S.) Om Mah&bhirat>'s SliUiag i den indiske Lileratnre. I. 

Fonog pa at udakille de sidste Bestanddele. KJabtnhavn, 1SE3. 8vo, 393 pp. 

FRENCH. 
Arbois de Jnbainville (H. d'). IntroductioB i I'etude de la litt^ature celtiqne. 
In-B. Tkarin. 8 fr. 

Br^toD (G.) Essai lar la po&ie philosopbiqne en Grice : X^DOphane, Par- 
m^nide, Emp^docle. In-B. Hachtltt. 5 fr. 

Chabai (F.) Cboix de textes ^pliens. Traductions inMiles de Fran(oii 
Chabas, publieei par P. J. de Horrack. Gr. in-B. KHniksitck. 6 fr. jo. 

Chodzko (A.) Grammaire de la langue persane. a <!d. augment^e de textes 
persans inMils et d'un glonaire. In-ia. Maiionnaaie. La i Mition est de 
1853. Cart.', 10 fr. 

Guyard (S.) Le Divan de Beba Ed-Din Zoheir. Variante* au texte arabe. 
In-ia. Maiscmtuftvi. a fr. jo. 

- — Melanges d'assyriologie. Notes de texicographie asiyrienne, saivie* 
d'nne Aude sur les JnBcripttons de Van. In-8. Maisfftttuuve, 5 fr. 

Joiet (C.) Des caractjres et de I'extensian du patoii nonnand. £tDde de 
phon^ique et d' ethnographic. In-S. Vieweg. 6 fr. 

Regnaud (P.) Nouvcaux aperjut sar le vocabulisine indo-eorop^n, prM- 
des d'une analyse critique de* systimes actuellement en vigueui. Gr. in-S. 
VitvHg. 1 fr. 50. 

Let incunables oiientanx et les impressions oneniales an 
du XVI sijcle. Rapport tur une mission en Bavifte et en Wnr- 
lifaitimtiatve. 6 fr. 

GERMAN, 
m f. die Knnde d. Morgenlandet. hrsg. v. der Deutschen Mor- 
Ischaft nnter der Red. d. Prof. Dr. E. Windisch. S Bd. Nr. a. 
Bnttliaui' Sort in Ctmm. Inhalt: Das Aupapitika SDtra, eiste* 
ina. I Thl. Eioleitnng, Text n. Gloisar. Von Dr. Ernst Len- 

n.6. 
.. G.) Ob. dentiche Volksetymologie. 4, ttaA Tcnn. Anfl. gi. 
Heilbronn, Httaangtr. n. %. 

Bedentung n. Gcbranch t. iti bei Homer. Inangnral-Disser- 
FQ S. Mnnchen, SlaU. n. 1.5a 

Ne attiscbeo Diaslen. Ein Beitng tnr griech. Heonolo^e. 
;tlin, Caertiur. n. i. 



RECENT PUBUCATIONS. 249 

Berber (F.) Ob. die Heerstrmssen d. r6misdien Reiches. II. Die Meilen- 
steine. gr. 4, 21 S. Berlin, Gaertner, (&) n. i. 

Bezold (C.) Die Schatzh6hle, syrisch u. deutsch hrsg. I TL A. u. d. T. : 
Die SchatzhOhle, aus dem syr. Texte dreier unedirten Handschriften in's 
Deutsche Qbers. u. m. Anmerkgn. veneheiu gr. 8, xi, 82 S. Leipzig, Hittrich^ 
VerL n.n. 8. 

Blass (F.) De Gemino et Posidonia gr. 4, 35 S. Kiel, Umvertitm-B^hh. 
n. 1.2a 

Bohn (O.) Ob. die Hexmat der Pnltorianer. gr. 4, 24 S. Berlin, Caertner. 
n. I. 

Bdhtlingk (O.) Sanskrit- WOrterbach in kilrzexer Fassung. 4 ThL i Lfg. 
gr. 4, S. 1-160. St. Petersbaig, 188a. Leipzig, Voss* Sort, n. 4.aa I — ^IV, 
I : n, 31. 

Brandscfaeid (Frdr.) Der '* Odipas Tyrannos " d. Sophokles u. seine neuesten 
Kritiker. gr. 8, 22 S. Wiesbaden, RtdrioM, — 75. 

Braumann (G.) Die Principes der Gallier a. Germanen bei C&sar a. Taci- 
tus, gr. 4, 44 S. Berlin, W, Weber, baar n. i. 

Ciceronis (M. TuUi). De legibas libri ex recognitione Job. Vahleni itemm 
editu gr. 8, xxit, 208 S. Berlin, VahUn, n. 4. 

*— Tnscnlanamm disputationnm lil^i V. Far den Scfanlgebrauch erklirt ▼. 
L. W. Hasper. i Bdchn. Buch 1 a. II. gr. 8, iy, 114 S. 1.20. 

Demosthenes, ausgewahlte Reden. Fur den Scbulgebrauch erklirt ▼. J. 
SOigeL I Bdchn. Die drei ol3mth. Reden a. die i Rede gegen Philippos 
enth. iv, 95 S. Gotha, F, A, Perthes, 1.20. 

Diels (H.) Theophrastea. gr. 4, 28 S. Berlin, Gaertner, n. I. 

Dissertationes philologicae Halenses. Vol. V, 2 partes, gr. 8. Halle, Nie- 
meyer, n. 4. I-V, 2: n. 35. Inbalt: I. De seriris Augustalibus. Scripsit 
Joa. Schmidt Adjecta est tabula (lith). viii, 132 S. 1878. 2. S. 135-364. 

Edler (O.) Quaestiones Sertorianae. Dissertatio inauguralis. 8, 42 S. Her- 
fordiae. 1880. Berlin, W. Weber, baar n. i. 

Ellger (G.) Die Zusfitze zu dem Pro5mium der Hesiodischen Theogonie. 
[Vers 36-115]. gr. 4, 20 S. Berlin, Gaertner, n. I. 

Enderlein (Osc.) De M. Antonio oratore. Diss. gr. 8, 46 S. Leipzig, 1882. 
Feek. baar n. i.aa 

Erbe (K.) Hermes. Vergleichende Wortknnde der latein. u. griech. Sprache. 
qu. 8, iy, 244 S. Stuttgart, Neff, geb, n. 1.35. 

Forcellini (A.) Totius latinitatis lexicon. Pars altera sive onomastioon 
totius latinitatis, opera et studio Dr. Vine De-Vit lucubratum. Distr. 21. gr. 4. 
3 Bd. S. 1-72. Prati. Leipzig, Brockhau^ Sort, (i) n. 2.50. 

Goebel (Ed.) Exegetische u. kritische Beitrage zu Platons Apologie u. 
Kriton. 4,2oS. Fulda, 1882. V^At^^m^ F, SchSninigh, n. — 80. 

Gruenwald (Eug.) Quae ratio intercedere yideatur inter Quintiliani institu- 
tionem oratoriam et Taciti dialogum. Diss. gr. 8, 57 S. Berlin, Mayer A* 
M Alter, baar n. 1.20. 

Herzfeld (L.) Einblicke in das Sprachliche der semitischen Urzeit. gr. 8, 
231 S. Hannoyer, Hakn, s. 6. 

Hitzig (H.) Studien zu Isaeus. 4, 29 S. 'BtTti.Jent &* Reinert, n. X. 

Ihm (Geo.) Quaestiones syntacticae de elocutione Tacitea, comparato 



250 AMERICAN JOURNAL OF PHILOLOGY, 

Caesaris, Sallnsti, Vellei nsa loquendi gr. 8. 77 S. Giessen, 1882. Richer. 
n. i.6a 

Isaei orationes cum fragmentis a Dionysio Halicamassensi senratis ed. H. 
Baennann* gr. 8, xiv, 156 S. Berlin, Weidmatm. 2.40. 

Jannaiakis (A.) Dentsch-neiigriechisclies HandwOrterbuch. Unter besond. 
BerQcksicht. der nengriech. Volkssprache bearb. 3 Abthlgn. 8, viii, 1372 S. 
Hannorer, Hakn, n. 8. 

Jordan (H.) De formae nrbis Romae fragmento novo disputatio. gr. 4, 10 
S. m. I Steintaf. Romae. Berlin, Asker 6f* Co, baar noi. 2. 

Karabacek (Jos.) Der Papjrmsfund ▼. El-Faijiim. Die textilen Graberfonde. 
Ein Voitrag. gr. 8, 43 S. Wien, GeroU 5* Co, im Comm. n. 1.2a 

Klatt (M.) Chronol<^ische Beitrage tax Geschichte d. achiischen Bandes. 
gr. 4, 42 S. Berlin, Gaertmer, n. i. 

Kleinschmidt (Max.) De Lncili Satnramm scriptoris genere dicendu Com* 
mentatio ab ordine philosophomm Marpoxgensium praemio omata. gr. 8, fiii, 
135 S. Marbnig. Elwerts VerL n. 2.8a 

Manns (P.) Die Lehre d. Aristoteles ▼. der tragischen Katharsis a. Hamartia, 
erklirt gr. 8, 86 S. Karlsruhe, Reuiher. n. i.8a 

Marx (Ant) HolfsbOchlein f. die Ausspracfae der lateiniscben Vokale in 
positionslangen Silben. Mit e. Vorwort ▼. Fn. BQcheler. gr. 8, xii, 8aS. 
Berlin, fVMmann, n. 240. 

Matthias (Frz.) Qaaestionum Blandiniananim capita III. Dissertatio in- 
flAgnralis philolc^ca. gr. 8, 72 S. Halis Sax. 1882. Berlin, Afayer &• MuiUr, 
baar n. 1.20. 

Meister (Rich.) Zor griechischen Dialektologie. I. Bemerknngen zur 
dorischen Accentnation. II. Die Excerpte irep2 diaXiKTuv, namentlich in Bezng 
anf die Abschnitte nepi Aupidoc. 4, 16 S. Gmtingen, VatuUmkoek ^ Rtipreckfs 
VerL n. — 8a 

Meibot (Rhold.) Aesthetische Stndien zar angelsichsischen Poesie. gr. 8. 
iii, 51 S. Breslan, Koebner. n. 1.5a 

HoU(Ladw.) De temporibas epistnlamm Tallianamm. Qoaestiones selectae. 
Diss. gr. 8, 57 S. Berlin, Mayer &* M&lkr, baar n. 1.2a 

MQlIer (Thdr.) Angels&chsische Grammatik. Aos dem handschriftl. 
Nachlasse d. Verf. hrsg. ▼. Dr. H. Hilmer. gr. 8, xi, 257 S. Gmtingen. 
Vandenhoeck 6f* Rttpreeht, n. 4.4a 

Munxel (Rob.) Qoaestiones mythographae. gr. 8, tI, 25 S. Berlin, Wdd* 
wuum, n. 1.2a 

Nttsche (Wilh.) Der Rhetor Menandros n. die Scholien zn Demosthenes, 
gr. 4, 26 S. Berlin, Gaertmer, n. I. 

Pannenborg (A.) Der Verfasser d. Ligorinns. Stndien zn den Schriften d. 
Magister Ganther. gr. 4, iv, 39 S. G<Mtingen, PefpmuUer, n. 2. 

Pancker (Carl). Supplementnm lexicomm latinonim. (In ca. 8 Fasc) 
Fasc I. gr. 8, 96 S. Berlin, Calvary ^ O. n. 3. 

Vorarbeiten znr lateinischen Sprachgeschichte. 3 Tie. i L^. gr. 8. X 

TL S. 1-64 n. 2 TL S. 1-16. Ebd. n. 3. 

— Bilaterialien znr lateinischen WOrterbildaiig^eschichte. I-III. gr. 8. 
Berlin, Cahfory 6* C?. n. 4.2a 

Periplos, der, d. Eiythriischen Meeres t. e. Unbekannten« Griediisch n. 




RECENT PUBLICA TIONS, 25 1 

deatsch m. krit. a. erklSr. Anmerkgn., nebst vollst&nd. WOrterverzeichnisse 
V. B. Fabricius. gr. &, iii, 188 S. Leipzig, Veit &* Co. n. 6. 

Plato's ausgewahlte Dialoge. Erkl&rt v. C. Schmelzer. 3 Bd. Phddo. gr. 
8, 118S. BtxXm, Weidmann, 1.20. (1-3, n. 3.20.) 

PUton's Apologie d. Sokrates a. Kriton. Fttr den Schulgebranch bearb. v. 
Dr. Ed. Goebel. gr. 8, xvi, iia S. Paderbom, F, Schdningh, n. 1.20. 

Preller (L.) R^mische Mythologie. 3 Aufl. v. H. Jordan. 3 Bd. gr. 8, 
xi« 490 S. Berlin, Wtidmann, (h) n. 5. 

Raehse (Hugo). Grammatisch geordnetes hebraisches Vocabularium. gr. 
8, iv. 42 S. Halle, Niemeyir, n. — 60. 

Roehl (H.) Imagines inscriptionum graecarum antiquissimarum in usum 
scholarum composuit H. R. gr. 4, iii, 72 S. Berlin, (7. Reirner, n. 4. 

Schiche (Th.) Za Ciceros Briefen an Atticus. II. gr. 4, 24 S. Berlin, 
Gaertner, n. I. 

Schroeter (Rob.) Quas fonnas nominam themata sigmatica in vocabulis 
compositis graecis induant. 8, 95 S. Cothene. Leipzig, Hinrichs* Sort, 
baar n. i. 

Sophoclis Oedipus rex, scholarum in usum ed. Frdr. Schubert. 8, xiv, 54 S. 
Prag, Tempsky, Leipzig, Freyiag. n. — ^40. 

Studien, altitalische. Hrsg. v. Dr. Carl Pauli. I Hft. Mit I lith. Taf. gr. 
8f viii, 72 S. Hannover, Hahn, n. 3. 

Studien, Wiener. Zeitschrift f. class. Philologie. Suppl. der Zeitschrift f. 
Osterr. Gymnasien Red. : W. v. Hartel, K. Schenkl. 5 Jahrg. 1883. 2 Hefte. 
gr. 8, 1 Hft. 174 S. m. i Lichtdr.-Taf. Wien, Gerokts Sohn, n. 10. 

Texte, rhktoromanische, hrsg. v. Jak. Ulrich. I. Vier Nidwaldische Texte. 
8, vii, 195 S. Halle. Niemeyer, n. 3.60. 

Tiedke (H.) Nonniana. gr. 4, 24 S. Berlin, Gaertmr, n. i. 

Tirocinium philologum sodalium regii seminarii Bonnensis. gr. 8, iv, 135 
S. Berlin, Weidmann, n. 3. 

Veigili Maronis (P.) Aeneis. FQr den Schulgebrauch erklart v. Oberlehr. 
Dr. Osk. Brosin. i Bdchn. Buch I-III. viii, 252 S. 2.40. 

Wagner (Rich.) Quaestiones de epigrammatis graecis ex lapidibus collectis 
grammaticae. gr. 8, vi, 127 S. Leipzig, HiruL n. 2. 

Westphal (Rud.) Die Musik d. griechischen Alterthumes. Nach den alten 
Quellen neu bearb. gr. 8, vi, 354 S. Leipzig, Veit &* Co, n. 9. 

Zeitschrift f. figyptische Sprache u. Alterthumskunde. Hrsg. v. C. R. Lepsius 
unter Mitwirkg. v. H. Brugsch, Ad. Ennan u. L. Stem. 21 Jahrg. 1883. 4 
Hfte. hoch 4. i Hft. 40 S. m. 2 Steintaf. Leipzig, Hinrichi Verl, n. 15. 

SPANISH. 

Picatoste (F.) Diccionario popular de la lengua castellana. Madrid, 1883. 
8vo, 1059 pp. 8 m. 

SWEDISH. 

Piehl (K.) Dictionnaire du papyrus Harris Nr. i, public par S. Birch d*apr^ 
Toriginal du British Museum. Upsala, 1883. 8vo, 116 pp. 22.20 m. 



2S2 AMERICAN JOURIfAL OF PHILOLOGY, 



BOOKS RECEIVED. 

Altenglische Bibliothek henasg. ▼. Eagen Kolbuig. Ecster Band. Osben 
BokenAin'f Legenden heraosg. ▼. C. Hontmans. Hellbcoiia, G^, Bemmimger^ 
1883. sm- 6opf. 

Aiistotle*s Psychology, Greek and English, with introdoction and notes. 
By Ed. Wallace. Cambridge UmverHiy Press, 1882. 

Baitholomae (Chr.) Handbnch der alteranischen Dialekte (Kongefasste 
' vergleichende Graromatik, Lesestttcke a. Glossar). Leipsig, Brtitkopfu. H§rtel^ 
1883. 

Branmann (Gnatay). Die Principes der Gallier a. Gennanen bei Caesar n. 
Tacitus. Johresber. ttberdas K6nigL Friedrich-Wilhelms Gymnasinm za Berlin. 
Ostern, 1883. 

Brugmann (K.) Ueber griechische dpa, dp, ^ a. litauisch ia. Berichte der 
k. sHchs. Gesellsch. der Wissenschaften, Phil, histor. Classe, Sitzung am 33 
April, 1883. 

Crane (T. F.) Mediaeval Sermon books and stories. Read before the 
American Philosophical Society, March 16, 1883. 

Kinch (C. F.) Quaestiones criticae Cortianae. Hanniae, In Libraria GyU 
dendaliana, 1883. 

Klussmann. Curae Africanae. Gera, R, KUuUrmam^ 1883, 

KOlbing*s Englische Studien VI, i, 2. Heilbronn, Gehr. Henmnger. 1883. 

Livy. Books XXI-XXV. Translated into English with notes. By A. J. 
Church and W. J. Brodribb. With Maps. London, AfaaniUan &* Co, Baltimore. 
Cushings &* Bailey, $2. 

Pindar. The Nemean and Isthmian Odes. By C. A. M. FennelL Cambridge 
University Press, 1882. 

Siiss (Eduard). Die Sintfluth. Eine geologische Studia. Sonderabdruck 
aus : Das Antlitz der Erde. Prag, F. Tempsky. Leipzig, G. Ffcytag, 1883. 

Wallace (Ed.) Outlines of the Philosophy of Aristotle. 3d ed. (enlarged). 
Cambridge University Press, 1882. 

Wiener Studien. Supplement der Zeitschrift fUr 6sterr. Gymnasien. IV 
X, 2. Wien, Carl GerokTs Sdhne, 1882. 




AMERICAN 

JOURNAL OF PHILOLOGY 



Vol. IV, 3. Whole No. 15. 

I.— THE NEW REVISION OF KING JAMES' REVISION 
\ OF THE NEW TESTAMENT. 

IIL 

Examination of the Revision of S. Matthew {continued). 

Ch. VIII. V. I. And (pi) whenyOk^T Sir John Cheke and Rh. ; 
But whanne, Wycl. ; When, A. V. omitting the introductory par- 
ticle, after Tynd., Cran., and Gen. — v. 2. there came to him^ by a 
change of text after Lachmann, Tischendorf, and Tregelles ; there 
came, A. V. after Wycl. and all the rest ; and so the Vulg. Cod, 
Am, — V. 3. he^ by an omission from the text after Lachmann, 
^ Tischendorf, and Tregelles ; Jesus, A. V. after Wycl. and all the 

rest ; and so the Vulg. Cod. Am, ; see on 4, 12. stretched forth, 
more appropriately, after Rh. ; put forth, A. V. after Tynd., Cran., 
and Gen. be — made clean^ after Wycl. and Rh., to conform to v. 2 ; 
be — clean^ A. V. after Tynd., Cran., and Gen. straightway ^ by a 
new rendering ; immediately, A. V. after Tynd., Cran., and Gen. ; 
and so Dr. Campbell, Dr. Noyes, Dean Alford, Mr. Darby, and 
Dr. Davidson ; this is a substitution of an English for a Romance 
word ; see on 4, 12. — v. 4. no man, after A. V. ; but closer to the 
Greek (jufitvl), no one, Dr. Noyes and Dr. Davidson ; and so in 
9, 30 ; 16, 20; 17, 9, and throughout the Gospels. — v. 5. he^ by 
an omission from the text after Lachmann, Tischendorf, and Tre- 
gelles ; and so Wycl. and Rh. after the Vulg. ; Jesus, A. V. after 
Tynd. and the rest ; see on 4, 12. — v. 6. in the house (Jv rji oUl^), 
after Wycl., to distinguish from cV ouy, at home, i Cor. 11, 34 ; 14, 



i 



AMERICAN JOURfTAL OF PHILOLOGY. 



BOOKS RECEIVED. 

Altenglische Bibliothek heransg. *. Eogen Kolbing. Enter B;i 
Bokenun'* Legenden heranig. t. C. Hontmann. Hellbronn, Cf!- 
1SS3. 5m. 6opf. 

Aiistotle'a Paychotogy, Greek and English, with iotroducti 
By Ed. Wallace. Cambridgt Univtnity Prtsj, 1883. 

BaTtholomac (Chr.) Handbach der alterBniKhen Dialekli 
veigleichende Grainmatik, LesestQcke u. Glossar). Leipiig, Bu 
1883. 

Braamann (Guilav). Die Priacipea der Gallier u. Germai 
Tacitus. Jahtesber. Ubetdas KOnigt. Friediicb-WUheliiii Gyn< 
Oitem, 1883. 

Brngmana (K.) UebcT griechische Apa, df>, }A n. liUuis' ' 
k. Mchs, GeselUch. der WiucDachaften, Ph'i t-""- ""i- 
April, 1883. 

Crane (T. F.) Mediaeval Sermon boot 
American Philosophical Society, March tG^ 

KiDch (C. F.) Quaestiones criticae Curt 
dendaliana, 18B3. 

Klustmann. Curae ACricanae. Gen, R. 

Kolbing's EngliBche Studien VI, i, 3. lA 

Livy. Books XXI-XXV. Translated ii 
Church and W. J. Brodribb. With Maps. L( 
Cuihings 6* Bailey. $1. 

Pindar. The Nemean and IsEhmiati Ode 
Univeniiy Pnti, i88j. 

Sliss (Eduard). Die Siniauth. Eine gc 
ani : Doi Anilill der Erde. Prag, F. Temf 

Wallace (Ed.) Outlines of the Philowpt 
Cambridgt Univeriity Press, 188a. 

Wiener Studien. Supplement der Zejti 
I, 3. Wien, CaH Gtrold's Sihnt, 1883. 



i space 

( xamples 

.1, Gibbon, 

' : the Show 

;)t and Ridi- 

, Hist. ch. I ; 

jiir, Macaulay, 

c case of words 

n, Spect. No. 8 ; 

; the Decline and 

Danube, ib. Hist. 

Hist. ch. I ; the 

: words of opposite 

.1 Nation, Addison, 

No. I ; the strength 

nporal and ecclesias- 

:ci ministrations, Macau- 

es, ib. (4) Where the 

o things : between the 

I ; this Society may be 

, ib. No. 21 ; between the 

I ; both of the prince and 

lling alike, Macaulay, Hist 

s feelings, ib. (5) Where 

> ing two or more: the same 

I. No. I ; all the Passions and 

^ and cities. Gibbon, Hist ch. 

. ; all the vigour and resolution, 

ne ancient confessions, suppli- 

6) When each noun has its own 

• Dignity of Thought and Sub- 

1. Spect. No. 42; the oppressed 

Gibbon, Hist ch. i ; the sublime 

lity, Macaulay, Hist ch. i. 

in Greek has about the same range 

cs of it are less numerous. Instances 

jTie heads as above, from Thucydides, 

niosthenes* Or. de Corona^ and Plato's 

• which are added all the instances that 

>r to Tischendorfs text 



256 



AMERICAN JOURNAL OF PHILOLOGY, 



(i) In the case of words synonymous or nearly so: rh d' iafkovv 

Kai aXrjSiSf Xen. An. 2, 6) 22 ; ariv rep diKai<^ ical koXw, ib. 2, 6, 18 ; 

Ttov TfXapatp Koi dfiafyraiXavj S. Matt. Q, II. (2) In the case of words 
of different meanings: irtpl ras yavs koi rpujpcir, Thuc. i, 13; njv 

Bifrrrpwrida Koi KtarpiifrjVf lb. I, 46 J roifs (rrparfjyovi Koi XoxayovSy Xen. 
An. I, 7> 2; row fiovs Koi SifovSf ib. I, I, 6; rav XcyotrroiP kqI iroXircv- 

ofji€pav, Dem. de Cor. 173 ; ra Xcx^ci/ra Kalnpax^ivra, Plat. Phaed. 58 C ; 

6 KpiTo^nvXos — ica\ ?Tt 'Epfwyivrji koi 'Eiriyci^r, etc., ib. 59 B ; koi (Xfyrnii) 
Tov yt rjXiop koI a'€Xr)vrjv Koi aarpa opacBaij lb. Ill C ; row ap\i€p(is koL 
ypappartiSf S. Matt. 2, 4 > ^^j ^^ > ^^i' ap\up€a^v xal npta-^vripiaVf ib. 26, 
47; 27, 3> 27» ^2; tS>p ypafifJMTiiov Kal ^apKraiaPf ib. 5, 20 ; 12, 38 ; top 
ypafJLfiaT€<op Ka\ irpfcfivripap, ib. 27) 4^ > ^^'^ ^apKrcuap koi Saddovjcatwy, ib. 
3, 7 I ^^» I > 16,6; 16, II ; 16, 12 ^.f / T«i» npta^mptop koi dpxt€p€<ap koX 
ypapfjLaTici>Pf lb. ,l6, 21 ; rov *Ialca>^ot; »cal *Io)<r^<^, lb. 27, 5^ » I7> ^ » ''1^ TaXi- 
Xalas Koi AricafroXrcos-, etc., ib. 4» 25 ; t6 ifiirai^ai koi paanySixTaiy etc., ib. 20, 

19; II, I (cf. Dem. 01. 3, 35; Plat. Charm. i6i E). (3) In the case 
of words of opposite meanings: ra^ dpaKpova^is kqI dUtarXov^f Thuc. 7, 

70; ras fAtylaras (vaCf) Koi /Xaxi<TTaSf lb. I, lO; rbXiyeip kcu irparrdpf 

Dem. de Cor. 59 > '*^*' *EXXrjpiKS>p koI (€PiKa)pf ib. ^11 ', ncpl rod ficlCopos 

Koi iXdrropoSi Plat. Euthyph. 7 C J Tovs ncaXovpras koi ayopdCopraSy S. 

Matt 21, 12. (4) Where the form of expression distincdy implies 

two things: rrjs rt 'iraXias koi 2t«Xiaff, ThuC. I, 36 ; I, 45; rovs re 
olKrjropas Koi (^povpouf, ib. I, 26 ; I, 28 ; t6p iroXffwp rStP n€X(mopprfaitAP 
Ka\ *A$i]pai<0Pt ib. I, I ; rqp re Kvpov Svpapnv kcu ;(^(opav, Xen. An. 2, 5> II ! 
2, 6, 22 ; rfjs vfi€T€pas €v<Tcfitias re koi 5<5f »/r, Dem. de Cor. I ; ib. 286 ; 
ir€p\ rov fiapmpov rt koi Kovffimpovj Plat. Euthyph. 7 C ; Phaed. 68 D. 
(5) Where the first noun has an adjective or other adjunct quali- 
fying two or more : ras eVciVwi/ Afiapriaf koI napa(rK€v^p, Thuc. 4, 29 ; 
ndvr€S ol ntpl avrop ^tXot Koi crvi^paircfbt, Xen. An. I, 9, 3I ; 17 tfi/j 
avptx^^a Koi nXdpoi koi raXaiinapiait Dem. de Cor. 2x8; ib. 297 J rijt 
nap* vpS>v tvpotas Ka\ ifkiXapSptanlaSy lb. 5 » 1D» 292 J rrjs rifupas cVctnjr mii 
w/xir, S. Matt. 24, 36 (in r^y arjs napovaias nal avrrtXdas rov aiotpos, S. 

Matt. 24, 3, the adjective standing with the first noun does not 
qualify the second, and therefore the article would regularly be 
repeated with the second noun; the article is used with, this 
phrase in three out of the four passages in which it occurs in S. 
Matt., namely, 13, 40 ; 13, 49 ; 28, 20, and the article is found in 
this passage also in four uncial MSS, including D). (6) When 
each noun has its own adjective : oi fUp KaroKrvaroi OerroXoi 

dpalaBijroi Brj^a7oi, Dem. de Cor. 43 » ^1^ raxifmiw dwaXXay^fP— 
a-tarrjpiav d(nf>a\rjf ib. 324* 



THE NEW REVISION OF THE NEW TESTAMENT, 257 

The omission of the article in these cases seems to be due partly 
to the principle of economy, and partly to the circumstance that 
the article is only a weaker demonstrative pronoun. In the Greek, 
as is well known, the article mostly appears a:5 a demonstrative in 
Homer, and its character as the true definite article is not fully 
developed till we reach the Attic period ; and in all the Romance 
languages the article was actually developed out of the demon- 
strative ille. Now the demonstrative pronoun itself is very rarely 
repeated. It is not repeated once in the Greek of S. Matt., but is 
repeated twice in the Revision, in 10, 14 and in 13, 54, and these 
instances will be considered in their places. It is not repeated in 
the two first Bks. of the Atiabasis of Xenophon, nor in Aeschines 
contra Ctesiphontem^ nor in the Phaedo of Plato; it is repeated 
only once in Demosthenes de Corona^ in § 172. And so likewise 
in English. It is not repeated in the first hundred pages of 
Hooker, Bk. V, nor in the two first chapters of Gibbon ; there is 
no simple case of it in the first or the twenty -second chapter of 
Macaulay's History, but in the ist ch. we find, for the sake of ful- 
ness of expression : both by that superstition and by that philoso- 
phy ; and in the 23d : all this clamour and all this wit ; and in 
three great speeches of Burke there are but four cases, in two of 
which it is repeated on account of a change of number : that time 
and these chances ; and, according to that nature and those circum- 
stances, Conciliation with America, p. 168 ; and two simple cases : 
these opportunities and these arguments , Present Discontents, p. 14; 
that sense and that reason J American Taxation, p. 97. 

This comparatively brief view of a large subject has been given 
here partly as a matter of reference when other like passages come 
to be considered, and pjartly to correct the impression -common 
even among scholars, if they have not given special attention to 
the subject, that this omission of the article in Greek and in 
English occurs mostly where the words are synonymous or are 
taken together to form one idea. But it is now seen what is the 
license of Greek in this matter, and this license is still greater in 
English. The Revisers of 161 1 seeming to understand this usage 
often availed themselves of the license of the English, as in 21, 
15; 21, 45, and 27, 62, and omitted the second article, while 
the Revisers of 1881 have in these cases and most others followed 
the Greek forms, thus converting the license of the Greek into a 
strict law for their English. 

v. II. shall sit downy after A. V. ; and so Gen. and Rh. ; Dr. 



!56 AMERICAN JO ' .ITT. 

(i) In the case of wonK -> ■ aaie. v. 13. sons, 

lai oXijA'c, Xen. An. 2, 6, 2.< ' after Tynd. and 

Dr. Davidson ; 
rendering, with 

so Mr. Darby; 
he outer dark- 

so Dr. Noyes, 
tuter darkness, 
there shall be 
irsh rendering ; 
re slial] be the 
d; Mr. Darby 
e shall be the 
noun is limited 
all be weeping 
if the articles ; 
:he rest seems 
ashing 0/ Uelh, 
)ean Alford, — 
vord from the 

bracketed by 
1 all the rest; 
le Greek, after 
matically after 

freely but ex- 
er Gen. neariy, 
e of text after 
.; them, A. V. 

Am. — V. 16. 
le, after Wycl. 

even, close to 
■- inserting the 
er Mr. Darby ; 
ien, possessed 

; that hadden 
ind Dr. David> 
r Tynd., Cran., 
eke. — V. IT, it 
nd Rh. ; that 

see on i, 22. 
rter Gen.; see 



: .:e xew revision of the new testament, 259 

.2. our, the Greek article here having the force of a posses- 

r.n\ A. V. ; see on I, 24 ; and so in v. 20. diseases^ after 

• . .sicknesses, A. V. after Wycl. and all the rest. — v. 19. a scribe, 

'^ser U) the Greek, after Wycl. and Tynd. ; a certain scribe, A. V. 

• L 1 Clan, and the rest there came a scribe y by a change of order 
.• ciifding to the Greek, after Tynd. and Gen. ; a certain scribe came, 
.1. \. after Wycl. and the rest — v. 20. the birds of the heaven, 

illy, after 2d Gen. ; the birds of the air, A. V. excellently, after 

. : jid., Cran., and ist Gen. ; see on 6, 26. — v. 21. the disciples, by 
. : omission from the text after Lachmann, Tischendorf, and Tre- 
. • lies ; his disciples, A. V. after Wycl. and the rest — v. 22. saith, 
' y a change of text after Lachmann, Tischendorf, and Tregelles ; 
ind so the Vulg. Cod. Am.; said, A. V. after Wycl. and all the 
rest, leave the dead to bury, after Dean Alford, Mr. Darby, and 
Dr. Davidson ; let the dead bury, A. V. after WycL and all the 
rest; and so Dr. Campbell and Dr. Noyes. their own (iavrav) 
dead, after Dr. Noyes, Dean Alford, Mr. Darby, and Dr. David- 
son ; their dead, A. V. after Wycl. and all the rest, and so Dr. 
Campbell ; and the Revisers themselves give this pronoim by an 
unemphatic form in 21, 8; 23, 37; 25, 3. — v. 23. boat, correctly, 
after Rh. ; ship, A. V. after Tynd. and the rest ; and so in v. 24. — 
V. 25. they, by an omission of words fi^om the text after Tischen- 
dorf and Tregelles, which are bracketed by Lachmann ; they are 
omitted also by the Vulg. Cod. Am. ; and so Rh. ; the disciples, 
A. V. after Wycl. and the rest. Save, Lard, by a new change of 
order according to the Greek, and by an omission from the text after 
Lachmann, Tischendorf, and Tregelles ; Lord, save us, A. V. after 
Wycl. and Rh. ; and so the Vulg. Cod. Am. ; but the pronoun, us, 
cannot be well omitted in the English. — v. 27. And (d«)» after 
Wycl., Tynd,, and Gen. ; and so HoU. Rev. ; But, A. V. after 
Cran. ; and so de Wette and Germ. Rev. — v. 28. Gadarenes, by 
a change of text after Tischendorf and Tregelles ; Gergesenes, 
A. V. after Wycl. and all the rest ; and so the Vulg. Cod. Am. and 
Lachmann. possessed with devils: so A. V. and Dean Alford; 
possessed of devylles, Tynd., Cran., and Gen. ; that hadden develis, 
Wycl. and Rh. ; demoniacs, Dr. Campbell and Dr. Davidson ; pos- 
sessed by demons. Dr. Noyes and Mr. Darby; see on 4, 24. could 
pass, after Rh. ; might pass, A. V. after Wycl. and all the rest — v. 
29. thou Son of God, by an omission from the text after Lachmann, 
Tischendorf, and Tregelles ; and so the Vulg. Cod, Am. ; Jesus, 
thou Son of God, A. V. after Wycl. and all the rest. — v. 30. Now 



258 AJ 

Noyes, doser 
close to the G 
the rest. £iu/, 
cast out, into o 
a conformity ol 
cast out into u 
ȣss, preservinj 
Dean Alford, 
A. V. omitting 
iAa weeping- an 
in the original : 
weeping and th 
and Dr. David: 
weeping and th 
by a genitive a 
and the gnash 
but therendei 
adequate and n 
and this satisB 
V. 13. as Ikou h 
text after Lad 
Tregdtes; and 
and so tbeVu 
Wycl.; the se 
Cran. and Gen. 
cellendy after 1 
iayde downe, a 
Lachmann, Tis. 
after Wycl. an 
And (B.) when, 
and Rh.; Wh. 
the Greek, aftei 
article after Tyi 
that were posst 
with devils : so 
develis, Wycl. ; 
son ; see on 4, 2. 
Gen., and Rh. ; 
might be fulfil 
which, Tynd- a 
Isaiah, to confc 



e( 



Y/':iV TESTAMENT, 263 

.. ; and so Dr. Noyes, Dean 
•iie-skiiis^ by a new rendering ; 
... and Rh. — v. 18. spake: so 
.., was speaking, which is bet- 
\r, closer to the Greek, after 
: Tynd., Cran., and Gen. is even 
. tter, preserving the Greek aorist, 
q ui valent to just now. Compare, 
■ hat some frogs were venomous? 
Vix.' is used in the same sense, ib. p. 
after Dr. Campbell, Dr. Noyes, and 
Wycl. ; a woman, which, A. V. after 
2, 6. had, after Wycl. ; was diseased 
m., and Gen.; but English idiom here 
had, and so Dr. Noyes and Mr. Darby ; 
I )r. Campbell, border^ after Dr. Noyes 
A. V. after Wycl. and all. — v. 21. If I 
' the Greek, and after Wycl. nearly. If I 
Ijut touch, A. V. after Tynd., Cran., Gen. 
luser to the Greek and to conform to v. 22, 
uill be made well. Dr. Noyes ; shall be whole, 
ring. — V. 22. turning and seeing her saidy 
ji about, and when he saw her he said, A. V. 
, ood cheer y after Sir John Cheke ; be of good 
- Tynd., Cran., and Gen. — v. 23. fluie-players^ 
. minstrels, A. V. after Wycl. and the rest. 
i>r. Campbell, Dr. Noyes, and Mr. Darby; the 
1 Wycl. and the rest except Rh., the multitude. 
Air. Darby ; a noise, A. V. after Wycl., Cran., and 
i^aid, by an omission from the text after Lachmann, 
.1 Tregelles; so Wycl. and Rh. after Cod, Am.; he 
, A. V. after Tynd. and the rest the damsel^ after 
.aid, A. V. after Tynd., Cran., and Gen. ; and so in 
of the substitution of a Romance for an English 
a I, 24. — V. 25. the crowds after Dr. Noyes and Mr. 
people, A. V. after Tynd., Cran., and Gen. he entered 
1., Cran., and Gen. ; he went in, A. V. after Wycl., a 
* substitution of a Romance for an English word; see 
— V. 26. weTii forth, after Rh. ; went abroad, A. V. 
1. — V. 27. as fesus passed by, closer to the Greek, and 
. nearly, as Jesus passed forth; when Jesus departed, 



' 



264 AMERICAN JOURNAL OF PHILOLOGY, 

A. V. after Cran. from thence^ after WycL, Sir John Cheke, 
and Rh. ; and so Dr. Noyes; thence, A. V. after Tynd., Cran., 
and Gen. ; and so Dr. Campbell, Dean Alford, Mr. Darby, and 
Dr. Davidson ; see on 4, 21. crying ouiy giving the Greek 
more fully, after Dr. Noyes ; crying, A. V. after Wycl. and the 
rest. Have mercy on us, thou Son of David^ by a change of 
order to conform to the Greek ; so Rh. after the Vulg. ; Thou Son 
of David, have mercy on us, A. V. after Wycl. and the rest — 
V. 28. They say, strictly according to the Greek, after Cran. and 
Rh. ; They said, A. V. freely, after Wycl. and the rest. — v. 29. 
be it done, after Wycl. and Rh. ; be it, A. V. after Tynd. and the 
rest ; and so in v. 29. — v. 30. strictly, after Dr. Campbell and 
Dean Alford; straidy, A. V. by a new rendering; this is the 
substitution of a modern for an archaic form, though the words are 
etymologically connected. See that no man know it : so A. V. 
supplying and italicizing that, after Rh., and the rest nearly; 
closely. See, let no man know (yiyvwo-Kcro)) it ! and so Mr. Darby 
except the punctuation. — v. 31. they went forth and spread, after 
Rh. ; they, when they were departed, spread, A. V. after Cran. ; a 
case of the substitution of an English for a Romance word ; and 
so again in this verse; see on 4, 12. land, after Wycl., Tynd., 
Cran., and Gen.; country, A. V. after Rh. — v. 32. And (d/) 
oj, preserving the introductory particle, after Wycl. and Rh. ; 
As, A. V. after Tynd. and the rest, went forth, after Rh. ; went 
out, A. V. after Wycl. and the rest, there was brought, by a new 
and free rendering to avoid the ambiguity of they ; so nearly, was 
presented, Dr. Campbell, possessed with a devil: so A. V. after 
Rh. ; demoniac, Dr. Campbell and Dr. Davidson ; possessed by 
a demon, Dr. Noyes and Mr. Darby ; see on 4, 24. — v. 33. the 
devil, after A. V. and all the rest ; the demon. Dr. Campbell, Dr. 
Noyes, Mr. Darby, and Dr. Davidson ; and so twice in v. 34 ; see 
on 4, 24. the dumb man, supplying but not italicizing man, after 
Wycl. and Rh., which accords with modern usage ; the dumb, 
A. V. after Tynd. and the rest, which form was once singular as 
well as plural (comp. A. V. Ps. 5, 12; 10, 2, &c. ; 22, 24; 
and so often), but this is now regularly used as plural. — v. 34. 
By the prince of the devils casteth he out devils, by a change 
of order according to the Greek ; and so Wycl. and Rh. after 
the Vulg. ; He casteth out devils through the prince of the devils, 
A. V. after Tynd., Cran., and Gen. casteth he out devils (ra 
daifiopia), omitting the Greek article after A. V., but Dr. Camp- 



THE NEW REVISION OF THE NEW TESTAMENT 265 

bell, Dr. Noyes, Dean Alford, Mr. Darby, and Dr. Davidson all 
retain it. — v. 35. all the cilies and the villages^ preserving the 
second Greek article against the English idiom, after Dean Alford, 
Mr. Darby, and Dr. Davidson ; all the cities and villages, A. V. 
correctly, after Wycl., Rh., and Sir John Cheke; and so Dr. 
Campbell and Dr. Noyes. When two or more nouns have a 
common adjective standing with the first, the article is regularly 
omitted with the second and following nouns ; as, trdvrai rovs 

ira>XoOi^ar koi dyopuCoyras 21, 12; iravras tovs ap)(i€p€h Koi ypafifiartis 
2, 4> ndaais rats cWoXatr koi diKaici)fjLa<n S. Luke I, 6: SO, rovs jci/icX^ 
aypovr teal KCi>fjLas S. Mark 6, 36 ; rhs kvkXc^ Ka>fjias koi dypovs S. Luke 

9, 12, where the article is repeated in A, C, D, and other uncials, 
but it is omitted by Tischendorf and bracketed by Tregelles. In 
24, 36, r^s fifitpas €K(i¥Tjs fj &paSf the second article is not found in k , 
B, D, and other uncials, and is omitted by Lachmann, Tischendorf, 
and Tregelles; and in the parallel passage, S. Mark 13, 32, the 
second article is wanting in A, E, F, G, and other uncials, in the 
important cursive No. 69, and in the title of the Kc<^aXo*oi/ in which 
it stands. 

But where the modification standing with the first noun does not 
affect the second, &c., the article is rightly repeated ; as, rovs hrra 

apTovs Koi rovs Ix'^vos I5i 3^» W^ ^^ napovaias Koi rrjs avvTfXtias rov 

alStvosy 24, 3, according to D, r and other uncials ; see on 8, 11 (5). 
There are three passages in S. Matthew which require notice 

here. In ndtn-ts ol dp^itptU kqI o\ irpia^xmpoi rov Xaov^ 27, I, the 

article is repeated against regular usage, and A. V. after Rh. rightly 
omits it, all the chief priests and elders, but the Revisers insert it ; 
in the similar expression in S. Mark 14, 53, A. V., against English 
usage, repeats the article after the Greek, and tlie Revisers do the 
same. In 11, 13, irdvr^^ ol irpo(l>fJTai KoX 6 v6posy all the prophets and 
the law ; and 22, 40, oKos 6 vdpos kqI ol irpo<f>rJTai, the whole law and 
the prophets y the adjective in each passage is used in two different 
senses, all the prophets and the (whole) law, &c., as appears in 
this corrected translation of the Revisers, and so the second article 
is properly repeated both in the Greek and in the English. 

For further examples of this omission of the article, in English 
as well as in Greek, see on 8, 11 under (5). 

all manner of —all manner of, repeated after the Greek ; all man- 
ner of, ist Gen., not repeating it ; all maner, Tynd. ; every^-every, 
A.V. after Wycl., Cran.,and Rh. disease — sickness, by a new render- 
ing ; and so 10, i ; sickness — disease, A.V. after Tynd., Cran., and 



266 AMERICAN JOURNAL OP PHILOLOGY. 

Gen. sickness^ by an omission from the text after Lachmann, Hsdi- 
endorf, and Tr^elles ; and so WycL and Rh. after the Vulg. ; disease 
among the people, A.V. after T)md. and the rest — v. 36. with com- 
passion far y closer to the Greek (irtp*' c. gen.), after Dr. Noyes, Dean 
Alford, and Mr. Darby ; with compassion on, A. V. by the old 
ft)rm, after Tynd. and Cran. ; thb noim in A. V. is commonly used 
with on or uporiy and sometimes with of^ as Heb. 10, 34; compare to 
have pity OKy S. Matt 18, 33, or upon^ Deut 7, 16, which is still the 
common form, they were distressed^ by a new rendering ; they were 
harassed, Dr. Noyes, Dean Alford, Mr. Darby, and Dr. Davidson ; 
they fainted, A. V. after Sir John Cheke nearly, thei weer fiunted ; 
they were pyned awaie, Tynd. scattered, after Sir John Cheke ; 
scattered abroad, A.V. freely and idiomatically, after Tynd., Cran., 
and Gen. not having, dose to the Greek and after Wycl. ; having 
no, A. V. after Tynd., Cran., and Gen., which is smoother, and like 
the form adopted by the Revisers themselves in v. 12. — v. 37. 
truly — iut (jutr^di); so A. V. after Cran.; indeed — ^but. Dr. 
Noyes, which seems preferable here, and the Revisers themselves 
have so rendered this formula in 3, 11 ; 13, 32 ; 17, 11 ; 20, 23 ; 26, 
41, and elsewhere, is — are; supplied, but not italicized ; is — are, 
A. V. ; supplied also by Tynd., Cran., Gen., and Rh. ; but Wycl. 
supplies only the first verb : there is myche ripe come, but fewe 
werke men ; and so Sir John Cheke : the hervest is great, and the 
woorkmen few ; this is more pleasing to the ear, and unifies the 
sentence, as in 5, 11. The suppression of the verb in the second 
clause is sometimes a great beauty ; as, the Lord shall be unto 
thee an everlasting light, and thy God thy glory, A. V. Isa. 60, 19 ; 
and this may take place not only when the verbs would be of 
different numbers ; as, in tragedy the design is weighty, and the 
persons great, Dryden, I, p. 183. Such was the state of the 
Roman frontiers, and such the maxims of Imperial policy. Gibbon, 
History, ch. I, p. 142 ; three hundred of his nobility were treacher- 
ously slaughtered, and himself detained captive, Hume, History, 
ch. I. Beyond this lies the Desert on one side, and on the other 
barbarous nations, Spedding*s Reviews, etc., p. 236; and when 
they would be of different persons ; as, the Sun is just rising, and 
I myself just now come to this place, Walton, Angler, p. 40 ; but 
even when they would be of different numbers and persons ; as, 
He (God) is above, and we upon earth. Hooker ; we are the clay, 
and thou our potter, A. V. Isa. 64, 8. — v. 38. that he send, closer 
to the Greek, after Wyd. and Rh. ; that he will send, A. V. after 
Cran. 






T32. J~1^ S± ■ — 







T 



oJSU L/T* LASn iJaUI {Tk-^z: -* -* S. i^t^*^ ^^ ■■■■ ■— ^ET^^r *>-..l*«.'~- 



the way. A. V. sicci7*3sr icr 3i:c Ta*:i z' l u r rii- aifT-*^ "^r^ 

cify^ scppnr::::^^ cr? bcc irx ^rs'i' -rm^ x xr^ cry A. \ 

OQ I, 17. — T. &. rEzj^- lie r>«r r^.cz^ rir jlT*:"^. rr x ^^:x^ 

and Trcgdjcs; aad sc Wj^ aad 5^ ar rr r Jj«£. .-it,.- ceisx 
the lepcis, raise the dead. A. V. arryr TjzlL ir\f ±e rest, ri, — 
/i^ — /4^, snppTacd bet act i&'irjfr: - sc A. V. ii:-r W'yd irxi 
aD ; see oa i, 17. dferus: so A. V. 2±*r WVd. ire al . xrvi $0 
Dean Aifard; deacxB, Dr- Cazrpbel, Dr/Nryes. Mr. rurt^y. 
and Dr. Davidsoc ; see oc 4, 24. je recr^JL to pnes*ervv^ the 
Gredc aorist, after Dr. Nc-ycs, Deas Alfzrd, asd Dr, Dji\?d<cc: 
ye have rece iv ed. A- V. zta WycL and aZ ; aad so EV* Cjuxv^>- 
ben and Mr. Darby; see on 2, 2- — t. 9. Gfi jc% Wm» /-aj. by a 
new rendering; Provide neither gold, A.Y*, by a new reixkiii\j;* 
brass / so A. V. after Tynd. and Gen. ; this requires the morvihvAl ; 
Gr. copper 4nr bronze ; Erz, de Wette ; Kupfcr, WeL^sicker. — 
V. 10. no waiki, by a new rendering; no scrip. A* \\ alWr \V^^^. 
and all the rest. neUher — nor — nor^ after Dr. Da\*id§vMi ; n^^n i. r 
— neUher'-nor yet, A. V. after Tynd. and Cran. staff ^^Wi^*r, wixji . 'i 
after Tynd. and Gen. ; all the rest hav-e a singular number hei^ 
except A. V., staves; Stephen's Greek text of 1550 and Bo«a*» 1^' 
1604 give the singular, but the plural OM;)dow) is the readii^ of 



268 AMERICAN JOURNAL OF PHILOLOGY, 

C, E, and several other uncial MSS. the labourer ^ after Dr. Noyes 
and Dr. Davidson ; the workman, A. V. after Wycl. and all the 
rest, food^ after Sir John Cheke ; and so Dean Alford and Dr. 
Davidson; meat, A. V. after Wycl. and all the rest — v. ii. 
village^ to conform to 9, 35, after Dr. Campbell, Dean Alford, Mr. 
Darby, and Dr. Davidson ; town, A. V. after Tynd. and all the 
rest except Wycl., castel. search out, by a new and closer render- 
ing of the Greek (cfcrao-arc) ; serch. Sir John Cheke ; enquire, 
A. V. after Tynd., Cran., Gen., and Rh. go forihy closer to the 
Greek, after Rh. ; go thence, A. V. after Tynd., Cran., and Gen. 
— V. 12. as ye enter, after Dr. Noyes and Dr. Davidson; when ye 
come into, A. V. after Tynd., Cran., and Gen. ; this is a case of the 
substitution of a Romance for an English word ; see on i, 24. the 
house, preserving the Greek article, after Rh. ; an house, A. V. 
after Wycl. and the rest. — v. 14. as ye go forth, after Mr. Darby ; 
going forth, Rh. ; when ye depart, A. V. after Tynd., Cran., and 
Gen. ; this is a case of the substitution of an English for a Romance 
word; see on 4, 12. that house or that city, incorrectly, after 
Tynd. and Gen., the demonstrative being used only once in the 
Greek and put with the last noun ; that house or city, A. V. cor- 
rectly, after Wycl. and Sir John Cheke; so Dr. Campbell, Dr. 
Noyes, and Dean Alford ; and so the Germ. Rev. ; see on 8, 1 1 . The 
demonstrative stands after the first noun in Cod. b, it is wanting 
altogether in Cod./, and in Cod. Am. and the Clem. ed. of the Vul- 
gate ; but the common modifier of two or more words sometimes 
stands with the last word; see on 24, 30. — v. 17. councils, 
close to the Greek, after Wycl. and Rh. ; the councils, A. V. 
inserting the article, after Tynd. and the rest, and in their 
synagogues they will scourge you, by a change of order accord- 
ing to the Greek, after Rh. and the Vulg. ; and they will scourge 
you in their synagogues, A. V. after Wycl. and the rest — 
V. 18. yea — and (koI — df). hy a new rendering, giving both par- 
ticles; And, A. V. omitting one, after Wycl. and all the rest; 
and so the Vulg. before governors and kings shall ye be brought, 
by a change of order according to the Greek, after Wycl. and Rh.; 
and so the Vulg. ; ye shall be brought before governors and kings, 
A. V. after Tynd. and the rest, for a testimony to them and to the 
Gentiles, after Wycl. and all except A. V., which has : for a testi- 
mony against them and the Gentiles; the Revisers here follow 
Meyer in making the reference general (/{?), although some great 
names make it specific {against), as S. Chrysostom and Theophy- 



THE NEW REVISION OF THE NEW TESTAMENT, 269 

lact among the ancients, Erasmus and Beza among the moderns; 
io them and to the Gentiles^ repeating the preposition for the simple 
dative in the Greek, after Wycl. and all except Rh., to them and 
the Gentiles ; in such cases the Revisers, like A. V., have com- 
monly exercised a wise liberty according to the occasion, employ- 
ing the preposition once only; as in 5, 20; 10, 14; 11, 22, etc., 
sixteen instances in all ; or repeating it, as here and in 22, 32 ; 
28, 19, etc., ten instances in all. — v. 19. be not anxious, after 
Dr. Campbell, Dr. Noyes, and Dr. Davidson; be not careful, 
Dean Alford and Mr. Darby ; take no thought, A. V. after Tynd. 
and all the rest ; nyle ye thenke, Wycl. ; see on 6, 24. in that 
hour, close to the Greek, after Wycl., Rh., and 2d Gen. ; in that 
same hour, A. V. by a new rendering; Tynd., Cran., and ist Gen. 
still stronger, even in that same houre; see on 5, 19. — v. 20. the 
Spirit— thai, after Wyd. and Rh. ; the Spirit — ^which, A. V. after 
Tynd. and the rest; see on 2, 6. — v. 21. brother shall deliver 
up brother, omitting the articles according to the Greek, and after 
Mr. Darby ; the brother — the brother, A. V. inserting the articles, 
after Wycl. and all the rest and the father his child, supplying 
but not italicizing the and his, after Dr. Noyes and Dean Alford ; 
and father child, close to the Greek, Mr. Darby and Dr. Davidson ; 
the father the child, A. V. by a new rendering ; the father lijkwijs 
the child. Sir John Cheke ; the fadir the sone, Wycl. and all the 
rest, children shall rise up against parents, close to the Greek, 
after Mr. Darby ; the children shall rise up against their parents, 
A. V. supplying the and their, but italicizing only the latter, after 
Gen. — V. 22. all men, supplying men, but not italicizing it ; all 
men, A. V. he — the same (pvroi) shall be saved, after Dean Alford 
and Dr. Davidson ; he shall be saved, Rh. and 2d Gen. ; and so 
Mr. Darby; shall be saved, A. V. omitting the pronoun after 
Wycl., Tynd., Cran., and ist Gen. — v. 23. the next, by a new 
and free rendering of the Greek (jriv htpav, after Lachmann, Tisch- 
endorf, and Tregelles ; tijv oKKrjv, Stephen's ed. 1550, Beza's, 1604) ; 
another, A. V., Wycl., and all the rest ; so the Vulg. ; and so Dr. 
Campbell, Dr. Noyes, and Dean Alford ; the other, close to the 
Greek, Mr. Darby and Dr. Davidson ; the formula, ovros — 6 mpos 
or 6 flfXXof , is a strange one, instead of 6 cw — 6 frc/aop, the one — the 
other, 6, 24 ; S. Luke 7, 41 ; or c^ — 6 h^^^, one — the other, 6, 24 ; 
S. Luke 16, 13. shall have gone through, after Dr. Campbell ; 
shall goe thorow, Cran. ; shall have gone over, A. V. by a new 
rendering. — v. 24. A disciple, close to the Greek, after Dr. 



