Skip to main content

Full text of "Proceedings at Boston, May 18th, 1881"

See other formats


STOP 



Early Journal Content on JSTOR, Free to Anyone in the World 

This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in 
the world by JSTOR. 

Known as the Early Journal Content, this set of works include research articles, news, letters, and other 
writings published in more than 200 of the oldest leading academic journals. The works date from the 
mid-seventeenth to the early twentieth centuries. 

We encourage people to read and share the Early Journal Content openly and to tell others that this 
resource exists. People may post this content online or redistribute in any way for non-commercial 
purposes. 

Read more about Early Journal Content at http://about.jstor.org/participate-jstor/individuals/early- 
journal-content . 



JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people 
discover, use, and build upon a wide range of content through a powerful research and teaching 
platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit 
organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please 
contact support@jstor.org. 



Proceedings at Boston, May, 1881. lv 



Proceedings at Boston, May 18th, 1881. 



The Society assembled at the usual place and time. The Pres- 
ident and all the Vice-Presidents being absent, the chair was 
taken by Prof. A. P. Peabody, of Cambridge, and later by Rev. 
W. H. Ward, of New York. 

The Treasurer's report for the last year was read, and his 
accounts audited by a committee appointed for the purpose, and 
accepted. The summary of accounts is as follows : 

RECEIPTS. 
Balance on hand, May 19th, 1880, .... $674.06 

Annual assessments paid in, ... $570.00 

Sale of the Journal, ..... 194.22 

Interest on deposit in Savings Bank, - - 27.57 



Total receipts of the year, .... 791.79 



$1,465.85 



EXPENDITURES. 



Printing of Proceedings and Journal, - - - $753.25 

Expenses of Library and Correspondence, - - 26.15 



Total expenditures of the year, .... $779.40 

Balance on hand, May 18th, 1881, .... 686.45 



$1,465.85 



Bills for printing will soon be due which will nearly or quite 
exhaust the balance now in the Treasury. 

The amount of the Bradley type-fund is at present $848.52. 

The report of the Librarian snowed the accessions to the Li- 
brary during the year to consist of forty-six volumes, sixty-three 
parts of volumes, forty-three pamphlets, and four manuscripts : 
the number of titles of printed books being now 4,046 ; of manu- 
scripts, 148. Among the gifts is a magnificent work, published 
at the expense of the Government of the Netherlands, and by it 
presented to the Society, on the Buddhist temple of B6r6-Bou- 
dour in the island of Java, consisting of 418 royal folio plates and 
a descriptive text in Dutch and French. 

The Committee of Publication reported that the twelfth vol- 
ume of the Journal, containing the Index Verborum to the 
Atharva-Veda, ordered published last year, was on the point of 
completion, and would be distributed to members doubtless within 
a month ; also, that progress had been made with the earlier- 



lvi American Oriental Society : 

begun eleventh volume, of which the first part might be expected 
to be finished in the course of the year. 

The Directors gave notice that they had appointed the autumn 
meeting of this year to be held in New Haven, on the last 
Wednesday (26th) of October. Also, that they had continued 
the Committee of Publication of last year for another year. 
Further, they recommended to the Society the election as Cor- 
porate Members of the following persons : 

Prof. Maurice Bloomfleld, of Baltimore, Md. ; 
Rev. P. F. Ellinwood, of New York ; 
Mr. B. W. Hopkins, of Bridge water, Mass. ; 
Rev. L. P. Mills, of Hanover, Germany. 

The gentlemen thus proposed were then balloted for, and de- 
clared duly elected. 

The election of officers for the ensuing year being next in order, 
a letter was read from Prof. Salisbury, of New Haven, positively 
declining to be a candidate for re-election as President. Prof. 
Abbot, of Cambridge, also requested to be relieved, after nearly 
thirty years of service, of the duties of Recording Secretary. 
These communications were referred to a Nominating Committee, 
which brought in and proposed the following Board of Officers, 
and it was elected without dissent : 

President — Prof. S. Wells Williams, LL.D., of New Haven. 

Vice-Presidents — Messrs. Clark, Parker, and Woolsey (as last 
year). 

Recording Secretary— -Prof. C. H. Toy, D.D., LL.D., of Cam- 
bridge. 

Corresponding and Classical Secretaries and Treasurer and 
Librarian, Messrs. Whitney, Goodwin, and Van Name (as last 
year). 

Directors — Messrs. Cotheal, Short, and Ward, of New York, 
Peabody and Lanman, of Cambridge, and Thayer, of Andover 
(as last year), and Prof. Isaac H. Hall, Ph.D., of Philadelphia. 

The presiding officer (Prof. Peabody) then communicated to the 
meeting the names of the members who had deceased during the 
preceding year: namely, of the Corporate Members — 

Rev. Rufus Anderson, of Boston ; 
Prof. J. L. Diman, of Providence, R. I.; 
Prof. W. C. Fowler, of Durham, Conn. ; 
Prof. S. S. Haldeman, of Chickies, Pa. ; 

and of the Corresponding member — 

Rev. S. F. Brown, of Japan. 

Prof. Peabody spoke at considerable length of the venerable 
Dr. Anderson, his own early teacher and life-long friend, describ- 
ing and extolling his many virtues of character, his long years of 
devoted service to the cause of Christian missions, his warm in- 
terest, in connection with that cause, in studies relating to East- 



Proceedings at Boston, May, 1881. lvii 

era language and history, and his contributions to them. The 
Corresponding Secretary called attention to the fact that he was 
the last survivor in the Society of its band of founders, having 
been a Director from the beginning and for many years a Vice- 
President, till age and infirmity led him to decline a re-election as 
such; and read extracts from the first records (1842), showing the 
active part taken by him in its earliest proceedings. 

At the invitation of the chair, Prof. Williams of Brown Univer- 
sity paid an eloquent tribute to the memory of his colleague Prof. 
Diman, dwelling upon the loss which American letters had sus- 
tained by the early death of this distinguished scholar and teacher. 

The Corresponding Secretary recounted the services of Prof. 
Fowler to the study of American history and of the English lan- 
guage ; and he gave a brief sketch of the life and works of Prof. 
Haldeman, who, from being a student of natural science, had 
passed to the study of phonetics, taking high rank by the produc- 
tion of his Trevely an Prize Essay (1860), and during the latter 
part of his life had devoted himself mainly to philology, publish- 
ing many works, and being especially active in connection with 
the American Philological Association. 

Dr. Ward gave some account of the long and efficient mission- 
ary labors of Dr. Brown, continued, with intermissions, for nearly 
forty years, in China and Japan. 

Extracts from the correspondence of the past half-year were 
read by the Corresponding Secretary. 

Mr. R. A. Guild, Librarian of Brown University, of Providence, 
R. I., communicates the information that the University has lately 
received from Burma a complete copy of the Buddhist sacred 
books, in Pali. The donor, Rev. J. N. Cushing, writes respect- 
ing them : 

"The set of books belonging to the Betagat (Tripitaka) is complete, as the 
Burmans accept them. Doubtless the text is imperfect, for there are always more 
or less errors in every palm-leaf book copied. All that I can say is that the books 
are such as any priest teaching Pali, in his Kyoung, would use. . . . Those having 
the bright gilding and vermilion covers come from Mandalay, where the art of 
palm-leaf book-making flourishes in its greatest perfection. These are new books. 
Some of the others have long been used in monasteries." . . . 

Prof. Isaac H. Hall writes from Philadelphia, in reference to 
the Greek Inscription from Beirut, communicated to the last 
meeting (see Proceedings for Oct. 1880, above, p. xli.), that the 
emendations then conjecturally made in it prove, on renewed ex- 
amination of the original by a friend on the spot, to be the true 
readings of the monument itself. 

