.~.--,.,. ....... _-~ ............. _ ........ ~ ............... -.

l l UNIVERSITY OF DELHI l l l l 1 l 1 1 1 l 1 l 1 1 1 1 1 " INSTITUTE OF POST GRADU- ,l l ATE (E) STUDIES LlBRARY ! 1 1 l 1 l 1 1 1 l l .............................................. ~ ....................................... ,..

EP5 1~.STnL TE CF FST-GRt-DUATE (E) SlUDIES

LIBRAR) (",, "0 G 1 ' 2..~'. ({ep '.503) \ \ , F,G, Ac. !'~O. E "'2. L.L~

r~\I~ 1)" ... _ho~l\l \" ,~ ... I~.:l or ('Il r'!oro t'. 1 (:. ,. j b hWJ. J"I "H'r:lue CII\~1 ~ '/5 , :. _' -, . - ,,' '1 \o,' b:,_!~C', '0 C":l:nc'a ~Hlkl]l..;1 c. ,'Q~2.&.IL71

--f---- - -- ---- --

- . --------1 -------

--~------!--------

-------- - ----- ---

-_._._-

~-~~~-+-- -- -~-----

I)INfO'OGERMANISOH E 81 8LIOTH EK

HERAUSGEGEBEN VON

H. HIRT UNO W. STREITBERG t

ER~TE ABTEILUNG

SAMMLUNG INOOGERMANISCHER LEHR- UNO HANOBCHER

l. REIH~: GRAMMATIKEN ACHTZEHNTER BANO

THE SYNTAX OF CASES IN THE NARRAilVE ANO OESCRIPTIVE

PROSE OF THE BRAHMAI':JAS BY

HANNS CERTEL

...

HEIOELBERG 1926 . CARL WINTER'S UNIVERSITATSBUCHHANOLUNG

THE SYNTAX OF OASES IN THE NARRATIVE ANO DESORIPTIVE PROS E OF THE BRHMANAS

BY

HANNS CERTEL

l.

THE DISJUNCT USE 0F CASES

HEIDELBERG 1926 CARL WINTER'S UNIVERSITATSBUCHHANDLUNG

Verlagtl-Nr. 1960.

Alle Rechte, besonders dlls ReoM der Ubersetzung In fremde Sprn.chen, werden vorbeha.lten

ro

Jacob lFackernagel this book is gratefully znscribed.

Foreword. The material presented in these volumes was originally collected

and primarily intended to serve as a basis for the extension of the 'compendious statement of the use of cases' ( 268-305, p. 89-103) in a proposed revised edition of Whitney's Sanskrit Grammar. If these syntactical paragraphs were to reCeIve a treat-ment comparable and conform to the statistical and historical treatment (Preface to the first edition, p. VII, 5-6 and 20-21) of the inflexional and morphological material in Whitney's Grammar a much more detailed and complete digest of the facts as ex-hibited in the literature o the earliest prose writings than could be gleaned from the admirable but necessarily brief paragraphs of Delbruck's Altindische Syntax ( 64-126, p. 103-188 and 216-217, p. 387-390) seemed indispensable.

The matenal is here given in extenso with a fulness of detail which would have been neither practicable nor admissible as part {)f the revised Grammar itself in the hope that it may not only save Sanskritists the time-consuming task of working through these extensive and often dreary texts but also be useful to those who are interested in Comparative Grammar.

The first two volumes (including the translations and many of the critical and exegetical notes) were substantially finished in manuscript about a decade ago aud intended for publication in the United States. This accounts for the English dress in which the book appears, for it seemed time and labour wasted to recast the wbole into German.

My thanks are due to tbe Notgemeinscbaft der deutschen Wissene schaft for a substantial grant wit\lout which the publication would have been impossible. But the greatest debt of gratitude 1 owe to Professor Jacob Wackernagel who kindly undertook the laborious work of reading one proof and favomed me with many valuable suggestions.

Mnchen, April 1926 . Hanns Oartel.

VII

Preface. 1. General Plan and Scope.' The present work is intended

to present, as completely as possible, the use of cases in tha-narrative and descriptive prose of the Brahma:r;.as.

The whole bulk of early .Sanskrit literature may, for syntactjc.al treatment, be eonveniently divided into the following seven groups ~

1. The metrical Mantras, 2. Tbe prose Mantras,

Both the metrICal and the prose Mantras are fully recorded in Bloom-field's Concordance. Witb regard to tbe latter the autbor of thls monu-mental dlgest rlgbtly remarks (p VIII, col. 2, 4). 'These prose formulas are m a dIalect or in a style tbat dIl'fers not a httle from the narratIve or descriptIve plOse of tbe Br.1bmanas and Stras. Tbe formulas abound at any rate in poetlc or otber archaisms and deserve to be collected ana treated by tbemselves.'

3. Oertain Ara:r;.yaka-like pr0se passages in the Atbarva V eda~ 4. Tbe narrative and descriptive prose of the Brahma1).RS proper. 5. The prose of the Ara:r;.yakas, 6. Tbe prose of tbe older Upanieads, and 7. The prose o tbe Stras. Each one of these seven groups deserves separa te and detaileq

treatment to furnish the material for a complete picture of tbe historieal development of early Sanskrt syntax.

It is then the fourth group, viz. the narrativa and descriptiva-prose of the Brabma1).as proper, which forms tbe subject of tbe present work.

In three respects only this restriction has not been strictly ad hered too (A) Certain Qlokas which are essentially narrative or descriptive in character, su eh as tbe Qlokas of the Quna"\ll(epa legend, bave been included although they do not faU within the scope of any of the seven groups enumerated above. This has been done for the sake of convenience, for there are relatively so few of them that a separate treatment of their case syntax would vrn

Preface.

not be worth while. They are distinctly marked as Qlokas. (B) For the same reason the prose passages of the AV. have been included. (O) The question whether the restriction of the present work to the BrhmaI;las in the strict sense of the word meant the exclusion of those Arawakas nd older Upan~ads which now form separate texts has been given careful consideration. As a matter o principIe, a restriction to the Bri~al}as proper seemed ad-visable. But other considerations favoured their inclusion: It is by a mere chance that e. g. the Brhad AraI;lyaka Upani~ad (now reckoned as an independent Upani~ad) happens to form an inte-gral part of the Qatapatha BrahmaI;la, while a text lIke the Ohan-dogya Upani;;ad is separately handed down by our tradition. More-over the prose of the AraI,lyakas and of the older group of Upani;;ads is so closely sImilar to that of the BrahmaI;las proper that it seemed inadvisable to omit the syntactical material furnished by them. And finally, the AraI;lyakas and the older Upani~ads antedate without doubt a late compilation like the Gopatha BrahmaI;la, however greatly the various parts of this last work may diifer in time o composition. Tbe grent weight of these considerations finaJIy induced me to include the syntactical matter of tbe AraI;lyakas and of the early prose Upani~nds 1ll the present work. These examples are plainly marked as 'Additions from the AraI,lyakas and UpaI,li;;ads', and included within r J brackets.

Tbe following texts are then included: 1. AB. The Aitareya BrahmaI,la; Aufrecht's edition. 2. KB. The Kau~Itaki or Qakbayana BrlihmaI,la;' Lindner's

edition. 3. PB. The Pancaviqa or TaI}Q.ya Maha BrahmaI,la; the Biblio-'

theca Indica edition. 4. SB. The SaQ.viuqa BrahmaI).a; Eelsingh's edition. 5. JB. The Jaiminfya or Talavakara BrahmaI,la .

..

Thls text, whlch is In a very corrupt state, has as yet not heen entirely .. puhlished. The pertinent material has, as far 3S posslble, been collected

from a manuscnpt copy mcludmg a collation of aH extant manuscripts made by Professor Wbltney and others (see JAOS 11, p. cxllv). Bemg unable to consult the original manuscripts 1 had to rely on the correctness of this transcnpt. A large number of extracts from tbe JB. has recently heen

IX

Preface.

pubhshed by Caland 1 and leferences to th1S paper have been added m theif proper place under the caption 'Caland, Auswahl.'

6. QS. The Qatapatha.BrahmsI).a; Weber's edition. 1 regret that the Kanva recenSlOn of thIS text was not accessIble to me.