270 AMERICAN JOURNAL OF PHILOLOGY, 

Campbell, Dr. Noyes, and Dr. Davidson ; The disciple, A. V, 
inserting the definite article, after Wycl. and all the rest ; and so 
Dean Alford and Mr. Darby ; and so again in this verse, a servant^ 
for the servant his {r6v) master^ after Tynd. and Gen., the article 
having the force of a possessive pronoun ; his master, A. V. by a 
new rendering ; see on i, 24. — v. 25. shall ihey call^ supplied and 
italicized ; and so A,V. after Tynd. and Cran. ; and so in substance 
Dr. Campbell and Dr. Noyes; it b better to omit these words 
after Wyd., Gen., and Rh. ; and so Dean Alford, Mr. Darby, and 
Dr. Davidson. — v. 26. nothing covered, — and hid: and so A. V. 
by a new rendering, but after Rh. nearly, and secrete ; and (wu) 
continues the negative here, and therefore nor hid is belter after 
2d Gen.; Cran, repeats the negative from the preceding, and 
nothynge hyd ; and so in substance Wycl., Tynd., and ist Gen. 
— V. 27. in the darkness — in the light, close to the Greek, but in 
violation of English idioip, after Dean Alford and Dr. Davidson; 
better, in the darke — in the light, with Sir John Cheke and Rh. ; 
and so Dr. Campbell, or in darknessis — in the light, with WycL ; 
and so Dr. Noyes and Mr. Darby, speak, close to the Greek, 
after Dr. Noyes, Mr. Darby, and Dr. Davidson ; thai speak, A. V. 
supplying that, after Tynd., Cran., and Gen. ; and so Dean Alford. 
proclaim, after Dr. Campbell and Dr. Noyes; that preach, A. V. 
after Tynd., Cran., and Gen. — v. 28. be not afraid of, to keep 
closer to the Greek (ji^ <f>offri$rjT€ arr6, a Hebraism, also in S. 
Luke 12, 4; after LXX, as Deut. i, 29; Ps. 21, 25 (23 A. V.), 
etc.), after A. V. in S. Luke 12, 4; and so Dean Alford, Mr. 
Darby, and Dr. Davidson ; fear not, A. V. after Tynd., Cran., 
Gen., and Rh. ; and so Dr. Campbell and Dr. Noyes. them which 
— him which: so A. V. after Tynd., Cran., and Gen. ; hem that — 
hym that, Wyd. ; them that — him that, Rh. ; see on 2, 6. — v. 29. 
and (koI) : and so A. V. after Wyd. and all ; better, ajidyet, with 
de Wette, Weizsa^cker, and Dr. Davidson ; yet. Dr. Campbell and 
Germ. Rev. ; see on i, 25. 7iot one of them shall fall, after Rh. ; 
one of them shall not fall, A. V. after Gen. ; and so Mr. Darby. — 
V. 31. Fear not, doser to the Greek, after Rh. ; Fear ye not, A.V. 
supplying the subject after Tynd., Cran., and Gen. ; and so Dean 
Alford; see on 4, 17. — v. 32. Every one — who, closer to the 
Greek form, and nearly after Wyd., Cran., and Rh. ; Whosoever, 
A. V. after Tynd. and Gen. him will I also confess, by a change 
of order nearer to the Greek (icay«) and after Rh. ; him will I 
confess also, A. V. after Tynd., Cran., and Gen. ; see on 2, 8. 



THE NEW REVISION OF THE NEW TESTAMENT. 27 1 

my Father which : so A. V. after Tynd. and all except Wycl., my 
fadir that ; and so in v. 33 ; see on 2, 6. — v, 34. / came, to 
preserve the Greek aorist, after Wycl. and Rh. ; I am come, A. V. 
after Tynd. and the rest ; and so in v. 35 ; see on 2, 2. — v. 35. 
the daughter — the daughter in law, supplying the definite articles : 
so A. V. after Wycl. and all the rest ; daughter— daughter-in-law, 
Dr. Campbell, strictiy after the Greek ; and so substantially Dr. 
Noyes and Dr. Davidson. — v. 38. that doth not take his cross and 
follow, to unify the expression, after Dr. Noyes and Mr. Darby ; 
that taketh not his cross and foUoweth, A. V. after Wycl. and all 
the rest ; see on 5, 11. — v. 39. his life — his life, with a new mar- 
ginal note, Or, soul, which would be absurd in this passage. — v. 42. 
cold water, supplying water, but not italicizing it, after Gen.; cold 
water, A. V. less correctiy ; for ^vxp(^v in Greek, like gelida in 
Latin, is used absolutely for cold water; aqua frigida, Vulg. freely. 
Ch. XL V. I. it came topccss — he departed : so A. V. against the 
English idiom, after Wyd. and Rh. ; it came to pass, that, Tynd., 
Cran., and Gen. ; and so Dr. Noyes ; see on 7, 28. and to teach and 
preach, after Wycl., Sir John Cheke, and Rh. ; to teach and to preach, 
A.V. with more dignity, after Tynd. and the rest ; and so Dr. Noyes, 
Dean Alford, and Dr. Davidson ; and the Rev. in Acts 5, 42. The 
sign of the infinitive where two or more infinitives come together 
may, according to good usage, be repeated or omitted with the 
second, etc., as suits the occasion. — v. 2. whenfohn heard (aKovirai), 
after Tynd., Cran., and Gen. ; had heard, A.V. more exactly, after 
Wycl. and Rh. the Christ, after the Holl. Rev. and Dr. Davidson ; 
the Messiah, Dr. Campbell ; Christ, A.V. omitting the article after 
Wycl. and all the rest ; and so de Wette, the Germ. Rev., Dr. 
Noyes, Dean Alford, and Mr. Darby ; see on i, 17. by his disciples, 
by a change of text (dw for tvo) after Lachmann, Tischendorf, and 
Tregelles; two of his disciples, A.V. after Wyd. and all ; and so Cod, 
Am, — V. 3. that cometh, dose to the Greek, after Dr. Campbell ; 
that should come, A.V. after Gen. look we, after Rh. ; do we look, 
A.V. after Cran. — v. 4. And(Kai)fesus, preserving the introductory 
particle, after Wycl., Gen., and Rh. ; Jesus, A. V. freely, after Tynd. 
and Cran. Go your way, to conform to S. Luke 7, 22, after Sir 
John Cheke nearly. Go iour wais ; Go, A. V. after Wycl. and all 
the rest ; this is a case of the substitution of an archaic for a common 
form ; see on 4, 1 2. and tell, to conform to S. Luke 7, 22, after Wycl. 
and Sir John Cheke ; and shew — again, A. V. after Cran. the 
things which, nearer the Greek (a), after Dr. Davidson; what 



272 AMERICAN JOURNAL OF PHILOLOGY. 

things, nearer still, Gen. ; those things which, A. V. after Wyd. 
nearly, tho thingis that. — v. 5. the blind — the lame — the lepers — 
the deaf— the dead — the poor ; so A. V. supplying the article in 
each case ; Mr. Darby and Dr. Davidson omit it in each case after 
Wycl. good tidings, after Dr. Noyes ; the glad tidinges, Tynd. 
and Cran. ; the Gospd, A. V. ed. 161 1, after Wycl., Gen., and 
Rh. ; and so Dean Alford and Dr. Davidson ; good tidings is the 
constant form of A. V. in O. T. ; as, 2 Sam. 18, 27 ; 2 Kings 7, 9, 
&c. ; the Revisers have used glad tidings twice, in Rom. 10, 
15 and I ; Thess. 3, 6. and the dead, by an addition to the text 
after Tischendorf and Tregelles ; the dead, A. V. after Wycl. and 
all. — V. 6. shall find none occasion of stumbling in me^ after 
Dr. Noyes {no occasion)-, shall not be offended in me, A. V. 
after Gen. ; so nearly Tynd., Cran., and Sir John Cheke ; and 
so Dean Alford and Mr. Darby; see on 5, 29. — v. 7. these 
went their way, after Rh. nearly, they went their way; they 
departed, A. V. after Tynd., Cran., and Gen. ; this is a case of the 
substitution of an English for a Romance expression ; see on 4^ 12. 
to behold, to distinguish between the Greek verb here and in the 
next verse, after Dr. Davidson ; to gaze upon. Dean Alford ; to 
see, A. V. after Wyd. and all the rest — v. 8. for to see ; so A. V. 
after Tynd., Cran., and Gen. ; to see, WycL and Rh. ; and so Dr. 
Campbell, Dr. Noyes, Dean Alford, Mr. Darby, and Dr. Davidson ; 
see on 24, i. in soft raiment, supplying and italidzing raiment; in 
soft raiment, A. V. after Gen. in soft raiment, as before ; in soft 
dothing, A. V. ed. 161 1, after Gen. — v. 9. wherefore went ye outf 
to see a prophet f by a change of punctuaticm after Tischendorf; 
and so Tregelles (marg.) ; what went ye out for to see ? A prophet ? 
A. V. after WycL and all the rest to see, after Wycl. and Rh. ; 
for to see, A. V. after Tynd. and the rest ; see on 24, i. miuh 
more than, to give the Greek (ntptavvrtpov) more fully, after de 
Wette, Germ. Rev., and HoU. Rev. — v. 10. This is he, by the 
omission of a word from the text by Tischendorf, which is 
bracketed by Lachmann and Tregelles; For this is he, A. V. 
after Wycl. and all the rest ; and so Cod, Am» my messenger — 
H^ho, after Dr. Campbell, Dr. Noyes, Mr. Darby, and Dr. David- 
son ; — that, Wyd. ; which, A. V. after Tynd. and the rest ; see on 
2, 6. — V. II. hath not arisen, to preserve the Greek perfect, after 
Dr. Campbell, Dean Alford, and Dr. Davidson ; arose there not, 
Tynd. and Gen. ; arose not, Cran. ; see on 2, 2. yet he that is 
but little, by a new rendering, but after Mr. Darby nearly, But 



THE NEW REVISION OF THE NEW TESTAMENT, 273 

the little one ; notwithstanding he that is least, A. V. by a new 
rendering, but nearly after Tynd., Cran., and Gen., notwithstanding 
he that ys lesse. — v. 12. mefi of violence y by a new rendering and 
nearer the Greek Oworai) ; violent men, WycL, and so Dr. Da- 
vidson ; the violent, A. V. after Cran., Gen., and Rh. — v. 14, if 
ye are willing (BtXtTt) to receive ity to avoid ambiguity, after 
Dr. Noyes and Dr. Davidson; if ye will (not an auxiliary here) 
receive it, A. V. after Wyd. and all the rest Elijah, to conform 
to the Hebrew ; Elias, A. V. after Gen. and Rh. ; see on i , 2. Elijah^ 
which; so A.V. after Tynd., Cran., and Gen. ; Elie that, Wycl. and 
Rh. ; see on 2, 6. is to come, after Wyd. ; was for to come, A. V. 
after Cran. — v. 16. the market-places, after Cran., and supplying, but 
not italicizing, the article; in the markets, A. V. after Gen. — w. 
16, 17. which call — and say, by a change of text after Lachmann, 
Tischendorf, and Tregelles ; and so in substance Wycl. and Rh. 
after Cod. Am. ; calling — And saying, A. V. by a new rendering. 
We piped— ye did not dance, to preserve the Greek aorist, after 
Rh. ; We have piped — ^ye have not danced, A. V. after Wycl. and 
the rest ; so also, we wailed— ye did not mourn, by a new rendering ; 
we hau morned — ye hau not weilid, Wycl. and Sir John Cheke ; we 
have mourned — ye have not lamented, A. V. after Gen. ; see, on 
2, 2. — V. 18. a devil; so A.V. after Wyd. and all; a demon, accord- 
ing to the Greek, Dr. Campbell, Dr. Noyes, Mr. Darby, and Dr. 
Davidson ; daemonium, Vulg. ; see on 4, 24. — v. ig. a gluttonous 
man, by a new order and freely; a man gluttonous, A.V. close to 
the Greek ; and so Dean Alford and Dr. Davidson. And(KaO, after 
Wycl., Cran., and Rh. ; But, A. V. more suitably to the passage, 
after Gen.; so Dr. Campbell and Dr. Noyes; And yet. Dean 
Alford; see on i, 25. dy her works, by a change of text after 
Tischendorf and Tregelles ; of her children, A. V. after Wycl. 
(Jones') and all the rest ; and so Cod. Am. — v. 21. had been done 
in Tyre and Sidon which were done in you, by a change of order 
according to the Greek, after Rh. ; and so Cod Am. ; which were 
done in you, had been done in Tyre and Sidon, A. V. after Wycl. 
and the rest ; and so nearly in v. 23. — v. 22. Howbeit, by a new 
rendering; the Revisers have so often introduced this old and 
awkward word, that it may be regarded as one of the character- 
istics of their work ; it does not serve to distinguish the Greek 
word used here (irX^v) ; for while they have rendered it howbeit 
again in v. 24, they have left the old rendering of it, but, in 18, 7, 
and nevertheless in 26, 39 and 26, 64 ; But, A. V. after Sir John 



274 AMERICAN JOURNAL OF PHILOLOGY. 

Cheke and 2d Gen. ; and so Dr. Noyes, Dean Alford, Mr. Darby, 
and Dr. Davidson; Netheless, Wyd., Tynd., Cran., and ist Gen. in 
the day 0/ judgement, to conform to v. 24, after Wyd. and Rh. ; at 
the day, etc, A. V. after Tynd. and the rest — v. 23. shalt thou 
be exalted unto heaven f after Dean Alford and Dr. Davidson, by 
a change of text after Lachmann, Tischendorf, and Tregelles ; and 
so in substance Wyd. and Rh. after Cod, Am, ; which art exalted 
unto heaven, A. V. by a new rendering, thou shalt go down, after 
Wyd. and Rh. (come), by a change of text after Lachmann and 
Tregdles ; thou — shalt be brought down, A. V. after Tynd., Cran., 
and Gen. unto Hades, leaving the Greek word untranslated, after 
Dr. Campbell, Mr. Darby, and Dr. Davidson ; to hell, A. V. after 
Wycl. and all the rest ; so Sir John Cheke, and so Dean Alford. 
— V. 24. Howbeii, by a new rendering ; But, A. V. by a new 
rendering ; see on v. 22. — v. 25. At thai season, by a new but 
not invariable rendering of the Greek word (xaipy) ; and so in 
12, I ; 14, I ; time, A. V. after Wyd. and all the rest; so Sir John 
Cheke; and so Dr. Noyes, Dean Alford, Mr. Darby, and Dr. 
Davidson, answered and said: so A. V. after Wyd. and all the 
rest; and so Dr. Noyes, Dean Alford, and Mr. Darby; said, Dr. 
Campbdl. The phrase here literally rendered answered and said 
is of very frequent occurrence in the Gospels, and commonly ap- 
pears under the form orok^B^ Attv, as here, in S. Matt, S. Mark, 
and S. Luke ; and in the form air€KpiBrj koI cfircy in S. John. It was 
derived from the LXX, and occurs in the first form in Gen. 18, 
27; 27, 37; 3i» 36; 3i» 43; 40» 18; 41, 16; 42, 22, &c.; and 
in the second form in Exod. 4, i; Numb. 22, 18; Deut i, 14; 
I, 41 ; Josh. I, 16, &c. ; and both also occur with some modifica- 
tions as well in the LXX as in the N. T. Now the Hebrew verb 
(Hf^), represented in these forms, does not in itself mean A? answer, 
but to strike ufi, speak out, say; often in answer to a question ; as, 
Gen. 18, 27 ; 23, 10 ; 27, 37, &c. which are rightly rendered by the 
LXX and by A. V. ; but sometimes without a question dther ex- 
pressed or implied ; as, Numb. 11, 28 ; Deut. 25, 9, &c. which are 
wrongly rendered, answered and said, both by the LXX and by 
A. V. ; and Deut 26, 5 and 27, 14, &c., which are wrongly ren- 
dered by the LXX but rightly by A. V. ; and Job 38, i, which is 
rightly but briefly rendered by the LXX, c^rcv, said, but wrongly 
by A. v., answered — and said. A like confusion may be found in 
the Latin Vulgate. The Greek Sri is both a demonstrative con- 
junction, that, and a causal, because, since. Now the Latin trans- 






THE NEW REVISION OF THE NEW TESTAMENT, 275 

lator after verbs of sayings thinkings &c., often took the wrong 
meaning of ort, and rendered for instance, X«y« 5ri in S. Matt. 5, 22 
by dico quiuy and the same in 3, 9 by dico quoniam ; and hoKovviv 
5rt in 6, 7 hy puiant quia; but no one would render this Latin by 
I say because^ I say since ^ they think because, but by I say that, &c» 
And this confusion of the LXX was carried over into N. T. 
usage, and the formula, answered and said^ was commonly used 
correctly, but sometimes, as in the O. T. where no question was 
asked or implied; as in the present passage and in 17, 4 
and 22, I and elsewhere, and therefore the proper rendering 
would be spake and said or the like, and so de Wette and Diodati 
in the passages just cited, and also in 12, 38 ; 28, 5 ; and 19, 27 
de Wette treats in the same way, but Diodati renders that ans- 
wered and said. I thank thee — that (5ti, that or because^, after 
Dr. Noyes, Dean Alford, and Dr. Davidson; because, A. V. after 
Tynd. and the rest ; for {^^for thai), Wycl. didst hide — didst re- 
veal, to preserve the Greek aorist, after Dr. Noyes and Dr. David- 
son ; hast hid — hast revealed, A. V. after Wycl. and all ; so Dr. 
Campbell and Mr. Darby ; see on 2, 2. the wise, etc. ; so A.V., sup- 
plying the article but not italicizing it, after Tynd. and all except 
Wycl., wise men and prudent ; wijs and witti men, Sir John Cheke ; 
men wise and of understanding, Dean Alford. understanding, 
after Gen. and Dean Alford nearly, of understanding ; prudent, 
A. V. after Wycl. and the rest. — v. 26. yea, Faiher, close to the 
Greek (vae), after Sir John Cheke and Rh. ; so Dr. Noyes, Mr. 
Darby, and Dr. Davidson; Even so, Father, A. V. freely but 
excellently, after Tynd. ; so Dean Alford. it was well pleasing, 
by a new rendering, but after Wycl. nearly, it was plesynge ; it 
seemed good, A. V. by a new rendering. — v. 27. have been deliv- 
ered, rendering the Greek aorist as perfect, after Dr. Campbell, 
hath imparted everything ; are delivered, A. V. inexactly, after 
Rh. ; ben govune, Wycl. ; and, are given, the rest ; were delivered, 
Dr. Noyes, preserving the aorist, and so Dr. Davidson ; see on 2, 2. 
no one, closer to the Greek (oWctp), after Dr. Noyes and Mr. Darby ; 
none, Dr. Campbell, Dean Alford, Dr. Davidson; no man, A.V. after 
Gen. and Rh. san)e the Father, by a new rendering to conform 
to the next sentence ; but — save, A. V. for the sake of variety, 
after Tynd., Cran., and Gen. ; but — but, Wycl. and Rh. save — he: 
so A. V. after Tynd., Cran., and Gen. ; so also the Revisers after 
A. V. in S. John 6, 46 ; Rev. 13, 17 ; but in Rev. 2, 17 they have 
altered saving he to but he. doth any know, closer to the Greek, 



276 AMERICAN JOURNAL OF PHILOLOGY. 

after Rh. ; knoweth any man, A. V. after Tynd., Cran., and Gen. 
the Son wilUth {fiovKr^raiL) to reveal, by a new rendering, to avoid 
ambiguity ; so, it shal pleese the Sonne to reveile, Rh. ; it is the 
will of the Son to reveal, Dr. Noyes ; the Son is minded, etc., 
Dean Alford ; the Son may be pleased, etc., Mr. Darby ; the Son 
may wish, etc., Dr. Davidson ; the Son will (not here an auxiliary) 
reveal, etc., A.V. after Tynd.. Cran., and Gen. {will open) ; see on 
V. 14. — V. 30. For my yoke is easy, and my burden is light : so 
A. V. after Tynd., Cran., and Gen. ; but the verb is expressed 
only once (and with the latter clause) in the Greek ; it would be 
closer therefore, and also better thus : For my yoke is easy, and my 
burden lighi ; and so Wycl., for my yok is softe, and my charge 
ligt ; and Rh., For my yoke is sweete, and my burden light ; this 
form unifies the sentence ; see on 5, 11 ; and on the suppression 
of the verb in the second clause, see on 9, 37. 

Ch. XII. V. I. season (natpij)), by a new rendering; time, A. V. 
after Wycl. and all; see on 11, 25. cornfields, closer to the 
Greek, after Dean Alford, Mr. Darby, and Dr. Davidson ; com, 
A. V. after Wycl. and all. ears, close to the Greek, after Wycl. ; 
the ears, A. V. after Tynd. and the rest. — v. 2. the Pharisees, 
when they saw it, said, by a new and closer rendering ; still closer, 
the Pharisees seing it said, Sir John Cheke ; when the Pharisees 
saw it, they said, A. V. after Cran. which it is not lawful to do, 
after Dr. Campbell and Dr. Noyes ; which is not lawful to do, 
A. V. after Wycl. and all, and so Dean Alford, Mr. Darby, and 
Dr. Davidson ; and so in v. 4. upon the Sabbath, after Gen., and 
so Dr. Campbell and Dr. Noyes ; still closer, upon a Sabbath, Dr. 
Davidson ; upon the Sabbath day, A. V. after Tynd., Cran., and 
1st Gen. — V. 4. them that, after Wycl. and Rh. ; those who, Dr. 
Noyes ; them which, A. V. after Tynd. and the rest ; see on 5, 44. 
— V. 5. how that, by a free rendering of the Greek (ort), after 
A. v., Tynd., Cran., and Gen. ; and so Dean Alford ; that, Wyd. 
and Rh. ; and so Dr. Campbell, Dr. Noyes, Mr. Darby, and Dr. 
Davidson. The Revisers have followed A. V. in rendering on 
{that) by how that also in 16, 12 and 16, 21 ; and o>r {how or thai) 
by how that for that in S. Mark 14, 72; and in S. Luke 22, 61 
they have changed how into how that, and have supplied how in & 
Mark 5, 19. It must be borne in mind that the particle in Greek 
in such cases is sometimes only an adverb of manner, and then is 
rightly rendered how; as wa>p in 12, 4; S. Mark 2, 26; 12, 26; 
12, 41 ; and so onm in S. Luke 24, 20; and sometimes the am- 



THE NEW REVISION OF THE NEW TESTAMENT, 2^^ 

biguous word, a»r, how or thai, may be rendered with propriety 
either way, when the idea of manner is not excluded; as in 
S. Luke 6, 4 ; 8, 47 ; and 24, 6, in which passages the Revisers, 
following A. v., render it by how. And in English the use oi how 
introducing narratives or the summary of a narrative, and carry- 
ing with it the idea of manner, is still good, as at the opening of 
Lord Macaulay's History, like ins at the opening of the 2d Bk. of 
Xenophon's Anabasis. 

In the development of how that, how was not prefixed to thai, 
but that was subjoined to how, after the analogy of where that, 
when that, where and when being originally interrogatives and 
afterward converted into relatives by this suffix. This usage was 
extended, and thus if thai, though that, and lest that became com- 
mon in the i6th and 17th centuries. In the first half of the i6th 
century how for that is found in Ellis, Original Letters, p. 310 bis; 
and how that for that on pp. 257, 258, 259, 267 and 315; but in the 
second half of the i6th century they are not found at all in the first 
hundred pages of Hooker, Bk. V., nor in the 17th century in the 
long Preface of the Translators of A. V,, nor in Walton's Angler, 
nor in the hundred and twelve pages of Dryden's Essay on Dra- 
matick Poesy, nor in the i8th century in Addison's own papers 
of the first hundred of the Spectator. How that, and how 
without the idea of manner, have long been disused in good 
English, whether written or spoken, and given up to common 
and illiterate life, and on account of their associations, they ought 
not to have been retained in the new revision, much less in 
additional cases to have been introduced into it. on the sab- 
bath day, suppl3dng clay, which they excluded in v. 4 and render- 
ing the Greek plural by a singular, after Tynd. ; on the sabbaths 
exactly, Dr. Noyes ; on the sabbath days, A. V. after Cran. and 
Gen. ; so in w. 10, 12. and (koi) ; so A. V. after Wycl. and all 
the rest except Tynd., and yet ; so Sir John Cheke ; and never- 
theless. Dr. Campbell ; see on i, 25. guiltless, by anew rendering 
for the sake of tmiformity ; see v. 7 ; blameless, A. V. after Tynd., 
Cran., and Gen. — v. 6. one greater than the temple is here, to 
conform to the order of the Greek, after Dr. Campbell, Dr. Noyes, 
and Dr. Davidson. — v. 7. this, supplied but not italicized ; this, 
A. V. I desire, after Dr. Noyes, Dean Alford, and Dr. Davidson ; 
I will have, A. V. after Gen. — v. 8. of, by an omission from the 
text after Lachmann, Tischendorf, and Tregelles ; even of, A. V. 
after Tynd., Cran., and Gen. — v. 9. he departed thence, and went. 



278 AMERICAl<r JOURNAL OF PHILOLOGY. 

after Tynd., Cran,, and Gen. ; when he was departed thence, he 

went, A. V. by a new rendering, behold, a man, by an omission 

from the text after Lachmann, Tischendorf, and Tregelles ; lo, a 

man, Wycl. also after the Vulg, ; behold there was a man, A.V. after 

Tynd. and the rest, having a withered hand, by an omission from 

the text, after Lachmann, Tischendorf, and Tregelles ; having his 

hand withered, A. V. by a new and incorrect rendering of their 

own text, but nearly after Wycl., Cran., and Gen., which had hia 

hande dryed up, — v. 11. of you, closer to the Greek, after Wycl. 

and Rh. ; among you, A. V. after Gen. if this, close to the Greek, 

after Mr. Darby ; if the same, Rh. ; if it, A. V. after Wycl., Cran., 

and Gen. will he not ; so A. V. after Rh. ; better, will not, after 

Cran., to give unity to the sentence ; so Dr. Noyes and Mr. Darby ; 

see on 5, 11. — v. 12. tf/^ »nt>« I'a^Ktf, by a new rendering, after Dr. 

Noyes nearly, of more worth ; greatly excel. Dr. Campbell, which is 

nearer the Greek form, torfu^oct/, after Wycl., and better suiting 

the context ; to do well, A.V. by a new rendering and exactly after 

the Greek form. — v. 13. thy hand, after Tynd., Cran., Gen., and 

Rh.; thine hand, A. V. after Wycl. ; see on 6, 17. m, after Wycl. 

and Gen. ; like as, A. V. after Cran. — v. 14. Bui (it), after Sir 

John Cheke, Dr. Campbell, Dean Alford, Mr. Darby, and Dr. 

Davidson ; And, Wycl. and Rh. ; Then, A. V. loosely after Tynd., 

Cran., and Gen. look counsel, after Sir John Cheke, took councel ; 

so Dean Alford, Mr. Darby, and Dr. Davidson; held a council 

'//, ed. 1611), A. V. after Tynd. and Cran. — v. 15. 

■ Wycl.; But, A. V. after Cran., Gen., and Rh. 

J- a withdrew, after the form of the Greek ; and so 

iving it, retired ; and Dr. Noyes, Jesus knowing it, 

n Jesus knew it, he withdrew, A. V. by a new ren- 

iving, by a new rendering, and substituting a 

I English word ; see on i, 24. many, by the omis- 

from the text after Lachmann and Tischendorf, 

:keted by Tregelles ; and so Wycl. and Rh. after 

at multitudes, A. V. after Gen. — v. 17. tluU it 

;d which ; so A.V. after Cran., Gen., and Rh. ; that 

fulfilled, that, Wycl.; see on 1,22. Isaiah; Esaias, 

I, 2. — V. 18. in whom my soul is well pleased 

.V. ; in whom it hath wel plesid to my soule, Wycl. ; 

lul hath wel liked, Rh., and both after the Vulg.; 

■eiare, after Dr. Noyes ; shew, A.V. after Tynd. and 

9. cry aloud, nearer to the Greek (xfiavydoti), after 



THE NEW REVISION OF THE NEW TESTAMENT, 279 

Dr. Noyes, Dean Alford, and Dr. Davidson ; crie out, Rh. ; cry, 
A. V. after Wycl. and the rest, any one^ close to the Greek 
(rlr), after Dr. Noyes, Mr. Darby, and Dr. Davidson ; any man, 
A. V. after Wyd. and all the rest. — v. 21. hope^ close to the 
Greek, after Wycl. and Rh. ; trust, A. V. after Tynd. and the rest. 

— V. 22. one possessed with a devil, after A. V. ; and so T)md., 
Gen., and Rh. ; that hadde a fende, Wycl. ; a demoniac, Dr. Camp- 
bell and Dr. Davidson ; one possessed by a demon, Dr. Noyes 
and Mr. Darby ; see on 4, 24. the dumb man spake and saw, by 
omissions from the text after L^chmann, Tischendorf, and Tre- 
gelles ; the blind and dumb both spake and saw, A. V. after Cran. 

— V. 23. ihe muliihideSy close to the Greek, after Rh. ; the people, 
A. V. after Wycl. and the rest. Is this, more in accordance with 
the Greek (Mjjrt ovtoO» after A. V. (ed. 161 1) by a new rendering ; 
Is not this, A. V. (ed. 1638) after Tynd., Cran., and Gen. — v. 24. 
This (ourof) man, supplying man, but not italicizing it, after 2d 
Gen. ; Th\s fellow, A. V. after all but Wycl., He this ; this pronoun 
is often used in classical as well as in Hellenistic Greek by way of 
contempt or aversion ; A. V. sometimes expressed this as here, and 
sometimes disregarded it, as in S. Matt 9, 3 ; S. Mark 2, 7 ; S. Luke 
I5» 2 ; S. John 7, 27 and elsewhere ; A.V. expressed it when the pro- 
noun designates our Lord in S. Matt. 12, 24; 26, 61 ; 26, 71 ; S. Luke 
22, 59 ; 23, 2 ; S. John 9, 29 ; and when it designates S. Paul in 
Acts 18, 13 ; but the Rev. have well changed the expression in all 
these passages to this man. devils — devils : so A. V. after Tynd. 
and all except Wycl., fendis — fendis ; demons— demons, Dr. Camp- 
bell, Dr. Noyes, Mr. Darby, and Dr. Davidson ; and so in vv. 27, 
28, except Wycl., develis — fendis ; see on 4, 24. — v. 25. he, by 
change of text after Lachmann, Tischendorf, and Tregelles ; Jesus, 
A. V. after Wycl. and all the rest, and so Cod. Am. ; see on 4, 12. 
knowing their thoughts he said, after Dr. Noyes and Mr. Darby ; 
knew their thoughts and said, A. V. after Tynd. and Gen. — v. 26. 
if Satan casteth (Greek indicative), after Dean Alford, Mr. Darby, 
and Dr. Davidson ; if Satan cast, A. V. after Wycl. and all the 
rest ; and so Dr. Campbell and Dr. Noyes ; see on 4, 3. how then 
shall, by a change of order after Rh. ; how shall then, A. V. after 
Tynd., Cran., and Gen. — v. 27. them, supplied but not italicized ; 
them, A. V. therefore shall they, by a change of order after Dn 
Noyes ; therefore they shall, A. V. after Wycl. and all the rest ; 
and so Dr. Campbell, Dean Alford, Mr. Darby, and Dr. David- 
son, which is better as giving the Greek more exactly (dia rovro avroO. 



28o AMERICAN JOURNAL OF PHILOLOGY. 

— V. 28. if I by the ^irii of God cast out devils ^ by a change of 
order to conform to the Greek ; and so Wycl. and Rh. after the 
Vulg. then is the kingdom of God come, by a change of order 
after Tynd., Cran., Gen., and Rh. ; then the kingdom of God is 
come, A.V. after Wycl. upon you, closer to the Greek (nri c. ace.), 
after Rh. ; unto you, A.V. after Cran. and Gen. — v. 29. Or, dose to 
the Greek, after Rh. ; Or else, A. V. freely and idiomatically, after 
Cran. and Gen. ; so Dean Alford ; the Revisers have made the 
same change in v. 33, but they have well left the old form in S. 
John 14, II ; Acts 24, 20; Ronv 2, 15; Rev. 2, 5 ; 2, 16; or else 
is thus used in Ellis, Original Letters, pp. 158, 180, 232, 327, etc.; 
Hooker, V. 13 bis, 17, 21, 48; Walton, Angler, pp. 83, 209,225, 227; 
Temple, I, pp. 93, 112, 120; Dryden, Dram. Poesy, p. 94; and 
is still in good use. the house of the strong mafi, after Rh. nearly, 
the house of the strong ; a strong man's house, A. V. after Tynd., 
Cran., and Gen. — v. 30. scattereth, after Rh. ; scattereth abroad, 
A. V. freely and idiomatically, after WycL and all the rest. — v. 31. 
Therefore, closer to the Greek (At^ tovto), after WycL and Rh.; 
Wherefore, A.V. after Tynd. and the rest Every sin, after Rh. and 2d 
Gen. ; All manner of sin, A.V. after Tynd., Cran., and ist Gen. the 
Spirit, close to the Greek, after Tynd., Cran., and Rh. ; the Holy 
Ghost, A. V. supplying Holy, after Gen ; see on v. 32. forgiven, by 
an omission from the text after Lachmann, Tischendorf, and Tre- 
gelles ; so WycL and Rh. after the Vulg, — v. 32. shall speak, more 
correctly, after Gen. and Rh. ; speaketh, A.V. after Tynd, and Craiv ; 
and so again in this verse. th€ Holy Spirit, after Dr. Campbell, Dr. 
Noyes, Dean Alford, Mr. Darby, and Dr. Davidson ; the Hoiy 
Ghost, A. V. after WycL and all the rest; the A. V., as is well 
known, employed both forms, the Holy Ghost and the Holy Spirit; 
the Rev. have changed the former into the latter in the following 
passages: S. Matt 12, 32; S. Mark 3, 29; 12, 36; S. Luke 2, 25 ; 
2, 26; 4, I ; 12, 10; 12, 12; S. John i, 33; 14, 26; Acts 2, 4; 6, 
5; I Cor. 12, 3; S. Jude 20; they have employed the form the 
Holy Ghost seventy-two times in all: and the form the Holy 
Spirit nineteen times, nor in that which is to come, after Dean 
Alford nearly, neither in that which is to come ; neither in the 
world to come, A. V. after Tynd., Cran., and Gen., repeating 
world from the forgoing, as it did in 5, 20, which was there 
followed by the Rev., and the effect is excellent here. — v. 33. its 
fruit, after Dr. Campbell, Dr. Noyes, Dean Alford, Mr. Darby, 
and Dr. Davidson ; his fruit, A. V. after Wycl. and all the rest ; 



THE NEW REVISION OF THE NEW TESTAMENT 28 1 

and so twice more in this verse ; see on 5, 13. ^r, after Rh. ; or 
else, A. V. after TyncL, Cran., and Gen. ; see on v. 29. — v. 34. 
Ye offsprings after Dr. Campbell and Mr. Darby nearly, Offspring ; 
O generation, A. V. after Tynd. and Cran. — v. 35. The good man, 
preserving the Greek article, after Dr. Campbell, Dr. Noyes, Dean 
Alford, Mr. Darby, and Dr. Davidson ; A good man, A. V. after 
Wycl. and all the rest; and so again in this verse, hts good 
treasure, by an omission from the text after Lachmann, Tischen- 
dorf, and Tregelles ; and so Wycl, and Rh. after the Vulg. ; the 
good treasure of the heart, A. V. after Tynd. and the rest his 
evil treasure, after Tynd. and Gen., the Greek article having a 
possessive force ; the evil treasure, A. V. by a new rendering, the 
rest neglecting the article; see on i, 24. — v. 36. And (bi), after 
Wycl. ; But, A. V. after the rest ; and so Dr. Campbell (^however'), 
de Wette, Germ. Rev., Holl. Rev., Dean Alford, Mr, Darby, and 
Dr. Davidson. — v. 38. 0/ the scribes and Pharisees, close to the 
Greek, after Rh. ; of the scribes and of the Pharisees, A. V. after 
Tynd., Cran., and Gen. ; see on 10, 18. ansivered him, by an ad- 
dition to the text after Lachmann, Tischendorf, and Tregelles ; and 
so WycL and Rh. after the Vulg. ; answered, A. V. after Tynd. 
and Gen. — v. 39. Jonah ; Jonas, A. V. ; and so in vv. 40 and 41 ; 
see on 1,2; Jonah the prophet, by a change of order to conform to 
the Greek ; and so Wycl. and Rh. after the Vulg. ; the prophet Jonas, 
A. V. after Tynd. and the rest, — v. 40. the belly oj the whale, 
after Dean Alford ; this is the form of Wycl., Dr. Campbell, Dr. 
Noyes, and Mr. Darby ; the whale's belly, A. V. after the rest — 
V. 41. shall stand up, after Dr. Campbell and Dr. Noyes, nearly, 
will stand up; shall rise, A. V. after WycL and all. in the judge- 
ment, preserving the Greek article, after Cran. and Rh. ; in judge- 
ment, A. V. after Gen. Jor (on here = ydp), after Wycl., Tynd., and 
Gen. ; because, A. V. after Cran. and Rh. — v. 42. the ends, after 
Wycl. and Rh. ; the uttermost parts, A.V. by a new rendering ; the 
utmost parties, Tynd., Cran., and Gen. — v. 43. But — when, pre- 
serving the introductory particle, after Dr. Noyes, Dean Alford, 
Mr. Darby, and Dr. Davidson ; And when, Rh., the rest neglecting 
this particle, the man, preserving the Greek article, after Mr. 
Darby and Dr. Davidson ; a man, A. V. after Wycl. and all the 
rest But the unclean spirit, when he is gone out oJ the man, by a 
new order ; when the unclean spirit is gone out of a man, A. V. 
according to the Greek, after Wycl. and all. passeth, after Sir 
John Cheke ; walketh, A. V. after Tynd., Cran., Gen., and Rh. 



282 AMERICAN JOURNAL OF PHILOLOGY, 

waterless places : and so in the parallel passage, in S. Luke ii, 24, 
by a new rendering and the use of a word not Biblical, but which 
Richardson cites twice : The sea (shall be) waierlesSy Poems of Sir 
Thomas Wyatt (d. 1552), Pickering's ed., p. 12 \ places barain 
and water lessi, S. Luke 11, 24, Nich. Udall, Trans, of Erasmus' 
Paraphrase on the Gospels and the Acts (c. 1550) ; dry places, 
A. v., excellently, which all the ancient and modem versions 
have, except Dr. Campbell, parched deserts, which is also good 
but free. Jindeth it not, closer to the Greek, after Dr. Noyes and 
Dr. Davidson; findeth not, Wycl. and Rh. still closer; findeth 
none, A.V. after Tynd., Cran., and Gen. — v. 44. whence, after Rh. ; 
from whence, A. V. after Wycl. and the rest, which is more rhyth- 
mical here. — v. 45. more evil, after Dr. Davidson ; more wicked, 
A. V. after Rh. becometh, after Dr. Campbell, Dr. Noyes, Dean 
Alford, Mr. Darby, and Dr. Davidson ; and so in effect Wycl and 
Rh., ben made. — v. 46. W/ule he was yet speaking, closer to the 
Greek, after Dr. Noyes and Mr. Darby ; and so nearly Rh., As he 
was yet speaking, multitudes, closer to the Greek, after Rh. ; 
people, A.V. after Wycl. and the rest, his (ij) mother, the Greek ar- 
ticle being used as a possessive ; his mother, A.V. ; see on i, 24. seek- 
ing to speak to him, after Wycl. and Rh. ; desiring to speak with 
him, A.V. excellently, after Tynd., Cran., and Gen. — v. 47. And (d«) 
one, after Rh. ; Then one, A. V. freely, after Tynd., Cran., and 
Gen. seeking to speak to thee, after Mr. Darby ; seeking to speak 
with thee. Dr. Noyes ; seeking thee, Wycl. and Rh. after the Vulg.; 
desiring to speak with thee, A. V. after Tynd., Cran., and Gen. — 
V. 49. towards his, after Dr. Campbell and Dr. Davidson ; toward 
his, A. V. after Cran. and Gen. ; and so Dr. Noyes and Dean 
Alford ; see on 5,' 25. — v. 5a he {a\rT6i) is my brother, and sister, 
and mother, after Wycl. and Rh. ; the harshness of this expression, 
unavoidable in Greek if the pronoun be expressed, is well avoided 
by A. v., the same is my brother, and sister, and mother, after 
Tynd., Cran., and Gen. ; and so Dean Alford and Dr. Davidson. 
This is merely allowing the translation the feiir advantage of an 
English form, which the Revisers themselves have done in ren- 
dering the pronoun (ovros) in the parallel passage S. Mark 3, 35 ; 

see on 5, 19. 

Charles Short. 



II.— ON THE ENGLISH DATIVE-NOMINATIVE OF 

THE PERSONAL PRONOUN. 

The syntax of the personal pronoun in the second person has 
been treated by many writers ; but, so far as I know, the history 
of the singular dative-nominative has never been made the subject of 
special investigation. The grammars, at any rate, have little or 
nothing to say on the subject. Koch actually confounds dative 
and accusative ; cf. Zupitza's correction in the new edition (Cassel, 
1878) of the second volume of the grammar, §324. Matzner is 
interesting with regard to the impersonal verbs with datives (cf. 
below), but does not treat our subject directly. I have not Zupitza's 
note to a passage in Guy of Warwick^ a note referred to in the 
grammar, treating of the dat.-nom. you. But with regard to the 
still earlier dat-nom. for the singular of the second pers., there is 
nothing, that I am acquainted with, which bears directly on the 
subject What here follows is mere sketch. The pressure of 
other work forbids any attempt on the writer's part to make a 
thorough study of the whole matter. Lack of material has pre- 
vented any consideration of the dialects. 

First, we must connect the later change (^ye to yoti^ u e, the 
formation of a plural dat-nom.) with the earlier like tendency in 
the singular, notably the Kentish sing, dat.-nom. f^ for fw. They 
result from the same cause. After the dative had driven out the 
ace. forms of the pers. pron. (cf. Koch, Saizlehre^ §313. §3i4i§324), 
it turned toward the nominative. The dative, in fact, had long 
performed nominative functions. There is the well-known con- 
struction with sel/y as in Koch's example, Leoma \te seolfa^ or in 
Andreas y 1348, 3^ \e syl/a id. Besides the pers. pronouns, the 
definite article itself undoubtedly takes its rise from the dative 
form : cf. mid \e king^ of \e temple ^ on \e circe ; and note to Old 
Eng. Homilies (Morris), p. xxxviii. In the Chronicle (ed. Earle, 
p. 260) we have the new nom. ^e kingy and the old nom. se kingy 
within five lines of each other. The dat in impersonal construc- 
tions, and the so-called ethical dative, we shall glance at below. 

We turn to the regular dat-nom. \e zxAyou. As to the second, 
Zupitza shows (in his edition of Koch) that the dat-nom. you 



282 



AME! . 



\ .-uT, 



waterless place: . 
by a new rendci 
Richardson cit« ^ \ 
Thomas Wyatt 
and wateriest. 
Paraphrase o\ 
A. v., excel^ 
have, excej)i . 
but free. /. 
Dr. David . 
none, A.W 
from whcij ' 
mical her- . 
A. V. aiV 
Alford, >: 
Rh., be: 
Greek, 
was y- * 
peopl* 
tide! 
inir f 

him, 

0V(' 

Gt 

\s': 



1 



V. 



. He 15th century : 

.- . jt does occur so 

1 England, the new 

,jst ibr example the 

^ the second pens, is 

»_r iv the dat.-nom., so 

_•/• rewarded. Take the 

1 ; ue. the two periods 

I ^5 to 1509. The first 

iaston Letters, shows as 

k almost no use) as the 

In 1449, Margaret 

M ye have another sone 

. * William Tailboys, about 

^snd to a prisoner* before all 

X hailed/' A litde later, 

^Na 159) sa)rs : ^^you have 

under Henry VII, the ye is 

:^ earliest letters. More than 

X ciriier sts^es of the language 

.= wm nominative constructions : cC 

ccer 929) with Chaucer's him itiste 

,r VoL III, p. 370) ; wkane ye like, 

^^J // thee well hadde likedr It 

^ jkX^ thus : " Sir, ther am XV jurores 

^ .BBi " I pray ye that," etc., I 7a 

.^<^. «e have the ace. " I pray ^w." The 

^ ^x^euskm, cited by Koch, is Shakspere. 

j^. J^oucer. C£ Troylas and Cryseide, 

.^ <^u<«eof Troylus to tellen . . . 
, ..^ ^ er |>at I parte frojte. 



Harleian 2280 has the S2JXi^,froye. 
^ jHijvtt foils to see this, and spoils the rime 



j^e ^ston Letters from beginning to end, 
^itt tbem alone the prospects of our now 
ptitfuOQC would find little encouragement 
%^ to 1509) there is not the slightest indica- 
Tbe most striking example of our dat.- 
^^ vntten about 1450 ; the later letters write 



THE ENGLISH DATIVE-NOMINATIVE, 285 

regularly ^'^ for nom. But if we turn to the Kentish A^enbiie of 
Imvyty written (1340) a century before these letters, we find a use 
of the dat-nom. in the second person of the singular pers. pron., 
that would justify the prophecy of a speedy substitution of \e for 
\tu It requires no very elaborate reasoning to prove that the 
singular was about to travel the same path that the plural trod 
later, was about to establish a dat.-nom. thee^ as it afterwards estab- 
lished ^'^w. But just here began the change from singular to plural 
form. '* Im Ne. (new English) gilt ye schon im 15 und 16 Jahrh. 
als das hoflichere," says Koch (p. 231). The singular was isolated. 
A hedge was set about it. It was reserved for solemn purposes, 
and was thus removed from the influence of linguistic change. It 
became one of those forms that men use consciously^ with effort ; 
just as we use ye with effort Most men use any case of the sin- 
gular pers. pron. second pers. only on especial occasions, and with 
this conscious effort. 

Most men, but not all. Koch is mistaken when he says (§299) 
that tkou is retained in " dem allgemeinen Gebrauch der Quaker." 
In point of fact, few members of the Society of Friends use thou 
in familiar speech. They use the singular in familiar speech, but, 
obedient to the tendency, it is the dat-nom. ihee^ not thou. Just 
2&you does service for all plural cases, so thee for the singular. 
This is well known to be the common household practice of 
Quakers. A few isolated exceptions only prove the rule. I have 
seen a familiar letter of an educated Friend, written in the early 
part of the i8th century, where the thee is used as nom., though 
any solemn passage calls out a formal thou. We shall see below 
what Dr. Abbott brings forward as reason for this Quaker prac- 
tice. Then we have the dialects — a field whence I am shut out 
through lack of material. But any reader of George Eliot*s Adam 
Bede^ of Tom Brown at Rugby (early chapters), or of any such 
books, will recall a host of instances of this sing, dat-nom. In the 
Eng. Gram, of Fiedler and Sachs (Leipz. 1877, p. 311) are noted 
such forms as The ^£y/ (Shropshire), Thee wart (Somerset and 
Wiltshire), as compared with the northern Thou is. 

Now let us glance at Dr. Abbott's explanation of certain forms 
with thee used in Shakspere. In his well-known Shakespearian 
Grammar ^ §212, he refers such expressions as " look thee," " hark 
thee," to the principle of euphony, " Thee^ thus used," he says, 
" follows imperatives, which, being themselves emphatic, require an 
unemphatic pronoun. The Elizabethans reduced thou to thee. 



288 AMERICAf/ JOURNAL OF PHILOLOGY. 

>e ssoldest ": whereas, " to ham Jiet fou ssddest "; (p. 52), " and 
|>es5ett conne"; (p. 24.1), "^sselthabbe"; whereas (p. 29), " foo 
sseitywyte"; (p. 54)1 "y^f* ^'V't "l whereas (p. loi), "yef fou 
wylt "; (p. 73), " ye woldest ^ ra|q% lete be ulaje quik "; but (p. 
146), " l>et \oa nddest "; (p. 90), " |>is ^-self |>e mijt yzy "; but (p. 
133). " fanne fou mi^ fitiyt gaderi"; further (p. 187), " Vayre 
zone hi merduol as ^ mijt, yef ^u best yiii^ of guode "; (p. 
232), " do hardeliche alsuo mocbe ase ^ mi}t . . . uor pou oe mi^t 
do no ^g "... (p. 224), " )>ench ase mocbe ase ^ mijt "; (p. 369), 
" be auenture \k my]t eft by onderuonge." The anak^ous confu- 
sion of \e and \ee wQl occur to every one. Had yu and ye con- 
tinued in common use, \e would have displaced \u, just as it 
previously displaced yec. 

The modern Quaker thee does not take a verb to correspond. 
Where the Kentish has ye sselt, ye multiplUst, the modem forms 
are thee shall, thee mulliplUs. The reason is perhaps twofold. 
The northern form of the second pers. sii^. of verbs (ending in 
-s) did its share. But the impersonal form of the verb with dative 
construction was a greater power. Thus the imp, form from P. 
Plow, already quoted : " If thee wel hadde liked," or Shoreham's 
" Levedy, the was wel wors." 

As 10 general reasons for the usurpation of nominative iiincdoas 

by the dative case, we have no space here for inquiry. Perb^is 

we may bring to bear on the question Schleicher's remark at the 

close of his chapter 00 the personal pronoun {^Comp. 4th ed. p. 

641 : " Es scheint als ob das deutlicbe hervmtreten dcr stamme 

I und n person in den sprachen vermieden sei : iiilleicbt 

wir hierin eine art euphemismus zu erkennen, mieja vil- 

% den sprachen eine scheu vor dem nennen des ' Uk ' tmd 

ich leigl." Thus one prefers " me seems " to " I see." We 

; personal part in an oblique case, rather than in the nom. 

latever the general reason, the tendency was helped by the 

y of the impersonal verbs in dative ocmstruction. 

: construction depends entirely on the inflexions. When 

>ecome less and less used or understood, which is the case 

he progress of Ei^lish, either the construction itself will 

ito disuse, or else it will be otherwise understood. Take the 

sion " (« bihoueC godes helpe." pf is dative, kelpe is geoi- 

But such syntactical rdations &1] out of use. The common 

landing is £iin to take ^ as a nominative, kelpe as direct 

of the verb. Thus arises a dative-nominative ; so lo-day. 