Dr. S. Merrill, of Andover, called the Society's' attention to the 
fact that the inscription in question had already been published, 
in Boeckh's Corpus, vol. iii., and also in the Bibliotheca Sacra, 
vol. v., p. 588. 

Prof. Hall also sends a brief account, with transcription and 
translation, of a charm picked up, a year or two ago, by an Ameri- 
can gentleman in Jerusalem, near the pool of Siloam. It was 
enclosed in a tightly sealed little tin box. 



lviii A merican Oriental Society .' 

"The paper contains one short titular line, and six other lines, written in a 
Hebrew character that is rather difficult to read, but which would be called 
Rabbinic ; but about two thirds of the last line are composed of Arabic numerals, 
carelessly written. Beneath is a square of sixteen spaces, with Arabic numerals 
in all the spaces, and an Arabic name written outside of each of the four sides. 
The language is a Chaldaised Hebrew, with at least one Arabic peculiarity, the 
use of the article. The following is the translation : 

" ' May the work of Satan prosper ! 

" ' I conjure you, ye the evil spirits of the evil spirits of Asmodai the King of 
the evil spirits and Rex Tartarotn, king, and Meimon and Zuba'h and Bflrkan and 
Murhab and Shemhoresh, and the red king and the white king, that ye shall put 
into the heart of Mehmed the son of 'Bllya fire and brimstone of mighty love, 
flame of Jah, that he may neither eat nor drink until he shall have done instantly 
the wish and will of Karmuz the son of Sugma, so that he may fulfil his request, 
and not delay in the least nor bring to naught, through the force of those names 
that are set over the moon, Llaklm, Llaki', Liakir, LialgO, LlarOth, Liarosh [each 
name twice], and in the Name, and the sons of Korah, Assir and Elkanah and 
Abiasaph and Elde'a.' 

" The numbers of the last line, when turned into Hebrew characters by their 
numerical values, seem to make no continuous sense. The numbers in ihe square, 
similarly treated, signify 'Love, mighty fire, flame mighty;' the words about it 
are the names Gabriel, Michael, 'Ursael, Asrafel." 

Rev. L. F. Mills, now residing at Hanover, in Germany, writes 
under date of March 6, 1881, giving an account of his labors on 
the Avestan Gathas, and of the publication of their results in 
which he is now engaged, and enclosing a few specimen pages 
of the latter. 

Mr. Mills's edition includes the Avestan text, with transliteration and verbatim 
and free translations (the former in Latin) ; the transliterated Pahlavi version with 
critical notes and translation ; Neriosengh's Sanskrit version in transliteration and 
translation; and the (transliterated) Persian Pahlavi described below. The Pah- 
lavi version of the Gathas, as of the rest of the Yacna, has hitherto rested on a 
single MS., published by Spiegel; Mr. Mills is placed, by the kindness of Dr. E. 
W. West, in possession of the collation of another MS. of about the same age, 
lent him by Destur Hoshangji Jamaspji in India ; and also had the loan from the 
Munich Library of a copy made for Haug just before leaving India from a Pah- 
lavi text in Persian characters, with interlinear Persian translation (mixed with 
Parsi and Arabic). It is not known from what source this latter text comes ; in 
the difficult task of its decipherment Mr. Mills has again had assistance from Dr. 
"West. It was found a valuable umpire between the other two texts, but so far 
independent that its own publication was deemed also desirable. The translation 
of the Pahlavi founded on these authorities has been revised by West, and in part 
by Spiegel ; the former's suggested alterations, where not accepted and incorpora- 
ted by Mr. Mills, the latter intends also to publish in full. For Neriosengh's San- 
skrit, Mr. Mills has received from Spiegel notes of a collation of another Copen- 
hagen MS. ; and the same scholar has revised his work. An elaborate commen- 
tary is to follow, in which will be reported the opinions on every point of the au- 
thor's predecessors, both Asiatic and European (except Anquetil) ; and there will 
be added glossaries of Pahlavi, Sanskrit, and Persian words, and a complete Index 
Verborum to the Gathas themselves, with references to the explanations of each 
word. It is hoped that the volume will appear in little more than six months. 

Mr. Mills's letter gives a succinct review of the condition of the Avestan field at 
the present moment, showing the timeliness of his undertaking. He was first 
drawn toward it by a desire to examine the connection between Zoroastrianism 
and orthodox Pharisaism. He has the approbation and counsel and aid of the lead- 
ing scholars of Europe in this department, and hopes to gain the sympathy and 
support of Americans also. 

Communications were now presented, as follows : 



Proceedings at Boston, May, 1881. lix 

1. Remarks on Guyard's theory of Semitic Internal Plurals, by 
Prof. C. H. Toy, of Cambridge. 

The Semitic broken or internal plurals have commonly been regarded as collec- 
tives (the language treats them as singular feminines), in which the numerical ex- 
tension is indicated by an inward extension of form, as it is in the external plural 
by an addition at the end. There are difficulties in this viev/, one of which is that 
some of the broken plurals show also additions at the end, and Derenbourg 
(Journal Asiatique, June 1867) held the form in an to be a real external plural. 

A few years later (1870), M. Stanislas Guyard extended this suggestion of De- 
renbourg's so as to include all the broken plurals, which he endeavors to show 
are nothing but remnants, more or less disguised, of the regular external plural, 
somewhat as from English man we have men for men-er. He makes the follow- 
ing classes : 1 . forms showing the regular plural ending, with or without nunation or 
mimation — as an, which is dual-ending in Arabic, and plural in Ethiopic and Ara- 
maic, ay, plural in Aramaic, and i, plural in Hebrew (as, deban-m); 2. such as 
have lost the termination, but preserve the internal vowel-modification consequent 
on the addition at the end — as kital (from which oMal by prosthetic Elif), for 
kitdli; 3. those which show the sequence a-a-i, occurring in plurals like aradi, and 
thence extended by analogy to all quadriliterals and to other forms ; and the se- 
quence u-u, imitated from biliteral plurals such as swnuna from sanat ; 4. those 
which have substituted for the plural termination the feminine ending t. All 
other forms called in the grammars internal plurals he regards as true collec- 
tives, and not plurals. 

This explanation is in many respects an attractive one. It accounts for a part 
of the facts in a satisfactory manner ; it gets rid of an apparent anomaly in South 
Semitic inflection ; and it is in accordance with what we know of the prevailing 
genesis of the plural (by addition at the end) in all families of languages. In its 
turn, however, it presents serious difficulties. 

It supposes that Arabic and the other Southern dialects have a double plural 
system, retaining the full Semitic form as a living inflection, and alongside of it 
the same plural in curtailed shape, and also living, except that its plural character 
has been forgotten and it is treated as a feminine singular. This seems to be 
highly improbable. Modern Arabic has not stood still in the path of phonetic 
degradation; it has dropt the nominative, using the old genitive ma for all cases; 
and further, has largely given up the external in favor of the broken plural. But 
it keeps the two classes distinctly apart. This theory supposes that long ago the 
language had not only already gone further in the same direction of phonetic 
change, but, after having produced a curtailed plural, had lost consciousness of its 
plural character and treated it as a singular. Such a transformation at such a 
time seems hardly credible. 

Further, the theory involves a non-Arabic system of internal vowel-change. 
The plural aradi from ard M. Guyard compares with Hebrew debarlm from dabar 
or melakim from malk, and sees in the two the same broadening of the pretonic 
vowel. This, however, is distinctively Hebrew, and not Arabic ; the latter shows 
no such vowel-movement. A similar objection holds to the comparison of Arabic 
nisa, 'women,' with Hebrew construct nese and Syriac nese. It is the transference 
of the phonetic usages of one dialect to another, without historical grounds. 