For syntactICal purposes a comparison of the Kanva version with the Ma dhyandina verswn would hare been of the greatest mterest and importance

7. TS. The prose parts of ~the Taittirya Sarilhita; Weber's edition.

8 TB. The prose parts of tbe Taittirya BrahmaI).a; the Biblio-theca Indica edltion.

9. MS. The prose parts ofthe Maitraya:l}.1 Samhita; v. Schroeder's edition.

10. K. The prose parts of the Kilthaka, v. Schroeder's edition 11. G B. Tbe Gopatha BrahmaJ)a.

The material was orlginally collected from the Bibhothera Indica editlOn. Fortunately the te::d could be revIsed on the basls of D. Gaastra's edItlOn.

The 'Additions from the AraJ)yakas and Upani~ads' present the material euUed from the foIlowing texts:

12. AA. The Aitareya Aranyaka; Keith's edition. This comprlses the AitareyaUpamsad.

13. Qankh Ar. The Qankhayana Aral).yaka, edited by FrIed-lander (Books I and II), and by Keith (Books VII-XV) 'in his edition of the Aitareya AraI).yaka; the fuH text in the Anandarama Series. _

14. TA. The Taittirya AraJ)yaka, the Bibliotheca Indica ediholl. This comprlses the TaIttiriya Upamsad and the M.ahanarayana Upam~ad.

J5 Kau~. Up. The Kau~taki Brahmal).a Upani~ad; the Bibli!3-theca Indica edition. ~

16. Chand. Up. The Cha~dogya Upani~ad; BoehtIingk's edition. 17. JUB. Tbe Jalminya or Talavakara Upani~ad Brahmal).a;

my edition in JAOS. 16. The- BAU {Brhad Aral).yaka Upani~ad}, the At. Up. (Atareya Upani~ad), the Taitt. Up. (Taittiriya Upalll~ad) and the Mahan. Up. (MahanarayaJ)a Upaniead) form portions of the texts already cited.

1 Das Jalminiya.Brahmana n Auswahl. Verhandelingen der KoninkhJke Aka demie van Wetenschappen te Amsterdam. Afdeebng LeUerkunde. Deel J, Nieuwe Reeks, Deel XIX, No 4 Amsterdam 1919.

X

Prefuce

2. The Manner of Presentation. The manner of presentabon is mtended to be essentially descriptive. It attempts to show the relative frequency of a given usage, and to exhibit peculiaritles which the syntactical usage of single texts or groups of texts may chance to exhlbit.

Historical prospects or retrospects have been purposely avoided. A really historicaI exposition of Sanskrit Syntax must await the working up in detall of the syntactical material of the texts which chronologically precede or follow the seventeen text enumerated aboye: But within the compass of these texts (unless otherwise stated) th~ collection of examples is intended to be as complete as possible.

A comparativa treatment, whether with reference to a cIosely allied language like the Iranian, or to the more remotely allied other Indo-European languages, lay entirely outside the plan of the present work.

The grouping of the material is, as far as practical, formal rather than logical. ..

A discusslOn of the formal treatment of eyntax may be conveniently fonnd in E. P. MOrrIs' and the author's paper on Agglutillution and Adap tabon (Harvard Studles m Classical Phllology XVI, 1905).

The categories and the criteria for grouping the material are therefore strictly grammatical, not logical 1 have tried to do justice to both scholastic rules of classification: 'Entia praeter necessitatem non esse multiplicands', and 'Entmm varietates non temere esse minuendas'; but, m case of doubt, 1 have preferred to err in favour of the second maxim.

In conformity with the ideas concerning the origin and deve-lopment of Indo-European inflection set forth in the paper on Agglutination and Adaptation referred to aboye \ no attempt has been made to determine basal meanings ('Grundbegriffe') for any case. Instead, the various types of usage ('Gebrauchsspharen') have been treated as fully and accurately as possible with special reference to their frequency and the peculiarities exhibited by single texts or text-groups.

1 Compare the discnSSlOn of the 'Origm of GrarnrnatiCal Elements' in chapter XIX of O. Jespersen's Langnage, its Nature, Development, and Origm, London and New York, 1922, p 367 ff

XI

Preface.

It also seemed worth while to note the beginnings of certain constructions, no matter' wbether they led by further development to established usage or whether their development was arrested without reacblng fruition.

The use of case-forms in compounds belongs to the chapter on word-formation. Partly for, this reason and partly because this subject has been very fully treated in tbe second volume of Wackernagel's Altindische Grammatik such cases have not been included in the present work.

3. Secondary Use o the Material here Collected1 As tbe Illustrative material for the case-syntax of the Brahma:o.a prose is here given in full, the present work may serve, to a certain and limited extent, not only as a Dictionary but also as a Con-cordance to the prose-sections of the Brahma:o.as, inasmuch as the verbal correspondences between the different BrahmaI}l\ texts will be found grouped together. Such juxtapositions may be a help to textual emendations; and the importance of such parallels for an eventual determination of the interdependence of the vanous BrahmaI}a texts has been briefly pointed out by the writer In the JAOS. 23, p. 325 with note 4.

Full Indices are intended to faCllitate such investigation.e.

1 Cf. tbe remarks 011 tbe secondary use of Bloomfield's Vedlc Concordance, p. VIII.

XII

Table of Contents. Gen e r a l In t r o d u e ti o n.

A. ConJunct and DISJunct Cases 1 B. Proleptic, Pendent. and Absolute Cases. 2

1. Prolepbc Cases. 3 2. Pendent Cases. 4 . 3. Absolute Ca;es. 5 .

page 1 1 2 5 6

e The Make-up of the DIsunct Case. 6 7 D.~The Signs of Severance of the DlsJunct Case from the Rest of the

Sentence. 7 7 The IndIvidual Cases.

1. The Disjunct Nommative I The Proleptic SubJect Nommahve

A. The Proleptic Subject NominatIve is Resumed by a Demon-strati ve Pronoun

10 10

10 1. The SImple Proleptic SubJect Nommahve wIthout Accom-

panying Infimte Verbforrn. 8-9 ..... 2. The ProleptlC SubJect Nominative IS accompanied by a Partl-

clple. 10-11 3. The ProleptJc SubJect NommatIve IS accompamed by a

Gerund. 12 4. The ProleptIc Subject NornmatIve 18 accornpamed by both a

Participle and a Gerund. 13 . . . . . . .'. . . B. The Proleptlc SubJect Nommative 15 not Resumed by a Demon-

strahve Pronoun 1 The Proleptic SubJect Nornmahve IS Accompamed by a

Participle. 14 2. The Prolepbc SubJect Nornmahve. 18 Accompamed by a

Gerund. 15 n. The Pendent NommatIve

A. The Pendent Nommabve is Resumed by a DemonstratIve Pro-noun m its Proper Obhque Case The Pendent Nominative Anbclpates 1. ,A Gemtlve. 16 . . 2. An Instrumental. 1 T 3 A Locative 18 . 4.-5A. An AccusatIve. 19-20A

"

10

18

27

, 28

28

28 29

29

29 30 36 36

XIII

6 An Ablahve 7 A Datlve

TabIe of Contents.