THE ENGLISH DATIVE-NOMINATIVE, 289 

ninety-nine people out of a hundred understand me thinks as / 
think. Only with the second person, however, did this dative- 
nom. take firm hold. There was a tendency to change the case 
as well as the general construction. An interesting example occurs 
in Hen. VI, Part II, 3, 2. The king says : ** Woe is me for Gloster, 
wretched man "; whereto the queen replies : " Be woe for me" 
understanding thou. So O. E, H. p. 31, "^if him is laf," but p. 
35, " he is laf," and p. 39, " ne beo eow noht la|>." Cf. also con- 
structions in Past Letters, quoted above. The much used imper- 
sonal constructions of Early English must have helped largely in 
the formation of a dative-nom. Matzner gives a number of these. 
Thus (Orm. 2050) " I>e birrf ec hire taelenn "; (Joh, 16, 7) " 6ow 
frema^ }>aBt ic fare *'; (Caedm. 3649) ** hd f 6 swefnade," etc. We 
may find expressions where a dat.-nom is still plainer. Thus 
Chaucer, Prol. to Wife of Bath's Tale, 329 : 

Have thou ynough, what thar the recch or care 
How merily that other folkes fare? 

Frere's Tale, 67 : 

The thar no more as in this cas travayle. 

Better still, Frere's Tale, 103 : 

If that the happe come into oure schire. 

In O. E. H. 195 {On god ureisun, etc.) we have, " |>et I>e ne 
wontep." These sound like regular nom. constructions, and were 
in time so regarded. The verb was put sometimes in second pers. 
sing., showing that the impersonal construction was forgotten. 
Instructive here is the reading of the folios in Hamlet, V 2 : 
"Does it not, thinkst thee^ stand me now upon," and this is 
strengthened by the usage of many dialects (as opposed to the 
usage of Friends), as well as by such expressions as Defoe*s 
"What ailest thee now?" (cf. Minto, Prose Manual, p. 409). 
Elworthy, in a paper in Trans. Phil. Soc. 1877-9, notes for West 
Somerset the form " Thee art" 

But there are other forms, e, g, thee are (addressed to one 
person), a compromise between sing, and plur. sometimes heard. 
The most remarkable case I ever observed was where a lady, not 
a Friend, extended to several visitors, who were of that sect, an 
invitation as follows : " Won't thee all walk into this room ? " 

The " ethical dative," finally, did its part to help the dat-nom. 
We know how &miliar the former was in O. E. poetry. Thus 



III.— PARTICIPIAL PERIPHRASES IN ATTIC PROSE. 

Not infrequently in classic Greek we find the combination of a 
participle with tivai or yiyvtoBai used when a finite form of the verb 
represented by the participle might rather have been expected. 
Such phenomena have nowhere received, I believe, any exhaustive 
treatment, so that the force of these periphrases has not been clearly 
determined, much less have the limits of these combinations been 
fixed either for the whole range of classic Greek or for individual 
authors. It is true the subject has been incidentally treated in the 
grammars and commentaries, but such treatment has been based 
on a range of examples at once too wide and too narrow, and 
without regard to the very different categories under which they 
fall. In consequence the deductions are uniformly loose and inade- 
quate, sometimes erroneous and contradictory.* Those who see 

' Ktihner (II, §353, 3) says : ** Urn dem Prftdikate ein grdsseres Gcwicht zu 
geben, zerlegt die Sprache zuweilen den einfachen Verbal-aasdruck desselben 
in das Partizip und die Kopula tlvai.^* Again, ** HSuBg ist sie auch in der 
Attischen Prosa wenn eine Handlung als bleibender Zustand bezeichnet werden 
soil." He also notes the frequency in Plato of Ix*^ tlvat. (He should rather 
have said elvai ^x^^* ^3 cases against 10.) Bernhardy, on the other hand (Syntax* 
P* 334)* considers such expressions are ** ohne eine gewSlhlteren Sinn (wie die 
Lateinische Formel dieser Art) oder den Ausdruck der Dauer den man in 
einzelnen Phrasen, worunter das Platonische ierlv ixov, zuweilen beabsich- 
tigte. Aber ein itv neben Participien vermied roan als zwecklose H&rte 
und so erkennt man in solchen Stellungen nicht sowohl die participiale 
als adjective Bedeutung.** KrOger (§56) cites indiscriminately a number 
of examples without offering any explanation. Madvig (I quote from the 
English translation) says: *'Some few present participles, viz., Siatftipov, 
IXtJV with an adverb, irpoc^Kuv, irptrrcw, cJ/ov, cfdv, avfu^pov, sometimes occur 
as adjective predicate nouns, with rt///, or yiyvofuut occasionally also others in 
connection with an actual adjective." Again, *'A participle of the present or 
aorist with tifil^ as a periphrasis of the simple tense of the verb (in like manner 
as the partic. pf. under certain circumstances is joined to ti^ii) is a poetical 
licence of not very frequent occurrence ; in the prose passages where it does 
occur there is apt to be a certain emphasis in the several and distinct expres- 
sion of the action (the partic.) and its existence (' <^V* Classen, in his Thucyd. 
(Anhang, Bk. I, i, i) draws attention to the different character of these combi- 
nations according to the position of tlvai, and observes that only adjectivized 
participles are used in this way in Thucydides. 



PARTICIPIAL PERIPHRASES IN ATTIC PROSE, 293 

Here then is the due to the differentiation of Xua and \v<ov iari. In 
the latter case we have an adjective and a copular verb, and hence a 
permanent quality predicated of the subject ; in the former we 
have an activity or series of activities predicated. This principle 
we shall find as we proceed to include what fragments of the 
truth the opinions already quoted contain, and will account for the 
phenomena they note. But to give definiteness to our results it 
will be well to throw the cases into their natural categories and 
then examine each group by itself. It is manifest, in the first place, 
that the participles of the several tenses do not lend themselves 
with equal readiness to this use. It will therefore be advanta- 
geous to treat separately periphrases into which present, aorist and 
perfect participles respectively enter.* Again, since efwu in an 
emphatic position may cease to be purely copular and come to 
contain the predicate within itself, we will separate cases in 
which the cu^oi (ylyyttrOai) follows the participle fi^om those in which 
it precedes, and, for convenience, will term the former periphrases 
of the Ftrsi Form, the latter of the Second Form. 

Present Participle. First Form. 

Under this head we have to consider such periphrases as Xvo>y 
fWi, which, according to the principle just laid down, asserts the 
existence of a certain quality in the subject in contradistinction to 
the predication of an action or series of actions on the part of the 
subject Supposing this to be true, is there any demand in the 
language for periphrases with such a function ? There is, provided 
the language has not an adjective already formed for the expres- 
sion of every conception which under any circumstances might 
possibly come to be regarded as a quality. A stock of adjectives 
sufficient to cover such a proviso, the language of course does not 
possess. There are, in the first place, certain verbs in Greek 
expressing the manifestation of a quality while at the same time 
no adjective exists for the expression of the quality itself. Such 
verbs, e. g., as npwr^Mip, nparuv (cf. Lat. convenit, decei, and Eng. 
' becomes,' ' befits '), dia^pccy, avful^ptiy and 6fioKoyti<r6ai. In each 
of these cases the quality in question has no existence in a single 
object but in relation to two objects, and the coming together of the 
two objects was, by the earlier language-users, regarded as the 

* The fut partic. scarcely occurs. I have noted one case, PI. Tim. 38 B, 
where the periphrasis results from symmetry. 



PHILOLOGY. 

>f this view is confirmed 
aiv,£tmveml, ' becomes.* 
e quality apart from its 
d to express this quality 
sloyed. Meanwhile the 
ad the distinct notion of 
It in the main the mani' 
m is equivalent to the 

ises, the language often 
1 of a verb as a quality, 
irb naturally lends itself 
Is an example ; 6^ i^yt 
opined is erroneous "/ 
is in the main a fiction, 
nee of a quality as of a 
L Isoc. Epist 9, 13 ; PI. 

f the type ina(iTa»6iur6r 

cond class, since, while 
[horoughly adjectivized 
be peculiarity of having 
inders the use of the 

pies as have nothing in 
) be used as adjectives, 
these periphrases : e. g. 

Itaiiff q/uv rk rav f6iutr 

itended to bring before 
a characteristic of the 

aiKOXiv tiiai/tOHimrrot xal 
SifliOTufif r« ■□! n-aXincqr 

jective dearly indicates 
loted that characteristic 
; of continuance, which 
tion or quality, for this 
lertion of the existence 

■n be diKovered in the uie 
irticipl«) an so (horoughljr 
ul it m>y occasioniUy be 
w; AA.Hai. IQ.Taikaia 



PARTICIPIAL PERIPHRASES IN ATTIC PROSE. 295 

of the corresponding characteristic or quality in the subject The 
finite form includes the periphrastic ; not conversely. The finite 
form has other functions and lacks definiteness ; the periphrastic 
is an accurate expression for the existence of an aptitude or ten- 
dency, not necessarily manifesting itself, but, it may be, merely 
potential. The conception, however, of a quality apart firom its 
manifestation is an abstraction, very necessary to the philosophic 
thinker, but not likely to be much used in practical life, where 
acts, not potentially existent tendencies, are of prime importance. 

In the following list of occurrences in the Orators of periphrases 
of the first form containing a present partic, such cases as Dem. 

3, 25 are kept by themselves : ovra> aoxf)poy€S ^aay xaX <r<t>6lipa iv T^ 

nokiTtiat ijBu /xcvovrcff &<rrc . . . Here the participle is parallel to an 
adjective' which precedes it, and by that p^allelism at the same 
time the function of the participle is clearly indicated and any 
harshness there may be in the combination is mitigated. 
wa^€x6iuvo^, Dem. 19, 37 ; dpjc»y (4 times), Ant. 2, ^, 2 ; 2, y, 3 ; 

2, d, 10 ; 4, y, 6 ;* drifiOKpaTovfi€yotf Dem. 24, 5 t iui<t>*p^v (7)1 Ant. 
5, 88; 6, 6; Isoc. 7, 45; 12, 120; Epist. 2, 3; Aesch. 3, 162; 3, 
168; f KCTTpaTcvcJfteKoi, Lycurg. 107 ; lx»*'(2), Isoc. Epist.9, 13; Dem. 
31, 11; 6fio\oyovfi€¥os (5), Isoc 6, 14 ,* Isae. 2, 40; Din. i, 90; 
Lyciu'g. 36 ; Dem. 20, 32 ; wotovfift^, Dem. 19, 37 ; wparav (6), 
Lys. 3, 9 ; 19, 59 ; Isoc. 6, 90 ; 15, 74 ; Epist 5, 3 ; 6, 7 ; npointKtiv 
(5), Isoc. 12, 124; Isae. 7, 14 ; Dem. 45, 49; 69 ; 48, 6; avfKJHpmv 
(8), Lys. 12, 7; Isoc. 14, 25; Epist 5, 3; Lycurg. 37; Dem. 16, 
10; 19, 75; 161 ; 24, 24.' 

Total number of cases 41, in which 11 different participles 
appear. In addition to these we have 25 cases where the parti- 
ciple is parallel to a preceding adjective : 

Isoc. 6, 72 {Apfi6TTorra) ; 8, 36 (frpoa^Mp) ; 12, 1 83 (irp€vmv) ; 1 5, 47 
(dvMi/xc»or) ; 77 (irpcjrttv) ; 9 1 (dvyd/icyor) ; 1 87 (Jiia<f>€pwv) ; Aesch. 

3, 28** (Xvciv); I, 141 (ir€pi<t>pop&v) ; Dem. 3, 25 (jutwv); 19, 25 

(yntpfiaXkwv) ; 20, 8 (irpcirov) : 55* (toiovitcp) ; 1 53 («aX«ff l;(«r) ; 1 57 
(icanttr ^X^O f 21, 66 (avfjLtf>€pov) ; 20I'*' (juya <f>pov&Vt fitya ^cyy($/iCVor) ; 

* In some cases a noun, with of course no essential difference {vid, PL Sym. 
191 D). I also include those cases in which a pf. participle precedes. 

* &pXi>nf is conjectured [And.] 4, 30. Since it has become a noun I have not 
noted all occurrences of this word. 

* In this and following lists Ktlfievo^ and its compds. are not included, inas- 
much as they were regarded and treated as pfs. 

^ In the cases marked thus * the copula it not expressed. 



PARTICIPIAL PERIPHRASES IN ATTIC PROSE. 297 

1 , 38, 4 is worthy of note, since there afUaKovrh iafLtv and afrapcVcoi/icy 
nv are used side by side, the former of an abiding quality which 
characterized the relations of the Corinthians and their colonies, 
whilst the latter refers to certain special acts. To the limitation of 
periphrases to adjectivized participles there may be one exception, 

unless Cobet's emendation of /ierairevrr/x/icVoi for MS fiercarffiirdfitpoi be 

adopted. As the passage stands, there is, according to the prin- 
ciples enunciated here, no fitness in the periphrasis ; but a slight 
change of the masc. to neuter ending, -01 to -a, would bring it into 
accord with them, and the meaning would be * such things as were 
imported/ But, as there is nowhere else in Thucy. so striking a 
case of periphrasis, I would prefer Cobet's reading.* 

I now pass to a list of the occurrences in Plato : alaOca^iiivot (4), 
Theaet. 159 E; 160 B (ier)] ay6ii€vov (2), Euthyph. 10 B {bis); 

dfxapToydfuvov, Phil. 37 E ; avofioKoyovfitvoSf Gorg. 495 A ; aTTOYrXi/pMy, 

Nom, 932 B ; ap/iorrw, Nom. 808 B ; &pxo>v (2), Rep. 558 D ; Tim, 
44 A ; dpxofuvosf Nom. 715 D ; ^Xcttwv, Nom. 963 A; yiyv6fi€yos (6), 
Euthyph. 10 C{dis) : Craty,4ii D ; Phil. 42 D ; Hip. Maj. 297 C ; 
Nom. 935 D ; dt6fi€vog, Nom. 768 E ; dtoy (7), Nom. 649 C (^w) ; 
793 E; 796 C; 800 E; 802 DE; Epin. 990 C; dia<f}€fmv (8), 719 
D ; 720 E ; 729 C ; 743 C ; 779 E ; 794 D ; 901 B ; Epin. 987 C ; 
dui^tip6fifpost Rep, 492 A ; ^«i<^^«p«v. Rep. 492 A ; dp&y, Tim. 33 
D ; cyytyvoftevop, Theaet. 187 D ; cWtinav (2), Theaet 157 E ; Nom. 
960 D ; €mTperrw, Nom. 932 B ; €tr6iitvoi (6), Pol. 271 B ; Rep. 412 
B; 461 E; Tim. 42 A; 54 D; Nom. 716 D; ^x'^v (lo), Craty. 
386 D ; 391 A ; Soph. 253 C; Parm. 165 A; Tim. 66 A ; Nom. 
713 B; 735 D; 798 E; 860 E; 967 A; ix6iifva, Nom. 828 A; 
rfy€iAovovVf Nom. 63 1 C; Brjpiovfitvos, Nom. 935 A; KOKovpyS>Vy Nom. 
933 A ; Kartxo^vy Tim. 52 B ; Kwovii^va, Soph. 249 B ; XrytJ/^i^or, Nom. 

719 C ; Xci9r($f(cvoy, Nom. 807 A ; XvcnrcX^v, Nom. 662 C ; iULiv6fuvoi 

(2), Prot. 350 B ; Nom. 934 C ; fi€T€xov (2), Tim. 58 D ; Nom. 859 
E ; fii(roOvrc9, Nom. 908 B ; ofioKoyovfityos (2), Craty. 387 D ; Phil. 
12 A; ofioifoStv, Nom. 759 B; ofmfitvos (4), Euthyph. 10 B (^w); 
Tim, 56 C; Epin. 984 E; Sp&y, Theaet. 164 A ; ^ciXofteror, Rep. 
332 A ; 54^Xo>v, Nom. 909 B ; nap€n6fi€Pos, Nom. 667 D ; naptx^y, 
Tim. 33 D ; ira«rx«v (4), Euthyph. 10 C {6is) ; Theaet. 157 A ; Tim. 
33 CD ; irotoiJv, Theaet. 157 A ; nptirmy (23), Lach. 188 D ; Gorg. 
504 A; Hip. Maj. 291 C; Tim. 21 A ; 33 B; Critias 112 B; 117 
A ; Nom. 665 D ; 670 D ; 755 C ; 7^4 C ; 779 C (*w) ; 796 C ; 

* Herbst (pp. 37-9) attacks Cobet*s emendation, but fails egregiously in making 
hit point. 



298 AMERICAN JOURNAL OF PHILOLOGY. 

800 E ; 801 E ; 804 E ; 818 A ; 855 A ; 917 A ; 920 E ; 950 C 
956 A ; vpwrqKtoy (10), Phil. 36 D ; Phaedr. 227 C ; Rep. 404 A 
442 D; 525 B; Norn. 713 C; 724 B; 751 B; 775 A; 904 E 
avfjuf)€fMv (6), Phil. 63 A; Alcib. I 114 E; 116 D ; Hip. Maj. 290 
C{6ts); Ion 540 C; trvv€n6fi€yos (2), Critias 117 A; Nom. 858-9 
A ; <n/vfx^^<^» Soph. 253 C ; riBifuvot, Nom. 822 E ; virapx»p (4), 
Rep. 458 A ; Nom. 774 C ; 846 E ; 923 D ; <t}tp6fi€vos (4), Euthyph. 
10 B (Jer) ; Soph. 249 B ; <t>tXovfitvo£ (4), Euthyph. 10 A ; C (dts) ; 
D ; xalfHov, Alcib. II 139 C ; &v (5), Soph. 240 B ; 245 C ; 256 D ; 
263 B ; Tim. 38 B. 

Total number of cases 151, in which 53 different partic. occur. 
In addition we have 58 cases where a parallel adj. precedes : 

Euthyph. 5 D (c^ov) ; Apol. 40 A (cVavriov/uvij) ; Phaedo 82 A 
(l6vTts); Craty. 408 C* (oImvv); 440 A (^xoO; Theaet 178 C 

(wvpmav); 182 B (^al<r0ap6fxfvos) ; Soph. 225 A* (irp€vrop); 229 D 
(JlXOiOl Parm. 144 A (jurtx^^v') ; 145 A* (j^xotdl 150 A (ntpuxovaa) ; 
159 A* (Kivovfifva) ; 166 B* (AnrSfitva) ; Symp. 175 E (Ixot/cra) ; 

206 D (ApfjL&rrov) ; Alcib. II 139 D (naiSfitvos Koi paKXofuyoi) ; Anter. 
135 B {npoafJKoy); Charm. 169 E (yiyyoKrieav) ^; Lach. 188 D 
(apf«JiTov); Prot 349 D (dco^'/Kov); Gorg. 493 E (cWo/xfrf/icwi) ; 
Meno 72 B (dia^'p«v) ; Hip. Maj. 289 E (ajrodtx6tt«voi) ; Rep. 423 C 

(doKovaa) ; 441 E (irparrani) ; 524 B (dfrfftcva) ; 552 E Ox^v) ; 568 A 
(dmi^'fwv) ; 571 C* (tipxoy); 577 E (n€v6fuvos') ; 596 D* (^ly/aovpyov- 
fi€vor) ; Tim. 36 E* (jiLtrix'»y) ; 44 ^ (decnroroCv) ; 87 C* (np€Wo>) ; 

Nom. 625 A* (vptnop); 666 E (dwdfuvosi) ; 716 D {vpiftov) ; 723 C 

(duiifHpov) ; 732 B (XcycJ^cva) ; 747 C (npoarjicov) ; 763 D (<rxoXaf«v) ; 

765 A (anoMovi); 775 D* (txdfttpa); 78 1 D (wpcVwv) ; 798 D 
(dfcJfwva) ; 821 A {iTviKt^tpov) ; 837 B* (?xov) ; 875 D (Jipx^i) ; 875 E 

(iia(f>€pov) ; 876 C* i^xov) ; 892 BC (^apx6iKva) ; 976 D (Spx^ap, apx6' 

fiMPoi) ; Epin. 981 E (^6p6fi€pa) ; 992 C* (furtxo^y)' 

It is Plato who affords the best field for the investigation of these 
forms, inasmuch as he employs them not only with accuracy but 
with freedom. In Euthyphro the argument hinges on the distinc- 
tion between <t}tXovfi(v6y iari and <^iXr(rai. Euthyphro has defined 
(9 E) rh oaiov as * that which the gods love.' Whereupon Socrates 
asks, *' Is that which is holy, loved by the gods, because it is holy, 
or holy, because it is loved by the gods ? " and illustrates his mean- 
ing thus ; t6 <t}€p6fi€yoVf di6Ti t^pcrai, if>€p6fi€y6if iariv • • • ovk apa Mri 
6p^fxtp6v yc cWi, di^ rovTO 6paTaip dXX^ rovvapriov Si&n Sparai^ dt^ rovro 

Sponfuyoy^ t, tf. * It is not because a thing possesses the quality of 
visibility, that it is seen, but because it is seen, the quality of visi- 



PARTICIPIAL PERIPHRASES IN ATTIC PROSE, 299 
bility is predicated of it.' He then makes the general statement : 

ci Ti yiyyrrcu ij n irdaxf^ «, olx ^* yiyt'6fi€v6if can, yiyvtrtu aXX* an yiyverai, 
yiyp6fifv6¥ eWiv I ov^ &ri ndaxov iarif ndax^h ^^* ^^ ndax^h i^daxov iarlv. 

This simply asserts that we predicate the general characteristic or 
quality in consequence of the several concrete manifestations, and 
not conversely. We have not space to pursue the argument, but 
this suffices to illustrate the use Plato makes of periphrases in 
philosophic discussion. And so frequently, e. f^. Soph. 249 B, 
<f>€p6fi€va Koi MvoviKpa ndvr cfpai, t. 6. 'All things are capable of 
motion'; Theaet 159 E foil. aloBawdfitvoi ylyvofuii, *I become per- 
ceptive.' Note especially Tim. 56 C where 6pd>fi€y6y [tariv] is used 
of what is capable of being seen, and 6paaBai of what is actually 
seen ; also the frequent employment of 5v to express the highest phil- 
osophic reality or absolute truth, e. g. Soph. 256 D ; again, itoiovv 
and itdaxov (agent and patient), Theaet 157 A. But, apart from 
the absolute needs of scientific accuracy, Plato employs these peri- 
phrases freely. There are, in the 151 cases, 49 into which parti- 
ciples of the first group enter. In the second group we may place 

^ov (7 times), d€Ofitvri (l), cXXetVo>y (2), virdpx<ov (4)1 Xf4ir<5/iew)y (l), 

Zxiov (10), fx^fAfva^ (i),and the following used in an adjective sense, 

ofiopo&v (l), avpnrdfJLtvos (2}, dyofJuiXoyovfitPos (l), Kaxovfrf&p (ij, \vaiTt\&v 
(l), fjMi¥6fjL€vo£ (2}, dfiapravdfifvov (l), Apfidrrav (l), x^P^^t 'sCOt-free ' 
(l), €ir6fi€¥os (6), &if* (5), BrjpiovfJktpot (l), alaOapofjktPos* (4)1 6po>iKyot:* (4), 

apx»¥(j2), dpx6fi€vo9 (i), \ty6fAfvos, 'traditional* (i).' For typical 
examples vtd, Critias 117 A ; Prot. 350 B ; Nom. 715 D. 

Some 40 cases still remain to be disposed of and these will fall 
into our third group, where, since the adj. character is not stamped 
on the partic. itself, it is necessary to quote the context in order to 
show the force of the periphrasis ; this space does not permit, and 
I will have to limit myself to one or two cases. Nom. 822 DE, 

ov yhp ^pprjrd <f>afAfv (tvai, \tyovT€s t6* avra ott v6fiovs oitaBai nBffiivovs 

ihfai voXKrjs dvoias y€iKiv, i, €. " For we do not Say that such subjects 
are not to be spoken of at all, when we say it is a great absurdity 
to consider them, like laws, to be matters such as are to be laid 
down in legislation^ Rep. 332 A, icmVoi yc o<^ciX<5/ifv<Jv vov /art 

Tovro, 6 irapaKaTiOtTO, Nom. 935 D, andrav Bvfi^ yiyv6p€vov ^, " when- 
ever it is of the kind that is done in anger." yiyvdfitvop is frequent 

* Nom. 828 A, where it is quite colorless. 

• Mentioned above. ' Nom. 719 C, cf. 782 D, Tim. 3i A. 
^So I read instead of re, which involves a clumsy anacoluthon (Stall, ad 

loc.) The variations in text do not aflfect the periphrasis. 



1 



300 AMERICAN JOURNAL OF PHILOLOGY. 

in periphrases in the sense of ' phenomenon/ vid. Hip. Maj. 297 C. 
There seems to be a tendency to use yiyyoficyoy where it is not 
absolutely required, vid. Theaet 187 D, Phil. 42 D. These two 
passages are» of all that I have found, the least satisfactorily 
accounted few by the theory. For further examples see Tim. 33 
C, 52 B, 58 D (c£ Nom. 859 E), Soph. 253 C, Rep. 492 D, Craty. 
41 1 D, Nom. 667 D, 631 C, 909 B, etc. In all cases it will be found 
that the main object of the writer is to describe or characterize, not 
to assert particular acts. It will be remembered that the peri- 
phrastic form is not absolutely necessary ; sometimes finite and 
periphr. appear side by side, Nom. 932 AB ; 908 B. Symmetry of 
structure is occasionally a factor, Theaet. 164 A ; in one long sen- 
tence (Parm. 157 AB), consisting of corresponding clauses, it has 
brought about the employment of the periphrasis where, according 
to the theory here advanced, the periphrasis ought not to be used. 
Two other passages which do not harmonize with the theory, Hip. 
Maj. 286 B and Cleit. 410 A, are accounted for by the post-clas- 
sical date of these dialogues (vid, Gildersleeve's Justin Martyr, 

P- 143).* 

Second Form. 

When wKik (yiyvtaOcu) is thrown into an emphatic position it 
ceases to be a colorless copula, and acquires a new force which 
varies according to circumstances. But the position of ctwu before 
the partic. may be merely incidental, so that it acquires no emphasis, 
and in this case (a) the periphrasis does not at all differ from 
those already examined ; nor need the second form differ from the 
first, when the cW becomes merely an emphatic copula. On the 
other hand, if full stress is given, the mind dwells on the ilvai rather 
than on the partic., and in it we must look for the raison d^Hre of 
the periphrasis. Characterization, if it exists, in such a case is 
wholly secondary. The resolution into periphrasis arises from the 
mind*s being occupied with the eW factor ; the participle comes in 
alterwards to complete the sense. Under this head we have two 
categories, either {!>) the Ahu, is an emphatic assertion, * is really,* 

* Not every passage where tlvax and a partic. come together is to be considered 
periphr.»<. ^. Gorg. 469 D (quotation), Craty. 405 (exigencies of etymology), 
l^hil. 48 D (corrupt). Elsewhere the partic. may be construed separately, and 
here UitTcrencei of opinion may arise. I give the excluded passages which 
i>rtVr most room for doubt : Rep. 502 D ; Pol. 29 C, cf. B ; Gorg. 523 B ; Nom. 
U% C J 871 C ; 909 E ; Phil. 33 B. 



PARTICIPIAL PERIPHRASES IN ATTIC PROSE. 301 

' actually/ or (jc) contains a predicate in itself, meaning ' exists/ or 
in the case of yiyv((r$ai ' comes into being.* The construction of the 
participle here is analogous to that which it has with <fiaivofiai, and 
shades off into the ordinary use of a participle of circumstance, so 
that it may be a matter of doubt whether the participle depends 
on the subject immediately or through the verb. It is manifest 
then that while cases under the head of a will have the same lim- 
itations as in the first form, those in ^, and more especially in c, 
will be used more freely without limitations as to the nature of the 
participle. We noted, under the first form, the tendency towards 
periphrasis in verbs like Hx^p which express a permanent condi- 
tion. Since the emphasizing of tiyai gives still more prominence 
to the idea of existence and permanence, we may expect to find 
such verbs even more frequentiy in periphr. of the second form. 
The following is the list of occurrences in the Orators : 0/1^^17- 

T&p, ISOC. 15, 57; ^iravfiyoyra, Dem. 3, 33; ?;t«ir (6)/ IsOC. 5, I ID; 
15,117; Din. 1,90; Dem. 20, 18; 113; 23,73; Xcyo/icyor,Isoc. 12,119; 
6/ioXayovfi«voff (3), Isae. 1, 38 ; 8, 2o; Dem. 55, 19 ; napoKoX&p, Isoc 15, 
57 ; ircpcoy, Dem. 36, 8 ; vpar^v, Isoc 5, I ID ; vpwntKintv (6), Isoc 5, 
no; 15, 188; Dem. 3, 24; 4, 38; 21, 196; 22, 33; trvf^pw (3), 
Isoc 5, 16; Lycurg. 140; Aesch. 2, 57 ; imapxov (3), Lys. 13, 91 ; 
Aesch. 3, 208 ; Dem. 20, 25 ; vKodtxoH^tpot, Dem. 55, 19 ; viroXfciro- 
fupog, Dem. 50, 24. 

Total 29, in which 13 different participles are used. 

There are 14 additional cases where an adjective precedes : 
Andoc I, 4 (virdpx»p) ;^ Isoc 12, 135 (jt/kkt^icw) ; Epist. 9, 19 

(jrviM/ftipow) ; Dem. 19, 202 (irpo(r$icoy) ; 294 (dfo/ic^a) ; 312 (jua&v) ; 
20, 94 (ovfM^'pov) ; 21, 70 {irpwTTiKov) \ 114 (dioptfoftv) ; 185 (Ac&y, 
vfipiCcnd ; 22, 73 (Ixof) ; 24, 181 (^xoO ; 29, 13 (irpwnrokovfuiwi). 

Since space is lacking to discuss cases which exhibit peculiarities 
similar to those of the first form, it will suffice to cite an example 
of the emphatic copula in connection with a partic which is clearly 
characterizing : Dem. 3, 33> <^^' ^^'^^ ravm Tfjv c«caoTov p^Bvftiap cVav(- 
(Syoira. Also Isoc 15, 57; 1 1 7. Let US rather note some cases 
where the periphrasis has arisen, not from the desire to charac- 
terize, but from the need of employing the other factor of the 
combination. Aesch. 3, 208, dvohf Bmpov iwdp^i dci, &y ovdcVcpoy 
/ore Afifioa6tv€i vtrdpxov. Cf. Dem. 20, 25 ; Lys. 13, 91* Again, Isae. 

I» 3^> ^IM£ d* oU f'oTiy ofKf^artpa ravra vap6, wdyntv 6fuXcyovfi€pa • • • 

' MSS read in this passage also diSofiivrj, which cannot stand. Some editors 
strike it out altogether, better read dsdofiivrf' 



J02 AMERICAN JOURNAL OF PHILOLOGY, 

Or Dem. 50, 241 o^k ^<m irp6<fMi<ns vwoKtinofitprit c£ 55> 19* Ii^ ttus 
last cited case the participle is quite secondary, but is yet predicate. 
Such a construction can of course be used with perfect freedom 
and we are far from the periphrases with which we started. In 
Thucydides there are some good illustrations of the secondary 
diaracter of the partic. ; 4, 109, 2 ; 8, 69, 3 ; 2, 80, 3. In 2, 67, i 
and 8, 92, 4, the participles may be taken independently of the 
verb as in the ordinary participial construction, or perhaps less 
naturally as predicates. In i, 99, 2, characterization predomi- 
nates. 

Let us turn now to Plato. The following is the list of occur- 
rences : ayvo&v, Phaedr. 239 B ; Syvv^ Phaedr. 237 D ; ala^avofAewot, 
Theaet. 159 E ; oKovofA^vos, Tim. 26 BC ; dvnnBffia^, Soph. 257 D ; 
awobM/ttyos, Nom. 768 A ; &pxwt Phaedr. 237 D ; yffuoy, Nom. 807 C ; 
yiy¥6/ifpof {16) f Euthyph. 5 E; Pol. 301 D ; Parm. 152 B; Phil. 
39 A; 39 C; 42 A; Phaedr. 274 A; Lys. 213 C; Prot. 356 A; 
Tim. 38 B ; Nom. 729 B ; 800 C ; 805 E ; 901 C ; 959 E ; Epin. 
985 B ; Mfuposy Nom. 913 A ; biov (2), Tim. 42 D ; Nom. 713 B ; 
duidMfitpoft Tmi. 77 ^ > dcoipov/fcyor, Nom. 895 ^ » BuuJKpw (7)1 Parm. 
154 D; Alcib. II 149 B; Gorg. 500 C; Nom. 696 B; 733 BC; 
861 B ; 963 B ; doxAy, Phil. 51 A ; do^dC^p, Soph. 240 D ; dwofntpot, 
Nom. 937 E ; AXctiroficvor, Soph. 258 B ; cWv, Rep. 431 E ; iwifu- 
\ovfxtpos, Epin. 980 D ; rVAn^dcvo/Acvof, Rep. 527 B ; ttro/upogf Nom. 
763 C; ?x»*' (23)> Phaedo 92 D; Soph. 258 B; 287 E; 297 E; 
Pol. 306 B ; Phil. 48 C ; 59 B ; Phaedr. 245 E ; Prot 330 E ; Gorg. 
484 A ; Meno 82 C (dis) ; Hip. Min. 368 E ; Rep. 397 B ; 602 C ; 
Nom. 663 D ; 743 A ; 747 D ; 770 C ; 857 B ; 876 E ; 892 C ; 
896 A; icoXovficyof, Nom.961 D; «Hroi;fi«vor, Theaet 153 D; Xtyofit^os 
(11), Soph. 257 D ; Pol. 302 C ; Phil. 11 B ; 26 E ; Rep. 490 A ; 
588 B; Tim. 90 E; Nom. 773 C; 855 A; 881 B ; Epin. 981 A; 
/uKov, Nom. 766 C ; iut^x^v (3), Pol. 273 B ; Parm. 141 A ; Rep. 396 

E; ficra/3aXX(Dy, Nom. 894 E ; yofioSrrovficvw, Nom. 834 1^ t Po/io$€Twrr€t, 

Nom. 692 B ; povovtms, Alcib. II 139 D ; olKovfAtvos, Rep. 521 A ; 6p»fU' 
ptn, Epin. 985 B ; impmcfuvos, Theag. 1 28 D ; wap»v, Phaedr. 27 2 A; ircpc- 
€%»Vf Parm. 138 A ; irtpi^poiwHtSt Rep. 402 D ; votovfitpost Theaet 143 

D ; froXircvoftfWf , Nom. 676 B ; iropiC6ft€Pos, Rep. 364 B ; npewmp (8), 

Tim. 17 B ; Nom. 627 C ; 756 B ; 767 B ; 931 D ; 944 E ; 945 B ; 
948 C ; irp€(ri3cvo^iw, Nom. 879 C ; npamiitpoi (2), Nom. 736 B ; 
870 D ; irpckT^iwir, Nom. 902 C ; <rviiil>ifHȴ, Rep. 338 D ; avMHicwr, 
Nom. 848 A ; -niimw (2), Pol. 308 E ; Meno 84 B ; nBifupot, Nom. 
963 A; vntpix^y Nom. 696 B; ^/N*y (2), Nom. 811 B; Aidb. II 



PARTICIPIAL PERIPHRASES IN ATTIC PROSE. 303 



142 B ; ^xo/*«w» Phaedo 118 A ; &¥ (15), Soph. 237 A ; 245 D ; 
256 E ; 259 A ; Parm. 141 A ; 162 A (Jer) ; Phil. 51 A ; Phaedr. 
247 E ; Tim. 38 BC (bis) ; 61 D ; Nom. 771 C ; 894 A. 

133 cases in which 53 different participles are used. 

There are 14 additional cases where an adj. precedes : Soph. 
258 C (^); Parm. 141 PL(txpv)\ 141 E (jirrtxw)', Symp. 191 D 
(fir»x<*pw); Alcib. I 114 B (jnnixjHpow) ; 116 DE (mfKlHpov) ; Meno 

99 E (wapaytypofUwrf) ; Rep. 374 E (dco/icvoy) ; 556 A (ayay«caCa»v) ; 

Critias 112 A (jUx^p) ; Nom. 649 D (yift^p) ; 840 AB (inf>piy&rrtt) ; 
918 C (Mfi^potO ; Epin. 981 E (tlx^)- 

The number of these latter cases is small as compared with the 
similar ones in the first form, an indication of the less uniformly 
adjective character of periphrases in the second form. A further 
confirmation of this is to be had from a comparison of the most 
frequently recurring participles in each form. These are for the 
first, the thoroughly adjectivized : 

irphruv trpwy^Kuv Staffpuv 

1st Form 23 10 8 

2d " 817 



ovfifipov Total. 

6 47 

I 17 



For the second : 

2d Form 
ist " 



23 16 15 II 65 

ID 6 5 I 22 



The recurrence of ?x»y illustrates the tendency of this form to 
express an abiding condition. The frequent employment of 
-ytyvo^icyor and &v arises fi'om the fact that in Plato's time these two 
words represented fundamentally opposing views of the universe, 
which Plato made it his business to reconcile. If it were needful 
to use yiy¥f<r6<u of Something which had real existence, the speaker, 
to avoid ambiguity, must employ ci^it in the emphatic position 
with yty^6fA€P0Sf e. g. Tim, 38 B; Parm. 152 B. Again, were it 
necessary to speak of a thing coming into real existence, yiyv€(r6ai 
would have to be used in the emphatic position with Xy; e,g. 
Soph. 237 A ; 245 D. Finally, even more frequently, lest ctnit 
should be taken in the loose popular sense, cfrac Sp is employed in 
reference to absolute existence, e. g, Phaedr. 247 D. The peri- 
phrases of these participles, however, are not confined to ques- 
tions of ' being '; Prot. 356 A, Pol. 301 D. The cases in which 
Xryofiow appears arise from the need of using a pregnant fiwu. 
Nom. 881 B, loTM dc \ty6\iM¥ov t6 fura rovro i^dc. Here there is no 



304 AMERICAN JOURNAL OF PHILOLOGY, 

emphasis on the aciy but to an abiding condition, " Let the next 
topic stand expressed in the following way." Nom. 773 F, trcpt 
ydfuiv dff ravT tar» napa/AvSia Xry((fi€vo, " Let these thin^ be granted 
to be," etc. Rep. 588 B, fp d€ nov \€y6fiftfoPt " It was laid down in 
the course of the discussion " (cC use of ^p in Aristotle, etc., of a 
definition which has been anived at and still holds). Rep. 490 A ; 
Pol. 302 C ; Tim. 90 E. Somewhat different are Phil. 11 B ; 26 E, 
cf. Nom. 961 D; Soph. 257 D. In Nom. 855 A X€y6fi€vo9 is added 
as an afterthought. Other examples of pregnant use of ctwt are 
Nom. 763 C and 770 Q(where ^p is used of a conclusion reached 
in a former part of the discussion) ; Nom. 959 E, foro) npartpa ^fup 
rit ntpl (ra>/ia leoi V^x^v fiprot ** Let US presuppose that bodily and 
spiritual things really exist," cf. 963 A. It is tan that gives the 
raison d'itre of the periphr. in Nom 895 E, ttm ww dixa duupovfA^pop 
..." There exists a twofold division," c£ Tim. 77 E. The parti- 
ciple is quite secondary in such passages as Rep. 431 E, rV ravrg t^ 
cTg rovTo (PoPf cC Soph. 257 D ; Nom. 870 D, To show the entirely 
secondary place which characterization may have in this form, we 

cite Nom. 692 B, rots Tore P0fio6tTatSt oiTiPtt Sp* ^aap pofioBerovpTts, for, 

if the idea of characterization were the main one, the relative 
clause would be tautological, but it is in ^aap we find the reason for 
the periphrasis, "the lawgivers who were aciucUly engaged in 
legislation." Again in Phaedo 118 A, jji* ^x^'M"^* 'was actually 
growing cold,* is used antithetically to the jailer's assertion as to 
what would take place. In one case, Nom. 768 A, I cannot satis- 
factorily account for the periphr. unless there be an anacoluthon. 
The other passages least easily accounted for are Euthyph. 5 E ; 
Nom. 729 B ; 800 C* 

I will conclude the discussion of periphrases containing the 
present partic. with a summary as regards the use of these combi- 
nations in the authors examined. Periphrases of the first form, as 
well as of the second form in as far as they resemble the others in 
function, were not employed in ordinary speech except in the case 
of certain thoroughly adjectivized participles, to which the few 
cases found in Lysias, Andocides and Isocrates are confined. In 
other cases the combination was felt to do some violence to the 
language, and becomes more harsh as the meaning of the parti- 
ciple lends itself less easily to being conceived as a quality ; hence 

'Again I cite excluded cases which may be open to doubt, Meno 84 A ; cf. 
Lys. 204 B; Cleit. 409 E ; Rep. 581 E ; cf. 478 C and Pann. 136 A ; Pol. 271 
A (Camp, ad lac.) 



PARTICIPIAL PERIPHRASES IN ATTIC PROSE, 305 

the three groups into which I divided the cases mark stages of 
increasing difficulty. Only four cases of the third group occur in 
the Orators, and three of these, as might be expected, in Demos- 
thenes. A parallel adjective preceding mediates the use of the 
participle, but even examples of this kind are not common and are 
confined to Isocrates, Aeschines and Demosthenes, the first-named 
employing only the easiest combinations. In the second form we 
find cases which do not differ essentially from those of the first 
form, and, having the same function, have the same limitations. 
There is further observable a tendency to use this form in the case 
of verbs which express an abiding condition, particularly ?x«ty In 
general the more emphatic and pregnant the cfww becomes, the 
easier is the combination, since there is an approximation to the 
ordinary use of a participle of circumstance. Hence in these cases 
there is greater freedom and a wider range of participles. Thucy- 
dides resembles the Orators in his usage. His periphrases of the 
first form are of the ordinary kind, into which adjectivized parti- 
ciples enter. In those of the second form he is bolder, but only 
where the stress of meaning is upon the verb. In the pregnant 
style of Thucydides we should have expected striking cases, but 
he has adopted another method for expressing characteristic, viz. 
the use of a periphrasis containing verbal substantives in -n/r. 
Plato exhibits a very free and accurate use of participial peri- 
phrases, as was to be expected in a philosophic writer. They are 
more fi'equent in the later and more scientific dialogues. Soph- 
istes, Politicus, Parmenides, Phileus and Timaeus, but the Nomoi 
and Epinomis (comprising about 20 per cent, of the whole of 
Plato examined) contain 41 per cent, of periphrasis, an increase 
bx beyond the demands of the subject-matter. 

AoRisT Participle. 

If the theory advanced in this paper be true, periphrases con- 
taining aorist participles are not a priori to be expected, unless in 
exceptional cases. Of the three tenses the aorist is fixed most 
closely to the expression of the actual occurrence of a definite act. 
It is particular and individual ; even the gnomic aorist generalizes 
through the particular. Accordingly, the aorist participle is not 
usually placed like the pres. and pf., as an attribute between article 
and noun. It is true, such phrases as oMr iarXv 6 vmoKtuva^ are 
characteristic, but the characterization consists in the identification 
of two individuals oJros and 6 airoxTciVoff, hence the article, ovroi 



306 AMERICAN JOURNAL OF PHILOLOGY. 

iarw iwoKT€ipaf, on the Other hand, does not normally occur, 
since here coFoierdvaf would have to be conceived as a gener- 
alized characteristic, a conception to which the aorist is in its 
essence opposed. I have noted a few cases of aorist periphrases 
which on examination, however, only serve to confirm the con- 
clusions already attained. In the Orators I have noted 4 cases. 
Two of these occur close together in Ant 3, b, 4 and 5 : 6 wtudarpl^ 

ly on-orrciyof avrop ciif and t6 fuv fJifipaKtop ovd€p6s /laXXor rmp avfifUkmuh- 

rmw iirri roO aKonov dfuipTw. In the former of these passages the 
insertion of the article with airocrtiWr is not difficult, but in the 
second the change is less easy. It is true r6 might easily have 
fallen out between eWl and rov, but the writer would scarcely have 
said t6 AfMfrrotf but 6 d/Mprcw^ and thus we should have to account 
for a double corruption. There is, moreover, good reason for the 
omission of the article, and I should be unwilling to make the 
change in either case. In this tetralogy, the prosecution seeks to 
show that the defendant is 6 afrotcrfivof and 6 Afiopr^p. Now the 
very use of these terms implies the existence of such a person or 
persons, and from the peculiar circumstances of th* case (for which 
from lack of space I must refer to the speeches themselves), had 
the defence admitted the existence of such a person, their client 
must have been that person. In the absence of an indefinite article, 
then, they are forced to employ the aor. partic. as a characterizing 
adjective.* The other two cases are in Dem. 21 and are intro- 
duced by a parallel noun or adjective: 156, «cayai fiip tB^Xowr^ wp, 

oirros dc Karairriit (( ayridoacosr rorc where Keeraards is for 6 Karwrrds, the 

article having been omitted under the influence of the parallel 

anarthrous noun. 114* ovt» toIpvp o^or tarip aatfi^t Koi fuaf>6s icoi war 

^p vnwrras €lnfip . . . affords Confirmation of the view that the aorist 
does not lend itself to the expression of characteristic. The nature 
of the signification of the verb brings about the use of the aor. in 
preference to the pres., but the writer feeling that vnoards was not 
fitted to express characteristic, annexed the ap and thus gave the 
requisite generalizing force. 

Let us now pass to Plato. The phrase pJi dirapyi^^cW y^pg^ Soph. 217 
C, is an evident imitation of tragic style (cf. Soph. Ajax 588 ; Philoct 
772) and need not detain us here. The remaining 7 cases are all 
of the second form and parallel to those periphrases of the pres. 
partic. where the raison d'itre lies in the so-called auxiliary and the 

* In 2 7 8, the h lacking before d7ro«re/vaf in the MSS has been inserted, 
rightly I believe, by the editors. 



PARTICIPIAL PERIPHRASES IN ATTIC PROSE, 307 

characterizing force of the participle is absent The partic bears 
the same relation to the ccmu {ylyptoOaC) as it would bear to <^vc- 
wBait and, as contrasted with the finite form, the periphrasis marks 
two stages, that of the action itself, by the participle, and that of 
the ascertainment, by the verb, e, g, Nom. 737 C, Syicor di) irX^^bvc 
Ittaw^ ovK SXKcK 6p$»s ylypoiT ^ X^x^is . . . ' would not turn out to 
have been rightly calculated.* So 866 D ; 867 C ; 739 E/ Agaiiv 
PoL 289 A draws more attention than the finite form would have 
done, to the position of matters subsequent to the act of ' placing.' 
So Nom. 957 C ; Tim. 47 CD. It will be noted that this form of 
expression is not common, no examples in the Orators, and 5 of 
all in the Nomoi. In addition there are 4 cases in Plato where 
an adjective precedes ; Nom. 711 D is a case with Sy, parallel to 
Dem. 21, 114 already explained. Nom. 913 C is similar to 737 C 
discussed above. Nom. 829 CD is a case of the aorist being used 
for a pf. under the influence of the negative, Phil. 51 A is excep- 
tional : npbt r^ Tivaf i^doyac cSroi doKovatKt oCtrat d< ovdafiWf Koi /ifycSXat 
Mpas Tipat dfta koi noKkhs if>arra«rBtia'as, tivai d*aur^ff avfAirt<f>vpfuy(u 

ofiov Xvir«iff . . *. Here the writer does violence to the language to 
express a special meaning. ^awraaStiaas is characteristic and the 
present participle would have been expected, but then the natural 
interpretation would have been, that these pleasures present many 
appearances at one and the same time, whereas he wishes to say 
that they underwent continual change and presented many appear- 
ances in succession. He therefore uses the ingressive aorist, the 
characterizing force being sufficientiy marked by the neighborhood 
of the participles and adjective.' 

Perfect Participle. / 

The perfect tense has a two-fold aspect ; in addition to predi- 
cating an activity, it predicates an abiding result of that activity. 
This latter factor may be regarded as an attribute of the thing 
affected, as a quality which has been generated in the thing. A 
pf. participle, we conclude then, approaches an adjective closely ; 
but they differ, inasmuch as the adjective presents a quality merely 

* The partic. in 739 E might be construed as an ordinary partic. with the 
subject, but the other is preferable. 

•I note as before some rejected cases; in [Lys.] 20, 1, avr«v is predicate, 
* for some, having plotted, joined them '; Phil. 64 B, cf. Phaedr. 24s E ; Pol. 
265 D ; Nom. 740 B ; 961 BC ; in Nom. 844 D the partic. is an afterthought ; 
Pol 372 D, corrupt ; Theag. 123 A, mark of late origin. 



308 AMERICAN JOURNAL OF PHILOLOGY. 

from the point of view of its existence, the pf. participle has regard 
also to its genesis. Of these two sides presented by the pf. tense, 
either may predominate ; but it must be noted that, while to express 
the first of these aspects simply we have another tense, the aorist, 
there is no tense which can assert the second simply. The conse- 
quence of this would be that in general, when the pf. itself is used, 
the main emphasis would lie on the second side — the existence of 
the result. But we have already seen that periphrasis is exactly 
fitted for the presentation of such a conception, and hence the very 
frequent use of periphrasis in the pf. is in accord with the theory 
all along maintained in this paper. Again, since the result of an 
action is more likely to be permanentiy manifest in the object than 
in the agent, we are not surprised to find periphrasis more common 
in the middle-passive than in the active ; and again, the use of finite 
forms for opt and subj. of p&. used as presents is also in accord 
with what has been laid down. From what has been said, any 
difference between finite and periphrastic forms in the pf must in 
any case be evanescent, and in addition to this we have a disinte- 
grating factor in the defectiveness of the p£ paradigm. When 
XcXvvrai and TTcirXry/ici'oi c^crt were used in exactly the same temporal 
sense, it was inevitable that any difference which might exist between 
XcXvvrat and XrXv/A€voi rtVt should be obliterated. Doubtless the 
emphatic pre-position of cZi^iu was often a determining factor. I do 
not propose then to examine in detail the cases of pf. periphrasis, 
it is sufficient to have shown that the phenomena here also are in 
accordance with the conclusions elsewhere reached. 

W. J. Alexander. 



IV.— STICHOMETRY. 
Part II. 

Extension of previous results to Bible-texts, 

It might almost be assumed that the previous investigations as 
to the nature and interpretation of stichometric data, comprehend- 
ing as they do writers of so many different centuries, and books 
of such different character, might be expected to apply without 
further examination to the texts of the Old and New Testaments. 
But as the subject reaches here its greatest importance, and has 
been attended by a good deal of confusion in consequence of the 
facility with which many of the books of the Bible are divisible 
into sense-lines, it becomes necessary to establish over again the 
fixity of the arlxoi, and other points connected with the development 
of the art of transcription. This we shall easily be able to do, for 
the examination of the texts after the manner previously explained 
will show that in almost every instance the verse of the ancient 
scribes is a hexameter, and is measured by a standard number of 
letters or syllables. 

Nature of stichometric data for Old and New Testaments. 

The MSS of the Old and New Testaments, but especially of the 
latter, provide us with a rich collection of stichometric references, 
both total and partial, which enable us to measure the text with 
very great accuracy from point to point, and are a very valuable 
addition to any critical apparatus which is aimed at the restoration 
of the text of the early centuries. The total subscriptions stand 
not only at the end of the separate books, but sometimes at the 
close of a group of books, as the Catholic Epistles ; the marginal 
subscriptions supply us with the successive fiftieth verses, and also 
with the number of verses proper to any particular lection in a 
book that has been divided for church or private use. 

The stichometric notes do not appear in the archaic numera- 
tion which we noted in Herodotus and Demosthenes, nor does the 
margmal stichometry present itself in the transitional form which 



3IO AMERICAN JOURNAL OF PHILOLOGY. 

uses the successive letters of the alphabet, but pays no regard to the 
decimal system, as we have seen it in some Plato and Demosthenes 
MSS ; there is, however, no doubt that these marks are of great 
antiquity, and in some cases we shall be able to fix an inferior 
limit to the date of their publication. 

Variations of sHchometric attestation. 

There are several hindrances that encounter us at this point of our 
inquiry ; and in particular the variety which is found amongst the 
stichometric subscriptions of any one book in different MSS seems 
to militate very strongly against the theory of a fixed and uniform 
verse-measure. A litde consideration, however, shows us that the 
same argument would hold against the hypothesis of sense-lines, 
unless we assume that these were perfectly arbitrary in their char- 
acter, and did not constitute a imiform system of division handed 
down by tradition as a convenience to the reader and a safeguard 

to the text 

The real reason of this variety lies in the following direction. 