There is nothing in the vowel-systems of these plurals that demands such a 
theory for its explanation. All the forms occur as infinitives, or as adjectives and 
nouns. The fact that quinqueliterals in making the plural reject one letter in 
order to have just space for the vowel-sequence a-a-i, on which M. Guyard is dis- 
posed to lay much stress, is not peculiar to the internal plural; a similar device is 
adopted in forming diminutives ami relative adjectives in ya, in both cases from a 
dislike to five-lettered words ; or, if the aim be to maintain a certain vowel-sequence, 
such sequence arises in the diminutive not through an external addition, but by 
a mere internal modification, and may so have arisen in the case of the plurals. 

Lastly, this theory fails entirely to explain certain of the internal plurals (mono- 
syllabic and dissyllabic triliterals), and these M. Guyard throws out of the cate- 
gory of plurals, and regards as singular collectives. The language, however, 
makes no distinction between them and the others, and so arbitrary a separation 



lx American Oriental Society : 

of the forms is unjustifiable, especially as collectives proper are in Arabic care- 
fully distinguished from those plurals. 

In spite, therefore, of the attractive simplicity of this explanation, and the in- 
genuity and learning with which it is presented by its author, it seems to labor 
under difficulties which, if not fatal, at least make it impossible for us to accept it 
till new light has been thrown on the facts. 

2. On Darmestetei's Translation of the Vendidad, by Prof. J. 
Lnquiens, of Boston. 

Prcf. Luquiens presented a review of this work of Darmesteter's, which consti- 
tutes the fourth volume of Miiller's series of Sacred Books of the Bast. His 
paper ended with the following conclusions: Considered from a literary point of 
view, the work leaves little or nothing to desire ; it is a bright and spirited 
rendering of a book which was not held to be either bright or spirited. If the 
chief aim of M. Darmesteter was to bring out in the strongest light the best sense 
to be elicited from the tradition, he has been eminently successful ; this result, 
however, seems an honor paid to the native commentators at the cost of a strict 
adherence to the text and to the most progressive methods of exegesis. As far 
as the coloring and subinterpretation of the Vendidad by the naturalistic myth 
are concerned, one must regret the hastiness, and yet admire the faith, which led 
him to thus irrevocably identify the fate of his work with that of theories not yet 
risen from the hypothetical stage. 

3. On the Metres of the Rig- Veda, by Mr. W. Haskell, of New 
Haven ; presented by the Corresponding Secretary. 

The object of Dr. Haskell's paper is to make a statistical exhibition of the fun- 
damental facts of Rig- Veda metric, as a necessary basis for future more detailed 
examination of the subject, having especially in view these three points : 1. what 
are the actual metres used, as opposed to those artificially distinguished and 
named by the Hindu commentators ; 2. what is their comparative frequency ; 3. 
what is the general metrical usage or law of each, as determined by an enumera- 
tion of quantities in a number of specimen verses. 

The metres are arranged on a (provisional) theory as to their historical relations, 
as follows : that the anustubh pada, of eight syllables, is the most primitive, and 
the anustubh metre, of four such equal padas, its normal form of occurrence, gayatrl 
and pankti etc. being the variations of this ; that the 8-syllabled pada is extended 
to one of twelve syllables more or less regularly alternating with the former, in the 
brhatl and other kindred metres ; that the jagatl is then made by putting together 
four 12-syllabled padas; that the tristubh pada, of eleven syllables, is a shortened 
jagatl; and that the 5-syllabled pada, of the dvipada viraj, is a syncopated 
tristubh. There are not, 'either in the Rig- Veda or in the Atharva-Veda, any 
other metrical elements than these ; all other so-called metres are various combi- 
nations of these elements, or imperfect and irregular verses, of varying degrees of 
irregularity, rising sometimes even to entire absence of traceable metrical form. 

The order of the metres in respect to frequency is a very different one from this. 
Here (omitting the minor variations and doubtful eases) the tristubh, of four 
1 1-syllabled padas, comes first, reckoning about 4200 verses, or over two fifths of 
the whole Rig- Veda ; the gayatrl, of three 8-syllabled padas, stands next, with near 
2450 verses (occurring especially in the 1st, 8lh, and 9th Books) ; then the jagatl, 
with near 1300 verses; the brhatl, satobrhatl, usnih, and other combinations of 
8-syllabled and 12-syllabled pfidas (especially in the 8th Book), near 1200 verses; 
the anustubh, over 800 verses; the pankti etc., of more than four 8-syllabled 
padas, about 250 verses. 

An enumeration of the heavy and liaht syllables, now, in fifty annstubh-verses 
(with omission, here as later, of a few syllables of doubtful value) gives the 
following results: 

vi. vii. viii. 

8 189 104 

187 1 92 



syllables, 


i. 


ii. 


iii. 


iv. 


v. 


light 
Anustubh: heavy 


87 
108 


33 
163 


52 
144 


31 
164 


186 
9 



Proceedings at Boston, May, 1881. lxi 

The pada of this type, accordingly, is one having a very marked iambic 
movement in its last half (the final syllable being, as in Greek and Latin, indiffer- 
ent), and a very weak iambic movement, consisting only in the greater preponder- 
ance of heavy syllables in the second and fourth places, in its former half. The 
different padas show no difference of structure that is worthy of remark — unless 
it be that at the end of the first and third padas the heavies are more frequent 
(namely, 54) than at the end of the second and fourth (only 38). The marked 
excess of heavy syllables throughout the whole former half of the pada is, as will 
be seen below, a feature shared by the 8-syllabled padas of all the other metres. 
The preponderance of lights in the concluding syllable of the pada belongs to all 
the metres without exception, and appears to indicate only the real indifference of 
that syllable, the greater natural frequency of light syllables showing itself there 
without hindrance. 

A similar enumeration for the other common padas of eight syllables — namely, 
the gayatrl, pankti (padas a-d), usnih (padas a, b), brhatl (padas a, b, d), and 
satobrhatt (padas b, d) — is as follows : 

syllables, i. ii. iii. iv. v. vi. vii. viii. 

Gayatrl : 

Pankti : 
Usnih : 
Brhatl : 

Satobrhatt : 

There seem to be no noteworthy differences of structure in these varieties of the 
8-syllabled pada: only the gayatrl shows a larger number of exceptional quantities 
than the others in its latter half. This is in accordance with the general greater 
irregularity of the gayatrl, rising even to a tolerably well-pronounced trochaic 
movement and cadence in certain hymns or parts of hymns ; such have been 
avoided in the enumeration here made. 

The total number of light and heavy syllables in the enumerated padas of the 
six metres is given below, along with a reduction to percentages, and statement 
of the limits within which the percentages vary (as between the different metres, 
as above reported) : 



light 


64 


38 


42 


34 


135 


17 


128 


96 


heavy 


85 


109 


105 


110 


14 


130 


21 


54 


light 


85 


61 


62 


44 


179 


12 


190 


138 


heavy 


113 


138 


137 


155 


20 


187 


9 


65 


light 


43 


20 


29 


18 


89 


10 


97 


63 


heavy 


55 


77 


70 


80 


10 


89 


2 


36 


light 


64 


41 


36 


49 


146 


6 


150 


98 


heavy 


86 


109 


114 


101 


4 


144 





52 


light 


42 


14 


42 


17 


92 


9 


96 


62 


heavy 


55 


83 


55 


80 


5 


87 


1 


35 



syllables, 


>. 


ii. 


iii. 


iv. 