} 21 . B Tbe Pendent Nommatlve IS not Resumed by a DemonstratJve

page 37

Pronoun 37 1 The Pendent NommatJve IS resumed by tbe RepetitlOn of tbe

Noun ID its PlOper Oblique Case. 22 37 2. Tbe Pendenl NominabV"e lS Resumed by a Past PartiClple

ID ItS Proper Obhque Case. 23 38 3 The Pendent Nominative wltbout ResumptlOn. 24 38

1lI. Tbe Absolute Nomlllabve. 25-26 . . . . 39 AppendlX l. Apparent Instances of DIsJunct Nommahves. 27 45 Appendlx n. Anacolouthic Constructons of RelatJve Clauses. 28-29 . . 55

2. The DisJunct Accusatlve . 74 1 Tbe Proleptic AccusatJve Resumed by a DemonstratJve Pronoun 74

A. The Simple Prolephc Accusative witbout Accompanying Infimte Verbform. 30-31.. . . . . . . . . . 74

B. The ProleptIc Accusative IS Accompanied by a Participle 78 1. A Present Particlple 32 . 78 2. An Active Perfect ParbClple. 33 80 3. A Future Parbclple. 34 . 80 4. A Past Particlple. 35. . 81

C. An AccusatlVe Dependmg on a Gerund IS Resumed by a De-monstratrve Pronoun. 35 A 81

n. The Absolute AccusatlVe 8 \ 1. Tbe SeDldlsJunct Use of the AccusatlVe 36 81 2. The Absolute Use of tbe AccusatJve. 37 -38 83

3. The Disjunct Instrumebtal. 39- 40 89 4. The DisJund Dative. 41-42 93 5 The Disjunct Ablatlve. 43-44. . 94 6. Th~ DisJunct Gemtive 100

l. The ProleptJc Gemtlve Resumed by a Demonstrabve Pronoun 100 A. The SImple Prolepbc GenitIve without Accompanying Infimle

Verb.form. 25 . . . . . . 100 B. The Proleptic Gemtive IS Accompanied by a Parbciple. 46 100

n. The Absolute GemtJve 101 1. The Germ" of the Absolute Gemtive. 47 . . 101

A. The Genit,jve IS a Possesslve Gemtive 101 a) The Mmn Verb oi the Clause is not a Verb of Se

parabon. 48 . . . . 102 b) The Main Verb of the Clause is a Verb of Separa.

tion. 49 110 B. The Gemtive is a Partltive Gembve 50 122

XIV

TabIe oL Contents.

page e The GemtIve depends on the Malll Verb of the Clause. 51 130 D The GemtJve depends on an AdJective. 52 139 E The Gemtlve depends on an Adverb. 53 140

2. The Caseform lS AmblguouS. 54 142 3. The True Absolule GemtIve. 55 142

7. The DIsjunct Locative 14& 146 147 147 154

A. The Proleptic LocatIve. 56 B The Absolute LocatIve

1. Germs of the Absolute LocatIve 57 2. The True Absolute LocatIve .

A. The Accompanymg Parbciple IS a Past,Parhclple 1. of alllntranslbve Verb 59 2. of a TransltJve Verb 60 . . . . . . .

B. The Accompanymg Pro ticiple is a Present PartJciple 1. of an Intranslve Verbo 61 . . . . . 2. of a TransItive Verbo 62

154 154 199 20S. 268 279

C The Accompanymg PartIciple IS a Future ParllcipIe. 63 292 The Meanmg oC the Absolnte LocatIve. 64 . . The Separabon of the Absolute Locative from the Rest

of the Sentence. 65 ..... The SubJeet of the Absolute Locave 66 The Predlcate of the Absolute Locative. 67

Appendix 1. Passages which are not to be construed as Loca-tIves Absolute. 68 .

Appendix n. ParatactIcal ConstructlOns SemantIcally Equi valent to an Absolute Locave. 69

Appendlx III. The Use of the Absolute LocatIve as a Means of Connecting Sentences 70 .

8 The DIsJunct Gerund . . . . The Absolute Use of the Gerund. 71

1. The Mam clause conslsts of a Verbal Adjectlve in -tavya or

29i>

298

299

302

304 307 307

-ya Used ImpersonaIly 72 .. ....... 30S. n. The Main Clause consists of a Verbal Ad]ectIve m -tavya

or ya Used Personally. 73 309 IlI. The Mam Clause consists of a Nornmal SubJect with a

FmIte Velb. 74 . . . . . . . 311 IV Tbe Main Clause conslsts of a Nommal Subject with or

wlthout the Substantive Verb 'to be' 75. . . . Appendix 1. Emendation of MS.3 7.5 (p. 81,13). 76

Corrections and Addltlon~ Indices.

315 315 317 332

xv

Introduction. l. A. Conjunct and Disjunct Cases. In t\ llormat Indo-

European sen ten ce a given case-form is directly related to sorne other element of the sentence; ID fact, the expression of such retatlou of oue member of a clause to auother lS the mam func-tlOn,-and from the point of Vlew of strict syntax the only func-tion-, of the IndoEuropean declensional endings.

In mtlectmg languages hke the IndoEuropean a clear distlochon between the strlCny syntactlcal function and otljer funchons of the mflectiOnal eCHlmgs IS not formally made; the case endings express he re also gender and number whlch have nothmg to do wlth the syntactlcal relatlon of words lo oneanother Cf J. Rles, Was 1st Syntax? (914; Sthr, AIgebra der Grarnrnahk, 1889, MOlrlS and Oertel, AgglutmatlOn and Adaptation (Harvard Studles m ClasslCal Phllology, 16, 1905, P 63-122) where the dill'erence of agglutmatlve and lntlectmg languages In th1S respect is emphaslzed.

They are the outward and visible signs of thls syntactical dove-taIllllg, and this normal use may be called the conj~nct use oC case-forms.

But there are instances in which this close knitting of a clause is loosened and a certam dlsintegration, the degree of WhlCh may vary, sets ID. When such a loosening of the syntactIcal Jomts, 1D whatever degree, has taken place we may speak of the disjunct use of case-forms.

The very general collective term 'Anacolouthon' comprIses al! mstances in whlCh the regular sequence of a clause, the 'conSec,utlO vetborum', 15 m any way dlsturbed, 'genenbus numens temponbu5 persoms pertubatur' (Clcero, Parttt. oral. 18).

2. B. The Disjunct U se of Cases more closely considered. Proleptic, Pendent, and Absolute Cases. The dlsJunct use of cases may be subdlvided accordmg to the degree of syntacticallsolation

Oertel, The Syntax of Cases. 1 1

3, a-e Introduction

which a given case-form exhibits, or, in other words. accordmg to the degree of syntactical disintegration of the slfutence Ulllt

3. 1. Proleptic Cases. a) Prolepsis with ResumptlOn by a. Demonstrative Pronoun. A noun may retam the case-form WhlCh is required by the syntaclical setting but be lifted out of its proper place, placed usually at- the head of tbe clause, and he tben resumed in its proper place by a demonstrative pronoun. Tbus, TS. 2. 4. 1. 1 deva manusyah pitara.'! te 'nyata asann asura

rak~asi pz~acas te 'nyatah, 'Gods, men, Fathers,-they were on one side; Asuras, Rak~ases, Piyacas,-they were on tbe other side'; K. 31. 5 (p. 6, 12) pranarh vyanam apanam tan eva yajarnrtne dadltatt, 'In - breathing, through -breatbing, out- breathing,-these he places into the sacrificer' j TB. 2. 6 9. 3 yaJaya yajamanaya 'tmane febhya eva , {{l$am ar;aste, 'For the sacrifice, for the sacrlficer, for hlmself,-for them be asks a blessing'; K. 34. 4 (p. 38, ] 2) antarik$ild dlVah prthivya ahoratrabhyarh tebhya enan sarvebhyo nzrbhajati, 'From the atmospbere, from the sky, from the earth, from day and night,-from all these he excludes them (VIZ. his rivals)' .

This type is represented in .glish by: 'My brother, he is in Elysium' (Twelftb Nigbt, Act r, Scene 2).

1>1 Prolepsis without Resumption by a Demonstrative Pronoun. Instead of the resumption o the proleptic case by a demonstratlve pronoun the incipient disintegration of the sentence may be out-wardly marked by the intrusion of a particle separating tbe pro-leptic case from the main verbo Thus, QB. 1. 8. 1. 25 sa vm catur upahvayamano 'tha nane 'vo 'pahvayate, 'Verily, h~ calling four times upon (this Iga), yet he calls upon her in different ways'.