First of all we must remember that we are dealing with books 
whose variety of reading is great, and where the importance 
attaching to the acceptance or rejection of a reading is likely to 
make the stichometry agree closely with the compass of the text, 
and change as the text changes. The insertion or rejection, for 
instance, of such a passage as the pericope de adultera would 
modify largely the stichometric count in the Gospel of St John. 
We must also bear in mind that these books are extant in various 
versions, and unless we adopt the hypothesis of sense-lines, the 
count will vary from version to version, even with a similar text 

We have further to observe that in the early Bible-texts we have 
certain conventional abbreviations which may in some cases even 
date fi'om the autographs, and which will certainly affect the reck- 
oning if a letter-line be used in the measurements, and probably 
also where the syllable-line is employed. Then there is a fre- 
quent corruption of the actual stichometric data, arising from care- 
lessness on the part of the scribe, and sometimes, perhaps, from 
an ignorance on his part as to the meaning of certain old symbols 
employed to designate the numbers 90 and 900, etc. Last of all, 
it is possible that we may have to admit in some cases a variety in 
the measuring-line, though we shall still see that the roost usual 
unit is the i6-syllabled hexameter. 



STICHOME TRY, 3 1 1 

TransiHon from space-Unes to sefise-lines. 

We shall also be able to trace that same law of degradation in 
the form of the transcription which we observed to hold in the 
adaptation of continuous uncial texts to public reading ; and it is 
possible that the first step towards this change of style in the early 
MSS consists in the exact numeration of the text from point to 
point by means of a suitable line-unit. 

This change of form is first apparent in the poetical books of the 
Old Testament, from which it seems to have spread gradually to 
the whole of the Bible. We have already seen from Jerome's 
preface to Isaiah, that the method of division by cola and com- 
maia was becoming general, and was reckoned by Jerome him- 
self to be as applicable to the Psalms as to the writings of Demos- 
thenes and Cicero, and to the prophets as to the Psalms and other 
distinctly poetical books. And it is almost inevitable that if two 
different systems of transcription, corresponding respectively to 
stichometry and colometry, are found in the same volume, that a 
degree of confusion will arise between the regular verses of the 
earlier and the irregular verses of the later system, and that in the 
end one of these systems will entirely supplant the other. This 
explains how it is that we find the term arixoi retained even when 
the fixed line to which it properly belongs has disappeared. It is 
in consequence of this degradation of form that we find the poetical 
books of the Old Testament in the earliest uncial MSS written in 
quite a different manner from the rest of the Bible. For example, 
the triple and quadruple columns of the Vatican and Sinaitic 
codices are replaced in these books by double columns of irregular 
verses, forming a remarkable contrast to the uniform writing of the 
remaining books. I regard it, however, as certain that this quasi- 
stichometry is not the original form of the books where it appears. 
The Song of Solomon, for example, is stated by Nicephorus and 
Anastasius to contain 280 verses ; and, by an actual enumeration, 
it may be seen to be 275 sixteen-syllabled hexameters, which is 
such a close agreement that we may conclude that the earlier 
mode of reckoning, and therefore, in all probability, of division of 
the text, must have been at some time applied to the book in 
question. A great deal of light is thrown upon these points by 
some remarks of Hesychius of Jerusalem, in the sixth century, 
introductory to the study of the twelve minor prophets. An 
examination of the following passage will show the progressive 



312 AMERICAN JOURNAL OF PHILOLOGY. 

encroachment of colon- writing upon the uniform text, and the con- 
sequait confusion between the trrixf^y properly so-called, and its 
substitute. 

Irixrjpov roy i^ irpo^iTrwy. 

Eon ftiiv apxoiop rovro rois B€o<f>6pois r6 avovdaa-fia, arixrfi^i^, «ff ra woXXa, 
Kpos r^ Twr ficXcrw/icMoy vcufnipttapy ra% vpo^yjr^ias ficnBtaBai. ovr» roiytip' 
ow Siyf^i itih riȴ Aavid iuBapi(oyTa, t6v Ilapoi/uaoT^y dc rat irapafioKat ical 
r^ EjucXj}Ota<m}y ras irpo^yjrwiat €K0€fi€POP, ovt» (rvyypa<f>€iaa¥ rqv rwi ry 
*Im0 |3t^Xor, cZt» iMpitrBtyra roit arixott ra t»¥ Aurfiarnp "Aurfurra I wX^p 
aXXa Mu rffp AyrooroXtic^v fiiffkov ovrta ripl avyypa<l>tla'€Uf cv/My, ov puarnv cV 
Ttus dvodcKQ fiiffkoig T&v frf}o<f>ifTS>p Koi avrhs ffKoXovBtjO'a ' aXX* rircid^ iroXXa 
fuv rmp a(ra<fwv 17 r&p aTix<»P (ratfujviCfi diatpccriff, dcdacrjcci dc rcov anyfiMP 
rmp €ar6ptd¥ nov dci rdrrciv ras wXfiovaSf S><rT€ Koi r^y IBioimjv xal r^y ayay 
rrtoT^/Aoya rpvyrjo'ai ri 7rdi^a>f ^ fUKp6v Ij fuya rov irovrmaros xph^^-f^^' 

It is evident from the foregoing passage that the first means 
employed to facilitate the reading of the continuous texts is iTiier- 
puncHan ; and that interpunction paves the way for colon-writing ; 
Hesychius himself extends the irregular verse-writing to the minor 
prophets, and informs us that some one else had edited the Pauline 
epistles in a similar manner ; and finally we notice that the new 
form of writing has the effect of restoring to the term arlxot some- 
what of its original indefiniteness, and deflecting it fi'om a space- 
line in the direction of a sense-line. 

Actual instance of numbered sense-lines. ^ 

An instance of this deflection may be seen in a MS Memphitic 
Psalter, referred to by Lagarde in his edition under the sign D, 
which has stichometric data to every psalm. An examination of 
these will show that the appended numbers are not proportional to 
the lengths of the psalms, neither in the Hebrew, the LXX, nor 
the Coptic The following table for the first ten psalms, based on 
Lagarde's edition and on the LXX, will make this apparent The 
arlxoi and psalm are measured in letters : 









Letters 




Utters 




^TlXOi, 


Memph. 


to verse. 


LXX, 


to verse. 


Psalm I 


15 


514 


34.3 


604 


40-3 


II 


27 


755 


28.0 


806 


29.8 


III 


15 


521 


34-7 


545 


363 


IV 


15 


619 


41-3 


651 


43-4 


V 


28 


880 


314 


911 


32.5 



^Migne, Patrol. Graec. 93, col. 1340. 







STICHOMETRY. 




3 








Letters 




Letters 




Xrtxou 


Mempk. 


to verse. 


LXX. 


to verse. 


Psalm VI 


21 


621 


29.6 


709 


33.7 


VII 


37 


Il8o 


31-9 


1289 


34.0 


VIII 


17 


619 


36.4 


646 


38 


IX 


82 


2559 


31-2 


2908 


35-4 


X 


17 


528 


31-0 


583 


34.3 



It is, however, easy to write the Psalms rhythmically in irregular 
sentences, so as to make the reckoning true. For instance, the 
119th Psalm, which has 176 verses in ordinary Bibles, has 170 in 
the Memphitic text It is even possible that the figure 6 has 
dropped. The remarkable point to notice is that the irregular 
verses are numbered just like the regular ones, a practice which 
leads to some confusion, though it has the advantage of giving the 
same reckoning for all the various versions. 

Euihalius and his work. 

We turn now to the stichometry of the New Testament. And 
here a fundamental misunderstanding seems to have prevailed for 
a length of time as to the connexion between Euthalius of Alex- 
andria and the stichometric divisions of the text 

Scholz, in his Prolegomena, I xxvii, states that " Euthalius in 
epistolis Paulinis, actubus apostolorum et epistolis catholicis, eos 
(sc. versus) ita distinxit in usum lectorum, ut singulae lineae singulas 
absolverent sententias ; qua distinctione observata scirent lectores 
quae continuo spiritu essent legenda, atque ubi intermissione opus 
esset. Exaratis in hunc modum epistolis adtexuit ad calcem cujus- 
que epistolae numerum versiculorum, qui in plurimos codices 
irrepsit." 

And the same statement somewhat modified seems to have been 
repeated right on to the present. According to Scrivener, Intro- 
duction to the N. T. p. 60,* " Euthalius is said to have been the 
author of that reckoning of the aTixo*- which is annexed in most 
copies io the Gospels^ as well as the Acts and Epistles "; and in the 
introduction to the American edition of Westcott and Hort's New 
Testament, Dr. SchafT remarks " that the stichometric divisions or 
lines (arlxoi) corresponding to sentences were introduced by Eutha- 
lius:'^ 

* P. 60, 2d Ed.; p. 62, 3d Ed. 

* Misled by the concurrence of these and other New Testament editors and 
critics, I endeavored to believe that in some way Euthalius and stichometry 
were inseparable; and for this reason stated in a former article that the division 
of the New Testament into numbered sense-lines was introduced by Euthalius. 



314 AMERICAN JOURNAL OF PHILOLOGY. 

But it will easily be seen that in no strict sense can Euthalius 
ever be regarded as the inventor of stichometry which is anterior 
in date to the Christian era, and by no means a peculiarity of the 
New Testament ; that he did not measure the Gospels at all ; nor 
will it be easy to prove that he broke up the text into sentences, 
nor are these sentences the <mx« which he enumerates. In fact, 
the New Testament text was reckoned by im'xM long before the 
time of Euthalius, as we find that Origen reckons the second and 
third epistles of John to be less than a hundred verses, and the 
first epistle to contain a very few ; and in Ae fourth century Eusta- 
thius of Antioch quotes two pass^es in the Gospel of John, with 
a remark that the interval betweeen them is 135 trnj^oi. Euthalius 
was a deacon of Alexandria somewhere about A. D. 458, and sub- 
sequendy became bishop of Sulca, supposed by some persons to 
' ~ity in upper Egypt. He describes his work in a dedication 
ounger Athanasius, in the following language : 

^ tHr immm)Mcmr ■y a f f* {pa Tf Ta» i M <^ T ii M i cnvnUr ifftaiiA 

..Vl-o-^-r**' ■ ■■ 

r r m ym p a i r ^tXAwftt Jji vn^;|W tw rpSmmr, . . . fa gy y t ^wyt 

Kar am ■«( wan^aXaMvaw^bi, ■■■ ftwlita r W T<w tuaifi rim «■«* 
tfit «- |L ii i » «' i ■ {■» •, JfcX^ 'ASa n i m w f tw^i Vi wr m i t , mm twnw iiami rym 

MV hA«i> j^ nil Aiaw ■yo^ii fgiyii n4n4t v 




Me coo&aoa seems to h>\« arisen in Ac text 01 the prcvioos 
^ between m^x^im and •n^^Mv. Of the dnce passives in 
dK vords ooruT. Zac^:Tu eiihs «mi«t^ in two pfaccs, while 
^aax wsA otbers reads vn^vk^ oniibnnlT. 
W pttsi^cs wfi, I thank. S^Kow tltK il is « 



STIC HOME TRY, 3 1 5 

prove that Euthalius introduced sticbometry into the New Testa- 
ment as to prove that he introduced reading and writing (({Myyovf 
rt Kot ypct^ff). The peculiar features of the arrangement of his 
text are prefaces, programmata, lists of quotations with reference 
to the authors, sacred and profane, from whom they come, and a 
complete system of convenient lections and chapters. The edition 
was also provided with a stichometric indication on the margin of 
every fiftieth verse and at the close of every complete lection. 
These annotations made reading and quotation a much easier busi- 
ness, but they are clearly only ancillary to the general arrange- 
ment of the work, though by a strange want of perspective the 
last feature has been made the most prominent one in the literary 
estimate of Euthalius. Neither must it be assumed that the lec- 
tions which Euthalius marked are of his own division ; in the Pau- 
line epbtles they have evidently been adopted from some earlier 
father, who gives his own date (A. D. 396 ?)' in a prologue to the 
work, which Euthalius merely corrects in an appended sentence. 
The chapters abo, at least in the Acts, are divided according to 
two totally distinct systems ; this fact alone shows that Euthalius is 
retailing the Masoretic efforts of earlier students." 

Importance of the Euthalian stichometry. 

The importance of the stichometric work done by Euthalius 
does not, however, diminish when we discount its originality ; on 
the contrary it increases. For in the first place he distincdy 
informs us that hb measurements were accurate ; and in the next 
place, the MSS which he employed, at least for the Acts and Cath- 
olic Epbtles,' were the celebrated copies preserved at Caesarea in 
the library of Pamphilus.* It is unfortunate that the word dxpifiAs 
which Euthalius employs, and which makes the weight of his work, 
has been so much overlooked. Accurate measurements made by 
reference to the best MSS provide us with critical data of immense 
value. It becomes interesting, then, to find out what the accu- 
rate measuring line is which Euthalius employs. 

In Zacagni's edition of Euthalius, or in the less complete one of 
Migne,' we have a rich vein of stichometric information which 

* Zacagni, p. 536. 

* cf. Tregelles, Canon Muratorianus, p. 104 ; Hug. Introduction to New Test. 
(English Tram.), i. p. 353. 

» Zacagni, p. 513. * Migne, 85. col. 691. » Patr. Graec. 85. 



1 



3i6 



AMERICAN JOURNAL OF PHILOLOGY. 



seems to have been very slightly worked. Not only is every pro- 
gramme, preface, and elenchus measured and the number of crrixoi 
appended, but there are so many intermediate stichometric data 
supplied for the text that we can measure from point to point with 
great accuracy, as soon as we know the measuring line employed. 
M. Graux examined casually the numeration of the separate 
lections for the Acts of the Aposdes, but he was perplexed at find- 
ing that the data supplied by Zacagni from the Vatican Codex 
Regius- Alexandrinus did not taUy with those given by a Madrid 
MS Codex Escorial. ^ — 1 1 1 — 6, and he seems to have given up the 
point in despair. The following table affords a comparison between 
the measures of the lections as given by the two MSS, and those 
given by actual division of Westcott and Hort's text into i6-syl- 
labled arlx^ : 





Chapter 
and Verse, 








Lection 


Cod. Esc, 


R,Al, 


SyUahU. 


I 


I I 


40 


••• 


40 


2 


I 15 


30 


30 


30 


3 


II I 


109 


109 


III 


4 


III I 


136 


136 


143 


5 


IV 32 


100 


100 


^121 
190 


6 


VI I 


88 


220 


7 


VII I (cV'i'fTo) 92 


120 


94 


8 


VIII I 


75 


95 


77 


9 


1X32 


216 


250 


210 


lO 


XI 27 


283 


300 


272 


II 


XV I 


193 


200 


201 


12 


XVII I 


164 


180 


164 


13 


XIX I 


239 


240 


242 


H 


XXI 15 


293 


293 


307 


15 


XXIV 27 


168 


268 


160 


i6 


XXVII I 


198 


? 


192 



The remarkable agreement between the first and third columns* 
leaves little room for doubt that Euthalius employed as his measure 
a rhythm of sixteen syllables. The data of the Madrid MS are 
better preserved than the other's : in the sixth lection the figure 
p has evidently dropped, and there are several other minor corrup- 
tions. 

1 Some trifling alterations have been made in correcting these figures from 
their first publication in Johns Hopkins University Circulars No. 35. The same 
remark applies to the tables which follow. 



STICHOMETRY. 



317 



Comparison between traditional and measured verses. 

A similar closeness of agreement is found between the other 
data supplied by Zacagni for the intermediate stichometry, and 
those furnished by actual measurement of the text; and the 
total is also found to be in remarkable agreement with the sub- 
scription of the best MSS and of Euthalius. The results are so 
good, in fact, that we are tempted to repeat Euthalius' work, and 
we shall divide the whole of the Acts and Epistles as given in 
Westcott and Hort into sixteen-syllabled hexameters. This being 
done, we exhibit the results, as in the subjoined table, and compare 
them with those deduced from Euthalius and from the majority of 
the codices of the New Testament in which any verse-measures 
have been preserved. 

by tradition, by measurement. 



Acts 


2556 


2559 


James 


237 or 242 


240 


I Peter 


232, 236 or 242* 


245 


II Peter 


154 


162 


I John 


274 


268 


II John 


30 


31 


III John 


32 


31 


Jude 


68 


70 


Total for Catholic 


Ep. 1047 


1047 


Romans 


920 


942 


I Corinthians 


870 


897 


II Corinthians 


590 


610 


Galatians 


293 


304 


Ephesians 


312 


325 


Philippians 


208 


218 


Colossians 


208 


215 


I Thessalonians 


193 


202 


II Thessalonians 


106 


112 


Hebrews 


703 


714 


I Timothy 


230 


239 


II Timothy 


172 


177 


Titus 


97 


98 


Philemon 


38 


42 



* Some confusion is apparent between th^ subscriptions in James and I Peter, 
which makes it necessary to record the principal variants; in other places 
these are not given, but may be found in Scholz and the ordinary critical 
apparatuses. 



3l8 AMERICAN JOURNAL OF PHILOLOGY. 

Correction of previous results for abbrevioHoiu 

The approximation of the results is very striking ; but there is 
almost always an excess in the second column, amounting in some 
cases to as much as 5 or 6 per cent. ; and this uniformity of effect 
implies some producing cause. Now it can scarcely be maintained 
that the text of Westcott and Hort is ever much in excess of the 
text of Pamphilus, and so we have only one hypothesis to &11 back 
upon : the text measured must have had abbreviations in it Let 
us then assume that the four words Bt6i^ Kvpwft /i/aow, xp^*rr6s, are 
abbreviated : we ought then on the average to deduct a syllable every 
time the woids $€6f, xp^*^^ occur, and two syllables for the other 
two words. The correction is easily made by means of a concor- 
dance with sufficient accuracy, and the result can be expressed 
at once in hexameters and so deducted : when this is done for the 
Epistles we have as follows : 

Traditional Measured 
verses, verses. 



James 237 or 


' 242 


237 


I Peter 232^ 


, 236 or 242 


240 


II Peter 


154 


158 


I John 


274 


262 


11 John 


30 


30 


III John 


32 


31 


Jude 


68 


68 


Romans 


920 


919 


I Corinthians 


870 


874 


II Corinthians 


590 


596 


Galadans 


293 


296 


Ephesians 


312 


314 


Philippians 


208 


209 


Colossians 


208 


209 


I Thessalonians 


193 


194 


II Thessalonians 


106 


106 


Hebrews 


703 


705 


I Timothy 


230 


234 


II Timothy 


172 


170 


Titus 


97 


97 


Philemon 


38 


40 



Allowing for the diversity of texts and for possible errors in the 
nvunbers copied, it would be unreasonable to expect a closer agree- 



STICHOME TRY. 3 1 9 

ment between results. We have now the direct comparison between 
the text of Westcott and Hort and the early codices, as well as a 
satisfactory conclusion with regard to the verse-unit employed by 
Euthalius. The importance of this discovery consists in the fact 
that the question of stichometry is now removed from the region 
of averages, and we are able to determine the length of any pas- 
sage to within a hexameter. The only difficulty of a practical 
character is the divination of the particular forms of abbrevia- 
tion employed in the copies to which Euthalius referred, and in the 
partial stichometry there is the difficulty of determining to what 
part of a line the numerical indication applies. It must also be 
borne in mind that in the statements made by Euthalius as to his 
own accuracy (a#cp4/3«s) the remark is in strictness limited to the 
Pauline episdes. 

A glance at the results already arrived at will show that the 
greatest inequality between the results is found in the first episde 
of John, where the traditional measure is 274 verses against 268 or 
262 according as we admit abbreviation or not At first sight 
this would seem to imply that the Euthalian texts contained a con- 
siderable passage which is not found in Westcott and Hort, and 
the celebrated passage I John v. 7 at once suggests itself. When, 
however, we examine the partial stichometric data which Zacagni 
collected fi-om his Vatican MSS, we find that the same inequality 
runs through the book. For instance, Zacagni directs us to put 
the mark for the first hundred verses against c. ii. 26, at which 
point the actual count has only reached 90. There is, therefore, 
some unexplained peculiarity to be dealt with before we can come 
to any critical conclusion as to the verse in question. 

Further verification of the length of the Euthalian verse. 

We may readily confirm the previous results by examining the 
prefaces, prologues, etc., of Euthalius which are prefixed to the 
separate books, a large proportion of which are numbered in otIxok, 
And although in some instances corruption has taken place in the 
figures, the majority of the data agree closely with the hexameter 
hypothesis. For example, the following table will give the com- 
parison between the data supplied for the Acts and Catholic 
Epistles and the numbers obtained by syllabic division. 



320 AMERICAN JOURNAL OF PHILOLOGY, 

Ads of the Apostles. 



Traditional. Calculated^ 



np6\oyo9 Toav Tlpd^ttdv 


(Migne, col 


.628) 


140* 


138 


* ApaK€<f>ci\ai(i)(Tit 


c 


u 


640) 


?« 


107 


*Eic^€(ris icc(/>aXma)v 


( 


(< 


652) 


?i7 


II 


Kc^aXaia Ta>v Tlpd^€<ov 


( 


(( 


652) 


172 


178 


Breviarium capitulorum( 


(( 


661) 


40 


40 




Catholic 


Epistles, 






*Ava«;c(/>aXata>a'i$' 


(Migne 


, col 


668) 


14' 


14 


K€ff>d\aia *laKa>^ov 






677) 


25 


26 


K€<f>d\aia Ilcr^ov a' 






680) 


25 


24 


Kcc^aXaia litrpov ff 






684) 


10 


10 


Kc(^aXa(a *Ia>ai'i'oi; a 






685) 


23 


23 


Ke^aXam *Ia>ai/vov ^ 






688) 


5 


5 


Kci^.iXata *lovba 






689) 


II 


II 



And in the same way we might count the text of Euthalius 
through the Pauline Epistles, and we should find our hypothesis 
fully confirmed. There is sometimes, as above, a little conftision 
in the figures, but this is precisely what we expect when figures are 
handed down by successive transcription. 

These then are some of the results of comparison between a 
measured selected text and the traditional verse-numberings. 
Although they are more irregular in the Gospels, to which we shall 
presently refer, than in the Epistles, it must be admitted that in 
both cases (but especially in the Epistles) they offer a new critical 
instrument to the student of the New Testament, by means of 
which to restore the text to the same compass as it occupied in 
early copies. 

The matter is, however, much complicated by those causes which 
produce diverse measurement, to which allusion has been already 
made. Corruption of the data is common, and frequendy affects 
the greater part of the testimony : for example, the number of 
verses in Romans is 920, as given by Euthalius and many MSS ; 
but a larger group gives the impossible Xk and XH, which are 
nothing more than a corruption of ^K. It is, perhaps, a reasonable 
prediction that the next edition of the New Testament will be 
accompanied by a marginal stichometry. 

' PN in Reg. Alex. PM in Cod. Esc. 
« PN in Reg. Al. PK in Cod. Esc. PZ or PH coir. 

2 The reading A I of R. Al. and I A of Cryptoferr. are evidently corruptions 
of this. 



STICHOMETRY, 



321 



Instances of partial stichometry, 

Zacagni, in his edition of Euthalius, has furnished us with a 
series of notes and various readings under the title *' Variae lectiones 
ex Regio Alexandrino Vaticanae Bibliothecae codice depromptae.'* 
Amongst these are found a great many instances of partial stich- 
ometry : some of these coincide with the close of the lections ; and 
others have reference to the measurement by fifties and hundreds, 
of which Euthalius speaks as having been a feature of his edition, 
though it is by no means certain that he introduced it. The fol- 
lowing intsances are given for the margin of the Acts : 



Chapter. 


No. of 
verses. 


No. of verses 
by count as 
before : 16- 
sylL abbr. 


Chapter. 


Xo. of 
verses. 


No of verses 
by count as 
before: 16- 
syll. abbr. 


I» 15 


40 


40 


15. 34 


1350 


1352 


x» 19 


50* 


50 


17, I 


1465 


1460 


2,36 


150 


150 


I7» 15 


1500 


1502 


3, II 


200 


201 


18. 4 (?) 


1550 


1570 


4t23 


300 


297 


18, II 


15901 


C?) 1580 


4>3i 


315 


319 


19, II 


1650 


1655 


6,1 


440 


438 


20, 7 


1750 


I75I 


6.5 


450 


449 


20, 28 


1800 


1803 


7, 10 


500 


501 


21, 8 


1850 


1852 


7.53 


600 


610 


21, 14 


1870 


1870 


7,60 


625 


625 


21, 28 


1900 


1905 


8,13 


650 


654 


22, 5 


1950 


1953 


8,34 


700 


703 


22, 26 


2000 


2004 


9» I 


717 


719 


23, 10 


2050 


2046 


9, 15 


750 


751 


23»30 


2100 


2102 


9,31 


792 


795 


24, 18 


2150 


2153 


9.36 


800 


804 


25»4 


2170 (?) 2187 


10, 12 


850 


851 


25, 12 


2200 


2210 


II. 7 


950 


954 


26, I 


2250 


2255 


11,27 


998 


1000 


27,1 


2325 


2336 


i3> II 


1 100 


1 102 


27, 10 


2350 


2360 


14, I 


1200 


1 201 


27, 29 


2400 


2406 


i5» I 


1271 


1271 


28, I 


2450 


2452 


15. II 


1300 


1301 


28, 17 


2500 (?) 2485 



And the completed reckoning gives us 2559, which must be cor- 
rected for abbreviations to 2527 ; results which agree very closely 
with the number given by Euthalius, 2556 ; and the number given 



322 AMERICAN JOURNAL OF PHILOLOGY. 

by Scholz from a large group of manuscripts, 2524. It wUl be 
noticed that our reckonings are 2 or 3 verses only in excess in 
either case. 

In the partial stichometry tabulated above, it will be noticed that 
the results (which I have done my best to keep clear of error) are 
very closely in harmony with one another : and it is conceivable 
that the adoption of a letter-line might niake the approximation 
even more close. In one or two of the data errors appear, as at 
c. XVJII 4, where we have 1550 verses ; and at XXV 4 we are told 
to put the figure 2170, where the scribe seems to have dropped a 
ten, and the defect shows itself in the subsequent figures. It must 
be remembered that a single printed verse will sometimes contain 
five or six <mxoi, so that we could hardly look for much better agree- 
ment, and we must defer a closer critical comparison until the text 
can be printed stichometrically with proper abbreviations, and an 
accurate marginal reckoning of the lines, suitable for comparison 
with a revised critical edition of Euthalius. I think we may con- 
clude also that the printed text of Westcott and Hort in the Acts 
is within three hexameters of the text circulated in the third 
century. 

The importance of these intermediate stichometric data is obvious ; 
and the only difficulty in applying them lies in the determination 
of the part of the verse to which the stichometric number belongs. 
Sometimes an intimation of this is given by Zacagni, at other times 
he does no more than designate the verse against the margin of 
which the mark stands. 

Let us apply the evidence supplied by these marks to the critical 
question of the authenticity of the passage Acts VIII 38. The 
doubtful sentence is about three hexameters long. Against the 
margin of VIII 34 stands the number 700: against the first verse 
of IX, which is also a new lection, the number 717. 

The 34th verse of the eighth chapter is 2I hexameters, from the 
35th to the end is 13 hexameters, omitting the doubtful words, and 
the first verse of the 9th chapter is a hexameter and a half. 

But since this first verse ought clearly not to be counted, for the 
beginning of the lection is the point noted, we have at the most 
i5i hexameters, with no allowance made for abbreviation. It 
requires, therefore, the disputed passage to make up the tale. 
The partial stichometr>% therefore, recognizes this passage. 

We shall now give in order for the Catholic Epistles, for con- 
venience of reference, the Euthalian measures, together with any 
partial stichometry supplied by Zacagni : 





STICHOMBTRY. 


3 


James 


Lection I 


Verses. 
112 


I Peter 


II (c 3, I) 
Lection I 


121 

58? 


II Peter 


II (c. 2, 9) 
Lection I 


149? 
154 


I John 

II John 

III John 
Jude 


Lection I 

II (c 3. 15) 
Lection I 

Lection I 

Lection I 


150 
140 

30 
31 

68 


James 


c. I, 26 


50 




c. 2, 21 


100 


I Peter 


ad fin. 
c I, 22 


230 (? 237) 
50 




c. 2, 9 


58 


II Peter 


c. 4, 19 
ad fin. 
c. 2, I 


200 
246 

50 




c. 2, 20 


100 


I John 


. c. 3, 17 
ad fin. 

c. 2, 26 


150 

154 
100 




c. 4, II 


200 


II John 

III John 
Jude 


ad fin. 
ad fin. 
V. 14 
ad fin. 


37 (?) 
32 

50 
. 68 


In the Pauline Epistles 


we have the following data 


• 
• 


Romans 


Lection I 


242 


* 


II c. 5, I 

III c. 9, I 

IV c. 12, I 


248 
185 
"5 




V c. 15, I 

Total 


125 
920 


I Corinthians 


Lection I 


250 




II c- 7, I 
III c. 8, I 


84 
116 




IV c. 12, I 


266 




V c. 15, I 

Total 


154 
870 



324 



AMERICAN JOURNAL OF PHILOLOGY. 



II Corinthians 



Galatians 



Ephesians 



Philippians 



Colossians 



Lection 
I 

II 
IV 

Lection 
I 



Lection 
I 



Lection 
I 



Lection 
I 



I Thessalonians Lection 

II Thessalonians Lection 



Hebrews 



Lection 
I 
II 



I Timothy Lection 

II Timothy Lection 
Titus Lection 
Philemon Lection 

Total for the Pauline epistles, 

The partial stichometry is as follows : 



Verses. 

c. 4, 7 156 

c. 8, I 94 

C ID, I 187 

Total 590 

130 
c 3» 15 163 

Total 293 

136 
c 4, I 176 

Total 312 

120 
c 3, I 88 

Total 208 

157 

c 3. 17 51 

Total 208 

193 
106 



c. 7, II 

C. II, I 

Total 



257 
232 

214 



703 



230 

179 (? 172) 

97 
37 
4936 



c I, 24 


50 


2, 14 


100 


3» 9 


150 


4. 9 


200 


5» \ 


240 


5.6 


250 


6, I 


300 


1> I 


350 


7, 21 


400 


8, 22 


450 



Romans c. 



9. 30 
II, I 

II, 
12, 
12, 

i3» 

I4» 

i5» 
16, 



24 
I 

II 

13 

23 

25 
18 



550 
600 
650 

675 
700 

750 
800 
850 
900 



STICHOMETRY, 



325 



I Cor. 



II Cor. 



Gal. 



Ephes. 



Phil. 
Colos. 



I Thess. 

II Thess. 
Hebrews 



I Tim. 



II Tim. 



c. I, 26 

3» 4 
4» 8 
5» 10 
7» 27 

8, I 

9, 16 
II, 10 

c 3» 2 

5» 10 
8, I 
8, 20 

c. 2, I 

2. 21 

3. 15 

c 3» 3 
3» 21 

4. 10 

c. I, 17 

2, 19 
c. I, 23 

2, 14 
3» 13 

c. 2, 10 
3> " 

c. 2, 9 

c. 2, 8 

3, 12 

4, 14 
5» 13 
7, 2 
7» 25 
9» I 

c. 4, I 

c. 2, 14 
3,6 



50 
100 

150 
200 
300 

331 
400 

500 

100 
200 
308 
350 

100 
130 

100 
136 

150 

50 
100 

50 
100 

150 

50 
100 

50 

50 
100 

150 
200 
250 
300 
350 

100 
150 

50 
100 



I Cor. 



II Cor. 



Gal. 



Ephes. 



Pha. 
Colos. 



c. 12, I 
12, 27 
i4» 29 

15, I 
i5i 16 

I5» 47 

16, 13 



c. 9, 14 
II, 4 

11, 26 

12, 18 

c* 3» 24 
4. 27 
5» 22 

c 5, 28 
6, 19 



c. 3» I 
4, 18 

c 3, 18 
4, 16 



I Thess. c. 5, 3 



Hebrews c. 9, 21 

II, 5 

11, 26 

12, 4 

13, I 
13, 23 

I Tim. c. 6, 10 

II Tim. c. 4, 16 



550 
600 
700 
720 

750 
800 

850 

400 

450 
500 

550 

150 
200 

250 

250 
300 

120 
200 

157 
200 



150 



400 
500 

550 
600 
650 
700 

200 
150 



326 AMERICAN JOURNAL OF PHILOLOGY, 

Extension of enquiry io th4 Gospels. 

When we turn to the Gospels we find a difiiculty arises from the 
fact that almost all the causes which tend to produce variety of stich- 
ometric subscription are in operation. In particular the variety of 
texts is great The Textus Receptus, for example, shows an 
excess of at least 50 hexameters in the Gospel of Matthew over 
the text of Westcott and Hort This makes our enquiry extremely 
interesting, for we begin at once to ask such questions as relate to 
the authenticity of the last twelve verses of Mark, the pericope de 
adultera^ and other important passages. Does the stichometry, 
which is certainly very ancient, recognize these disputed places as 
belonging to the texts of the New Testament on which its reckon- 
ing is based ? In the first place we have to face the diversity of the 
traditional measurements; the following tables are based upon 
numbers supplied by Scholz, Tischendorf and Scrivener. 

Matthew. 
MS, 
428 

421 

157 
161 

164, 262, 300, 376 

9, 13, 124, 163. 174, 175, 345, 346, 427 

G. H. S. 7, 18, 28, 41, 45. 46, 48, 50. 117, ^ 
122, 131, 153, 237, 241, 246, 252, 261, 263, i ^ , 
277, 280, 290, 292, 347, 348, 388, 435, and I '^^ 
1, m, n, w, (of Scr.) 

K. 6, 116, 387 ftylt' =2700 

339 A»f = 2860 

264, 273 ,yTt{^ =? 3397 

Mark. 

4 ^' = 1020 

164, 262, 300, 376 /w^' = 1506 

"7> i53» 157 ,o<^' = 1550 

A. ^if' = 1590 

G. H. S. 7, 18, 28, 41. 45» 48, 50, 128, 167, ] 

202, 237, 241, 246, 252, 261, 267, 277, 280, \ fix' = 1600 

290,292,301,347,388 J 

9, 13, 124. 163, 174, 175, 339, 346, 427, 435, fixes' = 1616 

K. 6, 116, 387, 128, 131 /t^t = 1700 

264, 273 fimtce^ = 1829 





Sri;r«- 




flvnif 




= 1474 


A' 




= 2400 


fivKif 




=2484 


fiH 




= 2500 


^^ = 


--Q^^: 


)=2554 


fi^^ 




= 2560 



= 2600 



STICHOMETRY. 


327 


Luke. 




fixf 


= 2606 


fixoC 


= 2676 


546 fi^ii' 


= 2740 


M> 


= 2750 


M^ 


= 2760 



20 

A. 164, 262, 300, 376 

124, 163, 174, 175, 345, 346 

9» I3» 427 

157 

G. H. K. S. 4, 6, 18, 28, 41, 45, 46, 48, 50, \ 

116, 117, 122, 128, 131, 153, 202, 237, 241, I ^, _. g^ 

246, 252, 261, 263, 267, 277, 280, 290, 292, ] *^ 

347» 343, 387* 388, 435» and 1, m, n, 

264, 273 ,yfl»«c{' = 3827 

John. 

4 /»'•'= 1300 

157 f^K' = 1930 

20 ^4' = 2010 

9, 13, 124. 163, 174, 175, 345, 367, 427 fiiJi = 2024 

A. 164, 262, 300, 376 fiai = 2210 

G. H. S. 4, 6, 7, 18, 28, 41, 45, 46, 48, 50, ^ 
122, 128, 131, 167, 202, 241, 252, 261. 263, I ^, _ 

267, 277, 280, 290, 292, 301, 347, 348, 387, \^ ^ 

388, and 1, m, n, 

These are the principal MSS data, and it must be owned that 
their discordance is a formidable objection to the assumption that 
the Gospels are measured in precisely the same way as the Epistles. 
A number of the data are evidently corruptions ; in Matthew fioi^ 
is probably altered from fi4>^\ in Luke fi^d is obtained by omission 
of a single letter from fix^dy and so on. 

In the Synoptic Gospek, the main body of the MSS divides into 
two groups, of which one gives the arixoi to the nearest hundred, 
and the other goes more into detail. When we find Matthew to 
consist of 2560 or 2600, Mark of 1616 or 1600, Luke of 2740 or 
2800, we may regard the larger group of MSS as less accurate 
than the other. The problem is now much simplified. 

In the Gospel of John the numbers are difficult to arrange ; it 
is almost impossible to believe that the book contains 2300 verses, 
and we may perhaps set the result again with the group of MSS 
that gives 2024. This is the number given by Scrivener. For the 
present, then, let us adopt the numbers 2560, 1616, 2750, 2024 for 
the four Gospels. We must now divide the text of Westcott and 
Hort and the Textus Receptus into i6-syllabled rhythms as before, 
firstly, without abbreviations of text, and secondly, with the same 
abbreviations as were previously noted. We have then : 



':'JILVAL OF PHILOLOGY, 

L.-.L ; cotnse be higher than the worth of the 
. -: z!t tradition. 

r=sr wiH arise in the working out of the 

.-^« :. die manner of syllabic division in early 

L-r. --2C question arises as to whether, in MSS, 

-:s-Jj.oie» etc 

*. s:? of the New Testament have now been 

.x:i.aon of the Apocalypse. For this there 

„^rtance in the MSS, but the stichometric 

=- ^ ves 1400. By actual enumeration we find 

. ^ .^joreviated, which does not agree with the 

^ Testament Siickomeiry, 

. -ae Septuagint and Apocryphal books, the 

^ _ tc -cchometric table of Nicephorus, previously 

, ^.;t; table is exhibited in a Latin translation of 

.. jobC every instance the number of verses is 

.^ iiatB hundreds. A stichometric table is also 

, . V Jtdromontanus. Other data referred to by M. 

. -V -^^ Catena in Heptateuchum, Codex Escori- 

^ . ,vc. M. Graux employs these numerical data 

. , . uienoe of the im'xoff and the average hexameter. 

,.^i>iiofus has been reprinted in Credner, Zur 

. ^^., in Migne's Patrologia 100, col. 1055 sqq., 

^ .. •* -JK Canon, p. 560-2. It is therefore unne- 

^ .^ . ,>ut it is well to notice that Westcott hardly 

^ .i-.endon of stichometry when he says (p. 520) 

v> u^ no more than tables of contents. If the 

^> .vmI been a little less approximate in its numbers 

^%.t ot preservation it would have been valuable 

.V- ^^r^erves a careful examination in the light of 

^,. ^^ ciws. As it stands, it sufficiendy verifies (which 

, ^ . cxiiKirts would do) the hypothesis of the hexam- 

. VL c :s incidentally interesting as throwmg light on 

>^ . >*. iiw lost apocryphal books. For instance, the 

.au .imi Modad, which is quoted in Hermas* Vision 

..V* -c >i ^'^^ verses, or almost as long as the Episde to 

^ . ^ So aiso the Apocryphal Ascension of Moses, to 

oa^ and Origen' refer the quotation in Jude 9, is a 



-♦'•, 



• Orig. dc Princip. iii 2. 



STICffOMETRY. 331 

work twice as long as the Epistle to the Hebrews. To the same 

source Euthalius' refers Gal. 6, 15, oiir€ vtptro^jaj n cWty oijT€ aKpopva- 

Tia aXXii Mtv^ KTiais, which throws light upon the reading of Codex 
B and allied documents which omit cV x^iv. I suppose we may 
assume the genuineness of these quotations, for either Euthalius 
verified them himself, or being, as he says, merely a novice, and 
having no originality beyond what we may call a printer's or 
editor's originality, he referred to some earlier writer ; a supposition 
which by no means detracts from the value of the quotations. And 
who shall say that the greater part of Euthalius' work does not 
date from the time and school of Origen himself? 

J. Rendel Harris. 

> Zacagni, p. 561. 



REVIEWS AND BOOK NOTICES. 

An Anglo-Saxon Reader in Prose and Verse. By Henry Sweet, M. A< Third 
edition, revised and enlarged. Oxford, at the Clarendon Press, i8Si. 

An Anglo-Saxon Primer. By Henry Sweet, M. A. Oxford, at the Clarendon 
Press, 1882. 

Specimens of Early English. By the Rer. Richard Morris, LL.D. Part I. 
A. D. II 50-1 300. Oxford, at the Clarendon Press, 1882. 

Mr. Sweet's well-known Anglo-Saxon Reader reached a third edition two 
years ago, but without " the complete revision promised in the preface to the 
second edition,'* and to make such revision he still awaited the publication of 
Sievers's Gertnanic Grammar and his own Oldest English Texts, Sievers's Angkh- 
Sctxon Grammar yfViS published in 1882, and Sweet's Oldest English Texts is 
referred to as authority for the gender of certain words in a recent number of 
the Anglia, so that we shall probably soon have a new edition of the Reader, 
which has already superseded most other Anglo-Saxon text-books. Meantime, 
however, to meet the want of a more elementary " introduction to the study of 
Old-English," Mr. Sweet has prepared his Anglo-Saxon Primer, and says, " My 
main principle has been to make the book the easiest possible introduction to 
the study of Old-English." It deserves notice in passing that, while Mr. 
Sweet says in his Reader (p. xi) : ** In this book the name ' Old English * will 
be used throughout," and repeats this in his Primer (p. i), his more practical 
publishers persist in using the name *Anglo -Saxon * as a title for each, so that 
Mr. Sweet will have to "reform it altogether" before the public will under- 
stand that *Anglo-Saxon * and • Old-English * mean the same thing. Mr. 
Sweet's Primer contains no poetry, and its 116 pages are made up as follows: 
Grammar, 54 pages. Texts, 36, Notes, 6, and Glossary, 20. The Grammar 
contains a useful synopsis of Anglo-Saxon inflection, but in the effort to be 
concise the phonology lacks clearness and has some omissions, as under muta- 
tions, a to a, to e, e to ie^ea to fV, and eo to ie, and the f -mutation is the only 
one treated. Mr. Sweet still prefers to use the terms mutation and gradatiffH 
rather than umlaut and ablaut^ but the German terms are decidedly preferable, 
and have now been used long enough in English to be regarded as thoroughly 
anglicized. He also discards the term breaking, thus losing a convenient term 
for a most important phonetic change, and classes breakings with other changes, 
as on page 5 ad Jin, 

In the Primer the symbol is discarded, and while in Reader the pro- 
nunciation of short is given as in S. G. stock on p. xiv, and in Primer as in F. 
beau^ p. 2 adimt.f the key-word not is given for this sound in Reader, p. xvii, 
and Primer, p. 2 ad fin. The English word 'wholly^ as given by Prof. March, 
expresses the sound more exactly. In both works Mr. Sweet has changed his 
former views with respect to some sounds, now saying, "^had the sound of 



REVIEWS AND BOOK NOTICES, 333 

E.^ before f, ^, j, both short and long** (R., p. xiv, cf. P., p. 3), and " j had 
the sound of s" (R., p. xv« P., p. 4). In Primer the character % is discarded, 
and p always written, but pronounced with *' the sound of our th (= dh) in 
then^'' except after '* hard ** (t. e, voiceless) consonants. Nothing seems gained 
by this forced consistency to suit beginners. Under gradation (p. 6) a . . . d^ 
as in nam^ ndmon^ is omitted, and a little fuller explanation here would have 
aided beginners. Mr. Sweet still adheres to the classification of nouns by 
plurals, though the classification by stems has many advantages, not the least 
of which is its aid to the student in comprehending the uniform scheme of 
Teutonic grammar. Under Ar-plurals in Primer they- and zc^-nouns of Reader 
(f. t,^ the/E>- and wa-, or according to Sievers they^and ux^stems) are omitted, 
and nothing is gained by changing the usual order of the cases in inflection, 
and, under adjectives, of the genders too. Some useful, though brief, remarks 
on the declension of proper names are added in Primer, but the explanation of 
the use of the weak declension in adjectives is omitted, which is the more 
noticeable as the inflection of the article with the adjective is also omitted in 
Primer : here too the genitive plural is given as gddena^ and nothing said, as in 
Reader, of the occurrence of the more usual gddra. Under the Demonstrative 
Mr. Sweet now states that ** j/as a demonstrative and personal pronoun has the 
vowel long," but this statement does not occur in Reader. He also writes in 
Primer \ids nUn vwrd{pp, 21, 44, 45, 62) for pds mitu 7oord of the Reader (pp. 
xc, xcii, 51), and says (pp. 43-4) the possessive pronouns always keep the strong 
form, but according to the Reader (p. 1), m/fu is strong as well as m/n. In 
Verbs (p. 22), " vowel-mutation " is written for " 'gradation,* and p. 23, " 1st and 
2d persons " for ** 2d and 3d,** and under the mutations (p. 23), atoa^^e to f/, 
and U to /, are omitted. On p. 25 we find weaxan now classed with the/a/t- 
conjugation (though placed in Reader under the J^^^-conjug^ation), f. ^., the 
reduplicating verbs, but no hint is given in either Reader or Primer that this 
is the redupiicating class. However easy for the memory it may be to class 
Anglo-Saxon verbs by using an English verb for a title, something more is 
needed to designate their systematic arrangement ; the arrangement by classes 
showing reduplication and ablaut, as in all the German grammars, is far 
preferable, and here again it aids the student in learning his Teutonic gram- 
mar. In the/ai7-conjugation examples are wanting of stems showing a before 
two consonants, as spannan, and those showing <i/, as Idttan^ the latter being 
an important omission, as this verb occurs in the texts. The order and name- 
verbs of some of the conjugations are different in Primer and Reader, an 
apparently unnecessary change, and the verb scacan, which gives the name to 
one conjugation, is omitted. Under weak verbs in Primer the zeva/t-conjuga- 
tion is classed as entirely separate from the i^^-conjugation, thus differing 
from the Reader, but for the better. The classification by stems would be 
preferable here too. In I (b), p. 32, tellan is omitted, though occurring in the 
texts. Under preterilive-presents, called by Mr. Sweet " strong-weak verbs,'* 
unmin and dugan are omitted in Primer; they might have been added for the 
sake of completeness. A few pages on derivation and a serviceable compen- 
dium of syntactical usages complete the grammar of the Primer. 

The texts consist of sentences *' gathered mainly from the Gospels, Aelfric's 
Homilies, and the Chronicle,'* selections from St. Matthew and the Old Testa- 



niOLOGY. 

the Chroniclt, and " King 
are well luited forelemen- 
nad«, il would be that Ihef 
tract! from the Chronicle; 
om King Alfred's tnnsU- 
for manj sladenU want an 
a basis for the historical 
>e lections for reading Mr. 
at interfering with the use 
itended coarse. In teacS- 
is not every school or col- 
as, howevn', done a scrrice 
: coald wish that he had 
and conjugations. There 
vers's Crunmar. The fol- 
p. 5S. I. 91, ansfnu {arnsitn 
an; 1. 139, utgaf; p.6[,t. 
es amctmudan) ; p. S3, 1. jg. 
ve capital p; p. 87, 1. 17S. 
a; should be p. 79 ; a note 
. 81, t^ pdn gtcieriem, after 
question whether Hekaman 
" instmmental dative of 



ilri. 

ntroduclion to the studjof 
( might be made still more 
Hr. Sweet " that this little 
rs, but also to some of our 
e, most of whom are now 
Ltary knowledge of 'Anglo- 

irt I, has at last appeared, 
vcring the fourteenth cen- 
riodi and these two books, 
le hand, and Prof. Skeat's 
ipply us with a complete 
Spenser. If they serve no 
icing people the continaicy 
: and colleges will be no 
Uy. Dr. Morris and Prof. 
if Part I, as in that of Part 
igations. Heretofore Prof. 
glish, which inclndes from 
cessible for this period to 
w. Prof. Znpiiia's Cbuogs- 



REVIEWS AND BOOK NOTICES. 335 

bttch is too meagre, Prof. Wttlcker*8 Lesebach does not begin early enough, 
and Maetzner's Altenglische Sprachproben, still the fullest and most useful for 
advanced students, is not so suitable for more elementary instruction, as it is 
without grammar or glossary, Maetzner's separate glossary being so extensive 
and so sl^ in its publication that it will be completed about the end of the 
century. The present volume consists of nearly 100 pages of Early English 
grammar based on the Southern dialect (in part contained also in Part II), and 
following as closely as possible Mr. Sweet's arrangement and classification of 
declensions and conjugations in his Anglo-Saxon Reader. Prof. Skeat has 
prefixed sections on pronunciation and metre, the former after Ellis, with a 
scheme of the vowels, chiefly after Stratmann in the Philological Society's 
Transactions for 1867, but ** with some re-arrangement and slight modifications." 
It is this " re-arrangement " that is objectionable from the historical point of 
view. A student will hardly take up this volume without an elementary 
knowledge of Anglo-Saxon, and Prof. Skeat acknowledges this in his preface, 
p. xvi. Therefore the vowels should have been arranged on the basis of the 
Anglo-Saxon vowels, showing under what different forms each Anglo-Saxon 
vowel appears in Early English. This arrangement is not only much simpler, 
bnt more logical, and is strictly historical. The list as given has some omis- 
sions, €, ^., short a = A. S. A» / a^ = A. S. a, /, ^, and ea j and long de^A. S. 
/ia, so common in Layamon ; also aem will hardly answer for an example of ae 
=s A. S. M, The phonology should be revised and rearranged. Prof. Skeat's 
section on metre, though brief, is useful, as text-books universally, one may 
say, neglect this important subject; his remarks on the metre of Layamon 
may be usefully compared with what Schipper says (III 7) on the same subject. 
In the forms of the pronouns there are several omissions, but perhaps it was 
not intended to give al/ the various forms which occur in the dialects. The 
strong verbs are classed as in Sweet, although the order differs from that in 
both Reader and Primer ; in a connected series of text -books we should expect 
it to be the same in each. There are, however, some wrong classifications : 
gnawtn (p. Ixvi) belongs to the shake-^ not to the /aiZ-conjugation ; drepen and 
tUken to the give-^ not to the ^r-conjugation ; and some omissions of verbs 
are found in the texts and glossary : laken^ under fall; wreken^ under give ; 
ripen (?), of which a past plural repen is found in I 196, and wiUHy under drive ^ 
and bruken under choose; also drapen^ II 28, past plural oidrepen^ is omitted in 
the list uf verbs ; and waxen is put under shake^ as in Sweet's Reader. 
Under anomalous verbs daren is given as infinitive where we should expect 
durren, and unnedsLS past participle of unnen. A few misprints in grammar are 
cuman (p. Ixix), skp (p. Ixxx), and node (p. Ixxxii). 

The selections for reading consist of nineteen pieces extending from Old 
English Homilies, before 11 50, to King Horn, before 1300, comprising 286 
pages, so that a very complete collection of specimens of English for this diffi- 
cult and little-known period in the history of the language is given. The pub- 
lications of the Early English Text Society have rendered such a series 
possible. These selections are in different dialects, and there are not many 
pieces that are valuable as literature in themselves, but it is as records of 
different stages of the language that we appreciate them, and the chronological 
study of those in the East-Midland dialect cannot but give us a clear idea of 
the historical development of our tongue. The selections from works in the 



T^ *P-. 



1 




^.** 



t D a- ii— 



ro^' tb* irjin.'il 



*-z »- — crc 



a TWT or 



f .•**!' 