V. 


vi. 


vii. 


viii. 


light 


385 


207 


263 


193 


827 


62 


850 


561 


heavy 


502 


679 


625 


690 


62 


824 


40 


334 


light 










93.0 




95.5 


62.7 


3ent heavy 


56.6 


76.6 


70.4 


78.1 




93.0 






limits 


55.4-573 


69.8-85.6 


56.7-76. 


61.3-84.1 


90.-97.8 


88.6-99. 


85.9-100. 


58.1-65.8 



Taking up, now. the padas of twelve syllables, there is a noteworthy difference 
between the usnih (8 + 8 + 12) on the one hand, and the brhatl (8 + 84-12 + 8) and 
satobrhatl (12 + 8 + 12 + 8) on the other (these three constituting more than four 
fifths of the whole number of mixed eight and twelve-syllabled padas, and the 
others being mainly extensions and variations of them). In the usnih, the 12- 
syllabled pada seems essentially an 8-syllabled one of the usual form, with four 
more syllables added at the end ; as will appear from the following enumeration 
of a hundred padas (half of them being those belonging to the 8-syllabled usnih 
padas already reported): 



lxii American Oriental Society . 



syllables, 


i. 


ii. 


iii. 


iv. 


V. 


vi. 


vii. 


viii. 


ix. 


X. 


xi. 


xii. 


n,. light 


49 


16 


38 


13 


75 


42 


76 


16 


94 


2 


95 


68 


in : , 

heavy 


50 


83 


61 


86 


24 


56 


23 


82 


5 


96 


4 


31 



The iambic movement of the middle quaternion of syllables is sufficiently 
marked, although by no means so cogent as that of the second quaternion in 
anustubh and gdyatrl etc.; it is especially faint in the sixth syllable, where the 
heavy do not very greatly exceed in number the light quantities. 

In brhatl and satobrhati, the middle quaternion has a quite other character : its 
first three syllables are prevailingly light, and the second of them (which in usnih 
was prevailingly heavy) is more uniformly light than either of the others, while 
the first is oftener heavy than the third. Thus : 



lyllables, i. 


ii. 


iii. 


iv. 


v. 


vi. 


vii. 


viii. 


ix. 


X. 


xi. 


xii. 


light 49 


15 


45 


12 


57 


91 


62 


10 


97 


1 


99 


61 


heavy 51 


85 


55 


88 


43 


8 


38 


90 


3 


99 


1 


39 


light 38 


15 


34 


13 


57 


76 


66 


5 


89 


5 


95 


70 


irhati : , KO 
heavy 58 


81 


62 


82 


39 


17 


31 


92 


8 


91 


2 


27 



This looks like an expansion of the ordinary 8-syllabled pada by an inserted 
element, tending toward the form — ■- — — (more nearly, in actual fact, — — — — ). 

The jagatl and tristubh padas agree quite closely in their metrical structure 
with this. As they are in all respects accordant with one another, save that the 
tristubh is catalectic, their enumeration may be presented together, thus : 



syllables, 


i. 


ii. 


iii. iv. 


v. vi. vii. viii. 


ix. x. 


xi. 


xii. 


light 


99 


18 


120 24 


104 164 138 2 


190 2 


193 


117 


ltl: heavy 


95 


175 


74 170 


89 23 56 192 


4 191 


1 


77 



light 115 26 105 24 116 166 100 5 186 4 120 

Tristubh: heayy g2 m 92 m 80 27 9 , m u 193 68 

The metrical movement of the second and third quaternions of syllables here is 
in no important degree different from what it was in the two preceding metres. 
On the other hand, the iambic character of the first quaternion is rather more 
marked, the light quantities even predominating over the heavy in the first and 
third syllables. No great stress, however, is to be laid upon this: in almost any 
set of verses examined, the preponderance will be found to be on the one side and 
on the other in different padas ; in another set of about 65 tristubh-verses whose 
syllables were enumerated, the heavy quantities were found to be, in all the padas 
together, slightly in excess of the light ; and in the 50 brhatl padas belonging with 
the 8-syllabled padas first reported, light syllables are in the majority in the first 
and third places. 

The summary of quantities, then, with percentages and limits of variation, for 
the 12-syllabled padas of brhatl and satobrhati. the jagatl pada, and the tristubh 
pada (counting its eleventh syllable with the twelfth of the others), is as follows : 



syllables, 


i. 


ii. 


iii. 


iv. 


V. 


vi. 


vii. 


viii. 


ix. 


X. 


xi. 


xii. 


light 


301 


74 


304 


73 


334 


497 


366 


22 


562 


12 


387 


368 


heavy 


286 


513 


283 


511 


251 


75 


222 


566 


26 


574 


4 


211 


, light 


51.3 




51.8 




57.1 


86.9 


62.2 




95.6 




99.0 


63 6 


P er oent - heavy 




87.4 




87.5 








96.3 




98.0 






limits, 1 


.58.4- 


84.3- 1 


. 61.9- 


86.S- 


53.9- 


81.7- 


50.8- 


90.0- 


91.7- 


94.7- 


97.9- 


60.3- 


h 


.59.2 


90.7 h.64.6 


88.0 


59.4 


91.9 


71.1 


99.0 


98.0 


99.0 


99.9 


72.2 



For the dvipada viraj, the thirty-one verses of i. 65-70 have been enumerated. 
The results are given for two successive padas, because the uniform and decided 
prevalence of heavy syllables at the end of the first pada of each pair (standing, 



Proceedings at Boston, May, 1881. lxiii 

if the provisional theory stated above be correct, in the place of the fifth and sixth 
syllables of a tristubh) appears to have a bearing of some importance on the view 
to be taken of the metre. Thus : 

syllables, 



u. 


ill. 


IV. 


v. 


VI. 


Vll. 


Vlll. 


!X. 


X. 


5 


93 


4 


21 


65 


7 


103 


6 


65 


111 


29 


118 


100 


56 


113 


18 


115 


56 



light 53 
Dvipada Viraj: heayy 69 

light 76.2 53.7 85.1 53.7 

per cent. heayy 56 6 g5 9 96 7 g2 6 94 2 95 Q 

The usual caasura after the fifth syllable is wanting in the fourth double pada 
of 68.1, and in the third of 70.5 ; and the same is the case in the first of vii. 
34.17 : a strong indication that the whole is essentially one pada. The occasional 
occurrence of an unsyneopated tristubh pada among dvipada viraj padas also 
helps to illustrate the transition : e. g. vii. 34.7 (second half) ; and, where the one 
metre changes to the other, vii. 34.21 (second half) ; 56.10 (do.). 

Any treatment of the other mixed metres, and of the irregular and defective 
verses, is reserved for a later communication. 

Dr. Haskell acknowledged the constant counsel and assistance of Professor 
Whitney in classifying and presenting the facts gathered by him. The sugges- 
tion of the true character of the dvipada viraj he owed to Professor Lanman. 

4. On the Sankhya Philosophy of the Hindus, by Prof. C. C. 
Everett, of Cambridge. 

It was maintained in this essay that the Hindu systems of philosophy differ 
among themselves fundamentally in regard to the view taken of the principle of 
subjectivity ; and that the difference in the accounts of the external world given 
by the various systems results from the difference in the conception of this prin- 
ciple. What may be called the Vedic system assumed the subjectivity of all 
knowledge and experience. Those early thinkers had discovered that man can- 
not get beyond himself. The world was to them a dream-world, and thus unreal. 
This view is implied in the Upanishads ; it is distinctly affirmed, and analyzed 
into certain proximate elements, in the Vedanta ; and by some later commentators 
is pushed to the logical extreme of an absolute solipsismus. The Sankhya system, 
on the other hand, affirmed the objective reality of the universe. It met the op- 
posing view with the only reply that could be logically effective. It found an 
element of objectivity necessarily present in the very form of subjectivity insisted 
upon by the Vedanta. It admitted in effect, at least in a certain sense, the dream- 
like nature of the world, but maintained that the dream as such was real and ob- 
jective. 