This kind of prolepsis is confined to the nominative (cf. below 14 and 15). .It contains the germ o an absolute case (cf. below 25-26). The noun is always accompanied by an infinite verb-form, either a participle or a gerund (cf. below 3, 1, Cj 4-6)

For atha and other separating particles see below 7, b e) Absence of any Outwa.rd Signa of Diaintegration. There are

numerous instances in the BrahmaI,la prose where such visible

2

Proleptlc Cases. 3, c".

signs as the resumption by a demonstrative pron0un (above 3. 1, a) or the separation of the no un from the main verb by atha (above 3, 1, b) are wanting. When the noun is accompamed by a participle and the ingredients of a clause are thus present (cf. below 4-6) the questlOn arises whether, even m the absence of any outward sign of disintegration, the noun with it8 accom-panying partiCiple may be assumed to have gained a certain degree of independence such as lS implied by tbe use of a hypotactical clause in an Ellglish rendering. For exa.mple, IS QB. 1. 6, 3 35 prajapater ha vat prajalJ, sasrjanasya parvant vtsasrasuf equivalent to 'Of Prajapati having created creatures the joints relaxed', or rather to 'When Prajapati had created creatures his joints relaxed'? The latter view is ad vocated by Gaedicke (Accusati v, p 4,7. note): 'Wenn es QB. 14. 3. 1. 2 h8113t yaJasya r,;ir~achinnasya r,;ug udakramat oder 13 1. 1. 3 ar,;vasya va alabdhasya 1'eta udakramat, so laJ~t zwar 13, 1. 1. 4 prajapatzr yaJam asrjata tasyu mahzma 'pakramat "Prajapati erschuf das Opfer, dessen Gr13e (Potenz) entwich (=yon dem wich dIe G.l" auch die Gepitive der eraten beiden Stze als vom Subjekt abhangig auffassen, dennoch aber ist zu beraetzen: "Ala dem Opfer der Kopf abgeschlagen war, ging seine Quc her-aus", "Al s da.s Ron geschlachtet war, gmg sem Same davon" ~ desgleichen 6. 1. 3. 9 tasmat putrasya jatasya flama kuryat, "Daher mge er, wenn ihm ein Sohn geboren ist, diesem emen Namen schaffen (eig. seinen N amen machen)". In diesen drei Beispielen druckt, der Genitiv mit dem nacbgestellten pradikativen Partlzip einen besonderen, temporalen oder konditionalell Satz aus. Aber die Zugehrigkeit des Genitivs zu dem Nomen ist noch eine der-artige, da13 wir ihn yor letzterem durch ein Possessivpronomen zu wiederholen haben, wenn Wlr ihn zum Subjekt eines besonderen Satzes machen'. Simllarly, but more guardedly as regards the Bindu conception of tbe syntactical relation, Delbrck (Vergl. Syntax, n, p. 494, 164): 'So wlrd Z. B. in der altind. Prosa erzllhIt, da13 Manu emen Stier opferte und dann fortgefahren: tasyu 'labdhasya sa vag apa cakrama (QB. 1. 1. 4. 15). Das heifit wortlich. "Dieses geopferten Stimme entwich", aber wir bersetzen: "AIs er geopfert war, entwich seine Stimme". DaIl wir mit dieser bersetzung das richtige (vom Spracbgebrauch des Inders aus

1* 3

3, c. InhoductoD.

gesprochen) getroffen haben, lallt sich freilich nur wahrscheinlich machen, nicht beweisen' In all these and kindred instances, it seems to me, considerations of loglC or of translation do not iouch the beart of the question which lS essentially and wbolly a question of Sanskrit syntax. There lS, of course, no doubt that in English or German the rendering by a hypotactical clause is possible, nay even preferable or necessary, because demanded by the English or German idiom. But tbe translation of such Sans-krit clauses into modern English or German, far from being a belp to tbeir correct interpretation, lS a so urce of danger, as trans-latlOns into a foreign idlOm frequently are j for such translatlOns lead us imperceptibly to view the ,syntactical relatlOns of one lan-guage through the dIsturbing medium of another. In the absence . of any visIble signa ther are really no cnteria WlllCh would enable us to say with regard to any mdIvidual passage of tbis char.wter whether the speaker or the hearer or both felt the noun wlth its accompanymg partrclple as (more 01' less) independent from the rest of the clause. All we can say is, that since the true absolute constructions (below 5) have th81r root and origm m such con-structions, some of them must at some time have been felt as more or les s independent syntactIcal uuits, and it is very probable 1hat frequently the speaker felt the case as conjunct whIle the hearer construed it as a disjunct case, or vice versa.

It lS instructIve to compare examples of such quasi-disjunct use of cases with parallel clauses in which the conJunct con struction of the case lS ,clear TIms, $8. 1. 7. 2 sadhyant:h vai dcvanam sattram astnanath r,arlcara alc$asu jajre 1 would naturally be rendered III Enghsh by 'When the Sadhya gods were engaged in a sacrificial session dust got into their eyes', but the un-doubtedly conjunct tesam III the immedlately following katham nu te~arh (= asma.~, Commentator) r,arkara ak$asu jayeran, 'How may dust get into our eyes?' shows that jt is impossible for us objectively to determine if, or to what degree, the con nection of sadhyanam .. aslnanam with ak$asu was loosened for the HIlldu speaker or hearer. The same is true for QS. 6. 1. 3. 9 cited by Gaedlcke (cf. also QB. 6. 1. 3. 20 tasya [agneh]

1 Cf below 48, Ex. 24-26. 4

Pendent Cases 4.

citasya nma karolt KB 7. 2 na va ajatasya garbhasya nama kurvanti) when compared with the clear conjunct genitives in QB. 6. 1. 3. 10-17 yad asya tan nama 'karot; 14. 9. 4. 25 (= BAU. 6. 4 25 Madhy. = 26 KaI).v) ath 'sya namadheyarh (KaIfv. mima) karoh KB. 6. 2-9 nama me kuru. In the passage TS. 5.2.3.4; 5.3.2.2; PB ~. 8. 13 devanm vm (TS. 5. 2. 3.4 te~am) suvargam (PB. svargarh) lokam yatam diQalt samavliyanta (PB. dit;o 'vliyanta) the Commentary to PE. expressly interprets dev nam ... yatm as an absolute genitiva and the English ldiom requires the translation:

5. IntroductlOn.

pztaro vir,;ah 'King Yama Vaivasvata, - hlS people are the Fathers'; 4. 4. 5. 1 athaz, 'tac char'iram tasmzn na raso 'sti, 'Now this" body, -there is no sap in it'.

When the noun lS accompamed by a partlcIple such conbtJ uchons verge very closely on the absolute nommatJve (below 9 25-26).

The psyehologlCal process whieh gives rIse to Pendent Nomina-tIves appears to be as follows: The speaker begins to utter hls statement before the syntactical mould mto WhlCh hlS thought lS to be cast is fully matured, I e before the syntactical integratlOll of its parts has fully taken place As he starts speakmg, his attentlOn is fixed upon a single item of the thought-complex ll'respeetive of the relation whieh lt will ultimately bear to the other elements of the clause The remainder of the thought-complex takes its syntactical shape after the pendent nominatIve has been uttered and thus the proper syntactical relation of thlS nominative to the rest of the sentence must be afterwards indicated by a resumptive pronoun. The nominatIve lS chosen as pendent case because it is the only case whleh not only expresses (as sub-ject case) syntactieal relation, but also implies the absence of aU syntactlCal relation (as in titles, headmgs, lists) 1 Conversely the attentlOn of the hearer is more strongly directed to sueh an ah-normal nominatl ve, so that the emphatlC attention of the speaker is commulllcated to the hea:rer.