-'.« ff 



« t.-.'^ 



/.jfT 




ttiv^-m i.*-!;. i/_--n'-'. n * Iokj^ -*.* Tut » a;n^]-!ra»-r 

ywr^t^\ .tji wVjl *iJt tai.^t Tw^ ■p-iiin it iirrsa. «c cstrrvc n: ttit ^ixssbtk. 

veu'.' rt ..Jt j^:.'^aj(fc.-7 ijfci: Zi HI* tirr^s Taj's ir *■ Zarrrmnnr gni: AddnJBBS.* 

iXA v» 1/v KJtan - vy"»rx*rt "tifTTT tZ- Ix Hail IT I :> ■•CT^*:^ It 

t»*4i**-« *-!.'. i't ' »"uitx t-?r-.i.j:-.-j BT-nt "iiit tmaismat "bencriLnc rvma$ tie 
*a u.v-<-^^» w -t t'^iu'^'fm ; ia I 4^ pt c xm s M ?» nn fi-r «ej' J^«« home k 
^A< ' i-x^ -iut i — t tiJ'A wfcfg lilt exr± irrt nir Saajf^" ss if 
•^ K^ i*/'^.'« ' W 'jUt ^.vMit'T f-TTct 'Here r3«*t:? = 

U*vt.4-*>y; ■ t«ncfnttMf *j^ f^rx>t»tt Ivaa*." far #7. 0« il ta, imiS isrrfem is 
t/|jn,i'^ " c».tfce4 il ^> I* »cocz>td witi kia^ia^^ fcCla^^Lssg sf^iuestly 
lUr*** jf>At*/f, U-t ti« glrAiATf fTTcs ^CTC rr /fli = " coTCT 13,"* mnd ProC 
^.ViTftil wzif*:^ tf.e fMx<^wry ofTcctioa to tlie noce ic£, ScruaumB); so II 74, 
/tm Uf * df« ft/:^v:4rtkt to** m note, bat "pertain to" in glossary; o« III 90, 
iStK ftrU f*:ut.t: \n n/Ac U wrong; perhaps L 3, p. 26 is meant; on V [163 tbere 
U « m it* 14 u%\;^Hf /tit Drihhun is datire, and another 00 1635, ** from** shoald be 
"of'; «/n VI ^A, the note is correct as against the glossary, aekUme^ 
"mj/hiei-n," wA. "good"; in Villa 73, Awf is wrongly explained of ••the 
devtl"; In ft3 ** I" i» omitted. and in V III h, leaf ttdez=^ leaf ede^ loTcd," in 
firilr, l;Hl more correctly "flattered," in glossary. On IX 13, we find " with- 
hold iny " for *• wiihholden "; 18, " knows " for " thinks'*; 90, segtm in quotation; 
1 33, " ( (til l« " in»crtcd ; 1 53 wrongly translated (cf. glossary s. y. don) ; 167, note 
and ylofctiaty fttfaln at variance, s. v. strapeles ; 331, "sternly" for "stem." On 
XII 7<». note and gIo»tary are at odds, s. v. swideS; so 102, s. v. TcvnrS, and simi- 
larly Xf V 6, e^Ucht ; on 338, reference should be XVI* or 16 (Jes.), and has 
little applicAtion; 331-3 Is mistranslated, and dwaUs, 414, is omitted in glos- 
sary, Inil tltene omintlont are numerous, as will be noted below. On XVI i, 
note and (;loii«ary are at variance, s. v. sumere^ but correction made on p. 538; 
on a6, ** when '* should be ** where"; on 39, speten^ cf. glossary, 170, bUnck^, 267, 



REVIEWS AND BOOK NOTICES. 337 

hut ; 34a. " she " for " it " (cf. " English as she is spoke ") ; 616, bleU, cf. glos- 
sary ; 744 is mistranslated ; on 84s. yne^ note and glossary are both wrong (cf. 
Stratmann's Dictionary), and in next line for pi/, corrected to vm in note and 
to Hu^ p. 536. read with Jesus MS Nu; 850, "goeth" for "go," plural ; 905, 
y^de =• ** people," as glossary, not •* land "; 1638, wear = •* cautious," not 
*• aware (sure)"; 1640, iwif/ = ** missett," not "mixest"; finally, l^st^ wis-dom 
would better be separated ; it is omitted in glossary, though occurring again in 
1766. There are so many errors in the notes to the ** Owl and Nightingale ** 
that one might readily believe that they were prepared by some foreigner, or 
tyro, and not by either of the editors. There are no notes on '* Havelok the 
Dane ** after 547, although there are 200 lines more of text, and in this line 
yW is plainly •* very " and not " foul " as in note, which would be mere tautology. 
The glossary consists of nearly 200 pages, and while quite full and very 
useful, has some mistakes in alphabetical arrangement and several omissions: 
bletcaed com^s after ^ibr, and under ^/f/x^, "See Bletsed'*^ should be bletcaed^ 
there being no bletsed. A serious mistake occurs under the preceding word : 
bleowut I 195, kit wex and ble&utu {ffleduw) in iudea, from blowen, to bloom, is put 
along with bleu^ from blowen^ to blow, under bleowen^ with a reference to blawe^ 
and the verb blawen^ to bloom, is omitted entirely. On p. 384, col. 2, from 
crabbe to cristendom is misplaced and should come after couerture^ col. i ; faeston^ 
2, 139, pt. pi. Uomfaesten^ is put undtr /aestned^ 2, 33. pt. part, from /aestnien, 
And /aesten is omitted; on p. 409, col. 2, iTom galwe-tre io gasteiich is misplaced, 
and should come after galun^ col. I ; seapeloris and scct^^ p. 476, col. i, are also 
misplaced; so welkede^ p. 519, col. i, wet^ wete^ p. 521, col. i, and on p. 534, col. 
2, from yescung to yt. The proof-reading is, therefore, seriously at fault, as 
these errors might easily have been avoided, and the book bears marks of 
haste in preparation. The following omissions in glossary have been observed, 
but I have not in all cases noted the reference : alle^ hall, 6, 188, unless we 
should read halle, but Layamon was one of the first cockneys ; arehXie^ 16, 1716 ; 
dpestre^, I, 168; bedesang, 5, 1450; belden^ verb; bi-smitted, <), 113; (the refer- 
ence 9, 139 is wrongly placed under bup^ is, instead of bup^ buys) ; chere^ 8 b, 19, 
though given under ^/f chere ; cherde, 16, 1658, though charen is given; dage^^ 
4 c, 60; deorcMn (reference under dierchin should be I, 52); derevmt^^ i, 161, 
'Hce, I, II ; drihtful\ drit, 17, 682 ; duhe^^ 8 a, 10; dtoaies, 14, 414; dwiUe, 5. 
1499 ; «>, 17, 606 ; erffe^ 5, 1068, but erue is given ; {JoUtdnge is wrongly placed 
undtx folc); fuUfremedd, 5, 1576; geme^ but yme is g\ycn \ goulen^ 17, 454; 
greue^ 9, 226 ; hefy, 5, 1442 ; (under kennes^ *' gen. s." should be ace. pi.) ; houe^^ 
12, 69; i-eundur^ 16, 85 ; iJHndUd, 12, 16 ; late, adv., 17, 691 ; lift, adj., 4 ^i 77 ; 
AV, adj.,4d, 13; mose, 16,69; f^^^t 9i 4^; on-walde, 3 b, 68, but anwalde is 
given ; purses, 9, 197 ; ran, 17, 691 ; under schote \schotte'\ we find " read sckolde, 
2 pt. s. subj. should," without reference, which should be 14, 411, and the 
explanation is wrong ; seholde here := scold (cf. Stratmann, s. v. scolde), as the 
connection plainly shows; sleate^, 9, 63; sU^rende, 4b, 100 ; sol. [= sondes'], 2, 
79, and no explanatory note ; strene, though referred to under istreoned; sum» 
chere, 8 a, 14; svhiI, 16, 7; tele\i, 14, 237; tobilimmpepp, 5, 1657; {to-she^yedd, 5, 
1498, is explained by White as " scattered" rather than ** scared away," though 
the former may be derived from the latter); to-tose, 16, 70; tukest, 16, 63; 
(under peo, 9, 23, peo pet = she that, is wrongly placed with pea pet =: those 



338 AMEKJCAJf JOVKNAL OF PUILOUXY. 

that); pru», V), %\\, aad in text cobiu ihoald be [daced after iM'ac amd 
omitted afi«T5tS; wJhto'. fl a, lO^aBd ■— trf. Ja. K^iboBld be — m<. aa in 
leu; ™-Wj*i, 16, hS (cL 16. 339); •wwor^; —j yt . 14, 444: ■^. 14. 
418 : uder vMe the mfiait " vkya, kjadi," as ui !&, 73, B gatittrd ; Bernard. 
14,438: arr [= Bu), 16, 743; acii, po«er. 6, t>6 ; a^cMS. 9. 6a ; a^-dbn, 16. 
1756, 1766; wdt, wowl, 16. 76; mui-mtV. 16. 1659; cC StratmaBB for tU« 
mud, and k t. frf, fcw ■ — Z " , Ba,4l, aniMi, Sh, 5^ wkicfc lut i* omitted ■■ 
^ass«i7; (for ymt%tt abo*C,aiid cC^riwci),- ;iTrr,I&.4iS; ^rftBiige, I&. 40u 
TKeaamber of vordsoaitixd in Na>.i6, 'Owlaad NitbtiBgale," eaafiiBi the 
opinion cxpiened above (hat a 'picatice kaad mail have had to do wilh its 
cditiiig. The jHTiniarj. ihefefore, needs revinoa as well as (he aoles, bat the 
difienlt; of prepating mth " ^"""J — -" *— ~— i-t— j .»J ».^— ;«fcrf.~i:— ^ 
sgne sixty oaissi^ ia it. and the conectiaa* needed tUc n hut , tcachen will 
be siatcfal for the book, hop*^ ihnt the scowd editiaa wiH tlwa 4 decked 

Juns H. GtMXm. 



Die EeCinichriften nnd das Alte Tes 

einoB Beiiiace ma Dm. Pacl Hacft. Zvctte itmlMiTili nad sdi 

Tcmehite Aaflacc Nrtot chwao knit^hta Bdcabea. sa«i Glcosaicn, 

EecisUTn and ciner Kaite. GkMea, J. Rkler'schc BachhandlB^. 1U3. 

This new editiaa of a wdt-kaova hook dnaicles Ik a di an ce which lea 

faasf Tcan haie aade ia a aew wieace. The iiiIsMI has pvwa \ufet hf 

■Kxe than 900 pages, aad the old Baiter has beca thuiiia^hly worked over, to 

thalthestKiealafthe Kbte historf aad nl ifioa wiD iad nock Aat is (ic^ 



d. Thee 

CO. bat there aa* * taife d i .cn,». tt aa c er ni at T- « a ty in part recoc- 
rn1ia[ the atce details t£ ttimafa^i aad sjatai. aad this anfiala 
e some Skesitic scholars adtstiast of :hewho>scieBce,whickjidds 
at ilowlj la the cAons t£ (hoc wte ne tr*^ lo sabttitBte More 
hods fat the lUMLohal hastr aad ha;^aaard stateveata of li^uMic 
1 bf which the endkt AmaiaB were Bwicd. As a eoaoribatiB 




f "« i|<iiM aie difcrcaa. SchavivT ^n^ d 
Ac dinBaa mta srlUUes uwn-ni— r..jj xa the a 

r: iha* the fits: EaccfrVCntCi^ciea-tshktBEiv^ 

l«:j-3>eif«- 

s. aad this IS Ae ane ax ici i«.-^aa( pnaai^k fiB^kafs «l 



t 



liE VIEWS AND BOOK NOTICES. 339 

cnneifonn signs; thus the first line of the Flood-tablet reads (p. 55): Izdubar 
ana sdsu-ma iszdkra ana Samas-napisti rUki ; (Schrader's method would give : 
1%'du'bar a^na sa-su-ma izzak-ra a-na Samas-napisti ru'U-^i). It is obvious that 
Haupt's method presupposes an accurate determination of syllabic laws and 
grammatical forms, and that it puts the reader one remove farther away from 
the original documents. If a uniform plan were to be followed we should 
have no hesitation in calling Schrader*s the safer, and therefore the preferable, 
but the combination of the two within the same book-covers is instructive. ^ 
should be said that Haupt, in his Glossary, follows Schrader's plan. Except in 
this respect the transcriptions are guided by the same principles. It is, how- 
ever, to be noted in detail that while Haupt represents the Assyrian n (=: Arab. 
(}a) by ^, Schrader retains ^, which seems less appropriate ; that Schrader's / is 
Haupt^s /, — the latter being ably defended, and having the great merit of sim- 
plicity and convenience of pronunciation ; and that Schrader's at is Haupt^s d, 
to which we are less inclined to agree. 

A prominent place is taken by the proper names which are discussed. The 
Assyrian form of borrowed names offers various matters of interest. Within 
certain limits we find scrupulous exactness, e, g, : Apku = Aphek, Ahabbu^z 
(Achab) Ahab (with the last consonant doubled by reason of the case-ending), 
Akubi =. Akzib, Dt^ru = Dor, Afenihimwuzz (Menachem) Menahem, Surru^: 
TsOr, I. /. Tyre, etc. 

Neither can it be attributed to inexactness, or to a defective ear, when we find 
Asdudu for Ashddd, Iskalluna for Ash\calon, ^amirina for Shdmerdn (Samaria), 
etc , for this results from the remarkable, but apparently secure fact, that the 
Assyrians (as distinguished from the Babylonians), had so far departed from 
the original values of their signs as to pronounce the j-signs with sh^ and the 
xA-signs with s} 

I For thU interchange see Schrader, "Ueber die Ausiprache der Zischlaute im Assyritchen." 
in the Monatsbcricht der K. p. Akademie der Wissenschaften su Berlin, March, 1877. We 
have observed, in the book before as many cases which bear on the qaestion : (i) The follow* 
Ing show t in Assyrian for th in Hebrew : p. 8^, Muskit later and more often Muski^ Meshek 
(see KGP. 155 ff-); P* 86, /r^M=Cush; p. 103, Paiattam, />i/tV/A = Philishtim ; p. 145. 
JSo^' s= Sheba : p. 161, {/r^^/riWMiM = Jerushalem ; p. 163, Asdudu =. K%Yid(t^ ; pp. 163,287, 
JLakinf= Lakish ; p. 163, Sauui'murmM =: Sh6mer6n-Merdn (T); p. 165, /s/^a//uma = Ashk al^n , 
p. 189, Minasi=z Menashsheh ; Ba*sa ^ Baesha ; p. 191, Samirtna = Sh6merdn : p. a6o, Attti' 
= Hofthea. (a) The following cases show s in Assyrian for fp in Hebrew : p. 148, Mas', Mas'ai, 
= Mas(h)t(h)a; p. 150, 5fr*Ai/ = It(h)rael : p. 158, 5iVar« = S(h)irydn; StMiVw = S(h)anir. 
(3) The following show t in Hebrew for tk in Assyrian : p. 340, Tiglathptleser = Tuk-Uit-/ai- 
isarra; p. 266, Sbalroaneser (last syU.) = .S<s/m4»»'Kfiir; p. )a7, Thelassar^ Til'As'sur ; p. 
^39, A8ar|}add6n = ^iiNr-o^-fV/f/rna .' p. 376, Asnapper = ^i>»r^»r>a/(?} ; p. 39a, Sarg6n=: 
Sarrukin: p. 411, segJLntm = jo^tf / .* p. 424, (ipsar = d^/iarrat (or du^sarruf), (4) The 
following show tk in Assyrian for f (D or jgr) in Hebrew : p. 138, £>imasi^i, Dtmaska=l>zmmtsck 
iW): PP« •69, 396, 64^'4 = Seve' (AV. So, with 0). (5) The following show sk in Hebrew for s 
in Assyrian : p> 366, Shemtram6th = SoMmuramai ; p. 384, Karkemish =^ Gar/^amis, A'ar/M< 
mis ; p. 266, Shalmaneser (first syll.) = 5a/m<f «M-«ii/V ; p. 319, Rabshakeh =.ra^saj^ (or roA- 
iaJ^ f) . (6) The following have s in both : p. 285, Sanherib = Stm-n^J-frSa ; p. 357, Sais =. Saai: p 
382, SAnh9MsL\=Sim-(M)Saiiif. (7) The following have sk in both : pp. 288, 294, Ushu =6'^jn (?) ; p. 
3a9f Shartser = Sar'Uptr; p. 4*7, Rteh = Xdsi (f ). 

A comparison of BabyUmian texts with the Hebrew leads, as is known, to an entirely different 
result. Here sk corresponds to sk, and s to s. It must be remembered, further, that the fore- 
going applies only to words directly borrotutd. The sibilants in roots common to the two lan- 
guages are represented by the corresponding characters, sk by sk and s by s. 

The above note was written before we had seen a discussion of the subject by Prof. Haupt 
(Nachrichten der Gdtt. Gesellschaft der Wiss., Apr. 25, 1883). in which a view divergent from 



Ajfutsz.^.' 'ZTiir^si zj ^*nzz:c7 











J* 2n inr:'cifiicn pneuaDeactu ^ ' lar -fie rzae j^ aor xc<;(U!ir±Iy riesz. It 




r.v;r IS Sjct- te rhc ermriuigj if x 5rw ttttcct Tantrs . 3^ jC, in rejxri to Asmr, 
^iacr -Vianiiirr r<E3caes Ae ■r-.^e cMiiiiu.K''n: diat tie Tiri as :t sasuis s to be 
tnnn^-rte-i wi tie rwc --cw "Tr * jw»i.'* atx fecu- -r^ die q:iescca v^edKr 
*r..i f-^ra yfiaw' j x --imccic* :f -^aiir Lotz. T. P, 7. 7^ . xad j eiiljj%^ o«t of 
'iut ttc-viat C-ilitTtcVi 3i2er3rKjnc« of .-Ixrjrxs ""rxrered plxia* i^inrfTi'r, 
:> 252 . wici 15 cerrxialT £ir frzm. ietag^ rrr^ti la tie iiscassic* v^ctker 
'I'.e j'vi Ais-ir TXi soaiec fr-na tiie ctr !♦? '. :r tie city L - j m tie joii • Sckzxd.) 
'9t xre .ncljut^i ta gc witi tie 'jizz^r. diim_;i r^ -^rtci's et7iBcI:>gT for Amar^ 
J}': XT, if iz 5i--iili be ja-iaare^ wculi be f-m-igrs^g » this viev. Oa pL 7« 
H-iiC' XB/i 2C5 "Scinieri. Jtxxr^JjK* ^jmjM3t, iPximcB =r Hebcev -T^ JKai> 
w'^ . ;i -itr-.r^ frzxa ^^irl-nm zj^ , " t3 tax^-fer,"^ so tiar Fiwiiam as die 
T'"=.i*r;;>L Two cr riree cimnsscxace* atue tiis doabtfkL (i* From 
rrimm we sh.-'ili expect, xccori;-^ to tie xbxI'Di^ of srat-Ur {bmxtioas in 
AxiyrixE, ^/^i«w iee DcL in 5ci :i*i 'riili. Gen. Gem. ed. p. 169V Hxnpt's 
ref IT, 'hat r£/w. - stcrm," *z= rriia Jirw, - wi=d," t=ririr») ct<x, afford snfi- 
',jeti? ^o^n'i for tie pr^cosed defTitijrc, $ee=is to orcriook the fxct thxt while 
'ht verb ** to thscier" is xm Assyr-ifi ••;«« or irsv, xnd the Bonn " thnndcr " 
•r.-r «xine, it w-.aM be Strang t:> hxr* s irpexr in the name of the Thunder- 
^^yi, (2; The Hebrew form X*t«ricHr. whether the rocxlLcatkm is wrong or not, 
indicates thxt the docb'ed m b cr ^.ciL This is snppocted br the *¥tMuaT of 
*hc LXX. *3i The appearance of the clxmse ^jmmSmm irmmm on omen-tab- 
• tn *a'^ge^t» the existence in AssTrian of a rcrb rewUmm^'^io thnndcr" from 
•wh;',h the name of the god might then be derired (Pinches, see p. 517). These 
cor.^ideTxtionf show thxt the qa est ion most still be regarded as an open one. 

Pp. 179, 180. the non-Shemitic origin of the nante I}iar\Jar in Akkad.=: 
" Af,fM'\on ") is mainta'ned ; the form *Athtar. in which it appears in Himyaritic, 
ir\f\ which gave Hanpt (ZDMG. XXXI Y, 4, i5So)arexsonforholdingtotheShe- 
mitic origin {th passing later, by a Shemitic law, into j), is explained as an irreg- 
ularity, similar to thxt by which the Aramaeans said Athor for .^xmt. Pp.240, 241, 

that held by Schra<!er aiwl, in cotnmcn vitk th« cootineatxl Assyriologists geoendly, by him- 
v^if hitherto, is presented and abiy maintained. An examinatioa of this new, which ve are 
n^A at preftent able 10 accept, is impo$5l>le here and now, and most be resenred. 



REVIEWS AND BOOK NOTICES. 34 ^ 

Anm., give the ctymologyof Tiglathpileser.i.^. TukulH-abal^sannir.^^Tvisi (ob- 
ject of trust) is the son of the Sanra-temple." Sarra is further explained as an 
Akkad. word = Assyr. tdbu^ asdru, " good." (We do not know on what grounds 
Haupt, Akkadische Sprache, 1883, p. xxxv, says : " Der Grundbegriff der akkad. 
Wurzelxtfr ist ' zusammenbringen.* Auch fsara heisst nicht * Haus der Gnade,' 
sondem * Haus der Versammlung.* ") We believe with Schrader that in the 
form tuiulti the suffix need not be found {iukuM is therefore unnecessary) ; but 
the suggestion (p. 241. Anm.) that tuMlat, st. constr. may be the better reading, 
and the sense be " Servant of the Son of the Sarra temple," is worthy of very care- 
ful consideration. (•• Diener des Sarra-Tempels," p. 241, last line, is of course 
a printer's error for ** Diener des Sohnes des Sarra-Tempels.") Salmdnu-uisir 
(Shalmaneser), p. 266. is explained as •* Shalman, be gracious!" St. Guyard 
(Journ. As. VII, 15, i88o, p. 49 ff.) is followed in the derivation of nJltV (Imper. 
Pa.) from masSru — a Pe Mem verb, like a Pe Nun, Without raising any objec- 
tion to the form, as such, we confess that the evidence for the meaning "be 
gracious," from the verb fnasarti, is not quite convincing, umasslr (Impf. Pa.) 
is used repeatedly in the sense of " to leave," •* abandon " ; only in connection 
with ana naptsH^ as far as we have observed, does it mean '* to leave alive," " to 
spare." The noun uYsuru doubtless has a meaning kindred with this, but the 
same question arises, as in regard to ussir — does it really come from masdru ? 
So in regard to useru (usseru), cited by Guyard, from IR XXII, 1. 113. If 
mussir is to be read on the Flood-tablet, I 21, that would look like Imper. Pa. 
of masdru, and so far render this root for «JJi> unlikely. On p. 284 we have a 
discussion of "Adrammelek " and "Anammelek." The former is explained as 
Adar-malik, the latter as Anu-malik ('*Adar or Anu is prince "). The difficulty 
here is not so much in the identification of Adar with Nin^b, which is very 
likely, although not yet absolutely demonstrated (Schrad. in Berichte Ober die 
Verhandlungen der k. s. Gesellschaft der Wissenschaften zu Leipzig, Phil. Hist. 
CI., 1880, pp. 19 ff.), but in the fact that these deities are called ** gods of Sep- 
harvaim ** (i. e. Sippara\ in I Kings, XVII. 31, while we know that the two 
deities especially worshipped at Sippara were Somas and Anunit (see pp. 279, 
280)— a point which Prof. Tiele has already urged (Theologisch Tijdschrift, 
March, 1883). If, then, i Kings, XVII, 31 is trustworthy— and there is cer- 
tainly no sufficient reason for questioning it-r-it seems doubtful whether the 
explanation above given of " Adrammelek " and **Anammelek " can be correct. 
On the other hand we are not aware of any grounds for believing Adru^ or 
Adru-malkUy to be a name oiSamas (Lenormant, Origines de THistoire, I, p. 
524, note. Eng. Trans, p. 514 f. note. Hommel, Semit. Volker u. Sprachen, 
I 2, p. 245), except the precarious ones that Soma's was worshipped at Sippara, 
and that Adra-kasis {Hasisadra\ the hero of the Flood, was called also Samas- 
napistim. There is perhaps more reason for following the scholars just named 
in regarding Anammelek as corrupted from Anunit-malkat We must be allowed 
to doubt the interpretation of // (p. 284) as a chief god. It is probably generic, 
and not a proper name at all. 

But most of the author's treatment of proper names seems judicious and 
right. E,g. Phai = /*«/« (p. 2yi),Safiherfb'=iSin-aht'Mb (or irba) (p. 285), 
—any other interpretation for M'^than ** multiply " can hardly be thought of in 
earnest (cf. G. Evans, Essay on Assyriology, 1883, p. 62)— Nisr6k^'A(T(Ta/)o;^; 



342 AMERICAN JOURNAL OF PHILOLOGY. 

^Aiur (p. 329) — as at least more plausible than anything yet suggested; 
Hal^v/s suggestion of " Nes6k " as the proper reading is ingenious, but not 
satisfactory (Doc. rel. de TAssyrie, etc., 1882, Commentary, p. 36.) Asnapper 
= Aiurbampal (p. 376), Sanballat =: Sin-(u)6aUif (p. 382), Sargon =.SarruBn 
(p. 392), Nebo = Nab4'^zNabium (?) (p. 4I2>— and many besides. We dismiss 
the subject of proper names, after calling attention to the testimony borne on 
p. 167 to the Shemitic character of the Philistine names, and the natural infer* 
ence as to the race to which the Philistines belonged. 

No Assyrian glossaries have yet been published which coiQpare in value with 
those appended to this book. Besides their obvious use for readers of the text, 
they afford material for testing Assyrian lexicography aad grammar at some of 
the most important points. 

(a) They give us new and abundant proof of the Akkadian influence manifest 
in the Assyrian vocabulary. We select a few cases almost at random : P. 493 
(Gloss. I, by Haupt) gives the following examples— omitting proper names : 
abkallu^ ** prince,'* ikallut *' temple,** ttda^ " daylight," ada^ur, as sacrificial vessel ; 
p. 544 (Gloss. II, .Schrad.),^'i«j^, " garden/' ^>ani, •* darkness "; p. 545, dupsarm, 
"tablet-writer"; p. 597, surmMu, a tree, etc, etc. This is, however, a fami- 
liar matter. The explanation of sarru^ ** king,'* from the Sumerian l/r (Akkad. 
li/r), which Haupt for the first time clearly proposed in his Sintfluthbericht, 
l88r, p. 25 f. (cf. Lot«, Tig. Pit, 1880, p. 99), is maintained here by himself (p. 
520), and not rejected by Schrader (p. 592, cf. p, S3). Certainly, no sufficient 
objection has yet been brought against it. 

(6) They bear witness to the right of the Assyrian vocabulary to claim from 
the Shemitic lexicographers equal attention with the other Shemitic languages. 
We will not pause to enumerate cases, where even th$ latest Hebrew lexicon 
(Gesenius, 9th ed., by Miihlau and Volck, 1882-83) has ignored Assyrian words 
entirely parallel with the Hebrew. In not a few of these the neglect has been 
fatal to a correct etymology. But this is a subject by itself. (Cf. 7 articles by 
Friedrich Delitzsch, on " The Importance of Assyriol(^ to Hebrew Lexico- 
graphy," in the Aihemtum^ May 5, 12, 26, June 9, July 21, 28, and August 25, 
1883), 

(c) But most interesting of all is the material offered us for constructing the 
Assyrian paradigms, and particularly those of verbs. The various species of 
the strong verb are fairly well represented.* The absence of an Aphel, whether 
in strong or in weak verbs, is clearly manifest. The most commonly used 
species are the Kal, Ifteal, and Piel. The Shafel is not infrequent, and the 
Nifal occurs repeatedly. There are numerous instances of Iftaal. Ishtafal, 
Ittafal and Iftanaal occur rarely, and we have observed no instance of Iftaneal, 
Ishtanafal, or Ittanafal. Of tenses, the imperfect and present are abundantly 
illustrated ; the permansive (perfect) occasionally. The precative occurs often. 

Of especial importance are the indications of the influence and treatment of 
weak letters in verbal inflection. It is well known that h is the only guttural 
which always appears. The others are reduced to the simple breathing, = k. 

> It is to be noticed with satis&ction that while Haupt has given up the nomenclature proposed 
by him for the Assyrian verbal species ($FG. 64, Anm. z), neither Schrader nor Haupt adopts 
the system of numbers employed by Lots (Tig. Pil. I, see especially p. vi). The names here 
tued are identical with those in use in Hebrew, as fiu* as this is possible. 



REVIEWS AND BOOK NOTICES, 345 

(The nomenclature first proposed by Haupt (SFG. 48, 3), and now becoming 
general is : Mi = ^AUph; Mt == Z^^/ Ms = weaker Chetk; K4 = ^Ayin ; Ma = 
Gain.) The following examples of impf. 3d pers. will illustrate the points 
referred to, and at the same time the disappearance of the prefonnative : Pe 
Guttural, K|, iUik (fr. aidku^ ist pers. o/ft'i, alik)^ but also imuTy imur (fr. amdrH\ 
Similarly Hi and k,. k* gives Hi (cf. Heb. nSp), (mid(ct Heb. ^op). Mt givea 
^mi (cf. Heb. an;). *Ayin Guttural, M|, t, •> /. /. wma*%r (fr. wuCAni)^ but alao 
fi/r (from sinsf). M4 gives ibil (<d, Heb. Spa). Lamed Guttural, Mi, tt ••'•/• 
ih\t (fr. fuit4, but cf./M = Heb. nn», and N4 and •, hir4,pih4, tiM). 

The final vowel is lengthened in all Lamed Aleph stems, thus : *}ama^u, 
** to hear,'* is sam^, *ifz'i#, ** to go out," is^ dpi, etc. Lamed Waw generally 
the same. Lamed Yod either the same or as follows : aimt (root ^oS), aM 
(root *2p). Of Pe Yod we have tdu, etc; of Pe Waw usii, ittuHk, etc. 'Ayin 
Waw and Yod give uktn (fr. k4nu = ^Jiawdnu) etc. 

It would be instructive to go very much more into detail, and to draw some 
of the obvious conclusions from the grammatical facts put before us in this 
valuable book. But enough has perhaps been said to indicate that it is indis- 
pensable not merely to the student of Biblical history, not merely even to the 
student of Assyrian, but also to those who care for general Shemitic grammar, 
and that it offers interesting and suggestive phenomena to all who have a taste 

for the study of language. 

Francis Brown. 



Wilhelm Gesenius* Hebrftisches und Chaldaisches Handw5rterbuch fiber das 
Alte Testament. Neunte Auflage, neu bearbeitet von F. MOhlau und 
W. VoLCK, ord. Professoren der Theologie an der Universit&t Dorpat. 
Leipzig, F. C. W. Vogel, 1883. 

The appearance of each successive edition of the time-honored Hebrew 
hand -lexicon, whose foundation was laid more than sixty years ago by the 
industry and common sense of Gesenius, should be not only of great moment 
for Hebrew and Old Testament study, but also of the very first importance for 
Semitic linguistic science generally. The Hebrew language, mainly on account 
of the literature which it embodies, has been, and it is safe to say always will 
be, studied in the Western World more than all the other Semitic dialects put 
together. But for the proper understanding of Hebrew, a knowledge of the 
related tongues is indispensable, and in every comprehensive Hebrew dictionary 
all the new facts that can be gained from any of them to illustrate Hebrew 
phonology, etymology, or sematology must be accurately and judiciously pre- 
sented. Hence it happens that the deservedly most popular of the Hebrew 
lexicons furnishes to its editors an unequalled opportunity of giving to the 
world a succinct record of comparative Semitic study. This fact they seem to 
have recognized, as they claim in the preface to have subjected the etymological 
and phonological matter of the previous editions to a thorough revision, and to 
have paid special attention to the latest results of Assyriological research. The 
importance of the subject demands that the representative work before us be 
subjected to a close and faithful scrutiny. 

The first thing to be noted in this latest edition is the improvement which 
has been effected in the increased number of citations of illustrative passages. 



344 AMERICAN JOURNAL OF PHILOLOGY. 

A great desideratum has also been supplied by the affixing of a cross (f ) to the 
articles in which all the proof passages are cited. The student is thus enabled 
to see at a glance how often the rarer words occur, and in what period of the 
history of the language. In the previous editions it was often impossible, 
without the aid of a concordance, to find out even whether words with but one 
citation were really ai^a^, Aey. The editors refer to their diligent use of concor- 
dances in this connection, and this is no doubt indispensable. But in Hebrew 
lexicography, at least, one has still to depend much on private collections, and 
we have to point out here a most singular instance of human fallibility in this 
kind of work. It might seem almost incredible, but it is nevertheless a fact, 
that a whole article is wanting in this and the earlier editions of the present 
work as well as in all other lexicons we have examined. No notice whatever 
is taken of the Chaldce word atr. '* elder," which occurs in no less than five 
places (Ezra 5, 5. 9; 6, 7. 8. 14), once in the emphatic and four times in the 
construct plural. An explanation of this phenomenon is perhaps to be found 
in the fact than in the Hebrew and Chaldee concordances of Baer and Fiirst 
the word is cited under air, while the related words in Hebrew are forms of 
l>v. However this may be, the fact itself is most instructive and suggestive 
for the makers and users of dictionaries generally. It would not, of course, be 
so worthy of remark if the Hebrew and Biblical Chaldee vocabulary were not 
so limited and so easily brought under control. 

Very little progress is to be recorded in the explanation of doubtful words 
and phrases. This is to be expected as long as the editors confine themselves 
to an elucidation of the traditional Massoretic text without considering any 
proposed emendations. One can understand why such a position should be 
taken and held in order to avoid confusion and to keep out a supposed new 
element of uncertainty. None the less is the position unscientific, and in the 
present state of these studies indefensible. If there are cases in which either 
the traditional forms of words or their traditional explanation are clearly or 
probably wrong, the facts ought to be mentioned along with the best attempts 
that have been made at emendation. It seems hardly proper that a reading 
which makes no sense should be retained as part of the Hebrew lexical material. 
For example, it is as plain as day that the text which lay before the LXX in 
I Sam. 20, 19 was the original, as Wellhausen has shown, but the editors still 
give Smn pK instead of rSn aiiw (cf. LXX with v. 41). and get rid of the addi- 
tional grammatical difficulty of the Massoretic text by omitting the article from 
before 73M. The effect of this is not simply to give a locality to Palestine 
which never existed, but also a new word to the vocabulary, for the noun *?tk, 
*' Weggang," does not elsewhere occur. It is hard to see, again, why the transb- 
tion of the LXX, aiudaei rovf b^^aXuovc tjftcjv for w>p S^xn in 2 Sam. 20, 6 should 
not be adopted. It is certain that the original form was Sxn in any case, as no 
vowel letters were written between two consonants in the same syllable (cf. 
Bleek- Wellhausen, Einleitung in das Alte Testament, p. 633 ff.), and we have 
to choose between the above natural and expressive image, and the following : 
'*(damit er nicht) unser Auge wegreisse, fUr: damit er sich nicht unserem Auge 
entziehe.*' These are only random specimens of the results to which an ignoring 
of the textual criticism of the Old Testament leads when the lexical treatment 
of the material is concerned. 



REVIEWS AND BOOK NOTICES, 345 

The article t\i^ is carried over from the preceding edition without alteration, 
but here again an important province of Semitic research is ignored ; the whole 
article should in fact be rewritten. A reference to Levy's little " PhOnixisches 
WOrterbuch " alone would show that rn «^m was the special name of a Phoe- 
nician divinity, a fact which is at least striking enough to be cited along with 
nrp ^fivn, Ps. 76, 4. Resep was, in fact, the Phoenician (and probably the 
CanaanilUh) Vulcan or Fire-God, 1. e, " fire " or " Hame ** personified. That it 
should have been mentioned in connection with the very obscure and difficult 
passage, Job 5, 7 ^, is obvious. It is also worth mentioning that in another 
interesting passage where the word occurs (Hab. 3. 5) -^ai , " pestilence," is 
employed as its parallel — the term which in the far East the Assyrians seem to 
have personified as the Pest-God Dibbara. 

Greater fullness and exactness are still to be desired in many articles. The 
treatment of ^]y and e^Mi may be noted as specially defective. For the former 
word the large '* Thesaurus *' of Gesenius may still be consulted with great 
advantage, and it would have been well to cite Neh. 3, 34. 5, 10. In general 
it may be said that the later books of the Canon are not sufficiently represented, 
either in the dictionaries or grammars. For the latter word, which has a much 
wider figurative use than would appear from our lexicon, such additional shades 
of meaning as are found in I Sam. 28, 2 ; i Chr. 16, 7 ; 24, 31 : 2 Chr. 13, 12 
should have been given. Of other words, defectively treated, we note the fol- 
lowing : No explanation of ma is given which suits the familiar phrase in i Sam. 
17, 12. Sjynxn, Gen. 33, i, means **he distributed among," but neither under 
the preposition nor the verb are these special meanings indicated. In Esth. 9, 
25 o]7 can only have an instrumental meaning, which is absent from the other- 
wise excellent article upon that preposition. The primary notion of p'Mn 
comparison with'* comes out in at least one passage, 2 S. 18, 3, where it is 
equivalent to nj^ Is. 40, 17, but this sense has been quite overiooked. 

Some errors in regard to the use of words are important enough to be cited. 
nf7j7 is said to be used in Deut. 21, 12 of trimming the beard, where the subject 
of the verse is a woman. It is really employed there of trimming the nails, as 
also in the original text of 2 S. 19, 25, according to the showing of the LXX* 
aiK, in Gen. 25, 30, is not used "of the reddish-brown color of a man," but 
of Jacob's famous mess of pottage. In 2 Sam. 18, 22 n^tr^ cannot mean *' mes- 
sage " as in V. 20, but only " a messenger's reward ' as in 4, 10. The statement 
that -on in the sense of ** love of man to God " is " very rare " is hardly true. 
Besides the passages quoted it occurs frequently in the later books, e, g, 2 Chr. 
6, 42 ; 32, 32 al., and is moreover the prevailing sense in the secondary deriva- 
tives. 

The chief claim the editors make for the present edition is the advance made 
in the etymological portion. That something has been done to make Semitic 
etymology more respectable cannot be denied ; but it is yet far from the level 
which might be reached upon thoroughly scientific methods. The advantages 
which the editors possess are the inherited labors of their predecessors, espe- 
cially of the late Franz Dietrich, who was certainly the most thorough Semitic 
etymologist of the last generation, and who edited the fifth, sixth, and seventh 
editions of the present work. Of scholars still living, they have drawn chiefly 
from the veteran Fleischer, who, from his stores of Arabic learning, has brought 



._...!>- -OURHAL OF PHILOLOGY. 

rm ..^ni 3p«B fclM prinary ideas of some important Semitic 

^ •=« ajivc tbe adrantage of the lecent researches into 

.*. .A«« uBtil lately been but little known, snch as local 

.\ Lou: aad Ethiopic that have been worked np by D. H. 

« , . -aeuxMs, Tnunpp and others, and above all the As^rrio- 

J :\Mft<3«»itic Akkadian, to which the Semitic vocabolaiy 

.. '.c\«. The main deficiencies to be remarked in this as 

. aa ^e» ^rst, the lack of a thorough and rational theory of 

c >caiittc roots, and of the characteristic principles of 

^uou. :1m want of consistent etymological and philological 

^ vt.iw ot direct controlling acquaintance with some of the 

-. 1....C ^laiccts, notably of the Assyrian. 

.c aru ;>oint it Bust be said that although the old theory of 

% ^ , as and not mere abstractions must be given up, the great 

... .1 . t« Semitic as well as in other linguistic sjrstems, certain 

*> .>! sounds stand for general ideas, and (broadly speaking) 

. . ^M^ 1 coaibinatioas by added sounds (" determinatives'*) gives 

^«,;nia\:;i]kGC to the respective forms. It is not necessary for 

^..> .o aave any special theory even as to whether the longer or 

^ Av.t: the earlier. The only postulate required is that there was 

^ w :\< Jevelopment of linguistic forms as well as of the corres- 

. . ;>i luiiive civilization. This being granted, it may be expected 

. .a ^o extraordinarily rich in significant sounds and so compara- 

.^ aouetk decay as the Semitic, the system of root -reduction 

.wiLU. M\ that is needed is a patient collection of facts, and 

^ . .^t.:» ;\v trained philologists. Such a system for the Proto-Semitic 

. .X H ;h« later history of the main dialects, was for the first time 

^ ..lU cou>isteatly and proved in detail in my book **Aryo-Semitic 

.. .V .i^o, and it is gratifying to note that the successive editions 

V .. v.c, through the accumulating force of individual objective 

> ^^ t-uoAvhiag the positions there taken. A few of the most 

.V .ca may be here recapitulated. First: Semitic "roots "(or 

> ^^ uoolic significant and independent combinations of sounds) 

.^ . v.ii. :>uri.>odcs viewed as primary or secondary, and the simplest 

'.. ^l^t v>f a consonant and vowel or of a vowel and consonant. 

. . ^.u \vAs not, as a system, characteristic of the earliest accessible 

'ui uetthec was any other special type of structure. 

^....>4e time before the separation of the Semitic tribes, the tri- 

\, ^ .ame as a simple matter of convenience to predominate and 

. V , .cv' vk« the ordinary inflective basis, but the tendency to conform 

,o.^ :o this standard was not thoroughly carried out either in 

^ ^ .u:^ auU a biconsonantal or even a monoconsonantal basis is 

^ .V a voauf of the so-called "weak" forms. Second, there is 

. ,, t'^Atf that the stronger gutturals and the ** emphatic** ezplo- 

.... >i ih9 original stock of Semitic sounds, though they were 

^ KiOic the breaking up of the family. Third, in the develop- 

^ vx, Ivih vowels and consonants entered as the secondary factors. 

. .,^i*jH of « and f, respectively, came i and \ to stand in the 



REVIEWS AND BOOK NOTICES, 2f^^ 

place of the first, second, or third radical when they were not primary, and k, 
sometimes changed to p, performed, though to a less extent, an analogous 
function for a. This is self-evidently always the case when the first and third 
radicals are *• strong ** letters. In many cases, however, ^ and ^ at least were 
independent radicals from the beginning, as true consonants. ** Determinative ^' 
letters, whether to form biliterals, triliterals, or multiliterals, were placed at 
the end, and any consonant whatever could fulfil that function. But z&prae- 
determinatives only a very few consonants were employed, and these were such 
as were also employed as formative inflective elements, above all n, j, /, the 
new roots thus formed being properly denominatives. Of the infixing of a letter 
between the first and third radical then seems to be no sure proof, n and j; 
alone can play this rOle, and here it is natural to suppose that they are modifi- 
cations of M. 

Etymologists have been a good deal ridiculed and ol^en deservedly. But all 
Semitic scholars must perforce become et3rmologists, even those who laugh at 
etymologists in general ; and in every Semitic lexicon, the etymological material 
must be pretty extensive. It is extremely desirable then that it should be 
accurate, and to secure this end it is necessary that such preliminary funda- 
mental work as that just indicated should be done once for all. The advan- 
tage of this systematic treatment is exemplified in the right theory of the root 
n">3, " to cut." The editors say, without mentioning any other view, that the 
simpler root is perhaps no, they do not give any other instance of an infixed r, 
and a great many instances would be necessary, since infixing is either rare or 
unknown. But, following the general principle above laid down, we soon find 
that a great number of secondary roots, with the idea of " cutting,'* •< dividing," 
contain ir as the main elements, and that that is in fact the fundamental notion 
of the combination from which the later senses of " digging," " hollowing out," 
etc., are plainly derived. A still more serious error is it when they combine 
(after Fleischer) a whole set of roots (at 7\yr() having strong letters at the 
beginning and end and ^ in the middle, with a common simpler root m whose 
alleged meaning, ** to twist," " turn," cannot be proved. One of the roots cited, 
nn, does not exist at all in Hebrew, and in the other dialects not in the sense 
referred to. The supposed Hebrew derivative imp occurs once, Ps. 107, 30, 
and as a comparison with the corresponding Tavg. and Assyr. word (see Lotz 
on Tigl. I, 52) shows, does not properly mean "haven" at all, but **town, 
village," and is of uncertain origin. 

The root Sar, whose true origin we shall mention later, is derived by the 
editors from the simpler St. They are greatly puzzled, in connection with the 
much-discussed Hebrew word for God, '^k by the discovery that in Assyrian 
ilu (and Sabaean i7) the vowel is short, and say that the root must then be some 
other than S^k, ** to be strong." The fact is that, correctly speaking, there is 
no root '1^/. The three consonants may be written without vowels to indicate 
in a symbolical (and in this case only approximate) way the form which lies at 
the basis of roost of the noun and verb stems usually associated with it. If the 
root is so written it must be understood to represent simply the vowel expansion 
of a simpler Sm, to which alone the Semitic il can be referred. The false theory 
that lies at the bottom of such uncertain etymologizing is the supposition that 
only fuch a form can be assumed for the " root " as lies, or is supposed to lie, 



348 AMERICAN JOURNAL OF PHILOLOGY. 

at the basis of verb stems. The Semitic lexicon really starts with nouns of 
action^ and all of its rerbs are, strictly speaking, a sort of denominatives. Most 
of the so^alled derivatives from Ayin Van verbs it is impossible, as in the case 
before us, to derive from them directly, and yet such is the condition of current 
Semitic etymology and its terminology that it is necessary to correct the editors 
here, and to say that after all they are wrong in supposing that '^m (tV) is not 
derived from Sw ! 

The main etymological defect of the whole work, however, is that inherited 
from Dietrich, namely, the tendency to overdo the et3rmologizing. It is amazing 
what combinations are sometimes made in order to bring an immense stock of 
roots with similar letters under the same primary root, when the meanings 
alleged to be related are only brought together by a stretch of fancy. Almost 
every ** Ayin Doubled " root gives occasion for an exercise of this kind, though 
occasionally one is more temperately treated, as inj, nnj. It is hard, again 
to understand what the editors mean when, after citing proofs at jn^ that the 
word is not the same as the Arabic wada^a^ ** to place," they say : **• nevertheless 
the Arabic wada^a can afford us the ground-meaning of jn\" But even if the 
comparison were admissible after they have proved that it is not, it would be 
hard to see how " placing ** can have anything to do with knowing. They say, 
after Schultens, that it means " placing in the mind "; but in this case, not to 
mention other objections, the essential idea has been interpolated. 

It will now be in place to note a few derivations of actual words, oi 
" blood," is derived from Tvory. This means *' to be like,*' and the problem is 
to bring these notions together. •* To be like " is " to be level,** i. /., of course, 
*• level with the ground." A thing is made level with the ground by ** pushing" 
or " pressing," from which we get the notion of ** pressing together," and so 
" making close." The notion of being ** dark" is connected with being " close " 
and " thick," and what is •* blood " but something "dark-red" and ''thick"? 
This is only a small part, however, of what the root on is made to yield. 
Nobody can disprove all this, but some of the transitions of meaning are, to 
say the least, rather violent. 

An error which, as far as we know, is found in all the dictionaries, is to be 
pointed out in the derivation of n^DB^, ** eye-lid," found once and in the plural 
(Ps. 77, 5), as the editors have forgotten to note. This is connected with nor, 
" to guard," as though it meant ** the guard ** of the eye. There are two fatal 
objections to this. First, the form is that of the part. pass, so that it must have 
meant ** the guarded thing," which is unsuitable. Second, the word comes out 
in Syriac with an original th sound : thimrd, and has the same consonants in its 
Targumic form. The root is, in fact, unknown, and as the H is written defec- 
tively, it is not certain that the Massoretic form is correct. The current 
etymology of the quadrilitcral ojnn, ** to interpret," " translate," is certainly 
wrong ; and the way in which it arose is an interesting study. The simpler 
root DJn in Hebr. Aram, and Arabic means " to stone,** as a capital punishment. 
Hence it has been supposed that it originally meant " to throw,*' then ** to 
throw over," and then ** to translate." But it never means " to tlirow " simply, 
but *• to throw stones^^ or " to stone.'* In Ezek. 23, 47; i K. la, 18 the word 
for ** stone *' is a '* cc^nate accusative,'* and the meaning ** jaculari," attributed 
in the lexicon to the Chaldee, is a mistake. It also never means *' to throw 



REVIEIVS AND BOOK NOTICES, 349 

over/' much \t%s " to translate.** It has been attempted to deduce the famous 
longer word from an Indo-European root which would bring it remotely into 
connection with our talk, but here the many necessary historical links are left 
to the imagination. If the word is Semitic, and if a derivation is insisted on, 
it is best to connect it with Assyr. ragdmu^ " to cry out," ** shout." Comparing 
this (Haupt, Sintfluth-Glossar in KAT^, p. 517) with the same root in Ethiopic 
and Arabic, where it means ** to curse/' ' it is plain that the idea of " speaking" 
must have intervened ; vgl. Hebr. aap, " to curse," with Assyr. qebiiy ** to speak," 
A. S. andswarian^ " to answer," with stu^rian, ** to swear." S«p, ** the Flood,'* 
is still derived from Sa% but the Eth. md^bal^ ** billow," " billows," from a root 
S^?, seems to throw doubt upon this view. The root Sss^, already mentioned, 
can only be a secondary from the kindred primitive Sa ; vgl. rh^vf^ ** stream," 
" flood." The origin of nitrp, ** pan," 2 Sam. 13, 9, is stated to be obscure, but 
Wellhausen has already acutely and satisfactorily explained it as = " dough- 
place," standing for mas*eret. The old derivation of n^,», " face," from the 
similar-sounding root meaning " to turn," reverses the true order, for the latter 
is a denominative from the former, as the Arab, derived V. conj. of wagaha^ 
*• to turn towards," is also a denominative from wagh^ " face." The problem 
is solved when we 6nd that the Assyr. pdnu, "face," is the strict plural form 
of/j^ (Arab./i^, = n^), "mouth," to which it bears the same logical relation as 
Lat. ora does to cs. 

These examples must suffice to suggest how much the etymological portion 
of the work requires to bring it up to the proper level of method and accuracy. 
A few words must still be said to show how indispensable it is for Semitic 
scholars to have a direct acquaintance with Assyrian or rather with the science 
of Assyriology in its widest sense. Many words, besides those already referred 
to, receive light, both as to their origin and meaning, from the Assyrian as well 
as from its local predecessor, the Akkadian. To distinguish between the last 
two sources and to control the material, generally needs a special training and 
preparation, and it is surprising how few, comparatively, have devoted them- 
selves to Assyriological studies. The editors have, for this last edition, relied 
almost entirely upon Schrader's KAT^ — a work which, with all its excel- 
lences, labors under many grammatical and lexical defects ; but this they have 
not used as fully and as intelligently as they might have done. The following 
are a few errors and deficiencies which we have observed. The now world- 
wide n^, " reed," is derived (without mention of any of its Semitic equivalents) 
from njp in its hypothetical sense of " standing upright." But, being the same 
word as the Assyr. fomi, it comes, as is now notorious, from the Akkad. gin = 
*• the bending thing." Kp? is derived from k03, *' to cover "; but the word is 
the Akkad. guza, the equivalent of the Assyr. Auss^t in the bilingual syllabaries, 
n^p, "a mina," is an old pre-Semitic Babylonian weight, Akkad. and Assyr. 
ptana (see, /. g,, Delitzsch, Assyr. Lesestilcke, 2 ed., 77, 36), and has only an 
accidental association of sound with the Semitic njD, " to reckon," " assign," 
from which the lexicon derives it. 

1 This meaning is well established for the Arabic along with the more common sense of 
"stoning," wtt,e.£., the Koran, Sura III, 31: "Satan the accursed." This is usually, but 
wrongly, regarded as being equivalent to " Satan who is worthy of stoning." 



3SO AMERICAN JOURNAL OP PHILOLOGY. 

The AuyriAn innit alio be often called in to reclif; or illnttrate wordi that 
rarely occur or are donblfiil in meaning and origin. For example, Sr* 
" banner," ii the Aisjr, Jiglti, eridently of a similar meaniriK, which ii properly 
"something conipicaoni," from Ass]rr. dagStu, "to see," "look at" (vgl- 
DelitischinLolz, TigL, p. 131 f.). The fact that Asifi.Aiwrf means "in front 
of" if. g. Sennacb* II, 77), calls for a new treatment of Hebr. thtt and d^'M. 
The Atijrr. equivalent of \r^, " thnmb," is uhSnu and this settle* the Proto- 
Semitic fonn of the root which the Arab, iihdm had put in donbt. The correct- 
ness of the traditional form and rendering of the an-, ijy. njpn is placed beyond 
donbt hy Asiyr. aUktu, " oreithrow," its exact equivalent. Aasjr. *ijff, " sharp 
point," show* that the Hebr, and Aram. >1PI, " to impale," " bang alofl," are 
denominatives. What the Assyrian has contributed to the nnderstandiiig of 
Hebrew proper names has become better known, and is more folly indicated 
in the lexicon. 

We have noted a few omissions and mistakes from oversight in sddiUon to 
those mentioned in the errata. At 3i>n, nr. 3, S shoald be omitted from np-ts^ 
in the important syntactical and theological passage cited from Gen. 15,6. The 
citation on p. 45 t, line 5 from bottom, should read Ri g. 37. The plural of tkj, 
" secret counsel," and of i;i, " omainent " (Jer. 3, ig), are omitted, though they 
arc both exceptional forms. The peculiar form* of the Hi61 inf. construct 
found in 3 Sam. 14, II are also wanting,ju well as the Semitic equivalents for 
the numeral " four," which are of great phonoli^cat importaDBe. All of these 
oversights are transferred from the preceding edition, as is the omission alreadj 
mentioned in connection with nici'. 

A word mast be said in conclusion inptaiseof the thorough manner in which 
the long introductory article, "Von den Quellen," has been worked over for 
the present edition, where a great deal of new bibliographical material has 
been added. The attention that has been bestowed upon the geographical and 
archaeological departments of the lexicon is evident not only here, bat through- 
out the work. 

It has been necessary, in the interest* of Hebrew and Semitic studies, to 
dwell upon the shortcomings rather than upon the merits of the work just 
reviewed. But, taken as a whole, the lexicon as it nowstaadsis,inouropinion, 
by far the best Hebrew dictionary in existence, and it ihould either In form or 
in substance be speedily done into English. If the leading defects to which 
we have called attention were to be remedied and, in addition, doe deference 
paid to all schools of exegesis and textoal criticism which are intelligently and 
conscientiously seeking to arrive at the truth, the demands of Hebrew students 
would be fully met by the next edition. 