To make clear this statement, we must examine the nature of the soul (purusha) 
according to the Sankhya system. The soul was, to it, pure intelligence, without 
emotion or causality. This view of the soul has been regarded as meaningless 
and absurd by all western commentators who, so far as known to the essayist, 
have expressed any opinion upon the subject. But the idea of the soul must fur- 
nish the key to the whole system ; and if this is not understood, the system can- 
not be understood. It is importaut then to ask how this view of the soul was 
reached. We find indications of the method used. The existence of the soul as 
distinct from the body is shown by ihe fact that I speak of " my body." " T " 
must then be something distinct from my body. If it is objected that we also 
speak of the body of a statue, the answer is that this is pure tautology, the 
statue and its body being one. This reasoning we may carry further. Just as we 
say "my body," so we can also say " my mind," "my thought," " my feeling." 
Mind and thought and feeling must then be as distinct from the " I " as the body 
is. This may be illustrated in another way. We can not only say "I know" 
we can also say " I know that I know." We may thus have a regressits into trie 
infinite. This regressits the Sankhya philosophers had too much common sense to 
admit; and the "I" is posited as lying behind all consciousness. A similar re- 
gressus into the infinite is possible in the opposite direction. We can ask of any- 



lxiv American Oriental Society : 

thing '' What is its cause? " and again, in regard to the cause assigned, "What is 
the cause of it ?" and so on forever. Here the common sense of the Sankhyans 
affirmed prakriti, which was sinply and avowedly to give the resting place 
needed. We must especially recognize the fact that in the search for the " I," and 
in that for the first cause of objective being, the movement is in opposite direc- 
tions ; and further, we must observe that all which we leave behind us in seeking 
the first cause is one of its effects, and thus belongs to it. Whatever on the con- 
trary we leave behind in seeking the ultimate ego is cast off from it, and thus is 
foreign to it. The subject flees from the object, and, as it flees, it flings off one 
covering after another, until it stands naked and alone. While these views are 
implied in the whole discussion of this subject in the Sankhyan literature, and es- 
pecially in the Aphorisms of Kapila, they are perhaps most distinctly stated in the 
Aphorisms of Patanjali. Here, two counter hypotheses are suggested to account 
for consciousness. One, that the "mind" is directly self-conscious and thus 
needs no ego behind it ; the other, that self-consciousness is produced by mem- 
ory. The first suggestion is rejected because " attention to two objects does not 
take place simultaneously" (Patanjali, v. 19). The other is rejected because " it 
would require a cognition of the cognition": that is, it would involve a regressus 
into the infinite (ibid. v. 20). 

It will thus be seen that a profound psychological analysis underlies the San- 
khya system. The subject, when we come to the last analysis, is but a single 
point over against the whole world beside. We understand also how real objec- 
tivity was reached, a result that had baffled the Tedantin. The nature of the 
difference in the views of the outward world held by the two systems is also ob- 
vious. To the Vedantin, the illusion which forms the essence of the universe 
exists in and through the subject. It is the soul that is the basis and sphere of all. 
On the other hand, the soul, according to the Sankhya system, being the one in- 
most point of subjectivity, "Intellect," which fills the place held by illusion to 
the Vedantin, becomes wholly objective. It cannot have its basis and support 
in the soul. The necessity of finding a basis and substance for it elsewhere leads to 
the notion of prakriti, which is merely this substantial basis of " Intellect." So, in 
the one system, we have the series of " sheaths " wrapping the soul, sheaths of 
"Ignorance," growing more and more dense as they overlie one another; and, on 
the other, we have the same forms of existence produced in a series by "Intel- 
lect," "the great one," or by prakriti, that works through it. 

We see also the hope of deliverance which this view of the soul was fitted to 
bring to these thinkers, burdened by the thought of the evils of existence. If the 
soul is a mere spectator, it can leave when the show grows wearisome. Or 
rather, if it has no organic relation with the objective world, it has only to become 
conscious of this fact, to know itself to be free. This " discrimination " (from 
which perhaps comes the name of the system) is all that is needed. Through it, 
the soul that fancied itself bound knows that it is free. 

The essay discussed, along with the views here presented, the nature of the 
three " qualities " (guna), the relation of the system to religion, and other points 
connected with it. 

5. On Relative Clauses in the Rig-Veda, by Prof. J. Avery, of 
Brunswick, Me. 

Prof. Avery discussed in a statistical way the subject of relative clauses in the 
Rig- Veda, so far as concerns their position with reference to the corresponding 
antecedent clauses, and also the various modes of treating the antecedent word. 
All passages had been collected and classified containing derivatives of the rela- 
tive root ya : excepting, however, yad, yadi, and yafha in clauses expressing con- 
dition or purpose ; yad as a conjunction meaning ' that, since, so that, although ;' 
and yafha, in the sense of iva. The passages are very nearly 4,000. 

I. The antecedent clause stands first more than 2,000 times, or 50.8 per cent, 
of the whole number of occurrences. The antecedent is fully expressed in its own 
clause alone near 1200 times (29.5 per cent.): e. g. term . . . gatarh rathena . . . 
yena cacvad uhafhur ddfuse vdsu (i. 47. 9), ' come with that chariot with which ye 
have constantly brought good things to the worshipper;' apo devl'r upa hvaye ya- 
tra ga'vah pibanti nah (i. 23. 18), 'I invoke the heavenly waters, where our kine 



Proceedings at Boston, May, 1881. lxv 

drink.' The antecedent is expressed in its own clause, and it or a synonym is re- 
peated in the relative clause, 60 times (1.5 per cent.): e. g. dmandan ma maruta 
stdmo dtra ydn me narah frutyam brdhma cakrd (i. 165. 11), 'the praise hath 
pleased me here, Maruts, what famous prayer ye have made for me, ye men.' 
The antecedent is expressed in the relative clause, and represented by a pronoun 
in its own clause, 51 times (1.3 per cent.): e. g. lyus te, ye pu'rvatardm dpacyan 
. . . mdrtydsah (i. 1 13. 11), ' gone are they, what mortals beheld her before.' The 
antecedent is found in the relative clause alone 69 times (1.7 per cent.) : e. g. dpa 
dahd' 'ratlr yebhis tdpobhir ddaho jdruiham (vii. 1. 7), ' burn away the grudgers 
with what heats thou didst burn the waster;' nd vi jdndmi ydd ive 'dam dsmi (i. 
164. 37), 'I do not understand quite what I am now.' The antecedent is not ox- 
pressed in either clause, except by a pronoun or adverb, 670 times (16.8 percent.): 
e. g. ihd bravltu yd u tdc ciketat (i. 35. 6), ' let there speak here whoever knows 
that;' aganma ydtra pratirdnta d'yuh (i. 113. 16), 'we have gone where they 
lengthen out life.' 