5. -3. Absolute Cases. In the precedmg two groups the dis-Junct case used eIther proleptically (nbove 3) or pendently (above 4) forms, syntactically consldered, an integral part wlth the rest of the clause and a medIate connection at least with another sentence-element is maintamed. This third group, on the other hand, lS charaeterized by the faet that the disjunct case no longer forms an integral part of the sentence, but, through the absence of even a medIate syntactical connectlOll with any other element' of the clause, has gained complete syntactical independence. Such

1 SmIlaI lS. use of the 'numberless' smgular III the so-called 0x'llla TIlV-' i'laplKv; 'der Erschemung hegt zugrunde, dal3 1m Moment der Setzung des Vel'bums der Sprechende sIch noch llIcht uber dIe Form klar st, dIe er dem SubJekt geben wlll, als llldIflerente Verbalform setzt er den Smgular' (Wacker-nagel, Vorlesungen ber Syntax, 1, 1920, p.21). 6

Absolute Cases. 6

independenee cannot be gained by a sImple noun but lt IS ne-cessary tbat tbe noun sbould be accompanied by an infimte verb-form (either a partimple 01' a gerund), in other words tbe dlS.1Ullct case, lf lt lS to be regarded as a true absolute case, must contalll the materIal for a sImple sentence, VlZ. subject and predi-cate (cf. aboye 3, 1, c). Thus, PB. 5. [) 15 devaQ ca va asuraQ ca 'dttye vyayacchantas tam deva abhyaJayan, 'The gods and the Asuras contendlllg about AdItya (= when they contended), the gods WOll him'; AB. 1. 15. 1 hal/lr attthyam mrupyate sorne raJany agate, 'The guest-offering lS spread klllg Soma havlllg arnved (= when he has arnved)'. In English thls type lS represented by. "And tell qualOt hes, How honourable ladles sought my love, Whch 1 denyng tbey feH slCk and dled' (Merchant of Venice, Act lII, Scene 4)

But whIle, on the one hand, the presence of an infinite verb-forro lS esselltIat for every true absolute case, the mere presenee of a partlClple tends, on the otber hand, to glve a eertam deglee of mdependence to proleptlC ( 3, 1, e), pendent ( 4) and even eonJullct cases ( 3, 1, c). Karl Kunst (Glotta, 12, p. 31) very Justly speaks of a 'seductIVe force' ('lockende Kraft') of tbe partl-eple and Delbruck (Vergl. Syntax n, 164, P 494) sketches the transltlOll from conjunct to absolute use wlth the words 'DIe EntWlCklllng verlauft be! diesen KonstruktlOnen [VIZ. the absolute gellltIve in Sanskrit aud Greek] so: Zu emem obliquen Kasus emes Satzes tritt em Partlzlp. Dasselbe ist fuI' den Smn VUll so tiberragender WlChtlgkeit, daU es den Kasus ins Sehlepptau mmmt, desseu Verhaltnis zu dem ihn regierenden SatzteIl slCh mfolge-dessen loekert. Bald geht das ursprungliche grammatische Ab-hangigke!tsverhaltms fr das Gefhl des Sprechenden. verloren, das Partlzlp nebst Kasus kann jedem beheblgen Satze angefugt werden, und so ist eme Ausdrucksform fur eine Nebensatzhand-lung gewonnen, welehe nicht zu dem Satzsubjekt in Bezwbung steht' C( 57, Ex. 5, Rem.

6. C. The Make-up of the Disjunct Case. The DlsJunct Case may conslst (a) either of a sImple noun 01' a serIes o' such sImple nouns, 01' (b) tbe noun may be aecompanied by an lU-:D.lllte verb-form, VIZ a participle 01' a gerund. The ImpOl'tauce

7

S 7, a, b. Introduction

of this difference has been discussed above 3, 1, c; 4-5: A simple noun can never reaeh the degree of independence necessnry for an absolute case; the presence of an infinite verb-forro, on tbe otbe1' band, gives even to a conjunct case a certam a11' of in-dependence

7. D. The Signs of Severance of the Disjunct Case froro the Rest of the Sentence. a) The fact that tbe eonnection of a dlsjunct case wlth the 1'est of the clause is mediate only is lU-dicated by the resumption of the disjunct case by lDeans of a demonstrative pronoun tbrough which the immediate conneetion wJtb another element of the sentence is established, cf. 8-1 D; 16-21; 30-35; 41, 43; 45.

Ir the case-form of the disJunct case is that required by lts syntacbcal rclallOn we speak of a ProleptJc Case ( 3); lf the dlSjUn case lS a nommatJ ve we sr Cdk of a Pendent Case ( 4 and below 7, e)

The place of the demonstrative pronoun may be taken by a noun (as at QB 2. . 4. 11 devanam, ef. below 22), or by ad-jectival partieiples (as at AB 8. 10. 9 upavistaya, anvarabdhaya, ef. below 23).

For a qUl1s1-prolepsls and resurnjJtion by (e)t/1smin kale cf. 56. b) The alienation of the disjunct case from the rest of the

sentence may be indicated by particles whieh intervene between tbe disjunct case and the sentence-element to WhlCb it beIongs. The most frequent particle is atha (ef. Delbrck, Altind. Syntax, p.539, 7, 263, II; Speijer, Sanskrit Syntax, 450, p. 348; Ved. und Sansk. Syntax, 286, p. 91, 20), see 14, Ex. 1 (QB. 1. 8. 1. 24-25); Ex. 2 (TS. 7. 1. 3. 1); 15, Ex 1 (QE. 1. 9. 2. 25); 37, Ex. 2 (TS. 7. 5. 6. -3); Ex. 6 (KB. 30. 5); 59, Ex. 69 (QB. 1. 6 4. 4); Ex. 212 (QB. 11. 2. 4. 10); 60, Ex. 4 (TB. 1. 2. 5. 1); Ex 23 QB. 5. 1 3. 13 = 5. 4. 33); Ex. 62 (QB. 7. 3. 1. 2; 3); Ex. 107 (QB. 1 8. 2. 7); Ex. 108 (QB. 1. 9. 2. 26 = 4. 4. 4. 3); Ex. 212 (QB 12.9. r.12-17); 61, Ex 6 (TE. 1. 6. 8. 9); Ex. 32 (GB.l. 3. 19); 62, Ex. 22 (QE. 1. 2. . 20 bis); and cf. the similar sepa-ration o the absolute gerunds from the rest of the sentence by atha 72, Ex. 5 (MS. 3. 9. 2, p. 114, 6); Ex. 6 (MS. 3. 10. 4, p. 134, 11); 73, Ex. 16 (MS. 1. 6. 10, p. 102,8); 76, Ex 3 (QE. 1. 9 3. 10). 8

Slgns ot Severance. 7, e, d.

Other parUeles of separatlOD are atra, tad, tad allU (see below 65, b-d) and tata!t, 26, Ex 6 (QB. 2. 3. 1. 10).

e) The severanee of the disjunet case lS indieated by the fact tJat, though being semanticlllly an mtegral part of the sentence, lt cannot be syntactically construed wlth any other element of the sentence; the necessary connectlon lS brought about through tbe mediation of a dernonstralve pronoun in tbe syntactlCal1y required case (see above 7, a). The result is the Pendent Nornma-tIve (above 4).

d) The severance of the disjunct case from the rest of the sen tenee has reached Its highest degree in tbe Absolute Case. Here tbe dlsjunct case is no longer, eIther mediately 01' immediately, construable with any element of the rest of the sentence nor does 1t form semanticaJ1y an integral pal't of it, out it playa tbe part of an adjunct and is equivalent to a paratactical (see below 69) or hypotactical clause.

9

s 8-9, Ex 1-2. ProleptIc NommatIve.

1. The Disjunct N ominative. I. The Proleptic Suhject Nominative.

A Tha Proleptic Subject Nominativa Resumad by a Demonstrative Pronoun.

8. The simple Proleptic Subject Nominative (without accom-panying infinite verb-form).

a) The Prolptic Subject NommatIve is usually placed at the head of the clause (e. g. Ex. 1-17, etc.); rarely 1mbedded in the sentence (e. g. Ex. 20) b) It consIsts most frequeutIy of a serIes of sllnpIe nouns (e g. Ex. 1, 5,6,9, 11-14, etc) which is occa-slOnally marked off by ttt (Ex. 3, 8, 22 B), more rarely of a smgle no un (e. g. Ex. 2,7, 21, 22G) WhlCh may be accompamed by an adJectival (Ex. 4) or by a nommal (Ex. 10) quahfier. c) As a rule the resumptive demonstratlVe pronoun follows immedlately upon the ProleptlC Nominative (e. g Ex. 1, 3-6, 8-16, etc.); rarely the former is separated form the latter by other words (Ex 2, 7, 18, 21). d) The Resumptive Pronoun agrees lU gender and number wIth the predicate no un (e g. Ex. 5, 6, 17, 19, 22 B, 22C, 22D, 221) Compare for this concord SpeiJer, Sanskrlt Syntax, 27, 1, p. 18.

9. Examples: 1. devah pttaro manu6yas (TS. deva manuf)ya} pztaras) te 'nyota

asann, asura rak~asz pzr;acas te 'nyatah, 'The gods, Fathers, men, -tbey were on one side, the Asuras, Rak$ases, PIQ3.cas,-they were on the other side'. K. 10.7 (p. 132,10); TS 2.4.1. 1; JB. 1. 154 (Caland, Auswahl, p. 54).