J, F. HcCinDY. 



iden nir den Elemenlarcursus de* Sanskrit mit UebongsstDcken nnd iwci 
[^lotsaren, von GtORC BOHI.EK. Wien, Verlag von Karl Kon^en, 1BS3. 
liter's bocdc may be said to be almost the fint practical introduction into 
cal Sanskrit, published outside of India. It i* written in the style of our 
luctory Latin and Greek book*, pre*enting alternately gnmmalica) leuons 
iractical exercises in which the student is from the very start brought face 
e with the living language. When BOhter, a few years ago, came to the 



REVIEWS AND BOOK NOTICES, 35 1 

University of Viennathe commenced with a beginner's class of over fifty students, 
and so effective and attractive was the easy practical way of his instruction 
that over thirty of these were still pursuing Sanskrit under him two semesters 
later. This will be significant to any one who knows how beginners* classes 
in Sanskrit thin out after the first semester in German universities. 

The very great value of the book lies in the graded practical exercises. 
These are not of the style of * Ollendorff's method,' but consist either of extracts 
from the literature, or close imitations of passages actually to be found. What- 
ever one may say of the grammatical part of the book, this collection of 
good Sanskrit sentences, undoubtedly the result of a good deal of labor, can 
always be utilized in recasting the book upon some other plan ; or they can 
be employed in practical exercises in prose composition, after the student has 
had some acquaintance with the language and can be made to judge the gram- 
matical part of the book for himself. 

Certainly the grammatical part will not commend itself to Western scholars 
generally. To the scholar who has been brought up in India, who believes 
that the grammatical rules and the forms of Hindu grammarians clear down to 
the Bhattikavya may be expected any day to receive practical illustration, or to 
turn up somewhere in the literature, the artificialities of the Hindu system seem 
an evil which can be tolerated. But he who believes only in phonetics, prac- 
tically capable of illustration in the MSS. and in * quotable* forms, will gladly 
and to his profit leave the Hindu grammar to the Hindus and to the special 
students of native grammar. He will prefer to take from the start the scientific 
and yet practical European view of Indian language, which certainly becomes 
unavoidable as soon as one leaves the domain of the classical language and 
turns to the Vedic Sahhitas, Brahmapas or sutras. Another general con- 
sideration militates against the introduction into the Hindu system which the 
book leads to. namely, the actually acquired position of Whitney's Sanskrit 
grammar. There can be no reasonable doubt that an overwhelming majority 
of European Sanskrit students, not to speak of the Americans, now actually use 
this book for daily reference to a large extent, to the exclusion of other gram- 
mars. The second edition of the book, with a list of all accessible verbal 
forms, may be expected at no remote future, and this appendix will enhance 
the value of Whitney's grammar-— one may fairly say it without being ac- 
cused of clannishness— out of reach of comparison with any other. Now if 
Btthler's exercises were combined with an introduction to Whitney, his book 
would be an unmodified blessing indeed. Those who teach Sanskrit according 
to Whitney's grammar know how difficult and unsatisfactory it is to make the 
necessary selections for beginners ; Buhler with his practical knowledge of how 
to teach Sanskrit would have guided him better than almost any other scholar. 

To be sure, the fault can even now be removed, certainly for English-speaking 
students. It would not be too difficult a task to employ Buhler's valuable 
selections and his equally valuable method of grading the lessons, but to trans- 
fer these into the framework of Whitney's method. Should there ever appear 
an English version of the book — and there is some reason to hope that an 
American scholar will undertake the work — it is hoped that the wish expressed 
above will not have been uttered in vain, and that Biihler's well-known 
liberality may permit this free rendering of his valuable book. 

Mauricb Bloomfield. 



352 AMERICAN JOURNAL OF PHILOLOGY. 

API2TOTEAH2: DBPI rrXHZ. Aristotle's Psychology in Greek and English, 
with Introduction and Notes. By Edwin Wallace, M. A., Fellow and 
Tutor of Worcester College, Oxford. Cambridge University Press, 1882. 

Hegel brought modem philosophy to the consciousness of the organic 
oneness of its largest results with the corresponding results of earlier, and 
especially of Greek, inquiry; and one of the most characteristic features of the 
philosophic morement in Germany since HegePs death has been the attempt 
to comprehend and teach philosophy in and through its history. Accordingly, 
and as a necessary incident of this attempt, the world— or at least the world of 
scholars — knows how assiduously and fruitfully German scholarship has been 
deroted, within the period mentioned, to the critical study of the texts of Greek 
philosophy, the preparation and publication of new editions and translations, 
and the elucidation of the same through note and commentary. 

English scholars will need a long time to catch up with their German cousins 
in this field of good works. Meanwhile, all symptoms of a disposition to enter 
it will be doubly welcome to a growing band of philosophic students, who 
consciously stand in need of nothing so much as of all fair aids to the ready 
comprehension of Greek philosophy. It is to such students that Mr. Wallace's 
work is chieBy addressed, and to them, we say advisedly, it will be invaluable. 
And for their benefit we may add that main stress is laid by the author on the 
exhibition of the nature and extent of the contribution made in Aristotle's vtpi 
i^XVi to the theory of cognition. 

While Mr. Wallace's aim has thus been (in his own words) " explanation, 

rather than textual criticism," he has not been unmindful of the requirements 

of the latter. In particular, the views of Torstrik respecting a double recension 

of the text (see Torstrik's edition of Arist. deAmma, Berlin, 1862) have been 

tested by him. ** in regard to several portions of the text," with results expressed 

by Mr. Wallace as follows : " Without denying the existence of repetition and 

disorder in much that Aristotle wrote, or rather left in notes, I have tried in 

several passages to maintain the general correctness of the ordinary text against 

Torstrik's objections and • emendations.* " 

G. S. M. 



REPORTS. 

Rheinischss Museum. 
XXXVI 2. 

1. Pp. 161-74. F> Reuss. King Arybbas of Epirus. An investigation of 
the obscurer historical facts of the time of Demosthenes. 

2. Pp. 175-7. !• Stich. In Marci Antonini Commentarios. Critical 
notes based on a recent examination of Italian manuscripts. 

3. Pp. 178-95. L. C. M. Aubert Adnotationes in Senecae Dialogum I. 
Critical and explanatory. 

4. Pp. 196-205. A. Ludwich. Notes on the Birds of Aristophanes. In 
the much discussed verse 492, L. points out very convincingly that much of the 
point must lie in the word vvKxup : there must be a contrast intended between 
the artisans who obey the seasonable crowing of the cock {bwdrav v6fiov bpdpiov 
(x'a^) and those who start out in the night when the cock crows too early. 
Kuelpides then breaks in with an example of what may happen to these latter 
unfortunates. In accordance with this view L. proposes to write hirode^dfievtu 
for vKodrfadfievoi in 492. But it is hard to see. that his explanation^ which seems 
clearly right, applies any better to the changed text than to that of the MSS. 
Why may we not understand: ot Se paSiiyOva* imodrjcdfievoi viKTup dKdrav ^ayl 
There is a loss of a comic touch in giving up the picture of the man who puts 
on his boots at midnight. In 525, L. writes Kal role lepolq instead of aav rolq 
lepoi^f a very neat correction. 

5. Pp. 206-14. A.. Riese. Geographica. Three notes. I. The traditional 
notion that the Chalybes were the inventors of the use of iron can be traced 
back to Pliny (H. N. 7, 197), who tells us, after quoting other views, that a/ii 
ascribed this invention to the Chalybes. Who were these alti^ Seemingly 
they were careless readers of Callimachus (Fr. 35 Schneider). Catullus (56, 
48 fT.) translates Callimachus as follows: Juppiter^ ut Chatybon omne genus 
pereal et qui principio sub terra quaerere venas insHtit ac ferri fingere dttritiem. 

And Catullas had not misunderstood his original : the fragment of Callima- 
chus must be written : XaXh^ov tif awdh^iro ykvo^ yetddeif avriXXov re KaKbv ^rbv 
01 irpiv i^rjvav. There is no authority for the ascription of the invention of the 
use of iron to the Chalybes. II. The modem word gari/la is due to a slip of 
the pen in the hand of some ancient copyist. We read in the Periplus of 
Hanno (118), of an island on the west coast of Africa: v^ffof ^ ficar^ dv6p67r(jv 
dypifjv, noXv dk itXelov^ ijcav ywaixec daaeUu role a6fiaaiv^ Af ol ipfirivhc CKdXovv 
TopiV.ac. Pliny (H. N. 6, 199), who is demonstrably following Hanno, calls 
these women Gorgades. The ippjpfkt^ elsewhere were careful to use names of 
Greek formation. The conclusion is almost irresistible that the name Gor- 
gades in Pliny is taken from a correct text of Hanno. III. In the fourth cen- 
tury B. C. (Ephoros, Skylax) the names lavpofidrai and Ivpfjidrcu (^pfidrai) 



REPORTS. 355 

I 1 8, the words ec raq vavc kapdvrec are Aglassema. I 19, the last sentence can 
only be understood to apply to the Athenians and their allies. I SS^r^iv Ko- 
ptvOiuv (after r^ iroXifu,}) must be stricken out as a nuatter of grammar as well as 
of historical truth. II 22, UoXvfi^Stf^ xal ^Apurrdvov^, dpxvy^ ^W oraaeuq iK&repo^, 
cf. Xen. Hell. 5, 2, 25. II 41, fi6v/f ohre t<j^ kneWdvri ayavdKTtfaiv ixet — rroXefiiu 
before lweW6vTi being an interpolation, see Classen on II 36. Ill 26, Sio Kai 
before TeaaapoKovra must be given up, see chapters 16, 29, 69, 76. Ill 82, u^ atf 
iKdoToic al furaPoXaX rCxv ^mrrvxtuv. IV 67 ft, an interesting note on the text, 
the topography, the military operations ; but without very distinct results. IV, 
102, bu ireptppiovToc Toit iTpvfi&tfo^, leaving out iff* dfji^drepa. V 27, oXlyov^ 
avrotcpdropoi : that is to say, dpx^v is a gloss on avroKpdropa^, V 60, kv "Sefd^ 
after iuc ^^ iv aOpdov must be stricken out. V 80, tov fpovpiov is a gloss on 
i^o TToi^jcac. There is still more in this article which will bear reading by the 
student of Thucydides. 

10. Pp. 260-301. W. Hoerschelmann. Investigations touching the History 
of the Greek Metrical Writers. An examination of the sources of the scholia 
to Hephaestion. 

11. Pp. 302.28. Miscellany. L. Mendelssohn gives a number of" Trifles.'* 
Babrius 75, 6, tj)v ahptov y&p, T?j^fiov, ovx imepp^y, Chariton, VII 5, 11, bXiyop 
re iirev6ow ovdiv, Polyaen. I I, 2, kvravda wtfyai TroXXei, IScu nvKvai. Caes. 
Bell. Civ. I 22, 5, ut tribuHos pUbis nefarie ex civitate expulsos, Cic de Leg. II 
5, II, esse laudabiUm quidam talibus argwfienHs docenU Val. Max. I i, 14, quam- 
que merito sibi infestos dominos. Veil. Paterc. II 17, 3, consukUum paene omnium 
civium suffragiis nactus est, 

A. Ludwich remarks on the metrical characteristics of the Gigantomachia 
of the Greek poet Klaudianos, and proposes several corrections of the text. 

H. Luckenbach, after an examination of the original stone at Verona, gives 
an exacter text of the epigram, Kaibel, 128. 

G. TeichmQUer discusses if^avaytjrfii^ itzayuyif and iirava^peiv, tTrt^peiv, 
Plato uses the compounds with two prepositions, Aristotle with only one. In 
Plato the words have hardly become technical ; in Aristotle they are fully so. 
Aristotle uses the two prepositions in familiar words, but in strictly scientific 
terms not grown popular he prefers to cast away useless elements. 

E. Hiller has examined the MSS of Schol. Av. Ran. 218. and shows that the 
only authoritative text is that given in Cod. Venetus 474. 

J. Sommerbrodt describes the Florentine MS (Laur. 77) of Lucian. It is a 
mixture of leaves written at different times by different hands and of very 
unequal value. 

H. Flach writes of the Lives of Roman Authors in Suidas. It appears 
that these authors either had written or were believed to have written in Greek. 
The source of the notices was probably the preface to Capito*s translation of 
Eutropius. 

O. Ribbeck calls attention to the expressions in Liv. V 21, which suggest 
that the historian in his account of the taking of Veii is only giving a para- 
phrase of z, f(Umla proitextaia, 

A. Biese defends velatumt Catull. 64, 64. He finds an echo of Catullus in 
Ovid, Ars. Am. I 525 ff. There (v. 529) we read tumea vektta recincta. Further, 



336 AMEJtICAX JOUBXAL OF PHILOLOGY. 

Citnllxu luaself, at r. 365 of :be mK poem, seats to iccsr to tike 
of 64. 

A. ScachiHsciieui ^tres a ca^Iarioa of tbe soCes im BeatleT^s copr of 
btiu :a c!ie Bnrish MasessL 

A. Mam arises brie^ dtat ae:Ui«r Rs^;;ieio nor t. Dolui kas 
prcmn^ aiSTthla^ new about the old covrse of tke rrrer or tke mIbii m m of tke 
docks at Poa&peii. 

XXX\T > 

I. Pp. 329-43. F. Bacheler. Cooiectxaea. Nine notes. L Pbeta %Botits. 
In the inscription CIL VI 7374. ib the diaJogtte betwcem H^driaa aad £ptctet«s, 
pu vUshed by Fabricius (BibL Gr. XIII, p. $61 >. and in tke Gcsta Roaaxnora^ 
dp. 36, are fcmnd the materials of a distich composed not mack later tkan tke 
time oi the Antooines^ which maj be thos restored : wulSa mtim mrimikwu pen- 
dmt^ sic corpora umtnt mut mtatmru cadumt out cita accrha rmuHt. A little farther 
00 tn the same dialogoe between Hadrian and Epictetos occur the words: mttda 
y^nns picia, nmdi pinpmtmr Amora : qmhus nmda placet^ nmdm £wattai tporttt. 
This is Ter»e, bat it is not qaite certain how many of the slips may be due to 
the aathor. Perhaps something like this may be restored : nudm Vrmms^ nudi 
pucn pinguntter Atnorcs : exihit nudus cm. dea nmda placet, — II. Victor gram- 
maticus. The grammarian referred to by Priscian (I 19) and Ra6nas Antioch- 
ensis (Gram, Lat. ed. Keil, VI, p. 573, 26) seems to be identical with the man 
so neatly praised in Anth. Pal. IX 71 1 : airr^ ypaftftarucyv 6 ^uypd^ igdeXe 
yp6fai^ hixropa dk ypa^^ "rdv awcyrrdv' eiirnt, **i;t"-" — HI* The Crispns 
Addressed in Anth. Plan. App. IV 40 is identical with the man addressed by 
Horace, Od. II 2. In the words TpioaoXlixai there is a local allusion. Vitm- 
vias (III 2, 2) designates the region of the Horti Sallnstiani by the words ad 
tres Fortunes. The sentence ri yap avdpi roa^e apxijet etc hnpuv faipiop rvfpo- 
ovvipf^ B. translates: gum huic satis erit infinitas ad itevandos asmcosf — IV. 
Ei'/c/^C. An Oscan inscription found near Agnone gives the names of the 
divinities to whom the altars and statues in the sacred enclosure of Ceres are 
consecrated as follows (the names being in the dative case) : Vczkei, Evkhi^ 
Kcrri, The second name corresponds naturally to the Greek Etw^ (EvicAof). 
Hesychius has this gloss : ev/cP^f • b (f<fw «ca2 bvofm<rrd^ Kal eieiS^^ in which there 
is nothing wrong but an accent. Further, Fiorelli has recently published 
(Notizie degli Scavi, 1880, April, p. 155) the inscriptions upon certain gold 
plates found in the tombs at Sybaris. One of these inscriptions has the lines: 

fpxofiat tK nadapCtv tcc^apd^ ;i^hviuv Paaiksia, 
EvkX^C Ei'^ouAff'f re Koi a6&vaTot deol dXAo<, 
Kal yap lyuv vfiCtv ykvo^^ dXptot, eixofiat elvat. 

The form Evkioi is to be translated into Latin Oreo. — V. Ennius et Gnipho. 
M. Antonius Gnipho, the teacher of Julius Caesar, was the author of a com- 
mentary on the Annals of Ennius. — VI. Explanatory notes on the anonymous 
poem addressed to C. Calpumius Piso. — VII. Propert. V 11, 72, B. proposes to 
write libera fama I'wjfwm.— VIII. Antipatri Tyrii. Fiorelli has published 
(Notizie degli Scavi, 1880, July, p. 250) the inscription upon the tomb of Philon 
Antas Antipatri Tyri /UtHS^ who was buried at Brundisiuip. There is reason for 



REPORTS. 357 

thinking this Philo the son of that Antipater Tyrius who held the first rank 
among Stoic philosophers in Cicero's time. That Philo was a merchant and 
a traveller may be inferred from an existing epigram (Kaibel. 779), found upon 
a votive tablet near Constantinople, in which the sentence w<5e rbv evdif-nfrov ael 
Btbv * Aj^TiTT&Tpov irdig ar^ae ^1?mv seems to contain a word-play upon the name 
Antas. The whole epigram is well written. Further, the man known always to 
Cicero by the name Antipater Sidonius was a facile versifier and a Stoic not with- 
out learning in philosophy. And we learn from a direct statement of Meleager 
(Anth. PaL VII 428, 14) that he was bom at Tyre. In spite of the common- 
ness of the name Antipater, it seems safe to infer that this earlier Antipater 
was of the same stock from which the distinguished philosopher and the poetical 
merchant descended. — IX. Anglosaxonum Latina aenigmata. 

2. Pp. 343-50. H. Diels. Stobaeus and Aetius. Remarks on various 
matters treated in the Doxographi Grtieci, for which £lter*s dissertation (De 
loannis Stobaei Codice Photiano, Bonn, 1880) affords new points of view. 

3. Pp. 351-61. L. Jeep. The Lacunae in the Chronicle of Malalas. 

4. Pp. 362-79. M. Schanz. The Writings of Cornelius Celsus. An 
attempt to find out precisely what may be known of the subject, apart from 
uncertain conjectures. The several paragraphs of the paper may be summar- 
ized as follows : I. The treatise on medicine must have been written later than 
B. C. 23. This is proved by an allusion to the cure of Augustus by cold water 
under the treatment of Antonius Musa (Cels. 3, 9). It must have been written 
earlier than A. D. 48, because Celsus (4, 7) says quamvis in monumentis 
midicorum non Ugerim of a recipe found in the work of Scribonias Largus, who 
published his collection not later than A. D. 48. 2. Graecinus, the father of 
Agricola, wrote a book on the care of vineyards, in which he followed Celsus 
(Plin. H. U. 14, 33). Graecinus died A. D. 38. Therefore, the De Re Rustica 
of Celsus was published before A. D. 38. 3. The De Re Rustica was written 
earlier than De Medicina^ as appears from references made in the later treatise. 
These two works were published together in the order of composition. The 
inference to this effect from the headings in the MSS is supported by the fact 
that the opening words of the De Medicina, ut alimenta sanis corparifms agricul- 
tttra^ sic sanitatem aegris medicina promitdt^ show the author's usual formula for 
transition from one subject to another. 4. It is certain that Celsus wrote upon 
agriculture, medicine, the art of war, rhetoric, and philosophy, and almost cer- 
tain that he wrote upon law. 5. It is probable that Celsus wrote only one 
work on philosophy. The statements of Quintilian (X i, 124; XII it, 24) 
support no other inference ; and Augustine, in the preface to his De Haeresibus, 
speaks in a way to make it plain that the Celsus whom he mentions wrote of 
heresies of later origin than the Christian religion. 6. The De Medicina suffices 
to prove the author's habit of referring to his own previous work. But it contains 
no references to any of the author's Artes except the De Re Rustica, It follows that 
the others were published later. The arrangement of the six Artes was probably 
this : De Re Rustica, De Medicina, De Re Militari, De Rhetorica, De Phihsophia, 
De lure Civili, 7. The title was Aries, in spite of the scholion published by 
Ritschl (Praef. ad Plant. Bacch. VI). 8. The De Philosophia followed the 
teaching of the Sextii. It is therefore impossible (Sen. Quaest. Nat. 7, 32, 2) 
to set its date much later than the death of Tiberius (A. D. 37). Connecting 



353 AMEJtlCAJf JOURNAL OF PHILOLOGY. 



^his Jate with thocB prrnoBslj finm. ve ittj ny dttt the encjdopftedim of 
Cdaos wtt pmblished boC Back later tkam tke death of Tibciias. 9. It is cer- 
tain tltat the aBBopapk «■ War with the FuthiaBS was aot writtea before 
A. D. 03, and at least pOK&le that it was written by Marias Celsas, a Roman 
geacnu who toakpart in theFuthian warinthe7car63. 10. Cohmella men- 
tions Ceisns as a aan of the past (3, I7f 4 ; >, l, 14), and Seneca as a man of 
the present ( J> 3» 5). Therefore Celsas mast hare died before Seneca (A. D. 65). 

5. Pp. 3^0-434. E. Robde. Stndies in the Chronology of Gredc Litera- 
mre. The brst instnlmcttt — for no more is giren in this namber— discusses the 
y ar ion s re Mn n f n gs and compntations which guided GredL chronologers in fixing 
the date of 



ix. Pp. 435-62. F. Hettner. The Discoreries at Nemmagen, with a plate of 
iUttstrattons. In the coone of certain excarations at Nenmagen on the Mosel, 
in the year 1878, Tcry extensive remains of Roman sepulchres were brought to 
lights— «tones which had been removed from their original places and built into 
the foandatiotts of mediaeval structures. The inscriptions and sculptnres on 
these stones, and the remains of decorative color as well, are remarkably fresh 
and well p re se rv e d^ but the restoration of the original architectural forms is 
diAcult. H. attempts only a preliminary account of the monuments, not a 
noal and complete description. The scenes represented in the sculptures are 
taken from common life, and show a great variety of occupations, utensils, 
focaas o£ dress, and the like, in gratifying detail. 

7. Pp. 463-30. Miscellany. W. Dittenberger gives a corrected restoration 
of a metrical iascription published in the Bull, de Corr. Hell. 1880, p. 150, as 

H>I!«>ws: 

'H/Midun^ Ne^fc/Vv )ra(r)pdf {p)rfiotv 
XoAkciov av6pUivTa warpidog ^f^^^ 
"^vvfufc re exari, fu'iTuxp^ ydp vjv iraatv^ 
TtpKvinf Tt fu{!)fiov oi)f l^pa^tev aareiuc. 

He idUs« by way of joke, this note of the original editor : *' celle-ci paralt ecrite 
cu trun^tres iambiques ; mais la rigle fondamentale de ce mdtre est vioUe k 
sh4*qtt« vers, puisqu* on y trouve r^guliirement un spondee ou un troch^ au 
^xi^me pied." 

K. INcheter repeats an epitaph on another author of Mimes from the Revue 
.\ivh<^ogique, 1 88 1 » p. 124. 

W. Hoefschelmann gives a note on the commentary to the hyx^ifn^tav of 
Hc{>h*e«tion. He there finds Alcaeus Fr. 5 in this form : xf"^ xvXAdMif i 
4«iK»\ 9c >ii^j UKH dvftbc vftvtiv, rhv Kopv^aatv airydtc fiaUt yiwa tm KpuviSti fuuiia 

A. tuUMrtch gives critical notes on the Bibliotheca of Apollodorus. 

H. H<»yde«Mtnn contributes several remarks on the interpretation of inscrip- 
.4x4 k^ v>u Uir«<k vases. 

V ("hiU^^M objects to BQdinger's method (Berichte d. Wiener Akad. 93, 
>v>* :t.> lu Uyxtig to restore the story of Solon and Croesus to its old place in 



REPORTS. 359 

A. Riese compares Anth. Lat. 901 with Serv. ad. Verg. Aen. VI 724 by way 
of showing that the fonner was not the production of a scholar ^i the Renais- 
sance. 

G. Voigt has collected a number of notices which warrant the hope that 
material for the restoration of the text of Cicero's letters ad Familiares may be 
found in France. 

M. Voigt writes aprici mergi {or apris mergis in the second glossary of Salerno. 
The explanation would then be intended for Verg. Aen. V 128. 

F. Biicheler writes of Petronius at the Hanoverian court in the year 1702. 
Leibnitz, in a letter to Princess J^uise of HohenzoUem, dated at Hanover, 25 
Feb. 1702, gave a minute description of the amusements of the carnival, which 
closed with a burlesque somewhat freely imitated from Petronius. The note 
with which B. concludes will not seem to readers of this Journal too long for 
reproduction here. It is as follows : ** Wenn dem geehrten Leser diese 
Mittheilung aus vergangenen Zeiten angenehm oder niitzlich scheint, so m5ge 
erzugleich erfahren, dass die Anregung dazu von einem Manne kam,den Bele- 
senheit, Geist und Geschmack zu einem Urtheil iiber philologische Fragen wie 
kaum einen anderen beffthigten, der abgesehen von dem Ehrenplatz, den er in 
der Geschichte unsrer Wissenschaft behaupten wird. um diese Zeitschrift 
besonders grosse Verdienste sich erworben hat, nicht allein durch die in der- 
selben gedruckten Beitrftge, gehaltreiche eindringliche anziehende Abhand- 
lungen vomehmlich aus frUheren Jahren, und feine treffende Monita die er 
geme ohne seinen Namen ausgehen liess, sondem durch seine hervorragende 
Betheiligung an der Redaction, von der auch der Titel mehrerer Bftnde des 
Nfuseums Zeugniss ablegt, mit und neben Ritschl und Welcker,und durch seine 
state Fiirsorge fUr das wissenschaftliche Gedeihen des von ihm miterzogenen 
Kin des, die sich bis in seine letzten Tage durch wohl bedachte RathschUge 
und wohlwollende Censur zu erkennen gab. Jacob Bemays starb am 26 Mai 
[1881], keine sechzig alt, ganz unerwartet, wenn auch zwischen Leben und Tod 
mehrere Tage der Bewusstlosigkeit und Aufldsung lagen, kaum hatte er fUr 
seinen * Phokion,* mit dem er eben die Freunde beschenkt, noch ein Wort des 
Danks entgegen nehmen kOnnen. Nicht mit alien Wegen und Mitteln der 
heutigen Philologie war er einverstanden, von den letzten Decennien lenkte 
er den Blick lieber zuriick zu der ersten Hftlfte dieses Jahrhunderts, zu Herren 
wie Scaliger und Casaubonns, seine Aeussemngen iiber die junge Sprachwissen- 
schaft und einige andere Theile des Gebiets das uns beiden angelegen war und 
das er in peripatetischen Gesprftchen zu behandeln liebte, entsprachen nicht 
den jetzt giltigen oder meinen Anschauungen ; aber ein grosser Kenner und 
ein Kenner des Grossen, getrftnkt aus den edelsten Quellen des Alterthums, 
bewandert in der Literatur modemer Vdlker,gewaltig unter den Mitforschenden 
und ein wQrdevolIer Tyrann der * Mitredenden * von sehr weitem und sehr 
scharfem Blick, weise und gerecht und frei von vielcn Banden, mit welchen 
ftusseres Leben oder eigene GelQste bestricken und das Urtheil der Machthaber 
auch in wissenschaftlichen Dingen gefangen nehmen, wog dieser Eine mehr als 
Hunderte. oImm fiheiv del rbv KaXuf evdalfiova pflegte er zu citiren. Hans und 
Universitftt waren ihm Eins fast ihm strengsten Wortsinne, das Weichbild 
Bonus hatte er seit 10 Jahren nicht verlassen (letztmals zu einem Besuch von 
Johannes Brandis in dessen nahe gelegener Villa am Rhein) ; der Einsamkeit 



36o AMERICA// JOURNAL OF PHILOLOGY. 

ergeben, las nnd bedachte er tmendlich viel, sinnend und rathend Qber PoHtik 
und Jadentham, Philosophic und gelehrte Welt, den Geist spannend ohne 
Nachlass bis das Him tddtlich geschlagen ward. Ein gnt Tbeil seiner Olkon* 
omia betraf dies Moseam, so stehe denn hier anch dies Gedenkzeichen fUr ihn 
in Dankbarkeit und Wehroath gesetzt von einem SchQler, Collegen nnd 
Frcnnde," 

XXXVI 4. 

I. Pp. 4S1-9. H. OsthofT. Notes on the Inscription of Dvenos (see R. 
M. XXXVI 235). O. regards love Sat deivos as accusatives of the end of motion. 
The following words he writes neited endo cosmis vir cosied asted. He then trans- 
lates: ** Whosoever shall offer me to the gods Jupiter and Saturn, let him take 
(^ins that a helpful man be with him and by him within, unless he wishes the 
offering of prayer to be brought to Ops Toitesia.'* It is not safe to identify 
c^fswds with c&mts ; rather we have here a confirmation of the view which con- 
nects cdmis with commodus. The form vols {vis) cannot be derived from the root 
vd: a parallel root vi, to which other indications point, must be assumed. 

a. Pp. 490-505. P. Egenolff. In Herodianum Technicum Critica. A 
continuation of the collations begun in the previous volume (XXXV 98 ff. and 
504 ff.) 

3. Pp. 506-23. P. Corssen. The Sources of Cic. Tusc. I. In his disserta- 
tion De Posidonio Rhodio{fionn, 1878), C. attempted to show that Cicero in the 
first part of Tusc, I followed his contemporary Posidonius of Rhodes. He now 
extends the scope of his reasoning. The two parts of Tusc. I, in spite of the 
contradictions between them, come from one and the same source. But the 
recognition of this fact makes the contradictions themselves all the more strik- 
ing. Cic. himself obviously wished the two parts to be regarded as indepen- 
dent and treated them accordingly. The whole arrangement is his own. But 
he found the substance of his material in some one work of an earlier author, 
and sought only to abbreviate what he borrowed and to state it from points of 
view largely of his own selection. The work thus used by Cicero we may fairly 
assume to have been of a popular rather than of a strictly scientific character : 
taking subject and treatment together, we may refer it to the class of Xoyot 
napa^rjTiKoi of which Krantor seems to have been the inventor. Such a con- 
jecture is strongly confirmed by a comparison of Cic. Tusc. I with Plutarch 
Cons, ad ApoUonium. The fact that both writers had drawn material from a 
common source was remarked long ago ; but C. points out for the first time that 
passages confirming this view are to be found in the first part of Tusc. I. If 
the views stated in C.'s earlier dissertation are adopted, we need not inquire 
further after the author followed by Cicero and Plutarch. Bat it has been 
generally assumed that Plutarch followed Krantor directly. C. therefore pro- 
ceeds to show that this view is untenable ; that the work used, although it made 
v\^e of Krantor^s material, was composed after Krantor's death. The article is 
a fine specimen of the better sort of *' Quellenuntersuchungen.*' 

4, Pp, 524-75. E. Rohde. Studies in the Chronology of Greek litera- 
tuve. Continuation of the investigations touching the date of Homer, begun at 



REPORTS. 361 

5. Pp. 576-96. W. Deecke. Notes on the Interpretation of the Messapian 
Inscriptions. ** Fassen wir die bisherigen Resultate zusammen, so kann das Mes- 
sapische fast eine altgriechische, wenn man will pelasgische Sprache genannt 
werden, wie der altepirotische Zeus von Dodona bei Homer der * pelasgische ' 
heisst. £s giebt demgemflss vielHlltige wichtige und interessante Aufschliisse 
aber die griechische Lautlehre, Flexion und Etymologie." 

6. Pp. 597-603. C. Wachsmuth. '0 en-t Aipniii,} ayov. The lines Ar. Ach. 
502-507 have very little of the Aristophanic flavor. It is generally agreed that 
the piece is not free from interpolations. Definite objections to the lines in 
question are the following: i. The connection of clauses in 504, indeed 
throughout the lines 504-506, is extremely awkward. 2. Unlike Aristophanes 
but very like an interpolator is the repetition of avroi ydp eofiev (504), in d^X* 
eafih aifTol (507). 3. Ik rwv ndXeuv ol ^ififiaxoi (506) is pleonastic. At that day 
in Athens al i:67^i^ alone sufficed to designate the confederacy ; a genuine 
way of saying what is here meant may be seen in 636. 5. The explanation 
given in 505-506 for the statement Koimiji ^tvoi irdpetaiv is simply silly. And 
finally, it is incredible that the poet — for this is his explanation with the audi- 
ence — should have gone to work to make it so very clear to them that the 
present was not the time of year for the visit of the ambassadors — as if any 
Athenian had been stupid enough not to know that ! In short, of what we 
have in 504-507, Aristophanes wrote only avrol ydp eofiev vvv ye irepuimafiivot. 
Having laid this foundation, W. goes on to show that 6 M Ar/vaiii) a>&)v is not 
an Attic name for the competition which the interpolator meant to designate. 
We find TOP dyCtva rCtv Arfvaiuv^ tv roi^ ArfvaiKoi^ dyoai. But we do not find the 
dyuve^ nor the x^P^^ t^ot the didaaKoklai nor the vIkoi called hn\ Afp^aitf), This 
can only be used with propriety to mark the locality of the thing designated. 
In that sense we find the official name Atoviaia to. ini Arfvait^ (from which the 
interpolator borrowed in fashioning his line), and if kn\ Arfvaict irofinrj. In our 
passage such a designation of the locality would have no sense, for at that 
period all the dramatic celebrations took place in the one Dionysiac theatre. 
In Plat Protag. 327 D. ctt* A/rvaiif> means nothing more than ** in the theatre,'* 
the preceding nepvai is quite enough to show that eirl Ajjfvaiifi is not used to dis- 
tinguish one festival from another. 

7. Pp. 604-40. Miscellany. F. Blass fills a dozen pages with notes on 
Greek inscriptions. These have to do mostly with the explanation of Boeotian 
and Doric forms, and do not admit of the compression necessary in these 
reports. A couple of remarks on Attic inscriptions may be reproduced. CIA. 
I 342 (CIG. 1 27) concludes : ArjfujrpdQ re x^P^"^ t*"*'] Ovyaripog [r]avwrf7r^v. 
G. Meyer (Gr. Gr. §111) makes two mistakes worthy of correction. The form 
T'h'TTe, which he cites as the earliest Attic example of e for at, is not to be 
found in the inscription in question (CIA. II 379, 18) : the genuine reading is 
bnu^ ytvrjf i^p6vTio\ev\, And aifidriov for elftdriov is nothing but a misprint 
copied out of Cauer. " Doch wftre es ungerecht, nicht anzuerkennen, dass die 
bei Cauer stehen gebliebenen Druckfehler weder seine zahlreichsten, noch 
seine unverzeihlichsten Siinden sind." 

H. Heydemann continues the notes to inscriptions on Greek vases begun at 

p. 465. 

F. BUcheler explains the inscriptions from Olympia, Nos. 38a and 383. 



362 AMERICAN JOURNAL OF PHILOLOGY. 

A. Ladwich writes Theocr. I 135-6: koi t«c icirvoc 6Aa^ ihmi i^ hpiuv^ ;jw 
OKcirreg hfj66ai S/fpUfatvrOf a correction which can hardly fail to find faror. 

H. Flach examines the biographical notices in Saidas which show a use of 
.the Greek translation of Jerome's De Viris Iliustribus made by Sophronias. 

S. Brandt offers six emendations to passages in Cicero^s Letters to Atticns. 

F. V. Duhn returns to the Harbor of Pompeii (see R. M. XXXV 127 ff.) 
Stricter investigation of the circumstances of the discovery of the articles found 
near the Molini de Rosa affords still clearer evidence that the mouth of the 
Samus and the docks of Pompeii were at that place. The space between the 
town and its harbor was occupied by a suburb. 

J. Klein fills up the lacunae in a recently discovered Roman inscription con- 
taining, the name of the curator locorum pttblicorum P. Catius Sabinus, and 
gives a list of such cwratores^ so far as known, from Augustus to Diocletian. 

J. H. Wheeler. 



Ancua. Herausgegeben von R. P. WClckbr und M. Trautmann. Blnde 
V und VI, I u. 2 Hefte. Halle, 1882-83. 

B. ten Brink begins volume V with brief remarks on the quantity of the first 
/ in the O. £. (s. e, A. S.) suffix rere^ and concludes from O. E. verse, M. E. 
accent, and M. E. rime, that it is long. 

G. Schleich furnishes some corrections to Vamhagen's text of the Proverbs 
of Hendyng, from the Cambridge and Oxford MSS, given in Anglia IV 180. 

Miss Lucy Toulmin Smith prints, for the first time, the text of a ballad by 
Thomas Occleve, addressed to Sir John Oldcastle, A. D. 141 St from Phillipps 
MS 8151, with an introduction, giving some account of Oldcastle and Occleve, 
and notes. Occleve appeals to Oldcastle **to renounce his opinions as a 
follower of Wiclif, warning him of his errors and of his danger.*' 

A. Fritzsche discusses the question : Is the O. E. [M. E.] story of Genesis and 
Exodus the work of one author ? ten Brink, who first called attention to the 
source of this thirteenth century poem, namely, the Historia Scholastica of the 
French priest Petrus Comestor (1169-75), suggested the possibility of two 
authors. Fritzsche makes a careful study of the poem, based on Dr. Morris's 
edition for the E. E. T. Society, and concludes (p. 84), from the use of the 
same source in the same way, the same versification in respect to metre, riipe 
and alliteration, and the same language in respect to phonology, grammar 
(referring to Morris and Hilmer), syntax and vocabulary, that there is but 
one author. He adds critical notes, agreeing in great part with K61bing 
Englische Studien, III 273-334, 

E. Einenkel discusses the question: Is the late Anglo-Saxon [thirteenth 
century] legend of St. Katharine of Alexandria a work of the author of St. 
Juliana and St. Margaret, or of the author of Hali Meidenhad? This essay is 
the third part of the author's work, Ueber die Verfasser einiger neuangels. 
Schriften, Leipzig, 1 881, in which he has shown that St. Juliana and St. Margaret 
are the work of the same author. He now compares the words and phrases, 
verse and style of St. Katharine with each of the above-mentioned works, and 
concludes that St. Katharine is not by the same author as the other two legends. 



REPORTS, 363 

but that the latter writer knew and used the legend of St. Katharine. The 
second question is more difficult, but cannot be answered affirmatively ; hence, 
the homily Hali Meidenhad is by a third hand and later than the others, or, at 
least, than St. Margaret. 

A. Ebert, in the studies for his History of Mediaeval Literature, has compared 
the Anglo-Saxon Genesis with the Vulgate, and communicates the additions 
and omissions which he has noticed from v. 852 to end. He concludes 
that this part of the Genesis cannot be by Caedmon, for the writer evidently 
had the Bible before his eyes, which does not correspond with what Beda says 
about Caedmon, and, moreover, the earmina mentioned by Beda were lyrical 
poems, hymns, and so of a different kind from the Genesis. 

K. J. Schr5er suggests two very plausible emendations to the text of 
Marlowe*s Fanstus ; the Latin words ' quod tumeraris ' should be * gmd tu 
morttris f *; and ' ignei^ aeri^ aquitani spiritus * should be * ignis ^ airis^ aqwu^ 
terrae spiritus I * Goethe*s text confirms these. 

A. SchrOer prints the full text of John Balers Comedy Concemynge Thre 
Lawes, Anno MDXXXyill, preceded by an introduction giving a full 
account of Bale and his works, and followed by notes on certain words and an 
excursus on the metre. Thb miracle play has in the colophon " lately inprented 
per Nicolaum Bamburgensem,*' o^ whom SchrOer knows nothing, but thinks 
the play was printed in Germany. Only three copies remain, two in the 
Bodleian Library at Oxford, one of them incomplete, and one in the British 
Museum. Hazlitt mentions a second imprint at London, 1562, but no copy of 
this is in London or Oxford. So far as known, the play has not been printed 
since. 

E. Einenkel, in his essay on An English Authoress of the beginning of the 
1 2th [13th, Anz. 64] century, labors to prove that the Wohunge of ure Louerd, 
the Ureisun of God Almihti, and the Lofsong of ure Louerde, are written by 
women, t. e, nuns, and that it is not improbable that our authoress is identical 
with one of the three maidens of the Ancren Riwle. He thinks that he has 
proved the first position, and that a great deal may be said for the second. 
The investigation seems to proceed from this comprehensive premise (p. 265): 
** Das geschlecht des verfassers aller dieser liebesschriften ergibt sich deshalb 
ganz von selbst aus dem gegenstande desselben,'* t. e, if the work praises the 
Virgin Mary, it roust be written by a man ; if Christ, by a woman. Once 
granted, this would settle the question, but, notwithstanding the " liebeskultus," 
it may be doubted whether the monks resigned to the nuns the privilege of 
writing all Lofsonges of ure Louerde, and the Scotch verdict may apply to all 
such investigations. 

D. Rohde writes an appreciative notice of W. Hertzberg, well known as an 
English scholar, and especially for his services to the study of Chaucer and 
Shakspeare. He died July 7, 1879. His life was nuurked by great literary 
activity. A complete list of his works closes the notice, which ends the first 
number of this volume. 

A. Schr6er opens the second number with a republication of the text of * The 
Grave * from the MS (Bodl. 343), correcting some errors in Thorpe*s text. 



364 AMERICAN JOURNAL OF PHILOLOGY. 

O. Lohmann contribates a critical essay on B3rron's Manfred and its relation 
to poems of like contents, i, e, to the Prometheus of Aeschylus, Don Juan of 
Molidre, and Faust of Goethe. Goethe and Byron have transferred much of 
their own personality to the characters of their heroes ; Moliire and Aeschylus 
stand apart from their works. Shelley's Prometheus might have been advan- 
tageously included in the comparison. ' 

M. Bech examines the Sources and Plan of the Legende of Goode Women 
and its relation to the * Confessio Amantis.' The sources have already been 
noticed by ten Brink, Sandras, Bartsch, and Hertzberg, but in no case exhaus- 
tively. Besides the Confessio Amantis, Boccaccio's De casibus virorum illustrium 
and De mulieribus claris liber are taken into consideration. Lack of space 
forbids following the investigation, but after a full discussion of the Sources 
and a brief comparison with Gower, the general result in respect to the Plan is 
summed up, that, as the Decamerone was the model for the Canterbury Tales, 
so the De mulieribus claris liber was the model for the Legende of Goode 
Women, which, however, was never finished. 

W. Sattler continues his Contributions on the use of Prepositions in Modem 
English, with XV, to die of, to die from ; XVI, the key of, the key to ; XVII, kind 
of kim, kind in kim ; XVIII, in a lottd voice, witk a (hud) voice, forming a very 
full collection of examples. 

A. Ebert on the Exodus (Anglo-Saxon), briefly states his reasons for regarding 
the so-called episode, 362-445, as an integral part of the poem and no interpo- 
lation, as recently treated in the Bonn dissertation of H. Balg on The Poet 
Caedmon and his Works. 

K. A. M. Hartmann discusses the question : Is King Alfred the author of 
the alliterating version of the Metres of Boethius ? He enumerates the opinions 
of scholars from Rawlinson (1698) down, showing that no one doubted the 
Alfredian authorship before Wright (1842), whose arguments are briefly stated, 
namely, that the writer omits to versify three metres, because they are not 
introduced by the usual formula in King Alfred's prose version; that the 
metres are very weak, considered from a poetical standpoint ; and that the 
author has a very deficient knowledge of classical antiquity, and commits 
errors where Alfred's prose has the correct translation. Hartmann combats 
each of these arguments, defends the Alfredian authorship of the Preface, 
which ten Brink had questioned, though he favored the authenticity of the 
metres, adds arguments drawn from the use of certain expressions and particular 
words, and concludes that Wright's arguments are ** pseudo^rguments,*' and 
that '■ King Alfred and nobody else is the author." 

R. P. Walcker, On the Vercelli-Book, states the results of his own examina- 
tion of the MS, giving the beginning and the end of each of the homilies con- 
tained in it, the thirteenth complete, and extracts from the Life of St. Guthlac 
compared with Goodwin's text from Cotton MS, Vesp. D XXI. 

Wttlcker prefixes an account of Blume's discovery, made in 1823, that the 
MS was written in Anglo-Saxon, the subsequent references to it, the first pub- 
lication of the poetry by Thorpe, from a copy of the MS made by Blume, in 
Appendix B to Cooper's Report on Rymer's Foedera, made for the Record 
Commission (1836), Grimm's Andreas and Elene (1840), from Cooper, Kemble's 



/REPORTS. 365 

edition of the Poetry, with translation, for the Aelfric Society (1844-46), and 
Zupitxa's edition of the Elene (1877) after a new collation made by Kn6il. 
The Homilies of the Vercelli-Book have never been published. 

T. Wissmann, in the last essay, On Middle-English Word-Accent, supports 
the yiews expressed in his King Horn against those of Schipper in his Old- 
English Metre. He examines the verse of the Ormulum and the Poema Morale, 
and argues that the * senkungen * and the * tonlose ' and ' stumme silben ' are 
treated in exact accordance with Lachmann*s rules for M. H. G. metre, and in 
direct opposition to Schipper's view (p. 476). He further examines again 
certain verses in King Horn, and claims that they correspond to the rules 
heretofore laid down. He argues, too, against Vetter^s view of Anglo-Saxon 
metre (supported by Schipper), and contends that by the acceptance of the 

* zweihebungstheorie * all intelligible rh3rthm is lost and the absolute formless- 
ness of Germanic metre is established (p. 481). 

The controversy is continued in the Anzeiger by Einenkel, in a review of 
Schipper's work, by Schipper in reply, and by Trautmann. T. grants the 

* zweihebungstheorie * for Anglo-Saxon verse, but his views on the metre of 
Layamon are similar to those of Wissmann on that of King Horn, and so are 
opposed to Schipper^s. It would prolong this report to unreasonable length to 
go into the controversy, even to give an outline of these lengthy articles, but 
to an ordinary English ear, Schipper^s view seems the more suitable to English 
Terse. Wissmann*s implied reproach to the ' zweihebungstheoretiker ' [he 
inadvertently writes rn^-], namely, *' Zwei gehobene silben in jeder halbzeile 
das ist so ziemlich das einzige [rather hauptsAchliche] erfordemiss das sie an 
den rhythmus der [des] verses stellen ** (p. 481), appears to be justified by the 
structure of Anglo-Saxon verse, and if so, historical consistency would lead us 
to expect it in the verse of Layamon and other Middle-English writers, and 
not to assume these works to be written on a different principle, however 
applicable that principle may be to Old and Middle High German v.erse. 
Also, there seems no good reason for assuming the final e in Icfri to be * tonlos ' 
but * tonf&hig,' while that of sp/Ae is * stumm,' and so * tonunf&hig,' according to 
Wissmann*s terminology. The question here is not one of quatUity of the 
root-syllable, but of accent ^ and so far from regarding ** Schipper's versuch, die 
absolute tonunOUiigkeit aller silben mit unbetontem e zu erweisen, als voll- 
stindig misglUckt *' (p. 476), I should take it as a simple statement of fact 
existing in English verse (cf. Schipper, section III, chapter 6, 226o~3)* 

Wissmann gives as Appendix a carefully prepared synopsis of the metre of 
eighteen M. E. works, illustrating the view which he is defending. 

R. P. Wulcker contributes a short obituary notice of L. Botkine, the young 
French scholar, known from his translation of** Beowulf" and the Rune-Song, 
who died in May, 1882, at the early age of twenty-nine. We can sympathize 
with Walcker*8 ejaculation : " Leider, wurde B. zu friih seiner wissenschaft 
entrissen und wer weiss wann wider jemand fiir verbreitung des Angelsftch- 
sischen in Frankreich wirken wird I * * French students of Anglo-Saxon have 
been all too few, and may be counted on the fingers of one hand. 

K. Elze adds an ' Entgegnung,' taking exception to certain remarks of Dr. 
Leo, in the Shakespeare Jahrbuch, which, however, seem quite complimentary. 



366 AMERICAN JOURNAL OF PHILOLOGY. 

It must suffice to add merely a summary of the contents of the Anzeiger to 
this volume. W. Muschacke reviews A. Brandos Thomas of Erceldoune; F. 
D6nne, G. LQdtcke's The Erl of Tolous and the Emperes of Almayn ; both of 
these works appear in the Weidmannsche Sammlung englischer Denkm^er 
in kritischen Ausgaben. G. Tanger replies to Fumivairs remarks on his 
edition of the First and Second Quartos, and the First Folio of Hamlet : Their 
Relation to Each Other, reprinted from the New Shakspere Society's Transac- 
tions, 1880-82 ; E. Einenkel reviews O. Zielke's Sir Orfeo,ein englisches Feen- 
mftrchen aus dem Mittelalter ; A. Brandl, R. H. Hutton's Sir Walter Scott, in 
English Men of Letters Series; M. Trautmann, C. Horstmann's Altenglische 
Legenden ; Miss L. T. Smith, the Catholicon Anglicum, edited by S. J. Herr- 
tage for the E. E. T. Society, 1881 : this is an English- Latin word-book, dated 
1483, and heretofore known in only two MSS; it contains about 8000 words, 
each provided by the editor with passages illustrative of its use ; we find, e, g,, 
^*'Loye : elegius. nomen proprium,*' and notes to it, which, as Miss Smith remarks, 
settle the question as to the Prioress's oath.* (See A. J. P., II 386.) E. Einenkel 
reviews at leng^th (23 pages) Schipper's Englische Metrik, and later (Anz. 139) 
adds a continuation (5 pages) of this review ; and M. Trautmann closes the first 
part of the Anzeiger with a review of D. Asher's pamphlet, Ueber den Unter- 
richt in den Neuern Sprachen, spezieller der Englischen, an unseren Universi- 
titen und hdheren Schulen, and of G. K5rting*s Gedanken und Bemerkungen 
liber das Studium der Neuern Sprachen auf den Deutschen Hochschulen, which 
review contains some valuable suggestions. 

G. Schleich begins the second part with a review of The Romaunce of the 
Sowdone of Babylone and of Ferumbras his Sone who conquerede Rome, 
re-edited by E. Hausknecht for the E. E. T. Society, 1881; R. P. Wtllcker 
reviews the Toller-Bosworth Anglo-Saxon Dictionary, Parts I and II, and A. 
Napier's Gdttingen Inaugural-dissertation, Ueber die Werke des altenglischen 
erzbischofs Wulfstan ; R. Boyle, on A. H. BuUen's Old Pla3rs, announces the 
completion of the reissue of Day's Plays, and asks for subscribers to four 
volumes of old dramas soon to be published by Mr. Bullen (Vol. I now ready), 
at one guinea per volume. 

F. Kluge reviews Sievers's Angels&chsische Grammatik, the latest and best 
Anglo-Saxon grammar that we have ; and W. Merkes, E. Einenkel's work, 
Ueber die Verfasser einiger neuangels&chsischer Schriften, t. ^., St Juliana, St. 
Margaret, and Hali Meidenhad. (See above.) 

Schipper, Zur altenglischen Wortbetonung, eine Entgegnung, follows^ in 
which reply he sustains well his previously expressed views against his oppo- . 
nents, Wissmann in particular, rejecting in toto Lachmann's rules as inapplicable 
to English verse ; Trautmann, Zur ait- und mittelenglischen Verslehre, combats 
Schipper, but, while accepting Lachmann's versregeln^ t. ^., adhering to the 
* Vierhebungstheorie ' for Layamon's verse, he rejects Lachmann's hetomrngs- 
gesette, and presents other rules of accent much more elastic, though still at 
variance with Schipper. He also takes exception to the measure assigned by 
Schipper to some other Middle-English poems. J. Koch reviews W. Eiler's 

* On the deficiencies of Heritage's ed. of the Catholicon Anglicnm, see J. H. Hessels, in 
Academy, No. 586, for July 28, 1883. 



REPORTS, 367 

Die Erz&hlung des Pfarrers in Chaucer's Canterbary-geschichten iind die 
Somme de Vices et de Vertus des Fr^re Lorens, an Erlangen dissertation, and 
Zupitza's edition of Chaucer's Prolog. Einenkel completes his review of 
Schipper, and Trautmann finishes the Anzeiger, and the volume, with a notice 
of the Jahresbericht 0ber die Erscheinungen auf dem Gebiete der German ischen 
Philologie, herausgegeben von der Gesellschaft fUr Deutsche Philologie in 
Berlin, dritter jahrgang, 1881 — a most useful work. 

VI I. 

B. Leonhardt begins the sixth volume of the Anglta with an article on the 
Sources of Cymbeline. Besides the acknowledged sources of Shakspere*s play, 
namely, Holinshed's Chronicle and Boccaccio's Decameron II 9, the following 
have been regarded by some critics as sources of the play : I. An English story, 
• The Tale told by the Fishwife of Standon-the-Green * [? Stand-on], found in 
Westward for Smelts, a book published in 1619 ; 2. Two Old-French romances, 
belonging probably to the first half of the thirteenth century ; 3. An Old-French 
miracle-play ; and 4. The German tale of Sneewitchen, compared with Cym- 
beline by K. Schenkl in Germania IV. After a detailed examination of each 
of these the writer concludes that there is no reason for thinking that Shakspere 
used any one of them, that Holinshed and Boccaccio were his only sources, 
the latter most probably in an English translation, although the earliest trans- 
lation of the whole Decameron was not published until 1620,' and that the 
union of the two stories in the drama is entirely his own work. 