II. The relative clause stands first more than 1850 times, or 46.8 per cent, of 
the whole number of occurrences. The antecedent is fully expressed only in its 
own clause over 900 times (22.6 per cent.), by noun, pronoun, or adverb : e. g. yd 
rdyd ' vdnir mahd'n . . . tdsma indrdya gdyata (i. 4. 10), ' who is a great stream 
of wealth, to that Indra sing ye:' ydtra grd'vd vddati tdlra gachatam (i. 135. 7), 
• where the pressing-stone is uttering its voice, thither go:' ydtra 'ham dsmi td'n 
ava (viii. 75. 15), 'on what side I am, them favor thou.' The antecedent is ex- 
pressed in the relative clause, and it or a synonym is repeated in the antecedent 
clause, 56 times (1.4 per cent.): e. g. ye te pdnthdh . . . tebhir no adyd pathibhih 
sugebhl rdksa (i. 35. 11), ' what paths are thine, by those easy paths guard us this 
day.' The antecedent is more fully expressed in the relative clause, and repre- 
sented in the antecedent clause by a pronoun or adverb, 276 times (6.9 per cent.) : 
e. g. yd' ta utih . . . tdyd no hinuhl rdtham (vi. 45. 14), ' what help is thine, with 
that urge on our chariot;' yd ha vHm mddhuno dr'tih . . . tdtah pibatam (viii. 5. 19), 
'what wine-skin of mead is yours, thence drink ye.' The antecedent is expressed 
in the relative clause alone 109 times (2.7 per cent): e. g. yd indra tusmo magha- 
van te dsti ciksa sdkhibhyah (vii. 27. 2), ' what might, magnificent Indra, is thine, 
bestow on thy friends.' The antecedent is not expressed in either clause, save 
by a pronoun or adverb, 526 times (13.2 per cent.): e. g. yd asmd'h abhidd'saty 
adhararii gamayd tdmah (x. 152. 4), 'whoso attacketh us, send thou to lowest 
darkness ;' iydm visrstir yata ababhu'va , . . so aitgd reda (x. 129. 7), ' whence this 
creation came into being here, he verily knoweth.' 

Til. As a third position, the relative clause stands within the antecedent clause 
94 times (2.4 per cent.). The same variety of treatment of the antecedent appears 
here as under the two preceding heads. Examples are : tdsya sddhvi'r isavo yd'- 
bhir dsyati nrcdksasah (ii. 24. 8), 'his are successful arrows, with which he shoots, 
men-beholding [ones]' etc.; catrum apa bddhasva durdm ugrd yah fdmbah puru- 
huta term, (x. 42 7), ' drive far away the enemy, O much invoked one —what 
weapon (?) is terrible, with that one (i. e. with whatever weapon is terrible) ;' 
muncdtam ydn no dsti ianu'su baddhdm krtdm mo asmdt (vi. 74. 3), ' put away what 
sin committed is bound to our bodies from us;' nahi nu yd'd adhlmdsl' 'ndram kd 
virya s pardh (i. 80. 15), 'for no one, surely, so far as we know, is beyond Indra in 
might.' 

IV. Once more, by a process the reverse of that just noticed, the relative clause 
takes the antecedent one wholly into itself. This singular arrangement occurs 
only twice, namely: yd djistha indra tarn su no da mddo vrsan (vi. 33. 1), ' what 
is the mightiest, Indra, do grant that to us, passion, O hero;'' yd eka it tdm u stu- 
hi krstlnd'm vicarsanih pdtir jajne (vi. 45. 16), ' who verily alone, him praise thou, 
is born the chief lord of men.' 

The natural position of the relative word seems to us to be at the head of its 
clause: and it is in truth found there in the Big-Veda about 2600 times (65 per 
cent.); but it has the second place near 1,000 times (24.4 per cent.), the third 
place over 250 times (6.5 per cent.), the fourth place 81 times (2 per cent.), and so 
on, in decreasing frequency, down to the ninth place. 

The preparation of this paper was suggested by certain brief statements made 
by Prof. Delbriick in his work on the Use of the Subjunctive and Optative in 



lxvi American Oriental Society: 

Latin and Greek. He there says, in substance, that while the nature of the rela- 
tive is such that the clause which it introduces should follow the principal clause, 
it in fact precedes it in most cases in Sanskrit— meaning, apparently, the Veda; or 
at least including the Veda, since that is the principal source of the examples 
quoted throughout his volume. In this usage, he declares, which is of secondary 
growth, the Sanskrit differs from the Greek of Homer. Now if my statistics are 
correct, it appears that in the Rig- Veda, at least, the relative clause retains its 
primitive position in a (small) majority of cases. He further states that the two 
forms of sentence, where the relative clause either precedes or follows the antece- 
dent clause and the antecedent word is expressed in its own clause only, are not 
very frequent. On the contrary, if we include in these forms the instances where 
a personal pronoun serves as antecedent, they are half the whole number of occur- 
rences. Again, we are told that the cases where the antecedent or a synonym is 
repeated in the relative clause are common. I find them uncommon, being less 
than 3 per cent, of the whole number. Yet further, it is maintained that by far 
the most frequent arrangement is that in which the antecedent word appears in 
the relative clause only ; that when the latter follows the principal clause, there 
is no reference in that clause to the antecedent; and that, when it precedes the 
principal clause, the antecedent is generally represented in the latter by a form of 
the demonstrative ta. The last only of these three statements seems to be cor- 
rect, so far as the Rig- Veda is concerned. It would appear that the author's 
views rested upon general impressions derived from reading, rather than upon any 
enumeration of instances. 

6. Studies on the Mahavana or Great Vehicle School of Bud- 
dhism, by Mr. Wm. W. iSoekhill, of Baltimore, Md. 

The object of Mr. Rockhill's paper was to set forth some of the principal fea- 
tures of the doctrines of the Mahayana school from hitherto unpublished Tibetan 
documents, and also to show the differences that exist between the older mahayana 
sutras, of which the Sutra in 42 Chapters (see Proceedings for Oct. 1880, above, 
p. 1.) is an example, and those of later dates. The following is a brief abstract. 

The oldest form in which we find the sutras of the Great Vehicle is furnished 
by the Sutra in 42 Chapters, in which the different points considered are set 
forth in unpretending, plain language, without any of the repetitions or embellish- 
ments of more recent works. The doctrine that is taught does not differ to any 
great extent from that of primitive Buddhism. 

The sutras on transcendental science (prajnd paramitd sutras) expose the more 
perfected form of teaching of the Mahayanists of the Madhyamika school. The 
object of all these works is thus defined by Eug. Burnouf (Intr. a l'hist. bud. ind., 
p. 483): "!Les livres de la pradjna paramitd sont consacres a l'exposition d'une 
doctrine dont le but est d'etablir que l'objet st connaitre ou la perfection de la 
sagesse n'a pas plus d'existenee reelle que le sujet qui doit connaitre ou le 
Bodhisattva, ni que le sujet qui eonnait ou le Bouddha." 

The Vqpachedika (Rdo-rje gchod pa) is a good sample of these works. It is 
quite short (18 folios in the Tibetan text), and may consequently be considered as 
older than the similar works in 100,000 and 8,000 clokas. This text differs in 
many respects from the Chinese, an English translation of which was given in 
1864 by Mr. S. Beal (Jour. Roy. As. Soc'y, new series, vol. i.). The " Histoire de 
la vie et des voyages de Hiouen Thsang," p. 310, gives some of the objections to 
Kumarajlva's Chinese version (the one followed by Mr. Beal). The Tibetan text 
approaches much nearer the Sanskrit original, of which a copy exists in the 
Bibliotheque Nationale (fonds Burnouf, No. 34). 

The founder of the Madhyamika school is said to be Nagarjuna (or Nagasena) ; 
but from different passages of Taranatha and of the work of the biographers of 
Hiouen Thsang (p. 274), " Kumarajiva was a contemporary of Acvaghosha, Deva, 
and Nagarjuna," etc., we conclude that he was the great representative of his 
school before it assumed its definite form. According to the above statements, he 
must have lived towards the end of the IVth and commencement of the Vth cen- 
turies A. D. 