2. devar; ca va asurar; ca samyatta asans, tan gayatry antara 'ttsthad ojo viryam annadyam pangrhya, sarnvatsaro vavat 'tan 60 10

Proleptic Nommattve. 9, Ex. 3-4

~ntara 'tisthat, 'The gods and the Asuras were engaged in combat; between them stood the GayatrI havlllg selzed strength, vlgour, {and) food eatJllg, the year indeed, -that stood between them.' K. 10. 7 (p. 133,8).

The parallel pa,,,ages wllhout prolepsl5' M8 2 1 11 (p 1.3, 8 and 14) deva~ ca va asurli~ ca 'spardhanta, tan gayatrI sarvam annam partgrhyli 'ntara 'tlsthat, ... samvatSal o vat gaya tri, samvatsaro wi tad atz8that, T8.2 4 3 1-2 dnusurah .samyattli asan, tesam giiyatry OJO balam tnd1'iyam V'iryam prajam pa~nt sam-grhIJa 'daya 'pakramya 'ttSthat, , samvatsaro vai gayatrI, samvatsaro Val tad apakramya 'tlsthat.

3. prano 'lJa1W9 caksuf, 9rotram lty etaln vat purusam akaran, 'In -breathmg, out- breathing, ;nght, (and) hearing,-these have made (= make) man'. GB. 2. 2 14.

For thls 'comprehenslve' u~e of %ti to mark the end of a senes er. PW. s tti, col. 790, 3-7, Boehtlmgk, Chrestomathle, 2d ed., p 356 = 3d ed. 396-.397 (note on ~B 1 8. 1 7), pw. s tti, p 20~, col 3, 43-48, Delbrck, Altind 8yntax, p 533, 7-17, and cf further the contIoversy artsmg out of Bhle's artlcle WZKM. 1 (1887) p.13-20 Boehtlmgk, ZDMG.41 (1887), P 516-519, Benchte der ph)-hlSt el d kgl. sachs. Ges. d. Wiss 1895, p. 152, Knauer, Festgru& an Boeht\wgk

(1~88),p 62-fi7, KeIlh, JRA8. 1910 (1I), p 1317-1321, t;A(OnentalTran,latlOn Fund, N 8 18) P 8, n G, p 40, n 6, p 51, n 6

Instances In the Brhmana prose are TB 3 8 20 5, t;B 1. 8, 1 7 (contrast 8, wllhout ztzl; 13.:l 2 2, 13 2 8.1 = 13 5.2 1 (contrast 13.5 1 13 ",here ~tt malks the pratika of V8. 24 1); 14 6 8 9 (= BAU. 3.8 9, where the Kanva

rencen~lOn has tt whlle the Madhyandinas omlt it); KB 6 15 (tel, p 27, 23 and 24 aud p 28, 6 where etli~ ftsm stayah stands In contrast to anya tstil/) , TA 1, 25. 1, Alt Ar.2 .3 1 (p 111,14- Kelthl. 2 3 3 (p. 113,8,15,16; 17); 2 6 (p 123. 6, 7) = AIt. Up 3 1. 2, Alt Ar 5 3 1 (p 146,10), JVB 3 16 2. Chiind Vp 4 11.1,412 1,413.1,5 1O!J, 7 ~4 2, t;A.l 4,2 1 =Alt Ar 3 2 2 (Kelth, p 134,1-3) 8 9,8.11 '

4. fas [apa!t] tatrat 'va 'bhya9ramyad abhyatapat samatapat, [ah frntas taptal} samtapta~ sardham eva retasa dva~dham abhavan, tdsam anyatara (Gaastra's emeud for anya anyatara) al1lavana apeya asvadvyas ta a9anta retalt samudram vrtva 't~~thanta, 'These (waters) he then tOlled over, he heated, he thoroughly heated, they havmg been tOlled over, heated, .thoroughly heated together wIth the seed became twofold: one (half) of thern bemg excessively salty, not dnnkable, not palatable,-they, not bemg appeased,' contmued to envelop the seed, the ocean'. GB. 1 1. 3

11

9, Ex 5-9 ProleptlC NOIDmahve.

The contmuabon of the sentence athe 'tarah peyii~ . samatapat is llsted below 19, Ex 1.

For .y 8th wlth gerund as amnliary verb ef Delbruek, Altmd. Syntax, p 407;, SpeJJer, Sansknt Synlax, 381, p. 298 Such gerundJal eonslructlOns should be added to WhItney, Gr,. 1075, e (ef 994, e).

5. dad7n madhu ghrtam dh(tnas ta?Zdulas tat samsr$tam syat, 'Curds, honey, ghee, grains, (and) rice-grains,-those should be the mix-ture'. K 11. 2 (p. 145, 7).

The neuter noun 8amsTbta 'mixture', whlCh should be added In pw., occurs also in the slllgular at K. 11. 2 (p. 145, 13): 36. 1 (p 68, 7) = MS 1. 10. 6 (p. 146,6" where the text should be emendend to sarhsTstarn malldam) , In the plural al K 11. 2 (p. 145, 4).

6. agnihotram darfaprnamasan cat1.trmasyany esa vava tarhi yaJna Ctsit, 'The Agnihotra, the New Moon and the Full Moon ceremonies, (and) tbe sea son al offerings,-they thell constituted the sacrifice'. K. 23 7 (p. 82, 19).

7 somahutlr ha va ebhyas [devebhyaf,] so 'ccakrama, 'The Soma-sacrifice,-that departed from these (gods)'. JB. 2. 158.

8. arUmr vajasaneyo barkur var~naf, pnyo janar;ruteyo bultla' uflvataraf{mr vaiyaghrapadya ~ty ete ha paea mahabrahma asuZt, 'AruI).i Vajasaneya, Barku VareI].U, Priya Janalfruteya, Bulila Ac;vatarayvi, Vaiyaghrapadya,-these were five great Brahman-priests'. JB. 1. 22 (Caland, Auswahl, p. 10).

Wlth mahabrahma- t B. 1. 1. 4. 14 asurabrahma-; Wackerna~el, Altmd. Gr. n, 1, 50, b, a, p. 117, 10-11

9. jivala9 ca ha kar"iradtr indradyumna9 ca bhallaveyas tau ha 'runer llcilryasya sabhaga(v) (the mss sabhagi, varo lect. sahagii)

ajagmatu~ cJivala Kariradi and Indradyumna Bhallaveya,-these two carne going for a discussion with the teacher AruI).i'. JB. 1. 271 (Caland, Auswahl, p. 103).

The mss of JB. vary here and elsewhere between kriradi and kalraj'l, ef. Wackemagel, AltIlld. Gr. 1, p. 163, 140, a, note, where the varant ApMB 2. 8. 2 Itccatrvdt for RVKh. 10. 128 3; HG. 1. 10.6 uccairva)i may be added. And ef. tIJe play between the v' l'Uj and rudra at K. 25. 1 (p. 103, 2) yat samaru)at tad l'ttdrasya rudratvam (Nlrukta 10. 5 yad arttdat tad rudrasya rudratvam itt Katha-kam seems to refer to this passage 2nd should probably be emended aeeordmgly).