P. Lange discusses Chaucer's Influence on Douglas. He compares at length 
Douglas's Palice of Honour with Chaucer's House of Fame and Prologue to the 
Legend of Good Women, and concludes that traces of this influence are plainly 
seen. Lange includes in his comparison some of the spurious poems, but, 
while this may show Douglas's acquaintance with these poems, it does not add 
any weight to the argument for Chaucer's influence. He notices also the 
sources of the Palice of Honour, and shows that the direct source is not the 
Tabula of Kebes,as thought by Warton, Irving, and Bishop Sage, the biographer 
of Douglas, but Le S^jour d'Honneur, by Octavien de St. Gelais, Bishop of 
AngouUme (1466-1502). Douglas's King Hart and Translation of Vergil are 
more briefly compared with Chaucer, and the conclusion drawn that in all his 
poems Douglas shows the influence of Chaucer. 

E. Hausknecht contributes Old-English Glosses on the Brussels MS (No. 
1650) of Aldhelm's De Laudibus Virginitatis. These have been already pub- 
lished by Mone (Quellen und Forschungen, 1830), and by Bouterwek (Haupt*s 
Zeitschrift IX), but with some errors in both publications. 

J. Koch supplies Chauceriana. I. ' Mother of God,' to the authorship of 
which he thinks Occleve has greater claims than Chaucer; and II. Canterbury 
Tales, Prologue 4S9-60, quoting a passage from Job! Ludolfi alias Leutholf 
dicti ad suam Historiam Aethiopicam Commentarius (Frankfort, 1691): 'De 

> Perry, Englbh Literature in Che Eighteenth Century, p. 987, note, states that "numy of his 
stories~BandeUo*s and Cinthio's^had been translated in William Paynter's 'Palace of 
Pteasur« ' (1566)." 



368 AMERICAN JOURNAL OF PHILOLOGY. 

copalatione conjugum ante fores tetnpli,' etc., which throws light upon the Wife 
of Bath's statement, ' Hoasbondes atte chirch dore I have had fyve.' 

E. Uhlemann compares Chaucer's House of Fame and Pope*s Temple of 
Fame, that is, those parts of the third book of Chaucer's poem which Pope has 
professedly followed. Uhlemann finds that Pope has completely changed 
many parts of his model, having taken only the leading thoughts from Chaucer's 
poem, and having worked the descriptive details to correspond to the taste of 
his time ; also, that Pope's style and metre differ greatly from Chaucer's. He 
thus confirms, in general, Pope's own statement of his indebtedness to Chaucer, 
but agrees rather with Warton than with Steele in his estimate of the poem ; 
the latter says : **the original vision of Chaucer was never denied to be much 
improved "; the former, *' He (Pope) has not only misrepresented the story, but 
marred the character of the poem." 

A. Leicht, in reply to Hartmann (Anglia V 411), discusses the question : Is 
King Alfred the author of the alliterating metres of Boethius ? He finds that 
this is but an indifferent paraphrase of King Alfred's prose version, padded 
out with stereotype poetical expressions drawn from the ancient epic poetry 
and inserted in a tasteless manner, and that the writer has occasionally mis- 
understood his prose model, especially two passages in Metre 26, IV 3, in one 
of which the prose version speaks of Ulysses as King of Ithaca (/p<za^), and 
for this the poetical version substitutes Brada, Moreover, this version is i>er- 
vaded not by a poetical, but by a pedantic spirit, and does not observe the 
laws of alliteration as seen in the older poetry. Therefore also, Alfred is not 
the author of the preface and introduction, but these are written by the poet- 
ical paraphrast, most probably a monk who desired to shield his unskilful 
work under Alfred's great name. Leicht thus controverts the view of Rawlin- 
son, Cardale, Fox, Tupper [!], and even ten Brink, and agrees with Wright, 
who (as stated above) first attacked the Alfredian authorship of the alliterating 
metres, though he takes exception to some of Wright's arguments. It is some 
consolation to have King Alfred relieved of the reputation of being a bad 
poet. Leicht will have another article in a future number on the prose Preface 
and the relation of the prose version to the Latin. 

J. Piatt contributes Angelsaechsisches^ as follows : I. A. S. genders, a list of 
words of different genders in older and later works. 2. Local names of the 
^<^-declension. Cent, Cert^ /', TentU IViht, 3. A true A. S. dual, i. e.sculdru [?]. 
4. A. S. u in feminine of the it-declension. $. A. S. feminine ?i^stems, mdtd^ 
Ides, 6. A. S. fetian^ feccan, 7. A. S. feminine termination -Uge, Some of 
Mr. Piatt's statements, especially under 2 and 3, might provoke discussion, and 
with regard to his assertion under 6 : '* So sprechen wir im engl. Tuaday 
nicht tjiiz sondem ckUt aus," I have only to say that, if this is the prevalent 
*' London" pronunciation, it is but another proof that the correct pronuncia- 
Hon of English is better preserved on this side of the water. 

C. Weiser furnishes an unpublished letter of Shelley's, of Nov. 32, 1817, sent 
him by Mr. R. Gamett, the editor of Shelley's letters. It is of interest 
only as showing Shelley's opinion of his " Queen Mab," and his professed 
determination to devote his life to inculcating " the doctrine of equality and 
liberty and disinterestedness, and entire unbelief in religion of any sort." The 



REPORTS. 369 

letter was foand in the library of Lord Lytton, and was sent to Weiser because 
he had published a translation of " Queen Mab/* 

2. — The Anteiger to Anglia VI opens with a lonjj review by W. Schu- 
mann of Dr. Morris's Genesis and Exodus, A. D. 1250 (E. E. T. Society, 1874). 
After some remarks on Morris's Introduction, with references also to Fritzsche's 
article in Anglia V 43, and to Kilmer's Gymnasialprogramm (Sondershausen, 
1876), Schumann gives nearly thirty pages of notes, both critical and explanatory, 
on about 150 passages of the poem. Lack of space will not permit a summary 
of these reviews and book-notices, but a mere enumeration must suffice. L. 
Morsbach reviews Thum's Anmerkungen zu Macaulay's History of England, 
and Hoppe*s edition of Dickens's Cricket on the Hearth ; H. L5schhom| J. 
Koch's Siebenschl&fer Legende, ihr Ursprung und ihre Verbreitung ; F. H. 
Stratmann, KSlbing's Sir Tristrem ; E. Peters, Holder's Baeda, in his German- 
ischer Bilcherschatz. Here is Baeda's complete history for 4.50 marks; the 
reviewer congratulates the public on this fact While mentioning the English 
editions of Stevenson and Giles, he omits those of Moberiy and Lumby. J. 
Koch excoriates Wihlidahl's Chaucer's ' Knightes Tale,' calling it ^* Ein in 
jeder beziehung elendes machwerk," and his notice abundantly substantiates 
this criticism. R. WUlcker reviews Mentzel's Geschichte der Schauspielkunst 
in Frankfurt am Main ; J. Koch, Der Sprachunterricht muss umkehren ! by 
Quousque Tandem, in which some ideas are presented that deserve attention 
in America, as well as in Germany; E. Peters, Wagner's Visio Tnugdali. 
lateinisch und altdeutsch. E. Einenkel adds an Erkl&rung gegen Schipper, 
still holding to his previously expressed views, but very unnecessarily reflecting 
upon a writer in 'The Nation' of Oct. 12, 1882 (No. 902), who supports 
Schipper's views, I cannot notice this controversy further. (See above.) 
M. Trautmann notices, with highly appreciative comments, Professor Child's 
new edition of English and Scottish Ballads, Part I, and gives a synopsis of 
the contents of its 28 pieces. L. Morsbach briefly dispatches Hierthes, W5r- 
terbuch des schottischen Dijilekts in den werken von Walter Scott und Bums, 
as ** das stiimperhafte machwerk," and sa3rs that it may be recommended to 
students only '* damit sie frUhe lemen, wie man es nicht machen soUe." R. 
WQlcker closes this number of the Anzeiger with a notice of Furnivall's edition 
of the Digby Mysteries (New Shakspere Society, 1882), containing three Mys- 
teries now published for the first time, though fifty copies of the Digby MS 
were once printed by the Abbotsford Club. 

JAMSS M. Garnitt. 



Hbrmbs, 1882. 
No. III. 

Wilamowitz, of Greifswald, A paper on the Heraclidae of Euripides. The 
most important of the four chapters in this paper is the first one. W. argues 
that there are imperfections in the extant drama so gross as to compel the critic 
to infer not merely that the piece is transmitted to us in bad shape, but that 
we have not before us the original Euripidean composition at all, and that we 
now read the play as recast by a later playwright of the histrionic profession. 

Wilamowitz's criticism in this part of the discussion consists mainly in con- 
tiderations bearing upon dramatic propriety. 



370 AMERICAN JOURNAL OF PHILOLOGY. 

Macaria, in order to insare victory for the hospitable arms of protecting 
Athens (in accordance with the condition of the oracle), announces her intention 
of sacrificing herself. She departs with the words, 595 (Kirchhoff) 

. . . rd yap daveiv 

KOKuv fikyiarov ^puaxov vofti^eTOi, 

One would expect to hear a report of her end from some &yyeh)c, with a monody 
of lamentation from the lips of her grandmother Alcmena, but we are met by 
an utter absence of these essential stages in the development of the action. 
W. argues earnestly and ingeniously against the assumption of a mere lacuna 
in the text. W. finds in Eurip. fragm. 848, baric 6k rove reKivTaf h pUf> oi^i icri 
the conclusion of the missing report of Macaria's death, assigns a place in the 
original drama to the three verses now classed as fragm. 219 (Antiope), and 
condemns the present parodos of the Heraclidae. This recasting of the play 
he assumes to have been made by a stage manager of the fourth century B. C, 
a man not actuated by literary vanity so much as by practical regard for stage 
uses.* 

0. Rossbach, De Senecae Dialogis. How far do the extant works of Seneca 
admit that title? Quintilian X 1, 128 quotes diaUgi. Generally editors have 
been rating Seneca's dialogi as libri deptrditi^ but Rossbach, from a special 
inspection of the Milan MS of S., infers that the word dicdogi really refers to a 
number of extant writings of Seneca. Of course the word dialogus must be 
taken not in the Platonic sense, but rather in the same way in which it applies 
to the discursive treatises, rhetorical and philosophical, of Cicero, opposing 
views^ being brought forward and combated, although represented by no defi- 
nite character or personality. Interesting remarks are subjoined on the 
estimate put upon Seneca*s books in subsequent times. 

Diels,* Stichometrisches. D. takes exception to some views advanced by 
Th. Birt (** Ueber das antike Buchwesen *'), and basing his own theory on a 
passage in Galei^(V 655 Ktthn, 656, 6, MUller) comes to the following conclu- 
sion : The ancient normal stichos, while substantially representing the bulk 
of an heroic hexameter, was based not on the counting of letters, but of syllables, 
the latter averaging sixteen in number. 

1. Yahlen, Varia. We have noted in former reports that in textual criticism 
this eminent scholar pursues a conservative tendency. How much his influence 

> Some of W/s sutements are far too positive, p. 341 : Wie das Sttick Yorliegt, ist die 
venrunderte Frage des lolaos ob denn die Enucheidung so nahe sei, nctkwendig, . . . Aber 
die Entscheidung ist hier minder leicht und sicher, weil so liemlich die ganse Scene 630-60 dem 
Bearbeiter xufallen mu9t. It nuiy be worth while to compare the following seateaces of W.'s 
with the lines of Euripides : " In der That bin ich (roh dass Euripides an den Geschmacklosig- 
keiten dieser Verse unschuldig ist. lolaos ruft die Alcmene nicht lauter heraus als ndthig ist, 
damit sie thn h5rt, ja wie es scheint nicht laut gcnug, denn die Begrandtmg seines Rufcs(644'S) 
hat sie offenbar uberhOrt. Sie poltert namlich heraus, behauptet der ganxe Tempd wire voU 
Geschrei gewesen, henrscht den unschuldigen Knappen an, im Glauben er w&re Kopreus, und 
droht ihm mit Thatlichkeiten." But as it not intrinsically probable that Alcmene, in her des> 
pondency and gloom, is prepared in advance for no message but one demanding extradition, for 
no news but bftd news, for no messenger but oi»e from Argos T The text really does not suggest 
her believing that she beholds Kopreus himself, for she says, 647 : 

«4pv{ «» 'Apyovc. E, G. S. 

s Compare Amer. Joura. of Philol. 1889, p. 108. 



REPORTS, 371 

will prevail against the long-e<ttablished bias of many German scholars towards 
reckless adir/fatCy time will show. Suffice it to say here that the ample learning, 
the circumspect use of analogy, the employment of sound methods of criticism, 
make the papers of Vahlen very instructive indeed. In the Brst note he proves, 
against Ritschl and Lachmann, that el in Plautus and in Lucretius do^s occur 
in the sense of etiam. In the second discussion (on Plato Phaedr. 236 ^), he 
maintains the MS reading ar&dirn against Cobet and Schanz. Also in the pas- 
sage a little below (236 c\ prrriov fih/ ydp aoi navrb^ /adXXov Kri^ he shows that 
the MS reading, with proper pointing, makes better sense and grammar than 
the text as changed by Cobet and Schanz. 

Th. Mordtmann (Pera) attacks the authenticity of a' number of inscriptions 
and MSS which Francois Lenormant has been publishing within the last 
twenty years. This is followed by H. Roehl (Berlin). In Franciscum Lenor- 
mant Inscriptionum Falsarium, in which Lenormant is called to account for a 
long string of forgeries. Among the scholars who condemn Lenormant's 
unsupported statements are mentioned Rirchhoff, Kumanudis, Koehler, Kaibel, 
Dittenberger, Near the end of his paper Roehl says : Dicet quispiam, peten- 
dum esse a Lenormanto ut duos catalogos antiquitatum a se editarum public! 
iuris faciat, alteram genuinaram, alteram fictaram. Nos hoc non petimus ; 
quis enim fidem ei habituras esset ? 

C. Robert. K5nig Philipp V und die Larisaeer. Robert reprints with slight 
critical alterations an Inscription of Larisa, Thessaly, discovered by H. G. Lol- 
ling. This inscription is a record of a certain transaction of the common- 
wealth of Larisa. The time is probably the earlier years of the Hannibalian 
epoch. King Philip (father of Perseus), last but one of the Macedonian dynasty, 
strongly urges Larisa to adopt into full citizenship their metics (f. e. those of 
Greek nationality), so as to strengthen both Larisa itself as well as the royal 
interest. This matter the king urged in two letters, the first of which (accord- 
ing to Mommsen's appended commentary) was written in 220 B, C, the second 
in 214. Both of these letters are incorporated in the record together with the 
decree asked for, as well as lists of those who consequently received the fran- 
chise. This inscription enables us to realize more vividly than could be done 
before the political situation as regards Rome and Greece in that age. The 
inscription on the grammatical side affords the student ample illustration of 
Thessalian phonetics and inflection, being of near kin to the Aeolian dialect 
of Lesbos, etc. A few specimens may here be given : /c/f =: rig ; 6ii ki = did ri ; 
TTox Ki = Trpdf ri \ ififiev = tlvat ; KarBifiev = Kardeivai ; dn"6 :^ oTrd ; ovvfia z= 
livofta, Attic w is invariably represented by ov: e, g, ^elSow, Klfiow^ rovv dXkow 
'EAA<iw)w, . . . rovv xp^tai^wv {xpijalfiov) ; noTedterw^.TTpocedtiTO ; rol pacihtog 
zz Tov jiaciXiug ; iaydvoiq =: CKydvoic ; cadofuv zz iKSovvai ; byypd^vrac = di?ay' 
pd\l;avTag, Literature probably was but little practiced amongst the Thessalian 
lairds (unless by hiring alien celebrities), and the consequent wearing away of 
inflection and the tendency towards apocop^ and assimilation are evident in 
the present inscription : noK kI^ ttot rbg raydc, icor rdv iTnarokdv, h takes the 
accus, (zr tig c. ace.) e. g, KarOefiev rhfi fitv lav kv to lepdv tov 'AffXowof . 

Kohlewein (Ilfeld) gives some specimens of a Latin version of Hippocrates* 
Aphorisms found in Codex 97 at Monte Cassino. This copy (it is evidently a 



374 AMERICAN JOURNAL OF PHILOLOGY. 

This interesting docoment prores that mumicipia^ in the nuUter of domestic 
celebration, made their own decrees independently of the Roman Senate. 
Thns, under date of Sept. 23, there is ** Immolatio Caesari hostia ** (a), whereas 
at the capital, divine honors were officially paid to Angostos only after his 
death. We notice, too, that the mnnicipiom of Cnmae had no official annual 
celebration of Actium, whereas the accession of Lepidns* army to the standards 
of OctaTianus is set down for Sep. 3. The historical value of the inscription 
is so considerable that the readers of the Journal will be glad to possess it 
entire. 
Aug. 19. [XIIII K. Septembr. Eo die Caesar pri]mum consnlatum in[iit. 

Supplicatio] . . . 
Sept. 3. [Ill Non. Septembr. Eo die exer]citus Lepidi tradidit se Caesari. 

Suppli[c]a[tio] . . . 
Sept. 23. [VIII K. Octobr. NJatalis Caesaris. Immolatio Caesari hostia; 

suppicatio [sic\ . . . 
Oct. 5. Nonis Octobr. Drusi Caesaris natalis. Supplicatio Vestae. 
Oct. 18. XV K. Novimbr. Eo die Caesar togam virilem sumpsit. Supplicatio 

Spei et Iuye[ntuti] . . . 
Nov. 16. XVI K. Dicimbr. NaUlis Ti. Caesaris. Supplicatio Vestae. 
Dec. 15. XVIII K. Januar. Eo die a[r]a Fortunae reducis dedicatast, quae 

Caesarem A[ugustum ex transmarijnis provinciis red[uxit]. Supplicatio 

Fortunae reduci. 
Jan. 7. VII idus Januar. E[o die Caesar] primum fasces sumpsit. Suppi- 
catio [sic] Iovi|sempi[terno]. 
Jan. 16. [XV]III [sic] K. Febr. Eo di[e Caesar Augustujs appellatus est. 

Supplicatio Au|gusto. 
Jan. 30. [Ill K. Febr. Eo die ara Pacis Aug. dedicata] est, Supplicatio 

impe|rio Caesaris Augusti cust[odis] [civium Romanorum orbisque ter- 

rar]um. 
Mart. 6. [pridie Non. Mart. Eo die Caesar pontifex majximus creatus est. 

Supplicato [sic] Vestae, dis pub(licis) P(enatibus) p(opuli) R(oman). 

Q(uiritium). 
Apr. 14. [XVIII Kal. Mai. Eo die Caesar primum vicit Suppli]catio Vi- 

ctoriae Augustae. 
Apr. 15. [XVII KaL Mai. Eo die Caesar primum imperator app]ellatus est. 

Supplicatio Felicitati imperi. 
Mai. 12. [IIII id. Mai. Eo die aedes Martis dedicatest. Supplicajtio Molibus 

MArtis. 
Mai. 24. [Vim K. Jun. German ici Caesaris natalis. Supp]licatio Vesue. 
Jul. 12, [IIII id. Jul. Natalis divi Juli. Supplicatio Iov]i, Marti ultori, 

Veneri [genetrici] . . . [Suppli]catio lovu 

£. G. SlHLSIL 



CORRESPONDENCE. 

Sebaste, Turkey in Asia, August lo, 1883. 

Sir : 

Those of your readers who are interested in the exploration of Greek antiquity 
in the East, are no doubt acquainted with the work in Asia Minor during the 
past three years, of Mr. W. M. Ramsay, of Oxford. But for the information of 
those who are not familiar with the archaeological doings of the day, it may be 
necessary to state that the object of Mr. Ramsay^s journeys is to illustrate the 
history of Asia Minor by a careful and comprehensive study of all the existing 
remains of antiquity. Accordingly copies and impressions of inscriptions are 
made, the sculptured monuments are drawn and described, coins of cities are 
collected as far as feasible, and in short, the aim is to do everything that will 
in any way illustrate or increase our knowledge of Greek and Roman antiquity 
in Asia Minor. Last winter in Athens it was, of course, well known in archae- 
ological circles that Mr. Ramsay contemplated an extensive tour in Asia Minor 
during the present summer, and it was also known to a few that two members 
of the French school at Athens had a similar archaeological journey in view. 
Mr. Ramsay was asked to give a minute description of the route he proposed 
to take, in order that the Frenchmen might traverse a different district and 
not interfere in the least with the English expedition. In explanation of what 
follows it is perhaps proper to state that the writer was invited to join the 
English expedition. 

The Frenchmen left Saraikieui, the present terminus of the Ottoman Rail- 
way, about ten days before we did. One of the two became ill and returned 
to Smyrna, leaving M. Paris to continue the journey alone. We had not pro- 
ceeded far, when we were astonished to find ourselves upon his trail. We 
could see what he had done, or more properly speaking what he had left 
undone. We noticed that he left whole series of villages unvisited and unex- 
plored. Even in places he had touched we saw that much had been left 
undone that might easily have been done with but small expenditure of time 
and energy. For instance, in one village we copied nine inscriptions which 
M. Paris had left untouched. This we knew because some digging and adjust- 
ment of fragments was necessary in order to read the^ inscriptions. All this 
very apparent hurry led us to believe that M. Paris was heading for the eastern 
country, and took cognizance of things in intermediate districts simply en 
passant. 

But it appears, as will be seen presently, that this was intended for earnest 
work in the line of archaeological research. After a time we passed through 
the villages Sevaslee and Seljikler, in the neighborhood of which the ancient 
Sebaste was situated, its name being preserved in the modem Sevaslee. Ten 
days after our visit to these villages the July number of the Bulletin de Corres- 
pondance Hell^nique— the publication of the French school at Athens — came 
to hand. This July number was published in advance of the May and June 



376 AMERICAN JOURNAL OF PHILOLOGY, 

numbers ; at least the latter had not yet reached the subscribers in Smyrna. In 
the July number we found a paper on the inscriptions of Sebaste, by M. Paris. 
He had brought his short excursion to an abrupt termination and had gone to 
Smyrna to publish the results of his journey. A glance at his paper sufficed to 
reveal to us errors in almost all of the inscriptions. We were thoroughly con- 
vinced of the accuracy of our readings, but when, in the course of our zigzags. 
we found ourselves once more in the neighborhood of Sevaslee and Seljikler, 
it was made convenient for the writer to revisit those villages in order to verify 
our own readings as well as those of M. Paris. Besides this, impressions of the 
stones were made, so that the accuracy of the statements which follow may be 
easily verified. The long inscription published by M. Paris bears the date 99 
A. D., and most probably marks an era in the Hellenisation of the city of 
Sebaste. The yepovoia was a feature of Greek cities, and as the cities of the 
interior became Hellenized they adopted, among other things, the institution 
of the yepovaia. So that on the whole it may be safely assumed that the stone 
was erected in commemoration of the organization of the ytpovaia in Sebaste. 
It may, however, be noted that the interpretation of M. Paris is altogether 
different 

In lines i-a, M. Paris reads *AaK?.ii7rtd6ov 'Epfioyhovg, It is true that here 
the stone is much worn, but nevertheless *A<rKXifinddw tov 'Epfioyivovc may be 
easily distinguished. In line 4, M. Paris reads [ — Jlva tov koI 'Avruviov, but 
the stone has IIoTrd tov koI *AvTuviov, In line 29, col. I, he reads Mip>6^tXog 
B^vridoc ^hoti EvirdTopo^, It is true that here M. Paris has the correct reading 
so far as the letters themselves are concerned, if I may except a distinct and 
unmistakable dot both before and after the B of his word SXtrctSog. Still that 
he understands the signification of the letters is a daring assumption. I need 
scarcely mention that Mrjvdfi^ ff is the short way of indicating that the man 
in question bore the same name as his father, or in other words it stands in 
place of VLnvd^OM^ Mtfuo^iXov, It was usual to write the second name of a man 
after that of his father, so that the passage under discussion is clearly M^v^Aoc 
<Jtf AlirwJof ^haei Evn-dropof. In line 31, col. I, M. Paris reads 'lAiytw ; the 
stone has ^^ytjv. The down-stroke of the ^ is bold, while the circular part is 
quite small, but it is very plain even in the impression. In lines 40-41, M. 
Paris gives up the contest and reads : 

*AX^fav<Jpof MeXiTuvof AON 
AidSupog Savdlmrov Teivoc, 

This apparent difficulty is easily explained. The stonecutter inserted the latter 
half of the word Aovyelvo^ under the AON in line 41, seeing that the space in 
line 40 was limited. The -.yetvttc is not horizontal, but runs at a small angle 
upwards, from which it is clear that line 40 must read *A2I^avdfH)g McX/r^woc 
Aovyelvoc, 

In line 46, coL i, M. Paris reads Ti/uof AdSiJv; the stone has Tifuo^ AdSiw, 
The name Tkpuo^ sounds queer enough, and may be a mistake on the part of the 
stonecutter for FiX^^of, but the M is certain. In lines 41-43, col. 3, M. Paris reads 
Ilairdf ^liTKOKp'tTov tov ndi Noi^dvov, and assures us that both his copy and his 
impression have Novrdvov, not Movrdwv. This is doubtless true, but the stone 
and my impression both read unmistakably Movrdixw, and just as unmistakably 



CORRESPONDENCE. 



377 



MtjvoKplrov instead of the 'iTnroxptrov of M. Paris ; that is, Uarrac 'bitivoKpiTov tov 
ical MovT&vov, 

For the sake of completeness and easy reference the text of the inscription 
is inserted here. 

'Ayad^ Ti'xv 
'Erowf piry enl Upiuv ^ AaK,1ffiridSov 
TOV '"Bpfioyhovi Koi AdvOov ^ Aprifiuv^oc] 
ol iatW6vTtq [e]<f r^ yepovaiav • 
QeoyivfK Tiana tov kqI 'AvTiJviov, 
5 *AptaTQVii TXbiunnCt Aiddupof 'IiriroSdfiov^ 



Column I. 
JAhavdpo^ Aioinxjiov, 

Ai66ijpoi Mrfvoi^iXov^ 

Auwboio^ ^Apurriov^ 
10 *HA<af *ATTo?.Xuviov, 

QeoyivTig Qeoyivovi *MTp6c, 

*IniT6vetJcoc *ApTefud6p0Vf 

Aiovifoto^ Niyepof , 

Add^ *AXcf dvdpov, 
15 ^ip/iitoc Jlarpo/cAiovf, 

Mdc^oc 'iTTtroveixov, 

J^etKoaiuv BhX^uvoc, 

"Eb^paaro^ Kaiaapo^, 

MdpKo^ OvaXtpto^ Kpia^ aTpaTi6T7f^, 
20 *A6avic 'Ajre^Aa, 

'Epfioyh/K Newtaa/wvof, 

Koi-apTO^ ^AnoX- 
Xuviov, 

M6cxoc "Hevdvdpov, 
25 llaTpokX^ AioSupov, 

MAjj^fof 'A^reP-Aa, 

*F6fiaii ^ AirnoTikuvioVf 

Hpurag 'Avrf^ovrof, 

M)7v<5^Xo( . B . AiniSoi 
30 ^Ujet Evndropoc, 

M. Ova?^ptoi ^XtyuVf 

'ArraXoc 'ApTefudCtpov^ 

rdiof OixtXiptoc A6[v]yoc OTpaTUniKt 

MeAircjv K<6xov, 
35 Xioivtf^ Mcvcerrpdrov, 

NeMoc TipoOiov, 

Tdioc KapfieiXic Tatov vld^, 
^pig, MtdpaddTffif 

Afffi^Tpeiof 'A^nra, 
40 *AXi(av6po^ MeAircnvc Aov- 



COLUMN II. 

A<ov('(7M)f AiowoloVt 
'AfficAdf Kl^ov, 
OevcJdf *Apf^aCov, 
Mo(7;ifdi' Meve(rrp4rov, 
Meveffdevc 8ewJd, 
^ ATjkqovdpo^ GcvcJd, 
fJLrjvdKpiTo^ fi.dvdov^ 
'AvSpuv AioddpoVf 

*lirir6viK0^ 'Avr/^wvrof, 
'AAi^avrfpof Mevdv<^/90V, 
rA{r«JV ^ApUJT^ldo^f 
'AvTt^Cw TlpuTopdxov, 

r. 'loC'XiOf np<5K?.of, 
KAavdm 1ei<6pavTi^^ 
r. *Io{fX/of ITpd/cAof vMf , 
'lovXIa *IovA/av7 OvydTtfpt 
r. 'loi'AiOf n/3<5/cAof AlAmv^, 
r, ^loiTuoc TepfMvdc, 
Moaxdc *A:rjrd, 
'AvrvXAof ^ihrrrdTopoc, 
Aiovhcioc AiovV' 

alov ^hioviavd^f 
'flXof 'AA^tof 'ATrep, 



^Aptarkaq AiowaioVf 

KpdTtK 'Itt- 

ffotJd^v, 
Neoco^df 2<iA(.rvoc, 

*A7reX?.df 

*AircAXd, 
EvfUv^ 'AffoX- 



37B AJtERICAX JOURNAL OF PHILOLOGY. 

Counm I. Column II. 

Me/^ruar MfAiruvof, *Av6piac Mdp- 

*It»-Ua Tm^Doiffrif J IlorpoitXiovf, 



IS modi doabt and uncertainty about the functions of the priests and 
p f ie & te ssc s of the Roman Emperors, as well as concerning the date of the 
mstitatioB of the office. Consequently all inscriptions which bear on this 
intercstiiig point are of great importance, especially if they are dated, as is the 
case with the following inscription. The date of an inscription is always im- 
pQftuit, and it is in regard to the date of this inscription that M Paris has 
siDBed grierously. The stone has BT0T2 • 2119 in large and perfectly preserved 
letters, with an unmistakable dot between the two Ts. This makes the date 
105 A. D., whereas M. Paris leaves out the 2 of the date and reads JIG, that is, 
$ A. D. Thus he misses the truth by a matter of 200 years. 

Kara rd iro?Xdiuf dd^avra 
1*9 PovX^ Koi r^ S^fUf} 
MefAfiiav 'Apiartfv TevOpaV' 
rtda dpxtipetav r^ 'Affiac 
ol iStoi OpeiTTol Trap* iavruVf 
hniif^hfoafihnv K. Mefi/tlov 
Kipov Tov Tpo^og ovn^f • 
irovg <nr^, fiJf(vdi) *«'» «' * 

Ttke inscription is in the wall of the minaret of the mosque. It is noteworthy 
that on a ground stone of this same minaret there is another inscription, not 
tes feet distant from the one given above. For this inscription we look in 
vaia in the paper of M. Paris. 

In the following inscription M. Paris reads iwtjjaafthnK instead of irpovoffoa- 
(uvffg. The letters votj are in ligature, so that the v and the 7 are combined, 
and above them is a small o ; the tr and p were also most probably in liga- 
ture, although in my copy I find n alone. 

[•Aya]^ rixv 

*H pov^ Kol 6 S^fioc irtlfojatv 
W^VTov) Mifi/uav XapiSrf/iov 
TeWpavra, ^Acia^ hpxttpkuv 
i/yovcVt npf^ ipufTov /t^opd, 
TiTf &vaaT6eeiJC n[p]ovoifoa/thnK 
IraretXiac KaXkty&tnfc nTf 
fOfTpdc airrov ' 
trove TKff, fuf(vdc) 9. 
Dat« 945 A. D. 



CORRESPONDENCE. 379 

In the following inscription in the yard of the Mussafir Odah of Seljikler. 
M. Paris wonders at the strange form TravroTrwAjyic, Unfortunately the stone 
has iravroi^lSht^ pare and simple. 

Not having the Bulletin beside me, I do not know what M. Paris has made 
of the praenomen in line i. The stone is broken at the commencement for 
the space of three letters. The first letter is gone ; the second is probably an 
N, the third is either a T or a F. The praenomen may be 'Avr., but cannot be 
Av/J. 

pAvr ?] n&^AAiuy travroTri&AjTf 
avr(^ Kai r$ ywaud Avp. 
*Afifu^ Ztjvod&rov koX rolq 
riKvoif avToii KoreaKthaafv 
^uv rd i^/i6)ov • ei 6k ti( irepov 
iniaevivK^ Tivd iare avr^ 
irpbg rdv 6e6v, 
irovg Tft\ fiti{vdg) ff, k\ 
Date 2S6 A. D. 

J. R. S. Stbrrbtt. 



RECENT PUBLICATIONS. 

TbttiiB are doe to Messrs. B. Westemunn & Co., New York, for material 

BRITISH. 

AristopliajMs. Acharnians. Translated into English Terse by Robert Yel- 
votiM TyrrelL (Dablin UniTersity Press Series.) Post 8vo, pp. 66. Hodges^ 
Dublin. Ltmgmuu. 3s. 6d. 

Aristode. Politics. Translated, with an Analysis and Critical Notes, by 
J. E. C Welldon. Post 8to, pp. 48a MacmiUan, los. 6d. 

OceiQL Pro Publio Sestio Oratio ad Judices. With Introduction, Explana- 
tory Notes, and Critical Appendix, by the Rey. Herbert A. Holden. i2mo, 
pp^ 55& Maamttan» 5s. 

Dobffee*s Adversaria cum Pnefatione G. Wagneri. Volumen 3, Miscellanex 
OiMervatiofies advarios Scriptores Grsecos. (Bohn's Coll^iate Series.) i2mo. 
B<a «Sr* iSwd. 5S. 

Euripides. Hercules Furens. By F. A. Paley. Fcap, Svo. Bell 6r* Sms, 
ts.6d. 

Gddart (E. M.) A Guide to Modem Greek. Post Sto, pp. 286. (Key, pp. 
a^ as. 6d.) TrAbner. 7s. 6d. 

Homer's Odjrssey. Book i, edited for the Use of Schools, by Rey. John 
Bond and A. S. Walpole, with Notes and Vocabulary. (Elementary Classics.) 
i3ao^pp.i2a MacmUkm, is. 6d. 

Horace. Odes. Edited, with Introduction and Notes, by T. E. Page. lamo, 
pp. 484. MQcmilkm, 6s. 

^— » Satires. Edited, with Notes, by Arthur Palmer. i8mo, pp. 436. 
Macmillan^ 65. 

King (Rey. James). Cleopatra's Needle : A History of the London Obelisk. 
Crown 8vo, pp. 128. ReUgious Tract Society, 2s. 6d. 

Ltvy. Book I. Edited, with Notes and Vocabulary, for the use of Schools, 
by Rtiv. H. M. Stephenson. (Elementary Classics.) i8mo, pp. 160. Macml- 
.'ojft. is. 6d. 

Outlines of the Philosophy of Aristotle. Compiled by Edwin Wallace. 3d 
qU. KcAp. Svo. Cambridge Warehtmse, 4s. 6d. 

Seliar ^. Y.) The Roman Poete of the Augustan Age— Virgil 2d ed. 
?o»t 8vv>, pp. 43a. Frowde, 9s. 

Souuen^cht^in (E. A.) Bentley's Plautine Emendations, from his Copy of 
Uioaovms. Small 4to. Oxford Warehouse, 2s. 6d. 

Whiuhcia (E. H.) The Quatrains of Omar Khayyam. The Persian Text, 
Hiih an English Verse Translation. (TrQbner's Oriental Scries.) 8yo, pp. 
\40. fyubfur. los. 6d. 

\cuophou*s Memorabilia. Book i. With Notes by Rey. C. E. Moberley. 
aU o4. V cup. 8vo» pp. 82. Rivingtons, as. 



RECENT PUBUCATIONS. 38 1 

DUTCH. 

Gericke (J. F. C), en Roorda (T.) JaTaansch-Nederduitsch woordenboek, 
venneerderd en verbeterd door A. C. Vreede. i Aft. Amsterdam, 1883. 8vo, 
p. 1-28. 5m. 50. 

FRENCH. 

Adam (Lucien). Les Idiomes n6gro-aryen et maleo-aryen, Essai d*hybrido- 
Ic^ie linguistique. In-8. Maisontuuve, 3 fr. 50. 

Arbois de Jubainville (H. d'). Essai d^un catalogue de la litt^rature ^pique 
de rirla'nde, precede d'une etude sur les manuscrits en lang^ue irlandaise con- 
serve dans les lies britanniques et sur le continent. In-8. Thorin. 12 fr. 

Baret (Adrien). l^tude sur la langue anglaise au XIV si^le. In-8. Cerf, 

Chandos. Le Prince noir; po^me du h^raut d*annes Chandos. Texte 
critique, suivi de notes, par Francisque- Michel. In-4. Texte en vieux fran- 
9ais ; traduction anglaise, preface et notes en anglais. Fotfuringham, 55 fr. 

Corpus inscriptionum semiticarum ab Academia inscriptionum et litterarum 
humaniorum conditum atque digestum. Pars prima, inscriptiones Phoenicias 
continens. Tomus I, fasciculus 2. In-4, avec album de planches in-folio. 
(Impr. Nationale.) Klifuksieck, 25 fr. 

Darmesteter (James). Essais de litt^rature anglaise. In-12. Delagrave, 
3 fr. 5a 

£tudes sur TAvesta; observations sur le Vendidftd. In-8. Extrait 

du Journal asiatique. MaUonneuve, 3 fr. 50. 

Duperr^ de Lisle (Mme. F.) ^tude sur la prononciation fran^aise. Avec 
deux letlres de M. Legouv^ et M. Got. In-12. Delagrave, 2 fr. 75. 

Henry (Victor). Etudes sur I'analogie en general et sur les formations ana- 
logues de la langue grecque. Or. in-8. Maisonneuve, 8 fr. 

Des Michels (Abel). Les Po^mes de I'Annam. Luc Vftn Tifin Ca Dien. 
Or. in-8. Leroux, 20 fr. 

Ptol^^e. Claudii Ptolemaei geographia. E codicibus recognovit, prole- 
gomenis, annotatione, indicibus, tabulis instruxit Carolus Mullerus. Voluminis 
primi pars prima. Gr. in-8. F, Didot, 15 fr. 

Schweisthal (Martin). Remarques sur le rdle de I'element franc dans la for- 
mation de la langue fran9aise. In-8. Leroux, 2 fr. 

GERMAN. 

Abel (Dr. Carl). Ob. den Gegensinn der Urworte. gr. 8, iii, 65 S. Leipzig, 
1884, Friedrich, n. 2. 

Aristophanis Thesmophoriazusae. Rec. Adph. v. Vclsen. gr. 8, 88 S. 
Leipzig, Teubner, n. 2. 

Aristotelis quae feruntur Magna Moralia. Recognovit Franc. Susemihl. 8, 
xix, 126 S. Leipzig, Teubner. 1.20. 

Beitr&ge zurKunde der indogermanischen Sprachen, hrsg. v. Dr. Alb. Bezzen- 
bcrger. 8 Bd. gr. 8. (i u. 2 Hft. 184 S. m. i autogr. Taf.) G5ttingen, Pepp* 
mUlUr, n. 10. 



RECENT PUBLICATIONS, 383 

Cicero's Rede Qb. das Imperium d. Cn. Pompejus. Hrsg. v. Fr. Richter. 3 
Aufl. V. Alfr. Eberhard. gr. 8, 76 S. Leipzig, Teubner, — 60. 

Cicero's Rede f. L. Flaccus. Erkl&rt ▼. Adf. du Mesnil. gr. 8, ti. 235 S. 
Leipzig, Teubner, 3.60, 

Clasen (Chm.) Historisch-kritische Untersochnngen Ub. Timaios v. Tauro- 
menion. gr. 8, 97 S. Kiel, Lipsius 6* Tischer, n. 2.40. 

Classiker, deuUche, d. Mittelalters. Mit Word- a. Sai^herkUrgn. BegrQndet 
V. Frz. Pfeiffer. 12 Bd. 8. Leipzig, Brockham, n. 3.50; geb. n. 4.50. In- 
halt : Erz&hlungen u. Schw&nke. Hrsg. t. Hans Lambel. 2 Aufl. xv, 374 S. 

Commentationes philologae Jenenses, edd. seminarii philologomm Jenensis 
professores. Vol. II. gr. 8, iii, 327 S. Leipzig, Te$ibner, n. 6. (I et II, 
n. II.) 

Corpus inscriptionum latinamm consilio et auctoritate academiae litteramm 
regiae Borussicae editum. Vol. IX. Fol. Berlin, (7. RHmer, cart n. 90. 
Inhalt : Inscriptiones Calabriae, Apuliae, SaniniL Sabinorum, Piceni latinae, 
ed. Thdr. Mommsen. (Ixix, 52 u. 847 S. m. eingedr. Fig. u. 4 chromolith. 
Karten.) 

- VoL X. Ed. Thdr. Mommsen. 2 partes. Fol. Berlin, G, RHmer, 
cart. n. 124. (I- VIII et X, n. 795.75.) Inhalt: I. Inscriptiones Bnittiomm, 
Lncaniae, Campaniae. Ixi, 712 S. 2. Inscriptiones Siciliae et Sardiniae, ed. 
Thdr. Mommsen. (S. 713-1229 m. I Lichtdr.-Taf. u. 5 chromolith. Karten.) 

Dehner (Seb.) Hadriani reliquiae. Particula I. Diss. gr. 8, 43 S. Bonn, 
Behrendt, n. 1.20. 

Dieterici (Fr.) Die Abhandlnngen der Ichwin Es-SafEl in AuswahL Zum 
ersten Mai aus arab. Handschriften hrsg. gr. 8, y, 172 S. Leipzig, Hinrichs* 
Veri, n. 6. 

Die sogenannte Theologie d. Aristoteles, aus dem Arab. Qbers. u. m. 

Anmerkgn. versehen. gr. 8, xviii, 224 S. Leipzig, Hinrich^ VerL n. 8. 

Dillon (Dr. Emil J. v.) Die Umschreibung der eranischen Sprachen. 8, z6 
S. Leipzig, Gerhard, n. — 60. 

Duncker (R.) De Paeanio Eutropii interprete. gr. 4, 21 S. Greiffenberg 
i. Pomm., 1880. Berlin, Mayer b*MuUer, baar n. I. 

Inter prinatarum causarum orationes Demosthenicas quae pro genuinis 

habendae sint quaeque pro falsis breuiter exponitur. Pars I. gr. 4, 17 S. 
Ebd. 1877. baar n. i. 

Engelbrecht (Aug. Gfr.) Studia Terentiana. gr. 8, 90 S. Wien, GeroUts 
Sohn, n. 3. 

Euclidis opera omnia. Edd. J. L. Heiberg et H. Menge. (Vol. I.) Ele- 
menta. Ed. et latine interpretatus est Dr. J. L. Heiberg. Vol. I. Libros 
I-IV continens. 8, x, 333 S. Leipzig, Teubner, 3.60. 

Ewald (Paulus), et Loewe (Gust.) Exempla scripturae visigoticae XL tabulis 
expressa. Fol. viii, 30 S. Heidelberg, KosUr, In Mappe. baar n.n. 50. 

Fellmann (Max.) De ^, liort particulis consecutivis earumque apud tragicos 
Graecorum poetas yi et osu. Diss. gr. 8, 80 S. Breslau, JCdhUr, baar n. i. 

Fokke (A.) Rettnngen d. Alkibiades. i Tl. Die sicil. Expedition, gr. 
8, IT, 87 S. Emden, MayneL n. 1.75. 

Frantz (Joh.) Die Kriege der Scipionen in Spanien 536-548 t. u. c. gr. 8, 
T, 77 S. MQnchen, Th, Ackermann, n. 1.60. 



^t$, -/•- i» r/. 



m^ '' 



(■*-. 




C'-itttm^r ?, iC.-r.'*-Ae 3«-aKrkanqea ana T^xe 1. L^^ibk. gn^^tzS. 
^/'>t:.i»Ann O.'' . Z«^ tarrxn ^JXSbc -vocaiuur ZjcsaraBae aadflBaat ToIHaii- 

^/•^•-mafin 'ri*"n • Ik? nr.nae nxuic Acschiiiia 0/:MpriontB»e anK oas caa- 
M^y\^iir,\* .t^xfre^iWf ric-'^a 11. Dvt*. jr. 5. 45 i Brtslaa, XJ4uier. bsarn.!. 

ftikf\e7, fC, tiey Im T'txrzhvt et -lit la. correctioii des testes ]vestxq;Bes. p- 
%, tv't. t*,h ft, I>e,ptf*it Cerkard. ■, 6, 

fttftm»n <f^, J. ].} 9\ta4-A AtiTr^hoctOL gr^^j^S. Logdiuu Bat. l3S2. 
t>f] fi, Otlvttry ^ ^, boar *, J, 

M'fvlMni ah **r^^<a di^i Marci libri VIII, ed. Laiw. Memiielssoha. gr. S, 
«f I, *^^ ft, I'^'J'^tfi Teubfur. ». 6.80. 

\inf(A<^tf%, ▼, K, Abicht, 3 Bd. Bach V b, VX 3 rob. AttiL gr. 8, ir, 224 
ft \^s\it\^, Ttuhur. I, So. 

M^Xftf (Max), Dc legation ibuf atttcU. Dis«. gr.8,73S. Cdttuges, i832, 
Vtipttf^nhifffk ^ ffuprtcht. baar n. I.3o. 

f f )ttm«nn (Frflr.) f>e arte critica in Orphei Aigonaattcts fiictitanda capiu 
111 I))m. gr S, 74 S. Leipzig, Matthts, baar n. I. 

IfirMbfi-M (Otto). Oaltitche Stadien. [Aus ** SiUnngtber. d. k. Akad. d. 
Wl^*/' j I.#'x,-I, ^K) S. WIcn, GeroUTs Sohn in Camm, n.n. —9a 

lli%(orl(oriim Romanorum fragmenta, collegit, diiposuit, rec Henn. Peter. 
H, RirvHI, 49fl H. Kfipxig, Teubtter, 4.50. 

Il(ifliii(l'nul). DrcodlceBtamlinloantiquissimo. Diss. gr.8,55S. Jena. 
/VA/''. n. I. 

llufTiitniin (Olto Adb.) l)e Imperatoris Titi temporibus recte definiendis. 
I>N*. Ht. R, 34 S. Marburg. Elwert^s Verl, n. i. 

Ilomfi** llirtu, V, J. La Roche. I Tl. Gesang I-IV. 3 verm. u. verb. Aufl. 
gi, 9, mIU. IB8 S, Lripylg, TtHhutv, t.50. 

tlub«rhmnnn (U.) Armeninche Studien. I. GrundzOge dcr armen. Etymo- 
lonU. I Thl. gr, 8« vl, loa S, Lcip/ig, Btrttkopf b* HdrUl n. 3. 

JiO»ibHih» biom»phi«ohts, f. AUcrthuintkundc. Hrsg. t. Prof. <3onr. Bursian. 
%, jMhig . I9{^f. |;u 8, (v. Ill S. Berlin, C*%h\iry ^ Co. n. 3. 



RECENT PUBLICATIONS. 3«5 

Jebb (R. C.) Die Reden d. Thukydides. Autoris. Obersetzg. v. J. Imel- 
mann. gr. 8, Hi, 65 S. Berlin, W, Weber, n. 1.60. 

Keller (Otto). Der satumische Vers als rythmisch erwiesen. gr. 8, 83 S. 
Prag, Tempsky, Leipzig, Freytag, n. 1.5a 

Kirchhoff(Fr. Chr.) Neue Messungen der Oberreste vom Theater d. Dio- 
njTsus zu Athen. gr. 4, 7 S. Altona, Schliiter, n. i. 

Klimek (Paul). Conjectanea in Jalianum et CyrilU Alexandrini contra ilium 
llbros. Diss. gr. 8, 42 S. Breslau, Kohler, baar n. I. 

Knaack (Dr. Geo.) Conjectanea. gr. 4, 11 S. Berlin, Wiidwtamt, n. 1.5a 

Koch (L. G.) De principe javentutis. Diss. gr. 8, 47 S. Leipzig, Lorentg, 
baar n. i. 

Koepp (Frdr.) De gigantomachiae in poeseos artisque monumentis usn. gr. 
8, 66 S. Bonn, Sirauss, n. 2. 

K6hler (Carl Sjlvio). Realconcordanz der Sprfiche u. Lehren in den Tra- 
gOdien d. Aeschjlos, Sophokles, Euripides, gr. 8, x, 200 S. Halle, HendeL 
n. 5. 

Kopp (Arth.) De Ammonii, Eranii, aliorum distinctionibus synonymicis 
earumque communi fonte. Diss. gr. 8, 108 S. KOnigsbeig, Gr&fe &* Clfuer, 
baar n. 2. 

Kraffert (Herm.) Beitrftge zur Kritik u. ErkUurung lateinischer Autoren. 
gr* 8, 153 S. Aurich, Reeuts in Comm, n. 3.60. 

Kremer (A. Frhr. v.) Beitriige zur arabischen Lexikographie. [Aus '* Sits- 
ungsber. d. k. Akad. d. Wiss.**] Lex.-8, 92 S. Wien, GerMt Sohn in Comm, 
n. 1.50. 

Kukula (Rich.) De tribus pseudacronianorum schoUorum recensionibut. 
gr. 8, V, 49 S. Wien, Konegen, n. i. 

Lehnerdt (Max.) De locis Plutarchi ad artem spectantibus. Diss. gr. 8, 
46 S. K6nigsberg, Gr&fe 6* Unur, baar n. i. 

Lemrn (Osc. v.) Aegyptische Lesestiicke zum Gebrauch bei Vorlesungen a. 
znm PriTatstudium. Mit Schrifttaf. u. Glossar. I Thl. Schrifttafel u. Lese- 
stQcke. I Hft hoch 4, 64 autogr. S. Leipzig, Hinrichs* VerL n. 8. 

Liebt (Hans). BeitHlge zu den Persius-Scholien. gr. 8, 54 S. Straubing, 
AUenAcfer, n. 1.50. 

Literatur-Blatt f. orientalische Philologie, unter Mitwirkg.v. Dr. Jobs. Klatt, 
hrsg. ▼. Prof. Dr. Ernst Kuhn. i Jahrg. Octbr. 1883 to Septbr. 1884. 12 Hfte. 
(k 2-3 B.) gr. 8. Leipzig, O, ScAulu. n. 1.5a 

Livii, Titi, ab urbe condita liber XXII. ErkUrt v. Ed. W6fflin. Mit e. 
(lith.) KArtchen. 2 Aufl. gr. 8, vi, 102 S. Leipzig, Teubner, 1.20. 

Lucian, ausgewfthlte Schriften. Fiir den Schulgebrauch erkUrt v. Dr. Karl 
Jacobitx. 2 Bdchn.: Die Totengesprftche. Ausgewilhlte G6tteigesprftche. 
Der Hahn. 2 mehrfach bericht. Aufl. gr. 8, 150 S. Leipzig, Teubner, 1.2a 

M&itrftyanl SamhitA, hrsg. von Dr. Leop. v. Schroeder. 2 Buch. Gedr. auf 
Kosten der deutschen morgenlftnd. GeselUchaft gr. 8, x, 169 S. Leipzig, 
Brockkaw^ Sort, in Comm, (&) n. 8. 

Marcks (Joh. Frdr.) Symbola critica ad epistolographos graecos. Diss. gr. 
8, 54 S. Bonn, Behrendt, n. 1.2a 

Matzat (Heinr.) ROmische Chronologic, i Bd. Grundlegende Untersuchgn. 
gr. 8, xii, 354 S. Berlin, Weidmann, n. 8. 



386 AMERICAN JOURNAL OP PHILOLOGY. 

MAyerfa&fer (Ant) Die BrOcken im alten Rom. gr. 8, iii, g6 S. Exlaagea, 
DeicherU n. 2. 

Mirsdi (Panl). De M. Terend Vairoois antiqnitatsfli rentm Tinmamram 
libris XXV. Diss. gr. 8, 144 S. Leipzig. HirteL buu- a. 2. 

Mittheilmigen d. deatschen mrchlologischen Institutes in Athcn. 8 Jmhig. 
gr. 8. (I Hft 80 S. m. 3 Stein-, i Lichtdr.-Ttt a. 2 BeiUgen.) Athen. Wil- 
herg in Comm» n. 15. 

Mittheilnngen, archaeologisch-epignpliische, ans Oesterreicb, hisg. t. O. 
Benndorf a. O. Hirschfeld. 7 Jahrg^ 1883. gr. 8. (i Hft. 152 S. m. 3 Tal) 
Wien, G<roUfs Sohn im Camm. n. 9. 

Mommsen (Ang.) Cbronologie. Untenochungen Qb. das Kalendenresen 
der Griechen, insonderheit der Athener. gr. 8, riii, 532 S. Leipzig, TetAner, 
n.T4. 

Mommsen (Tycfao). Griechische Formenlehre. gr. 4, 48 S. Frankfaita. 
M. Leipzig, Teubner, n. i. 

Mordtmann (J. H.), u. Mailer (D. H., DD.) Sabftische Denkmftler. Mit 8 
photozinkogr. Taf. [Ans **' Denkschr. d. k. Akad. d. Wiss.'*] Imp.-4, 114 S. 
Wien, Ceroid* s Sohn in Comm, n. 9. 