The Kayatraya sutra, which belongs probably to the Togachfirya sect of the 
Mahayana school, is a short text taken from vol. xxii., mdo section (fol. 8la-b), 



Proceedings at Boston, May, 1881. Ixvii 

of the Bkah-hgyw. It teaches that all Buddhas are endowed with three bodies, 
the dharmakaya ' the body of the Law,' the samVhogakaya ' the body of perfect 
acquirement,' and the nwmanakdya ' the body incarnate.' When they have 
finished their ministry in this world, they divest themselves of the nirmdnakaya, 
but retain in the Parinirvana the two other purer forms, of which they have 
become possessed on account of their omniscience and many perfections. This 
text differs considerably from the one mentioned in Julien's Si-yu-ki (liv. iv., p. 
240, note). 

The third and last sutra, the aparimita dyurjndna sutra, seems to be of very 
recent origin. The text that has here been used was published by the Baron 
Schilling von Cannstadt. The Buddha does not instruct Kumarabhuta Manjucrl 
on any point of the doctrine. He simply tells him that, if the present work is 
copied, recited, or even kept in the house, it will greatly prolong life. The sutra 
or sermon has here become a magical formula, the simple repetition of which is 
all that is necessary to salvation. 

The aparimita dyurjndna hrdayand dhdran'i, the sequel to this sutra, professes 
to contain in a charm of a few words all the virtue and power of the sutra itself. 

The tantrika school, to which the last sutra belongs, was introduced into Tibet 
in the Xlth century, and has been predominant there since that time. 

7. On Lepsius's Views of African Languages, by Prof. W. D. 
Whitney, of New Haven. 

Prof. Lepsius has recently (1880) published a Nubian Grammar, the fruit of 
studies begun during his celebrated expedition to Egypt and Ethiopia, in 1 842-6, 
and afterward continued under favorable circumstances in Germany. It is 
worked out with the thoroughness, and in the clear and attractive style, which 
are characteristic of its author. Besides the grammar itself (200 pages), there is 
a body of Nubian texts (60 pages), a Nubian-German and German-Nubian vocabu- 
lary (180 pages), and an appendix (60 pages) on the dialects of the language, 
including also a criticism of Reinisch's work on the Nubian. To the whole is 
prefixed an Introduction (126 pages), on the classification and relationships of 
African languages in general ; this will interest, of course, a wider circle than the 
rest of the volume, and is worthy of the most careful attention. 

Lepsius believes all the African races proper to exhibit only a single physical 
type; and in addition to its ordinarily recognized characteristics he calls attention 
to a forward tilt of the pelvis, which gives a peculiar bearing to the body. But 
he regards the northern and northeastern peoples, the so-called Hamitic races, as 
early intruders from Asia, followed later by the Semites, these two divisions being 
ultimately related with one another. The whole southern peninsula of the conti- 
nent, now, from 7° or 8° N. L. nearly to the Cape, being filled (with the insignifi- 
cant exception of the Hottentot and Bushman) with the dialects of a single well- 
defined family, the South-African or Bantu, and there being between these and 
the Hamitic a broad band of heterogeneous tongues, falling into numerous and 
discordant groups or families, he holds the Bantu and the Hamitic to be the 
two original language-types, and the others to be the product of their mutual 
modification and mixture. The generalization is a grand and striking one: and if 
it be true, its demonstration in detail will constitute a highly important division 
of linguistic history. Without laying any claim to the detailed knowledge that 
would enable him to criticise it with authority, Prof. Whitney reported succinctly 
the author's views and arguments, and commented on them, especially on those 
to which he was obliged to take exception. In his opinion, there were too many 
questionable points involved in it to allow of our accepting it otherwise than pro- 
visionally, as a basis for further investigation. 

There is, in the first place, the capital question whether the influence of one 
language can so metamorphose the structure of another as the theory would imply. 
The prevalent views as to language-mixture are called in this work an "assump- 
tion" and "prejudice;" but they appear rather to be the best induction thus far 
possible from the known and indisputable facts of mutual influence of languages, 
and cannot be put down except by actual proof of their inapplicability to a given 
case ; if an offered solution of the African problem simply takes for granted their 



lxviii American Oriental Society : 

falsity, we are driven to inquire whether some other solution is not possible. 
Prof. Lepsius draws up a list of twelve leading particulars in which the Bautu 
and Hamitic tongues differ, and by them tests the intermediate tongues, ascribing 
the agreements and disagreements of the latter to the influence of the one or of 
the other element. The method is not without its dangers, since the differences 
of an}' two languages may be taken as test, and other tongues will be found to 
stand upon the side either of the first or of the second with regard to each point 
of difference (for a door must be either shut or open) ; the question of origin of 
the discordance is still left to be settled. Two of the adopted criteria are of 
wholly indecisive value, because even the Hamitic dialects themselves differ in 
regard to them ; two or three more are such phonetic matters as even nearly rela- 
ted tongues of other continents are sometimes found to differ upon ; the rest ar- 
range themselves mostly under two heads: prefix or suffix structure, and gender 
founded on sex. As to the first, the intermediate tongues are very discordant, 
and many of the facts brought to notice by Lepsius are in the highest degree 
curious and interesting; but it seems still to be open to question whether more of 
it all than he is inclined to allow, in Bantu and elsewhere, may not be the product 
of positive growth out of a less developed general condition, and not mere decay 
and metamorphosis of an original structure most nearly represented by the Bantu. 
We should not limit too narrowty the possibilities of new production in agglutina- 
tive tongues: our author himself gives a very notable example of this, in exhibit- 
ing the acquisition by certain Upper Nile dialects, not under Hamitic influence, of 
an apparent sexual gender distinction, growing, as he believes, out of an earlier, 
grosser and more material, distinction between stout and puny. Perhaps the wide 
territorial domain of the Bantu gives a false impression of its predominant impor- 
tance as a factor in the history of African language ; there is nothing in its pres- 
ent extension to prove that it might not have been originally a coordinate mem- 
ber of the congeries of Central African groups, to which favoring circumstances, 
along with the superior capacities of its speakers, have given a very exceptional 
growth; whether there is anything in the language itself to show the contrary, 
remains to be ascertained. 

The subject of gender is one of leading interest in the Introduction, and the 
highest degree of value as a criterion is attributed by the author to this grammat- 
ical element. He holds, for example, the absence of gender in Nubian to be a 
sufficient indication that that language is fundamentally Central African ; though 
in all the other respects considered by him it agrees with the Hamitic. He holds 
the Hottentot to he Hamitic solely because it has gender, while in other points of 
structure and in material no trace of anything Hamitic is discoverable about it, 
and while the physical type of the race is purely, if not exaggeratedly, African ; 
he believes the Hottentots to represent a branch of Hamitic stock, severed from 
the rest by the crowding outward of the Bantu peoples, and pushed southward, 
with an ever-increasing admixture of African blood, till its Hamitic characteristics 
were completely swamped. And this, although he has shown us an example, as 
noticed above, of the virtual acquisition of gender by a body of African dialects, 
and the Persian offers a familiar example of a language of our own family that 
has utterly lost the distinction. He regards the common (and nearly exclusive) 
possession of gender by the Indo-European, Semitic, and Hamitic families as 
proving their ultimate relationship : the fact is certainly a very striking one, and 
that it may have so decisive a bearing need not be too dogmatically denied; 
while at the same time we are justified in regarding this as unproved, and even 
in the highest degree questionable, considering how probably the distinction ap- 
pears to have been worked out in the course of the structural growth of each 
division of language. Prof. Lepsius endeavors to find a psychological basis for 
the African classes, on the one band, in the attitude of African savage man 
toward nature, and for the genders of the higher races, on the other hand, in the 
regulation of the relations of the sexes which made family organization the start- 
ing-point of the superiority of those races. Various considerations were adduced, 
however, to cast doubt upon the sufficiency of either explanation. Thus, as re- 
gards the latter, it does not seem clear that a moral organization of the family, in 
our sense, any more than the virtues of benevolence and justice, are what ad- 
vances a race that is struggling upward toward power ; then, all languages have 



Proceedings at Boston, May, 1881. lxix 

distinct names for human beings in all their various relations, and can by help of 
these constitute the family as purely as they have moral sense for; and it is no 
honor done to the element of sex to extend it fancifully to everything in crea- 
tion, any more than it would show a keen sense for form to call birds and 
the weather square, aud goodness and headaches round; and the most important 
words designating gender in Indo-European, father, mother, brother, sister, daugh- 
ter, have no gender characteristic, either in derivation or in inflection. On the 
whole, gender remains still the same difficult and trying problem as hitherto : un- 
less we are to see in the special gender-development out of a distinction of size 
and dignity on the part of the group of Nile languages referred to above a valu- 
able hint as to what the history of the same thing may have been in our own 
language. 