1 have adopled Caland's emendatIOn (WZKM., 27, p. 77) of the mss readmg sabhaga (val'. lect. sahiiga) to sabhaga(v). Tbe adJective sabhaga occurs besldes the two passages B 3. 3. 4 14 (where, however, aecordmg to Eggeling'~ note, J2

Proleptic NommatIve 9, Ex. 10-11.

the Kaova recension omlts the clause m whICh sabhiiga occurs) and Chand Up. 5 3. 6, both cIted m PW., also at JUB. 4. 6 4 (where, of COUlse, sabhiiga should be read as one word) = JB. 1. 22 (Caland, Auswahl, p. 13) atha hai 'salh sabhiiga iivavraJa Wtth sabhiiga the PW compares VS. 30. 6 = TB 3. 4. 1. 2 siwhacara the exact llleanmg of WlllCh 15 uncertam (WebeI, ZDMG 18, P 277 = Ind Strelfen, 1, p 77, Eggelmg, SBE.44, P 413) In as much a~ sabha and samlti are closely related m sense (Edgerton, KZ, 46, P 174,17) the adJectIve samtttmgama QB. 14.9 4.17 (=BAU 64.17 Ma:dhy. = 18 Ka:ov) may be com-pared wlth sabhiiga The gemhves QB. 3. 3 4 14 asya, JB. 1. 271 atlener iicaryasya and JB. 1. 22 = JUB. 4. 6. 4 tesam depend 00 the prior memher of the compound (WIlhoey, Gr. 1316; Wackernagel, Altmd Gr. 11, 1, P 31, 12, a, SpelJer Sanso knt Syntax, p 178, 231) cf. AV. 8. 10.5 yanty asya sabhiim ,6 yanty asya samltlm .. The prIor memhl'r sabha of the compound sabhaga must not be taken in the nalroW meanmg of 'Ratsvelsalllmlung' hul as 'cOlloquIUlll or con-c!lIUm, a common (sa) meeting for the eXlhange of speeche~ or conversahon' (EdgeIton, KZ, 46, p. 175, 17). Edgerton's etymologlC

9, Ex 12-16 Ploleptic NomlOative.

asatuf, 'Datva Sautemanasa and Mitravid Dam~trad'yumna,these two were puplls of king Pratidar

Proleptlc Nommahve. 9, Ex 17-22.

17 trini gayatru;atani te dve brhat'i(;ate, saptatim anustubhak sap-tattm patikti(; eatvari(;aeehatam brhatmam, trayanam trlstupehatanam gayatri(;atam uddhrtya tani tr'ini brhati(;atani, tae ca gayatr't(;atam jagati(;atam ca te dve brhatl(;ate, paea(;at trirtuMah paea(;ad usmhak (;atam brhatyak sampadyante, 'Three hundred GayatrIs,-they are (equivalent to) two hundred BrhatIs, seventy Anul';ltubhs (and) seventy PaktIs are (eqUlval~nt to) one hundred and forty Brha-tIs; subtracting one hundred GayatrIs from three hundred 'Tril';l-tubhs,-that lS (equivalent to) three hundred Brhatis, these one hundred Gayatrls and 011e hundred JagatIs,-they are (equivalent to) two hundred Brhatis; fifty Tri;;tubhs (and) fifty U;;l].ihs malee up one hundren Brbatls'. KE. 18. 3

18. praJapatzr ycw't prathamam ahutzm aJuhot sa hutva yatra nyamr~ta tato mkakatak samabhavat, 'Prajapati,--the first oblatlOll wbich he offered,-where he, having made the oblation, washed (his hands) thence the Vlkakata-tree sprang up'. QB. 6. 6. 3 lo

Tbe resumptlve sa IS due to the mtervening relative clause ylJm ... aJuhof: er. belolV 11, Ex. 1 A, Rem., Ex 8 B, Rem.

19. a(;va rsabho vrstl-zr bastak sa dak~na, vr~atvaya, 'A horse, a bull, a ram, a he-goat,-these are the sacrificial fee, for the sake of (attammg) VIrIlity' TS. 2. 3. 7. 4.

20. etena Lsaktena] ha va maro 'gastyo marutas te samajanata, 'By means of this (hymn) lndra, Agastya, (and) the Maruts,-they agreed'. AB. 5. 16. 14.

21. tasmad dha 'py etarh't bhayan wa naktam sa yavanrnatram lVa 'pakramya (Bib!. Ind. wlth the mss. prakramya) bibketz, 'There-fore even now, even a strong (man) at night,-he, going off even for a short distance, is afraid' GB. 2. 5. 1.

Tbe palallel pas~age AB 4.5 1 wlthout plOlepsls' tasmad dhii 'py etarh~ naktam yavanmlJtram wai 'va 'pakt'amya bibheti.

For iva = eva el. Aufrecht, AB., p 430, 26-28, Eggehng, SBE, 44, p. 152, note 3 on. QB. 12. 2 2 11 j Keith, JRAS. Fl08, p. 1193, note 1; 190

9, Ex 22 A-C. ProleptJc NommatlVe. ,

tha ' dlto bhamau nidlgdham fad amuya syad evam tat, tasmad u saha vai 'va vasatkare'!la juhuyad t:a$atkrte va 11 tad yatha yonau rc-tah siced eva'l1l tad, atha yat pura va~atkaraj juhuyad yatha 'yonau (Weber's text omlts the avagraha) 1'etah s1ktarn tad amuya syad evam taf, tasmild u saha 'l"a~ 'va t'asatk(trena jHhuyd vasatkrte va, 'Tbe Vll~at-call is tbe vessel of tbG- gOdR; it lS just as if one were to offer (food) ladhng it out of a vessel, now if he were to make the offermg before the Va~at-call, as (food) stickmg down on the ground,-(as) that would he useless, even so (ihe offering would te useless); therefore he should make the offermg either along with the VaQat-call or after the Val;lat-call has been uttered: jt lS Just as lf one were to pour seed jnto a womb, now Ir he were to make tbe offerlllg before the Vaeat-call, as seed poured mto .a non-womb,-(as) that would be useless, even so (the offenng would be useless); therefore he should make the offering elther along wIth the Val;lat-call or after the Va$at-call has been uttered'. QB. 1. 7. 2. 13-14.

r Additions froro the Aral,lyakas and Upaml;lads: 22 A. cakf;tth !(1'ot1'am mano vak pl'anas ta eta paca devata ma/u

V?5tah purusam, paco ha~ 'vat 'ta ayam 'V1~falJ pltru~a!t, 'Slght, hellrmg, mind, speech, breath,-these are the five dlVinities which have entered man, and thlS man has entered tbese five (divmities)' Alt. Ar 1. 3 8 (p. 92, 13 Keith).

22 B. prthwi vayur a7car;a apo Jyotiris~ 'ty e~a va atmo '7ctham pacamdham, "fhe Earth, the \Vind, the Ethereal Space, the Waters, the Lights,-these are tbe Self, tbe five-fold Uktha'. Alt Ar. 2. 3. 1 (p. 111, 13-14 Keith)

22 C. tasya [samnaM trayy eta vidya hzlika1'o, 'gnzr vayur asav udLtya esa prastava, ima eva loka adi~,. ", r;raddha yajio dakrtna e~a ttdgttho, dlr;o 'vantaradi!(a a7c!(a esa pratthara, pah praJa osa-dhaya e;a upadrava!(, candrama nak,atruni pttara etan nidhanam,

ProleptJc NommatJve. S 9, Ex 2~ D-H.

(aod) ihe plaots,-they are tbe Upadrava; tbe Mooo, tbe Coostel-latioos, (aod) the Fathers,-they are the Nidhana'. JUB. 1. 19. 2.

Note that the prolepsis IS confined to subJect nornmatIves whlch consist of a series of nouns, whlle single nouns (trayi vidya, ime lokah) show lhe normal constructlon Contrast tma eva loka, a,dth wlth Chand Up '1.. '1.1. 1 trayIJ ~me lokh sa prastiivah.

22 D. tray;;-vidya hikara~, traya zme lokah sa pray[avo, 'gnzr vayur adityah su udgtho, nakJatram vayas~ marIcayaZt sa prat~haralJ" sarpa gandharvah pitaras tan nidhanam, 'The threefold knowledge is tbe Hnikara; tbese three worlds,-tbey are tbo Prastava, Agni, Vayu, (aod) Aditya, -tbey are tbe Udgitba; the coostel1atioos, the birds, (o.nd) the rays,-tbey are the Pratihara; tbe serpents, tba Gandbarvas, (aod) tbe Fatbers,- tbey are tbe Nidhaoa.' Chand. Up. 2. 21. 1.

See tbe note to the preceding example, 22 C. 22 E so 'savadityas sa e$a eva ud, agnir eva gI, candram eV,l

tham, silmany eva ud, rca eva gi, yaja$!f eva tham ity adhidevatam I atha 'dhylitmarh : prtJ1}a eva ud, vag eva gi, mana eva tham, 'Yonder Sun,-that same is (tp.e syllable) ud, Agni is (the syllable) gi, the Mooo lS (the syIlable) tham; the Samaos are (the syllable) ud, the Rig-verses are (the syllable) gl, the Yajuses are (the syllable) tbam; so witb regard to tbe divinities; oow with regard to the Self: Breath is (the syllable) ud, speech is (the syllable) gI, rumd is (tbe syIlable) tham'. JUB. 1. 57. 7.