Miiller (Job.) Der Stil d. &lteren Plinios. gr. 8, xi, 158 S. Innsbrack, 
Wagner, n. 4. 

Mnnier (Mor.) Die Palftographie als Wissenscbaft u. die Inscbriften d. 
Mainzer Mosenms. 4, 29 S. m. i Taf. Mainz, Diemtr. n. i. 

Neubaenser (Prof. Dr. Jos.) Anaximander Milesius siye vetnstissima quae- 
dam remm uniyersitatis conceptio restituta. Cum I tab. (litb.). gr. 8, xri, 428 
S. Bonn, Cohen ^ Sohn, n. 14. 

Neumann (Dr. Karl Jobs.) Strabons Landesknnde ▼. Kankasien. Eine 
Quellennntersocbg. [Aus "Jahrbb. f. d. class. Pbilol. 13 Suppl.-Bd."] gr, 
8. 36 S. Leipzig, Teubner, n. z. 

Noetbe (Heinr.) De pugna Maratbonia quaestiones. Diss. gr. 8, 71 S. 
Susati, 1 88 1. Leipzig, Pock, baar n. 1.20. 

Nottrott (Miss. A.) Grammatik der Kolb-Spracbe. gr. 8, 104 S. Gaterslob, 
1882. Berlin, Buchh. der Gosmei^schen Mission, baar 1.50. 

Oikonomos (S. A.) 'H v^oof Wtn&pffioq, Diss. gr. 8, 32 S. Jena, Pokle. 
baar n. — 75. 

Pappageorg (Pet. N.) Codex Laurentianos v. Sopbokles u. e. neue Kollation 
im Scholien-texte. [Aus ** Jabrbb. f. class. Pbilol. 13 Suppl.-Bd."J gr. 8, 40 
S. Leipzig, Teubner, n. i. 

Paucker (C.) Supplementum lexicorum latinorum. Fasc. 2. gr. 8, S. 97- 
192. Berlin, Calvary ^ Co, (it) n. 3. 

Plato's ausgewiUilte Dialoge. ErkUrt v. C. Scbmelzer. 4 Bd. Apologie. 
Krito. gr. 8, 92 S. Berlin, Weidmann. 1.20. 

Platonis opera quae feruntur omnia. Ad codices denuo collatos ed. Mart. 
Scbanz. Vol. VI, fasc. 2. Charmides, Laches, Lysis, gr. 8, viii, 90 S. Leipzig, 
B, Tauchmiz, n. 2. 

— ^ Charmides, Laches, Lysis. Ed. Mart. Scbanz. gr. 8, 72 S. Ebd. — ^45. 

Plauti (T. Macci). Comoediae. Rec. et enarrayit Joa. Ludoy. Ussing. 
Vol. IV pars 2, Pseudolum et Poenulum continens. gr. 8,yiii, 362 S. Hayniae. 
Leipzig, T, O, Weigel n.n. 10. (I-III, I, IV, i, 2, n.n. S6>50.) 



RECENT PUBLICATIONS, 387 

Plotini Enneades praemisso Porphyrii de vita Plotini deqae ordine librorum 
ems libello, ed. Ric. Volkmann. Vol. I. 8, xxxiv, 350 S. Leipzig, Teubner, 
3.60. 

Poetae latini minores. Rec. et emendavit Aemilius Baehrens. Vol. V. 8, 
446 S. Leipzig, Teubner, 4.20. (I~V, 15.90.) 

Porta linguarum orientaliam. Pars I. Ed. Ill u. d. T. : Hebr&ische Gram- 
matik m. ObungsstOcken, Litteratur u. Vokabalar. v. Prof. Herm. L. Strack. 8, 
XV, 163 S. Karlsruhe, Reuther, n. a.70. 

Primer in the Ga or Akra language, Gold Coast, W. Africa. 8, viii, 40 S. 
Basel, Missionsbuckh. geb. n. — 60. 

Probst (Arth.) Beitrftge zur lateinischen Grammatik. I. Zur Lehre vom 
Verbum. gr. 8, 104 S. Leipzig, Zangenberg 6* Himly, n. 3. 

— — Beitrftge zur lateinischen Grammatik. IL Zur Lehre v. den Partikeln 
u. Konjunktionen. gr. 8, S. 105-72. Leipzig, Zangenberg &* Himly, n. 2. 
(I u. II, n. 5.) 

Radloff (Dr. W.) Phonetik der n5rdlichen TQrksprachen. Anderes Hft. 
Consonanten. gr. 8, xxi-xlv u. S. 101-318. Leipzig, T, O, WeigeL n. 6. 
(cplt. n. 9.) 

Reiflferscheid (Aug.) Anecdotum Fulgentianum. 4, 10 S. Breslau, KdhUr, 
baar n. — 50. 

Reinisch (Leo). Die Bilin-Sprache. i Bd. A. u. d. T. : Teste der Bilin- 
Sprache. Mit Unterstatzg. der kais. Akademie der Wissenschaften in Wien. 
gr. 8, viii, 322 S. Leipzig, Grieben, n. 10. 

Rheinhard (Aug.) C. Jul. Caesar*s RheinbrQcke. Eine technisch-krit. 
Studie. Mit 3 (eingedr.) Abbildgn. Lex.-8, 16 S. Stuttgart, Neff. n. —50. 

Rosenstiel (Frdr.) De Xenophontis historiae graecae parte bis edita. Diss, 
gr. 8, 54 S. Jenae, 1882. G5ttingen, Vandenhoeck 6* Ruprecht, baar n. 1.40. 

Ruete (Edm.) Die Correspondenz Ciceros in den J. 44 u. 43. Historische 
Dissertation, gr. 8, v, 122 S. Marburg, Elwerfs VerL n. 2.40. 

Rumpel (Joa.) Lexicon Pindaricum. gr. 8, 498 S. Leipzig, Teubner, n.i2. 

Rutherford (W. Gunion, M. A.) Zur Geschichte d. Atticismus. 2 Abhandlgn. 
tJbers. v. Lehr. Dr. A. Funck. [Aus " Jahrbb. f. class. Philol. 13 Suppl.-Bd."] 
gr. 8, 45 S. Leipzig, Teubner, n. i. 

Ruzicka-Ostoic (Camilla). Transcription d. tQrkisch bearbeiteten Lustspieles 
Ajjar-i Hamza. 8, 90 S. Wien, Lechner's Verl, in Comm, n. 2.40. 

Sallusti Crispi (C.) Bellum Catilinae, bellum Jug^rthinum. Ex historiis 
quae extant orationes et epistolae. Rec. Aug. Schneider, gr. 8, xvi, 130 u. vi 
S. Prag, Tempsky, Leipzig, Freytag, n. I. 

Sammlung englischer Denkm&ler in kritischen Ausgaben. 4 Bd. gr. 8. 
Berlin, Weidmann, n. 7; Ausg. auf Kupferdr.-Pap. n. 10. (1-4, n. 23.60.) 
Inhalt : Wulfstan. Sammlung der ihm zugeschriebenen Homilien, nebst Un- 
tersuchgn. Qb. ihre Echtheit hrsg. v. Arth. Napier, i Abtlg. Text u. Varianten. 
X, 318 S. 

Sammlung der griechischen Dialekt-Inschriften, v. F. Bechtel, A. Bezzen- 
berger, F. Blass, H. Collilz, W. Deecke, A. Fick, G. Hinrichs, R. Meister, hrsg. 
V. Dr. Herm. CoUiU. I Hft. gr. 8. Gdttingen, PeppmUUer, n. 2.50. Inhalt : 
I>ie griechisch-kyprischen Inschriften in epichorischer Schrift. Text u. Um- 
schreibg. m. e. (autogr.) Schrifttaf. v. Lyc.-Dir. Dr. Wilh. Deecke. 80 S. 



3^8 AMERICAN JOURNAL OF PHILOLOGY. 

Sartorius (Max). Die Entwicklung der Astronomie bei den Griechen bis 
Anaxagoras n. Empedokles, in besond. Anschluss am TbeophrasL Diss. gr. 
8, 39 S. Halle. Breslau. KdhUr, baar u. i. 

Sauppe (Herm.) Emendationes Plutarcheae. 4,158. QXAWn^tn^DieterUKs 
Verl, baar n. — 80. 

Schack (Gr. v.) Die Unterweisung d. KOnigs Amenemhat I. I HfllAe, hoch 

4. a I antogr. S. Paris, Viewtg, n. 4. 

Schmitt (Heinr. Ludw.) Quaestiones chronologicae ad Thucydidem perti* 
nentes. Diss. gr. 8, 105 S. Leipzig, 1882, Teubner, n. 1.60. 

Schradcr (Dr. O.) Sprachverglcichung a. Urgeschichte. gr. 8, x, 490 S. 
Jena, C^sknohU, a. 11. 

Schuchardt (Hugo). Ob. die Benguelasprmche. [Aas ** Sitxangsber. d. k. 
Akad. d. Wiss."] I«ex.^, 14 S. Wien, GeroUts Sdkn in Comm. n.n. —3a 

Schort (Dr. Wilh.) De mutation ibns in imperio Romano ordinando ab im- 
peratore Hadriano factis. gr. 8, vi, 68 S. Bonn, Strauss, n. 2. 

Scemann (Heinr.) De asyndeto Sophocleo quaestiones. Diss, gr. 8. 57 S. 
Breslau, 1882, IC^AIfr, baar n. i. 

Sellge (Jul.) De studiis in Sallustio Crispo a Pompeio Trogo et Justino 
epitomatore collocatit. Diss. gr. 8, 96 S. Sagani, 1882. Breslau, KohUr, 
baar n. i. 

Semi Gfammatici in Veigilii carmina commentarii, recensuemnt Geo. Thilo 
et Herm. Hagen. Vol. II, fasc. I. In Aeneidos libros VI-VIII commentarii. 
gr. 8, 306 S. Leipzig, Teubner, n. 10. (I et II, I, n. 34.) 

Sittl (Karl). Geschichte der gnechischen Literatur bis tuf Alexander den 
Grossen, i Tl. gr. 8, Yi, 359 S. Mflnchen, 1884, Tk, Ackemiann, n. 4.8a 

Sophokles. Erklart v. Gust. Wolff. 5 Tl. : Oidipus auf Kolonos v. Ludw. 
Bellermann. gr. 8. ri, 202 S, Leipzig, Teubner, 1.50. (1-5, 6.30). 

Stem (Ernst). Catilina u. die Parteikflmpfe in Rom der J. 66-63. Abhand- 
lung. gr. 8, 178 S. Dorpat, Karow in Omm, n. 3.60. 

Stoll (H. H.) Anthologie griechischer Lyriker. 2 Abtlg. Melische u. chor- 
iscke Lieder u. Idyllen. 5 Aufl. gr. 8, iv, 200 S. Halle, Gtsenius, n. 2.25. 

Studien, indische. Beitrtge f. die Kunde d. ind. Alterthums. Hrsg. y. 
Prof. Dr. Alb. Weber. 16 Bd. gr. 8, 480 S. Leipzig, Broekkam, n. 15. 

Studien, Leipsiger, znr classischen Philologie. Hrsg. ▼. G. Curtius, L. Lange, 
O. Ribbeck, H. Lipsius. 6 Bd. i Hft gr. 8, T92 S. Leipzig, Hirtet, n. 4.5a 

Thiemann (C.) WOrterbuch zu Xenophons Hellenika m. besond. Rflcksicht 
aaf Sprachgebrauch u. Phraseologie. Fttr den Schnlgebrauch bearb. gr. 8, iv, 
112 S. Leipzig, Teubner, 1.5a 

Thiofridi Eptemacensis vita Willibrordi metrica. Ed. Konr. Rossberg. 8, 
xxTii. 128 S. Leipzig, Teubner, 1.80. 

Tobler (Adf.) Die altTenezianische Obersetzung der SprQche d. Dionysius 
Cato. [Aus ** Abhandlgn. d. k. preuss. Akad. d. Wiss. zu Berlin.**] gr. 4, 87 

5. Berlin, DUmsnUf^s Verl, n. 3.5a 

Untersuchungen, philologische, hrsg. v. A. Kiessling u. U. ▼. Wilamowitz- 
MoellendorC 6 Hft gr. 8. Berlin, IVeidmann, n. 3. (1-6, n. 23.40.) In- 
halt: Analecta Eratosthenica. Scripsit Ernst Maas. 153 S. 

Urlichs (L. v.) Pergamenisdie Inschriften. 16 Programm d. v. Wagnerschen 
Kunstinstitutes. gr. 8, 31 S. Wflrzburg, StaMel in Cnnm. n. — 80. 



BOOKS RECEIVED. 389 

Verhandlungen der 36 Versammlnng deutscher Philologen a. Schulm9.nner 
in Karlsruhe vom 27 bis 30 Septbr., 1882. Mit 2 lith. Taf. gr. 4, iy, 334 S. 
Leipzig, Teubner, n. 12. 

Wagner (Jos.) Zur Athetese d. Dialogs Euthyphron. gr. 8. 46 S. Briinn, 
1882, WinkUr, n. I. 

Wassner (Jul.) De heroum apud Graecos cultu. Diss. gr. 8, 53 S. Kiel, 
Lipsiut &* TischiT, n. 1.60. 

Weise (Paul). De Bacchidum Plautinae retractatione quae fertur. Diss, 
gr. 8, 62 S. Berlin, Mayer b* MulUr, baar n. 1.20. 

Wex (Jos.) Die Metra der alten Griechcn u. Rdmer, im Umriss erkUrt u. 
Qbersichtlich dargestelk. 2 Bearbeitg. gr.8,iv,94S. Leipsig, TVf^^M^. 1.50. 

Xenophontis institutio Cyri, rec. et praefatus est Amoldus Hug. Ed. maior. 
8, c, 344 S. Leipzig, Teubfur, 1.50. 

Ziemer (Henn.) Junggrammatische StreifzUge im Gebiete der S3mtax. In 
Abschnitten. 2 Aufl. gr. 8, x, 158 S. Kolberg, Post, n. 2.70. 

ITALIAN. 

Bencivenni (J.) Question! ardenti di riforma scolastica. Torino, 1883. 8vo. 
4m. 80. 

Cima (Ant.) Breye teoria dello stilo latino. Torino, 1883. 8yo, viii, 135 
pp. im. 80. 

Fanfani (Pt.). e Frizzi (Gius.) Nuoyo Vocabolario roetodico della lingua 
italiana. Parte I. Vocabolario domestico. Milano, 1883. 8vo, viii, 874 pp. 
6m. 

Massaroli (Gius.) Phul e Tuklapalasar II, Salmanasar V e Sargon. Roma, 
1882. 8yo. 2m. 50. 

SPANISH. 

Conto (C.) Apuntaciones sobre la lengua inglese. Madrid, 1883. 4to, 342 
pp. 14m. 40. 



BOOKS RECEIVED. 

Altfranzdsische Bibliothek, herausg. yon Dr. Wendelin F5rster. Heilbronn, 
Gebr, Hmmnger, 1883. 

2ter Bd. Karls des Grosten Reise nach Jerusalem, ein altfranzdsisches 

Heldengedicht, herausg. von E. Koschwitz. 4m. 40pf. 

— 6ter Bd. Das altfranzOsische Rolandslied. Text von Chauteauroux, 
nach Venedig VIII, herausg. von Wendelin F5rster, lom. 

Anecdota Oxoniensia. Classical Series, Vol. I, Part IV. Bentley's Plautine 
Emendations, by E. A. Sonnenschein. 

Be6wulf and the Fight at Finnsburgh. With Text and Glossary on the 
basis of M. Heyne, edited, corrected, and enlarged. By James A. Harrison 
and Robert Sharp. Boston, Ginn^ Neath &* Co,^ 1883. 

C<^p (H. N.) United Sutes Salary List and the Civil Service Law. Wash- 
ington, D. C, 1883. 35c. 

Deutsche Liiteraturdcnkmale des 18 und 19 Jahrh. HeilbronUi Gebr, Hen" 
ninger^ 1883. 

— 8. Frankfurter Gelehrte Anzeigen vom Jahre 1772. 3.8a. 

— 12. Vier kritische Gedichte, von J. J. Bodmer. 1.2a 



392 AMERICA 1/ JOURNAL OF PHILOLOGY, 

^yxtipidia et napa(i<l>iias. Est qui ^ MemoricUes^ titulum fecerit^ est 
qui fTpayfjuxTiKa et ndptpya et didao-icaXiied» est item qui ^ Historiae Naiu- 
ralis^ est practerea qui ^Pratum* et itidem qui irdyKapirw, est qui 
rdnov scrifisit. Sunt item multi qui * Coniectanea^ neque item nan 
sunt qui indices libris suis fecerint aut ^Epistularum Mbralium * 
aut ^ Epistulicarum quaestionum^ aut ^ Con/usarum^ et quaedam 
alia inscripta nimis lepida multasque prorsus concinnitates redo- 
leniia. The authors of some of these works are known. The 
*Afia\B€ias Ktpas or Cornu Copiae was by Sotion, the Antiquae 
Lectiones by Caesellius Vindex, the Historia Naturalis by Pliny, 
the Pratum by Suetonius, the navUtcraK by Tullius Tiro. The 
reference to a Silvae may possibly be explained as an allusion to 
the Silva Observationum Sermonis Anliqui by Valerius Probus : 
possibly *hy6fipd may be the Florida of Apuleius. Epistulicae 
Quaestiones was the title of a work by Varro, thrice quoted by 
Gellius {Nodes Atticae, 14, 8, 2) ; Quaestiones Con/usae was the 
name given to his miscellaneous collections by Julius Modestus ; a 
book of Coniectanea was written by Ateius Capito. 

The gentile name of Aulus Gellius shows that he belonged to a 
very old Italian family. All that is known of his life and career 
may be briefly put together from his Nodes Allicae, He nowhere 
mentions his birthplace, but he was at Rome when he assumed the 
toga virilis in his sixteenth or seventeenth year (18, 4, i). The 
date of his birth is only a matter of approximate inference. His 
residence as a student at Athens fell after the consulship of Herodes 
Atticus (143 B. C), for Atticus is spoken of as consularis vir at 
the time {Nodes Atticae, 19, 12; i, 2, i). Gellius calls himself 
iuvenis while at Athens (15, 2, 3, and elsewhere): a term which it 
is surely unnecessary, with Teuffel, to press so far as to make it 
imply that Gellius was a man of thirty or so in these student years. 
Supposing him to have resided at Athens from the age of nineteen 
to that of twenty- three, he must have been born A. D. 123 or there- 
abouts. 

The ordinary educational course in his day began with grammar, 
and passed through rhetoric to philosophy (10, 19, i^adulescentem 
a rhetoribus et afacundiae studio ad disciplinas philosophiae trans- 
gressuin)» In grammar he attended, among other lectures, those 
of the learned Carthaginian scholar Sulpicius Apollinaris, also the 
master of the emperor Pertinax.* In rhetoric one of his favorite 

> 7, 6, 12, quern inprimis sectabar; comp. 20, 6, x^cum ewn RamaiaduUsctntuIus 
sectanr^ 



THE NOCTES ATTIC AE OF AULUS GELUUS, 393 

teachers was Antonius Julianus, described (19, 9, 2) as docendis 
publice iuvenibus magister^ in whose company he seems to have 
spent many pleasant hours (9, 15). Another was Titus Castricius, 
a man gravi atque firmo iudicio (11, 13, i), the chief professor of 
rhetoric in Rome.* Gellius also heard Fronto in Rome during his 
early youth.* 

In philosophy his tutors were mainly Favorinus and Calvisius 
Taurus — Calvisius Taurus he heard at Athens, whither he went 
from Rome after finishing his course of rhetoric," and appears, 
though to what extent is uncertain, to have studied Aristode and 
Plato with him.* 

Gellius also saw a great deal at Athens of the enigmatical phil- 
osopher Peregrinus, sumamed or nicknamed Proteus, of whom he 
gives a very different account from that of Lucian.* Had the 
eighth book of the Nodes Alticae survived we might have heard 
more of this interesting personage, who figured in the dialogue of 
the third chapter. During the same time he saw and heard the 
celebrated rhetorician Tiberius Claudius Atticus Herodes.* 

There are several pleasant allusions, scattered up and down the 
Nodes Atticae, to Gellius*s student life at Athens ; to his boating- 
trips to Aegina and back (2, 21, i); his excursion to Delphi (12, 
5, i); the monthly gatherings of students (15, 2, 3, in conviviis 
iuvenuniy quae agitare Athenis hebdomadibus lunae sollemne nobis 
JuiL^ 

It was after his return from Athens to Rome that Gellius became 
intimate with Favorinus,* and thus fell under a philosophical influ- 

'13, 22, I, rhitoricae disciplinae doctor ^ qui habuit Romtu locum pHncipem dech' 
mandi ac docendi^ summa vir atutoritate gravitcUeque et a divo Hadriano in mores 
aique litUras spectatus, 

* 19, 8, I, aduUscentulus Romae^ priusquam Athenas concederem, 

'17,8. I , Philosophus Taurus accipiebat nos Athenis, 7,13,1, factitatum obser- 
vatumque hoc Athenis est ab his qui erant philosopho Tauro iunctiores. 19, 6, 2, 
hoc ego Athenis cwn Tauro nostro legissem, 

* 7, 10, 1, Taurus^ vir tnemoria ttostra in disciplina Platonica ceiebraius. 17, 20, 
I, Symposium Platonis apud phihsophum Taurum legebatur, 19, 6, 2, problemata 
Aristotelis, 

•12, II, r, Philosophum nomine Peregrinum^ cui postea cognomentum Proteus 
Jactwn est^ virum gravem atque constantem^ vidimus^ cum apud Athenas essemus, 
deversantem in quodam tugttrio extra urbem, Cumque ad eum frequenter ventitare- 
mus, multa hercle dicere eum utiliter et honeste audivimus, 

*I9. 12; comp. I, 2, I. 

^14, I, I, Audivimus quondam Favorinum philosophum Romae Graece disseren^ 
iem egregia atque inlustri oratione» i, 21, 4, cum Favorino Hygini commentariwn 



394 AMERICAN JOURNAL OF PHILOLOGY. 

ence which extended at least beyond the time at which he entered 
upon professional life.* If we may trust the impression left by the 
Nodes Attic ae, Favorinus was not merely a technical metaphy- 
sician, but also an acute and learned scholar. As is well known, 
he was the author of works entitled anofivtifioptvfAara and iraiTod<nn) 
Itrropia, the latter of which most probably suggested the form, if 
indeed it did not supply much of the contents, of the Noctes Atticae, 

Once returned to Rome, Gellius seems to have entered upon 
active life, of what kind he does not tell us explicitly ; but he was» 
homo aduiescens as he says (14, 2, i), chosen a judge for the deci- 
sion of private causes. He can hardly have bien older than 25 at 
this time.* In one other passage (12, 13, i) he alludes to his 
undertaking judicial functions ; but in other places his accounts of 
his life are somewhat vague, though they refer generally to a 1^^ 
career.* There is no mention of elevation to any high office; 
perhaps the mediocrity which stamps his literary work may have 
been also obvious in the discharge of his judicial functions. 

I now come to the most important and difficult part of my task, 
which is to give some account, and attempt some analysis, of the 
Noctes Atticae. It appears from the author's preface that before 
he published this work in its final shape he had laid the foundation 
for it in a number of excerpts. Praef. 2; usi atUcm sumus ordine 
return fortuito, quern antca in excerpendo feceramus. Nam 
proinde ut librum quemgue in manus ceperam scu Graecum seu 
Lalinum^ vel quid memoratu dignum audieram, ita^ quae libitum 
erat, cuius cunque generis erant, indistincte atque promisee adno- 
tabam^ eaque mihi ad subsidium memoriae quasi quoddam liitera^ 
rumpenus recondebam^ etc, 

egissem. 10, 12, 9, Favorinus phiiosophus^ memoriarum vettrttm exequentissimwo. 
16, 3, I. cum Favorino dies pUruntque totos eramus^ tenebatque attimos nostras komto 
ilU fandi dulcissimus^ atque eum^ quoquo iret^ quasi lingua eiusprorsus capti prose- 
quebamur, 

'14, 2, I, II. qtto primum tempore a praetoribus Uctus in indices essem , . . a 
subseliiis pergo ad Favcrinum phihsophum^ quem in eo tempore Romae plurisnum 
sectabar. Comp. 2, 22, i ; 17, 10, i ; 18, i, i. 

* Digest 42, I, 571, Quidam consulebat^ an valeret sententia a minore vigisUi 
quinque annis iudice data, 50, 4, 8 , ^ rem publicam administrandam ante vicen- 
simum quintum annum, vet ad munera quae non patrimonii sunt vel konorrs^ 
admitti minores non oporiet. 

* 12, 13, 1, cum Romae a consulibus index extra ordinem datus . . . pronuuHetre 
iussus sum, 13, 13, I, cum ex angulis secretisque librorum ae magistrorum in 
medium iam hcminum et in lucem fori prodissem, zi, 3, 1, quando ok ctrHtris 
negotiisque otium est, 16, 10, I, otium erai quodam die Romme in foro a rngotiU, 
Praef. 12^ per omnia semper negotiorum intervalld. 



THE NOCTES ATTICAE OF AULUS CELL/ US. 395 

The tide Nodes Atticae was given to the book simply as a record 
of the fact that Gellius began to make his collections during the 
long winter evenings of his student years at Athens. It is profes- 
sedly a handbook of miscellaneous information, but aims, as its 
author expressly says, at being comparatively popular, and regards 
quality more than quantity in the facts presented. For the pres- 
ence of some few specimens of recondite learning the author thinks 
it necessary to apologize.* 

Gellius does not tell us what is sufficiently obvious to a reader 
of his book, that he has taken great pains to enliven his lessons by 
the form in which his scraps of information are presented. Often 
indeed an extract is simply copied from an older author, and given 
in its naked simplicity without introduction or citation of authority ; 
but quite as often an attempt is made to set it in the frame of an 
imaginary dialogue, a description, or an anecdote. The uniformity 
of the devices employed is amusing. Certain individuals, as Favo- 
• rinus, Fronto, Castricius, Calvisius Taurus, Sulpicius Apollinaris, 
figure as the interlocutors in the dialogue ; but it is hardly to be 
supposed that the scenes into which they are introduced are other 
than fictitious. They may, of course, be taken as giving a general 
idea of the life of Gellius, his pursuits, and the sphere in which he 
moved ; but they are, in all probability, no more historical than 
the introductory scenes of Plato's or Cicero's dialogues. As a foil 
to the instructed scholar or philosopher there often appears a con- 
ceited or affected or generally unseasonable individual* whose 
delusions are exposed by the light of superior wisdom. Sometimes 
the devil's advocate appears in another shape, as in 19, i, 7, where 
a rich Asiatic Greek is disagreeable enough, on a sea-voyage, to 
ask a Stoic philosopher who has shown signs of alarm at a tempest, 
to explain to him how it is that he has been pale and trembling all 
the while, while the speaker has given no indication of fear. 

' Praef. 11, 12, 13. 

* X, 2, 3, adulescens philosophiae sectator . . . sed loquacior impendio et promptior, 
I. 10, I, adulescfftti vtterum vtHwrum cupidissimo. 4, i, i, ostentabat quispiam 
^grammaticae rei ditior scholica qiuudam ntigalia. 5, 21, 4, reprehetfsor audaculus 
verborum. 6, 17, i^t^grammaticum primae . . . ceUbntatis, , . . insoUntis h^minis 
iMstitiani, 7, 16, i, eiusmodi quispiam ^ qui tumult uariis et inconditis Hnguoi 
txercitationibus ad famam sese facundiae promiserat, 8, \0^ grammaticus quidam 
praestigioius, ib. 14, intempestivus quidam de ambiguitatt verborwn disserens. 9, 
15,1, introit aduUsctns et praef atur arrogantius et elatius, 11,7,3, t'^'w-f cetebratusque 
homo in causis^ sed repentina et quasi tumultuaria doctrina praeditus. 18,4, i 
iautator quispiam et venditatar Sallustianae lectionis, 20, 10, 2, HU me despiciens. 



396 AMERICAN JOURNAL OF PHILOLOGY. 

Were these loquacious or ignorant or conceited individuals to 
be taken seriously, we should have reason to hold up our hands in 
horror at the social condition of the second century A. D. ; but 
they are in all probability mere men of straw. In any case tbe^ 
are tedious enough ; nor is their constant introduction the only in- 
stance of want of skill shown in the composition of the Noctes AUUae. 
Sometimes, as Mercklin and Kretzschmer' have pointed out, the 
fbnn of the dialogue is not consistently maintained through a whole 
chapter; thus in i, 7 Gellius starts by quoting a passage from 
Cicero's hflh speech against Verres; no indication of time or place 
is given, yet in §3 the writer proceeds videbcUur comphtribus m 
extreme verba menda esse, and in §4 aierat forte amicus naster. 
In 2, 22 an elaborate account of the winds is put into the mouth of 
Favonnus ; the dialogue is continued to the end of §26, yet in %y> 
Gellius quotes something which he has already attributed to Favo- 
nnus as if he had said it himsel£ There is a similar awkwardness 
at the end of 5, 21, where an opinion of Sinnius Capito, having 
been originally introduced in the course of a supposed dialogue, is 
treated as if it had been cited by Gellius. In 13, 21, 9 it is quite 
clear that the passage discussed by Gellius had really been treated 
by Probus in the work from which the first part of the chapter is 
quoted, and this fact is enough to raise a suspicion that the anec- 
dote about Probus is mere padding. A similar remark applies to 
the end of 19, 8, where there is no real distinction between the 
observations offered by Gellius himself and those previously put 
into the mouth of Fromo. 

There are other marks of carelessness in composition. Gellius 

is apt, for instance, to introduce one of his interlocutors twice over, 

thus Herodes Attlcus is described (i, a) as vir et Graeca facvndta 

et cotisulari honore praeditus, and so 9, 2 Herodem AttUum, con- 

su/arem vtntm ingenioque amoeno el Graeca facundia celebrem, 

-—'-■■— '-'ianus ('. 4^ i). rhetor perquam fuii honesti atque 

it; doctrina quoque ista utiliore {subtiliore, Madvig) 

■ veterumque elegarUiarum cura et memoria multa 

■cripta omnia tarn curiose spectabat,etc. 19, 9, i An- 

isrlietor,docendispubUceiuvenibusmagisler,Hispano 

ue homo facundiae el rerum Utlerarumque vetertmt 

IS Castricius, 11, 13, 1, disciplinae rhetortcae doctor, 

xrmo iudicio vir. 13, 22, rheloricae disciplinae doctor, 

klin and Kretzaclimer are quoted, th« reference is to the 
lin, in Ihc JakrbU<hir Jar Ctatsitehi Philologit, Suppl. Ill 
lat of Kretiichner, Dt Auli Gellii Jenliiui, Greifswald, 1S60. 



THE NOCTES ATTICAE OF AULUS GELLIUS, 397 

qui habuit Romae locum principem declamandi ac docendiy summa 
vir atuioritate j^ravitateque. Apion, 5, 14, qui nXeurromKijr appel- 
iaius est^ litteris homo muliis praediius rerumque Graecarum 
plurima atque varia scientia fuit. 7, 8, i, Grtucus homo qui 
lXK€urTou€iKiii appellatus est^facili aique alacri facundia fuit Tul- 
lius Tiro, 6, 3, 8, M. Ciceronis liberius, sane quidem fuit ingenio 
homo eleganii et haudquaquam return litterarumque veterum 
indociuSy eoque ab ineunte aetate liberaliier insHtuto adminiculatore 
et quasi administro in studiis Utter arum Cicero usus est 13, 9, i» 
Tullius TirOy M. Ciceronis alumnus et libertus adiutorque in 
litteris studiorum eius fuit. 

An extract is sometimes so carelessly torn from its context that 
marks of the rent are still visible. Thus the epitome of 3, 17 begins 
id quoque esse a gravissimus viris memoriae mandatum^ where 
there is nothing in the previous chapter to lead up to the quoque. 
Exactly i;i the same way 10, 8, ijfuit haec quoque antiquitus mili- 
taris animadversio. 12, 12, i, haec quoque discipiina rheiorica 
(? disciplinae rhetor icae ?) est. 18, 12, 1, id quoque habitum est in 
oratione facienda clegantiae genus. 

Sometimes Gellius alludes or seems to allude to things which he 
has nowhere said, or proposes discussions which are nowhere 
started : thus 2, 22, 31, considerandum igitur est quid sit secundo 
sole^ a question whicl) is not treated anywhere else ; and so it is 
with 12, 14, 7, censuimus igitur amplius quaerendum. 13, 7, 6, 
in quibuSy quod super ipsa re scriptum invencrimuSy cum ipsius 
Aristotelis verbis in his commentariis scribemus. 14, 7, 13, de hac 
omni re alio in loco pienius accuratiusque nos memini scribere (a 
discussion on the forms of the senatus consul/urn j which occurs 
nowhere else, not even in the epitomes of the eighth book). 18, 
4, II, quos notavi et intulisse iam me aliquo in loco commentatio- 
nibus istis existtmo. 

It should further be observed that the same point is sometimes 
treated twice in much the same words : compare 2, 26, 9 ; 3, 9, 9, 
palmae termes ex arbore cuntfructu evulsus * spadix* dicitur : 
awa^iKa dopiari vocant avulsum e palma termitem cum fructu. 3, 
16, §§18-19, i5» 5» 5» O'dfecta . . . eaproprie dicebantur quae non 
ad finem ipsum sed proxime finem progressa deductave erant. 
Hoc verbum ad banc sententiam Cicero in hac fecit quam dixit de 
provinciis consularibus. The same quotation, with others, is given 

in 15, 5» 5- 

We may now approach the central question, from what authors 

and from what works does Gellius mainly derive his information ? 



THE NOCTES ATTIC AE OF AULUS GELLIUS. 399- 

In 17, 15 Gellius borrows hb whole account of the two kinds of 
hellebore from Pliny 25, 47 foil. But Pliny's name is not mentioned 
until the sixth section, and then only in such a way as to put the 
reader off the scent. The two following chapters, however, which 
contain stories of Mithridates and his knowledge of medicine and 
of languages, although they may be found in Pliny (25, 6 ; 29, 24) 
in a shorter form, contain some information which is absent from 
his text, and must therefore be taken from some common authority, 
perhaps the memoirs of Pompeius Lenaeus. 

The instance of 17, 15 will serve as a specimen of what we must 
look for throughout the whole of the Nodes Atticae. Gellius often 
alludes to his authority, but gives the false impression that only a 
part of the chapter in which it is mentioned is borrowed from him. 

It sometimes, to all appearance, happens that Gellius makes 
extracts from more than one work in the same chapter. At the 
end of 3, 9, for instance, after speaking of some proverbial expres- 
sions, he goes out of his way to inform us that spadix and poeni- 
ceus mean one and the same thing ; at the end of 9, i there is a 
remark of a lexicographical character on the word defendo ; so at 
the end of 10, 3 on Bruttiani, of 13, 11 on bellaria, of 13, 22 on 
crtpidariuSy of 20, 5 on cognobilis, Mercklin thinks the same was 
the case in other places. 

Perhaps the best way of getting an approximate idea of the 
character of the works consulted by Gellius will be to analyse his 
whole book according to the subjects of which it treats. In this 
way we shall obtain a conspectus of its general scope, and shall also 
be able to establish a visible connection, not only between some 
neighboring chapters, but between distant parts of the Nodes 
Atlicae, This connection is sometimes so close as to lead irresist- 
ibly to the conclusion that the kindred sections belong to the same 
original work. 

The ^octes Atticae is a work of such miscellaneous contents 
that it is impossible to make an entirely satisfactory table of them. 
A margin of unclassified matter must remain, whatever principle of 
arrangement be adopted. A rough distribution of the main bulk 
into certain great divisions is however possible. We may take as 
the first branch that of philosophy, understanding that term to 
include metaphysics, psychology, logic, and morals. 

The true as distinguished from the false study of philosophy is 
touched upon briefly in 1, 2, and 10, 22; but there is nothing in 
these chapters which should lead us to connect them. 5, 15. cor- 



400 AMERICAN JOURNAL OF PHILOLOGY. 

pume sit vox an dir^fiaTop, varias esse phUosophorum senUniias^ is 
evidently from the same source as the following chapter, d€ vi 
acuhrum deque videndi rationibus. The authority is at least as 
late as the Ciceronian age, and almost certainly Latin, as Lucretius 
and Ennius are quoted. The first and second chapters of the 
teventh book, in which the opinions of Chrysippus on Providence 
and on Fate are discussed, are no less obviously akin, and probably 
from the same source ; which, if we may press the fact that Cicero 
is quoted at the end of the second chapter, was presumably a late 
one. The first chapter of the fourteenth book, dissertatio Favorini 
philosophi adversus eos qui Chaldaei appellantur et ex coeiu moti- 
busque siderum et stellarum fata hominum diciuros pollicentur^ 
deals with a cognate subject 

Turning to ethics, we find a discussion as to the nature of the 
summum bonum between a Stoic and Peripatetic in the first chapter 
of the eighteenth book ; the doctrine of Chrysippus trc/ii koKov au 
^oif^ff as applied to the character of justice is expounded in 14, 4. 
Connected in subject with the latter is 9, 5, in which various philo- 
sophical views of pleasure, concluding with that of the Stoic Hiero- 
cles, are presented. Three chapters on the relation of reason to 
passion (i, 26; 12, 5 ; 19, i) are closely connected, and may come 
from the same manual (a very late one), or set of lectures. The 
first, on anger, purports to be from Taurus and Plutarch; the 
second, which is also professedly from Taurus, deals with the Stoic 
theory of bearing pain ; the third gives the opinion of Epictetus 00 
the subject of fear. We may mention in this connection the dis- 
course of Herodes Atticus against cnra^cca (19, 12). 

The following chapters touch on various points of logic: 11, 12 
(Chrysippus on ambiguous terms); 15, 26 (a proposed Latin 
translation of Aristotle's definition of a syllc^ism) ; 16, 8 (Latin 
equivalents for several Greek technical terms); 5, 10, 11 (the 
argument called dprun-fHiftw, again treated in 9, 16) ; 18, 13 Ca story 
of a fallacy tried unsuccessfully upon Diogenes). Of these 16, 8 
deserves the most attention ; I am tempKed to think that it comes 
from Varro, whether from the lost twenty-fourth bock of the de 
Lingua Latinay quoted in the fourth section, or fixxn the Discip^ 
linae. 

The eighth and ninth chapters of the second book are finom 
Plutarch ; the second, fourth, fifth and sixth of th^ niDetecnth book 
from the Problemaia of Aristotie, though in the fifth chaplcr tbc 
debt is not quite directly admowledged. 



THE NOCTES ATTICA E OF AULUS GELUUS. 401 

The ninth chapter of the first book, the eleventh of the fourth, 
and the fifth of the eleventh, touch on points connected with the 
history of philosophy : the first two treating of the Pythagorean 
discipline; the last, of the difference between Pyrrhonists and 
Academics. 

We may now pass on to another head, that of ethical principles 
applied. Here some sort of classification is possible, though there 
are hardly any data for inference as to authorities. Four chapters 
(9, 2; 12, II ; 13, 8; 13, 24) treat of the relation of philosophy 
to conduct ; of these, one (12, 11) contains a dictum of Peregrinus, 
virum sapitntem non peccaiurum esse^ eiiamsi peccasse eum di 
aiqu€ homines ignoraturi forent ; the other three are protests 
lodged in various forms against dilettantism and hypocrisy in the 
philosophical profession. Two of these (13, 8; 13. 24) have a 
distinctly Roman tinge. 

Four chapters are devoted to questions of casuistry. In i, 3 
Favorinus, quoting Theophrastus and Cicero, starts the problem 
an pro utilitate amicorum delinqiundum aliquando sit. The 
second and third (1,13 and 2, 7) open in very much the same way, 
in officiis capiendiSy censendiSy iudicandisque^ quae KaBi\KovTa philo- 
sophi appellant y quaeri solely etc, Quaeri solitum est in philoso- 
phorum disceptationibusy an semper^ etc. Does this fact point to 
identity of source ? The first discusses the question whether the 
letter or the spirit of an order is to be taken as the more important ; 
the second, how far a parent's commands are to be taken as binding. 
Both questions are approached from a Roman point of view. The 
remaining casuistical chapter is 14, 2, where Gellius consults Favo- 
rinus de officio iudicis, 

A number of exhortations to particular virtues and warnings 
against particular vices should be mentioned here, i, 17 (from 
Varro), de tollendis viiiis uxoris, 2, 1 2, Solon's law enforcing the 
duty of taking a part in political dissensions, and Favorinus's view 
about a similar duty in private life. 12, i, Favorinus suadet nobili 
feminae uti liber os quos peperisset non nutricum aliarum sed suo 
sibi lacte aleret. 13, 28, Panaetius de cavendis iniuriis. 17, 19, 
Epictelus (quoted by Favorinus) avixov Koi ani'xov. i, 15, Favorinus 
against the vice of loquacity. 6, 16 ; 15, 19, Varro (ntpX f^tafuimp) 
against luxury. 9, 8 (Favorinus), qui multa habet, multis eget, 15, 
8, an ancient orator de cenarum atque luxuriae opprobratione. 7, 
XI (Metellus Numidicus), cum inquinatissimis hominibus non esse 
convicio decertandum. 8, 6 (Taurus, from Theophrastus and 



402 AMERICAN^ JOURNAL OF PHILOLOGY. 

Gcero), cum post offensiunctUas in greUiam redeatur^ expasbda^ 
tumes fieri muiucu minime utile esse, lo, 19 (Taums), mm fmr^ 
gari neqtu levari peccaium cum prtietenditur peccaiorum^ quae 
alii quoque peccaverunlj similiiudo, 

A transition is natural from the last head to the exempla or 
remarkable instances of praiseworthy conduct cited in the Nodes 
Ailicae. Among these two only are from Greek history, the story 
of the habits of Socrates put into the mouth of Favorinus (2, i), 
and that told by Taurus (7, 10) about the youth of Euclides. The 
rest are Roman, and are as follows: i, 14, the story of Fabridus 
and the Samnites ; 2, 2, the two Fabii, father and son ; 4, 8, Fabri- 
dus Lusdnus and the avaricious Rufinus ; 6, 18, the sanctity of 
oaths among the ancient Romans ; 6, 19, Ti. Gracchus and Sdpio 
Asiaticus; 7, 8, Scipio*s continence; 7, 9. Cn. Flavius the scriba ; 
12, 4, Ennius's character as sketched by himself; 12, 8. the recon- 
ciliation of P. Scipio and Ti. Gracchus; 15, 12, C. Gracchus on his 
own quaestorship. 

Of the authorities for some of these stories something certain 
may be said, at least on the negative side. Gellius has not borrowed 
anything from Valerius Maximus, natural as it would seem that he 
should have done so. In 12, 7, §8, after relating the story of Cn. 
Dolabella and the woman who was brought before him at Sm3rma 
on the charge of poisoning her son and husband, he says scripia 
haec historia est in libro Valeri Maximi factorum et dictarusm. 
memorabilium octavo. Yet any careful reader of Gellius's narrative 
must see that although he has read Valerius Maxinius (8, i, 2 
damn.), he has not copied him, for he adds two details about which 
Valerius is silent : venenis clam daiis, of the one murder ; exceptum 
insidiis, of the other. Again, let us compare Gellius i, 14 with 
Valerius Maximus 4, 3, 6. The story (of C. Fabridus and the 
Samnite envoys) is told by Gellius in a far fuller and more char- 
acteristic manner than by Valerius. Gellius professes to take it 
from Hyginus, de vita r^busque inlustrium virorum, which was 
probably the common authority for both writers. Both writeis 
again have a story about Fabridus Luscinus and Cornelius Rufinos 
(Gellius 4, 8, Valerius Maximus 2, 9, 4) which occurs in a different 
context in Gellius from that in which it is set by Maxmius. The 
style of Gellius*s narrati\*e in this case tempts roe to suppose that 
it is from the hand of a classical writer, such as Hyginus or Nepos. 
The work of Nepos called exempla is quoted cxjJidtly by Geilios 
when, in the eighteenth chapter of his sixth book, he is umxiDg 



THE NOCTES ATTICA E OF AULUS GELLIUS. 403 

the history of the ten captives who returned to Rome after Cannae ; 
indeed it is not impossible that the whole of the chapter comes 
from this work. The same may be the case with the story in the 
following chapter about TL Gracchus and Scipio Asiaticus, which 
is given in an abridged form by Valerius Maximus (4, i, 8). For 
Gellius begins by saying pulchrum atque liberale aique magnanu 
mum factum Ti, Sempronii Gracchi in exemplis repositum est 
The story of Scipio*s continence (7, 8) is apparently drawn 'from 
an older source than the version given by Valerius Maximus (4, 3, 
i). The relation between the two writers is, I think, the same 
with regard to the two narratives given by Gellius 12, 8, and 
Valerius Maximus 4, 2, 3; 4, 2, 1/ Thus in six instances it is 
highly probable that Gellius follows an authority older than Valerius 
Maximus ; In one of them he expressly cites Hyginus, in another 
Nepos ; and it is therefore not rash to infer that he is indebted to 
these two writers for a considerable part of his information under 
the head which we have been discussing. 

Five chapters of the Nodes Atticae are devoted to natural phil- 
osophy; these are 2, 22, on the winds; 2, 26, on the names of 
colors ; 2. 30, on the effect of different winds on the motion of the 
waves ; 9, i, on the direction of blows as influencing their strength. 
Of these, chapters 2, 22 and 2, 30 must be derived from the same 
sources as the corresponding passages in Pliny (2, 1 26 foil.) 

There are abo four chapters on points of human pathology ; 3, 
16 (temporis varietal in puerpcris mulierum)^ partiy from Varro. 
4i 19, again from Varro, de moderando victu puerorum inpubium, 
17, II, from Plutarch, de habiiu aique natura stomachi. 18, 10, 
errare isios^ qui in explorandafebri venarum pulsus periemptari 
puianti nan arteriarum. 

The department of rhetoric is not very fully represented in the 
Nodes Aiiicae, The notes which fall under this head consist 
mainly of criticisms on passages in the ancient orators from Cato 
to Cicero, and exhibit a considerable similarity ; but it is hardly 
possible to infer anything as to their source. Indeed it is not 
impossible that they come, as they profess to do, from the contem- 
poraries of Gellius himself. We may notice as kindred in spirit 

* Mercklin thinks that the story of Aemilius Lepidus and Fulvius Flaccus 
comes directly from Valerius Maximus. There seems, however, to be nothing 
in the language to necessitate such a conclusion, while of the preceding story 
about the older Africanus and Ti. Gracchus, Gellius gives a fuller, and therefore 
probably an older, version. 



404 AMERICAN JOURNAL OF PHILOLOGY, 

the remarks of Castricius upon Metellus Numidicus (i,6), and the 
defence of Cato against the strictures of TuUius Tiro (6, 3). In 
both chapters the point insisted on is the difference between the 
manner suitable to an advocate and that suitable to a statesman. 
Perhaps we may also connect 12, 12 and 16, 2, which deal with the 
art of reply. Cicero is defended against captious criticism in i, 4 
and 17, 5. The remaining chapters do not admit of any classifica- 
tion; they are 9, 15 (a case of oiropoF or inexplicabile) \ 17, 12 
(jnaieriac infantes) ; 17, 20 (a translation of a passage in Plato's 
Symposium), 

If the contributions of Gellius to the art of rhetoric are scanty, 
the same cannot be said of the passages of ancient literary criticism 
which he has preserved. Twenty-eight chapters are devoted to 
this subject, some of which may be easily arranged together as 
containing similar matter. Nine are given to the question of trans- 
lation or adaptation from Greek into Latin. These are 2, 23, where 
Caecilius is compared, much to his disadvantage, with Menander ; 
II, 4, a criticism of a translation from Euripides by Ennius. 2, 27, 
where Castricius is represented as contrasting Sallust's description 
of Sertorius with that of Philip by Demosthenes. 8, 8 and 17, 20, 
which touch upon Gellius*s own efforts at rendering Plato. 9, 9 ; 
I3i 27 ; 17, 10, comparisons with the originals of Vergil's renderings 
or adaptations of Theocritus, Homer, Parthenius, and Pindar. 19, 
II, a translation by a friend of Gellius of some erotic verses by 
Plato. It is natural to suppose that some of these criticisms are 
taken from a manual in which the whole question of translation 
was discussed. Such a work, in all probability, was the o^jM/^/nfm 
oi Octavius Avitus, mentioned by Suetonius in his life of VergiL 

I, 10 and II, 7 contain protests against the affectation of anti- 
quarianism in writing. General remarks on style will be found in 
2. 5 (a short comparison between Plato and Lysias) ; 10, 3, where 
the styles of Gracchus and Cicero are contrasted; 16, i (the 
expression of the same thought by Cato and Musonius). 11, 13, 
and 14, which contain the praises of C. Gracchus and the historian 
L. Piso, seem to be intimately connected. Other chapters falling 
under this head are 15, 24 (the metrical criticism of Roman come- 
dians by Volcatius Sedigitus); 6, 14, where Varro's distinction 
between the three styles {ubery gracilis, mediocris) is quoted ; 18, 
8 (Lucilius on o/xotorAcvra) ; and 12, 2 (Seneca upon Ennius and 
Cicero). 

Three chapters (3, i ; 4, 15 ; 10, 26) are devoted to criticism, 



THE NOCTES ATTICA E OF AULUS GELLIUS. 405 

mainly defensive, of Sallust, whose abrupt and antiquarian style 
appears to have attracted a great deal of attention on the part of 
scholars and literary men. Three again (5, 8 ; 9, 10 ; 10, 16) con- 
tain pleadings in defence of Vergil against strictures of Hyginus 
and Cornutus, taken possibly from the work of Asconius contra 
obtrectaiores Vergilii, In 15, 6 attention is drawn to a mistake of 
Cicero's. Finally, 3, 3 deals (after Varro) with the question of the 
genuine and spurious plays of Plautus. 

History and biography absorb thirty-six chapters. Among 
these we may fairly distinguish the following groups : (i) i, 23 ; 9, 
II, 13, on Roman cognomina (Praetextatus, Corvinus, and Tor- 
quatus). These notices are so similar in tone and composition as 
to suggest the inference that they come from the same source, 
which may have been perhaps the work of Cornelius Epicadus, 
Sulla's freedman, on cognomina. It should be observed that the 
twenty-third chapter of the first book is verbally identical with a 
passage in the first book of Macrobius's Saturnalia (i, 6, 18 foil.). 
It has been of course assumed that Macrobius borrowed from 
Gellius ; but against this hypothesis it may be urged that Macro- 
bius goes on to supplement the story about Praetextatus by further 
information respecting other cognomina unknown to Gellius, and 
this in such a natural and easy way as to lead us to suppose that 
the whole passage is taken from some book which dealt in a com- 
prehensive way with the whole subject. We should probably have 
known more of this work and its contents had the last book of 
Nonius been preserved. (2) Six chapters (i, 24; 3, 3; 8, 15 ; 12, 
4; i3» 2; 17, 14) are devoted to interesting passages in the lives 
of Latin poets. In one of these cases the relation between Gellius 
and Macrobius is precisely the same as that which has just been 
considered. I allude to the notice of Publilius Syrus, which is 
fuller in Macrobius 2, 7 than in Gellius 17, 14. Is Varro the 
authority for these fragments of biography ? he is expressly quoted 
in I, 24, and 3, 3. (3) Fragments of biographies of Greek poets 
are preserved 3. 11; 15, 20; 17, 4. The last of these comes 
ostensibly from the Chronicon of Apollodorus, but may well have 
been taken from Varro's adaptations from that work ; for Varro is 
actually cited in 3, 11. (4) Another group of chapters (5, 3; 13, 
5 ; I4» 3 ; 20, 5) deals with lives of Greek philosophers ; while (5) 
a large number contains notes of remarkable facts fi-om Roman 
history (i, 13 end; 2, 11; 3, 7, 8; 4, 14, 18; 7, 3, 4; 10, 27, 28; 
15, 4, II ; 18, 22). We are here brought back to the question of 



THE NOCTES ATTIC AE OF AULUS CELL! US. 407 

ac quo fwmine a pontifice maxima capiahir^ et quo statim iure esse 
incipiat simul atque capta est ; quodque^ ui Labeo dicit^ nee hites- 
taio cuiquam nee eius iniestatae quisquam iure heres est, 10, 15, 
de fiaminis Dialis deque flaminicae caerimoniis ; verbaque ex 
edicio praeioris apposita quibus dicit non coacturum se ad iuran- 
dum neque virgines Vesiae neque Dialem ; this chapter bears the 
names of Varro and Masurius Sabinus. 2, 28, apparently from 
Varro, non esse compertum cui deo rem divinam fieri oporieat^ cum 
terra movet, (2) On social customs. 2, 15, quod antiquitus aetati 
senectae potissimum habiti sunt ampii honores^ et cur postea ad 
mariios et ad patres idem isti honores delati sini ; the authority is 
uncertain, but not older than the leges luiiae, 5, 13, de ojfficiorum 
gradu atque ordine moribus populi Romani observato. This 
chapter quotes from Masurius Sabinus. 6, 4, cui