Other of Prof. Lepsius's general views laid down in this work were reported : 
thus, for example, his repudiation of "Turanian" affinity for the race that laid 
the foundation of Mesopotamian culture, and his reduction of the latter to an 
Egyptian origin through Cushite mediation. The hope was expressed that he 
would take occasion to write himself out more fully on this subject, with state- 
ment of his reasons. 

8. On a Manuscript Fragment of the Samaritan Pentateuch, 
by Prof. Isaac H. Hall, of Philadelphia ; presented by Prof. Toy. 

Some days ago, through the kind offices of Rev. Dr. W. Hayes Ward, I came 
into possession of a parchment folio, or pair of leaves, written in the Samaritan 
character, quite old, and somewhat obscure. Tt was obtained from a Jew, who 
stated that he brought it from Jerusalem fifty years ago. 

The size of each leaf is 4J x 3£ inches ; of the written page, 3 x 2£ inches. 
It is written with twenty-four lines to the page, except that a word is pushed 
into the twenty-fifth line on two of the pages, and on another the same is true of 
the punctuation at the end of a chapter. The style of writing is that of ordinary 
Samaritan manuscripts, with a fine point or dot to separate the words, here and 
there replaced by a punctuation mark like a colon. At the end of a chapter the 
punctuation is like that seen in Petermann's edition of the Book of Genesis in 
Samaritan characters. Spaces are left between the letters toward the end of a 
line when necessary, so that the last letters of the lines may stand in an upright, 
even column. No words are divided at the end of a line. A hole in the parch- 
ment, older than the writing, divides some words, in one case separating the 
letters by more than half an inch. Paragraphs are marked by leaving a whole 
line blank. 

The manuscript is a fragment of the Samaritan Pentateuch, containing Numbers 
xxvii. 24 (beginning at 7X&)) T^K J1K) — xxviii. 16; xxxii. 23-42. An easy com- 
putation shows that just eight pages, or four leaves, or two folios, were inside 
this folio in the quire when the MS. was complete. It was therefore the middle 
folio, or one of the outer folios, of the quire : if the quire was a ternio, then it 
was the outer one, which I do not think was the case. 

The writing begins in a verse which I have called 24, above ; but it is a verse 
not there in the Hebrew, added after verse 23 from Deuteronomy iii. 21, 22, 
slightly altered. The paragraph and chapter end with this extra verse in the 
MS.; and the next paragraph ends with verse 10. Another paragraph ends with 
verse 15 ; and the page ends with the third word of verse 16, [rTJ7]a*TSO, of which 
last word only the first 3 can be read without a lens, and the last two letters are 
hopelessly defaced. The previous word is interrupted by the hole : thus, Jlty CNTT. 

The next leaf begins with Numbers xxxii. 23, and has paragraph divisions at the 
end of verses 28 and 33. The last page ends with the chapter, at verse 42. 

In connection with the following collation with Blayney's edition of the 
Samaritan Pentateuch (Oxonii, 1790) are noted the ehirographical peculiarities 
not mentioned above. "When not otherwise stated, or a parenthesis not used, the 
variation from Blayney's text is to be found in his lower margin as a manuscript 
reading. I have not thought it worth while to repeat them from his edition. 

Chap, xxviii. 2, 'tyxS for ntyxS. Here the ' seems to be a re-inking of a faded 
D, and not a correction, or change of mind of the original scribe. The two letters 



lxx American Oriental Society ; 

resemble each other very nearly in the script. — "ITi for TttVJ- This is slightly 
different from Blayney's variant, which is TTJ. 

Verse 5. rWBflN for n'TBjn. 

JiflUTI for nS'XTI. Here the ]} is written over a faded N. It is difficult to 
account for this change in restoration except by ignorance. This variant is of 
course not given by Blayney. 

Terse 7. VXJ1 for 1X31. 

Verse 8. [UDJDl for J3DJ31. 

Verse 9. Same as above in verse 7. 

(Verse 12. "WW! is omitted by error of scribe, but added by a later hand above 
the line.) 

Numbers xxxii. 24. (nnjl for niTUI by mere error, but the T is added above 
the line prima manu.) 

(Verse 26. UBBT1 for USB, but the superfluous n has a stroke drawn obliquely 
across it by a later hand in token of erasure.) 

Verse 28. nUX for ni3Nn. 

(Verse 29. yeHTI JIN for pNnN; but the n has a horizontal stroke drawn 
above it prima manu, in token of erasure.) 

(Verse 33. pK for pXTI ; but the n is added above the line prima manu.) 

Verse 38. nUDID for niDlD- 
HDty for DB>. 

So far as can be seen from this comparison, the manuscript appears to be a 
very respectable one. It is also evidently ancient ; but how ancient, I have no 
means of determining. The collation discloses only one real variation from 
Blayney's text or margin ; and that of no great importance. Its real interest lies 
in its disclosing the fact of a partial re-inking, and a correction both by the 
original scribe and a later hand, and the manner of so doing. It is worth while 
to remark that there is one vacant space, in one of the lines, large enough for a 
whole word. I am unable to determine whether this is an actual erasure, or left 
blank originally because of a defect in the surface, or to make the line come out 
even. In some cases the spacing seems to be done for the latter purpose through- 
out a whole liue, sometimes only through the last half, but oftener Only in the 
last word or two. One line leaves wide spaces between both the words and the 
letters of a word for that purpose. 

9. On the Assyrian Monuments in the Museum of Fine Arts 
at Boston, by Rev. Selah Merrill, of Andover, Mass. 

These monuments consist of seals, a number of casts of important relics, and 
one very fine slab, recently received, of Assur-nazir-pal, B. C. 885-860. This is 
similar to other slabs of this king that have previously been brought to the 
country, and from its perfect preservation it may be classed among the very best 
of them. The inscription upon it is clear, and is generally known as the "Stan- 
dard Inscription." A detailed account of all the Assyrian monuments then known 
as having been brought to America, accompanied by translations, was presented 
to the Society by Mr. Merrill at its meeting in October, 1874. The design of the 
present paper was first to call attention to the desirability of supplying our 
museums with casts of these valuable relics and records from Nineveh and Baby- 
lon ; secondly, to point out some new features in the slab here mentioned ; and 
thirdly, to describe briefly some new inscriptions of Assur-nazir-pal that have 
lately been discovered. 

Dr. Ward, of New York, had brought with him copies of all 
the recently discovered Hittite inscriptions, but the lateness of 
the hour rendered their exhibition impracticable. 

After passing a vote of thanks to the American Academy for 
the use of its room, the Society adjourned until Wednesday, Oct. 
26th, 1881.