Note that only the first Item (so 'sl1v adyas sa esa eva ut) shows prolepsIs 22 F. sudak$zno ha vai k{laimih pracina~alzr jbalau te ha

sabrahmacaritta asu~, 'SudakeiI).u Keairui, Pracloa~li, land) tbe two Jabalas,-they were fellow-students'. JUB. 3. 7. 2.

22 G. praclna~ala aupamanyavah satyayajnal; paulu$zr mdra-dyttmno bhallaveyo janal; ~ilrkarak$yo burJ,ila ~vatara~zs te hai 'te mahar;ala mahar;rotriyal,t sametya mimlsarh cakrul,t, Pracioa~llla Aupamanyava, SatyayaJa Paulu~i, Indradyumna Bhallaveya, Jana Qarkarkeya, (and) Buido. A~vatara~vi-these same great bouse-bolders (aud) great scholars, bavmg come togethel', deliberated'. Chand. Up, 5. 11. 1.

22 H. sac ca 'sac ca 'sac ca sac ca vk ca manar; ca manar; ca (the mss. omlt tbe second manar; ca) vak ca cahpu9 ca rrotram ca

Oertel, The Syntax of Cases. 2 17

9, Ex. 221, J, 10 Proleptic Nommabve.

rrotram ca caksur ca fraddha ca tapar ca tapar ca rraddhd ca tanz o(lara 'Existence and non-existence, non-existence and eXIsten ce, speech and mind, mind and speech, sight and hearing, hearmg and sight, faith and penance, penance and faith,-these are SlX-teen'. JUB. 4. 2. lo

221. rg gatha kumbya tan :nztam yaJU?' mgado vrthavak iad am~tam, 'A ~c, a Gatha, (and) a Kumbya,-these are measured a Yajus, a Nigada, (and) casual speech,-these are unmeasured' Ait. r 2. 3. 6 (p. 115, 9 Kmtb).

22 J. tat pttrttO 'manavao, I sa enan brahma gamayati, 'Then a non-human person,-he makes them go to the Brahman'. Chanclo Up.4.15 5-6; 5.10.2.

WIth the excephon of the Anandac,rama edItlOn al! the editlOns read manavah wlthout avagraha, and the Commentary of RlLgavendra also omJts the avagraha Itankara's and RamanuJa's Commentanes both read amiinavah. Boehtlmgk m bis edItlOn of the Chand. Up. suggested the emendatlOn to miinasa~, comparmg BAU. ~ 5.7 and 6. 1. 18 M. = 6.2 15 K. but abandons thls emendation m favour of puruso 'miinavah, Ber. sachs. Ges. WISS. 1897, p.87 Deussen regarded the puruso miinasah of BAU. 6. 1 18 M. (= ItB 14.9. l. 18) = 6 2 15 K. as 'Correctur des llIcht mehr verstandenen puruso 'miinavah' of Chand. Up. and It is note-worthy that both Itankara and RamanuJa on Vedanta Stra 4.3.4-6 have amiinavah not only for Chand. Up but also m the quotatlOn BAU. 6 2 15 K.

Boehthngk's obJectlOn to the resumptJve sa 'das hier gar mchts zu thun hat' (p. 102, 29) appears to be groundless III vlew of tbe examples he! e col!ected

J 10. The Proleptic Subject Nominative is accompanied by

a Participle. Tbe examples where the Participle is not further quahfied are

given in Group A (Ex. 1-8). In Group B (Ex. 9-18) are listed the instances where the Participle is further qualified, eIther (a) by a dependent object accusative (Ex. 9-16), or (b) by an lD-struroental expressing tbe agent (Ex. 17). In Group C (Ex. 18) tbe Participle is the present participle of tbe y man with a predI-cative nominatIve.

As regards the tense of the Participle, the present tense pre-ponderates (Ex. 1-9; 11; 14-18); once a future participle occurs (Ex 10); twice the middle perfect participle is found (Ex. 12; 13), once ihe past participle (Ex. 8 B) 18

PlOlephc NommatJve. S 10.>

Caland, Auswahl, p. 48, no. 40, note 2 (to JB. 1. 147, see 11, Ex. 8 below) regards the present participle caran in passages hke-Ex. 3-8 below as having the functlOn of a finite verb; but 1 cannot persuade myself that, with the exception of the past parti ciple III -ta, participles occur in the BrahmaJ;la prose in the clear role of finite. verbs, cf. Delbruck, Altmd Syntax 219, p. 393 (last line)-394, 7; SpeiJer, Ved. und Sanskrit Syntax, 287, p.92, 4; Sanskrit Syntax, 9, Remark, p. 5, Keith, ZDMG.63 (1909), p. 346-349. Delbruck nghtly emphasizes the close simIlarity of such proleptlC nominative constructions with the proleptlC accu satlVe constructions 1ike QB. 1. 1. 1. 16 (below 32, Ex. 6). A proleptic constructlOn like PB 15. 3. 25 dirghar,;rava rajanya nnr jyog aparuddho 'r;;anayar;; caran sa etad dairghar;;ravasam [sama} aparyat (below Ex. 6) is closeIy paralle1ed by the normal conjunct arrangement PB. 9. 2 19 devattth~k saputro 'r,;anayanr;; carann aranya.. urvarny amndat, 'Devatithi, accompanied by l11s son, waIkmg hungry m the forest found cucumbers', ef. also the proleptic eon-structlOn K. 21. 6 (p. 45, 2) with its normal conjunct parallels, given > below Ex. 14

At PB. 13. 4 17 the text should read tan adhinidhaya pancary acarad (Blbl. Ind. paricaryyacaran) vmdhayans, tan vardhayltva 'bravU ., cf JAOS., 19, P 125-and the Cornmentary' tan prthural)mytid'in adh~mdhya rathe ropya vardhayan posayan paryacarat (Blbl. Ind. paryacaran) pancaryam krtavan . Partcar'i 1S-adJectJve as taken by PW. s par1-carin, and acmat may well have Its full verbal force: 'He went as theh care-taker, tendIng them', cf Chand. Up 4.4. 2, bah/} a~at carant partcnn; yawane tvam alabhe 'Wandermg a good deal about 10 rny youth as a servant-woman 1 concelved thee'. The order of words makes the cnstrucLlOn of acarat as auxlhary verb Wlth vardhayon (Delbruck, Altmd Syntax, p. 390; Whltney, Gr. 1075, b) less likely.

The perfect middle parhcip1e ijanrrft at TS. 6. 6. 7. 4 yatayamani va etasya chandas~ ya ijana~, 'The metres of hIm who has sa-crifieed are exhausted', does not stand in Jieu of a finite verb . but is a predicate particip1e wlth the substantIve verb bhavatt or as ti omitted, ef. Delbruck, Altind Syntax, 218, 1, p. 392 wbere instances of the future particip1e with -y bh and 1"as are given.

For the finite use of parhclples 10 Mantras cf. Delbruck, Altmd Syntax, 21!!. P 393 f.; OIdenberg, ZDMG. 54, P 170, Anrn. 1, RIgveda Noten, 1, p. 428 col. 2 s. 'PartJzlp statt Verbo fimtum', n, p. 379, col. 1 s. 'PartlZIp ohne Hauptverb'; Neisser, BB 27, p.274, KeIth, ZDMG., 63, p.346.

ti 11, Ex. 1-3 ProleptlC Nommatlve For a pendent nominatIve wrth present palttclple antIclpatIng an accusative

(JUB 3. 29 7) see below AddHlon from te Aranyakas and Upan~' sads, 20A, 5A, Ex 1 A.

For an absolute nommatlve wlth present parhClple before the exclamatory att (= etl, AB. 2. 13. 6 see below 26, Ex. 8.

Do Examples' A. 1. par,ur val, myamana{t sa mrtyum praparoat, cThe victim

bemg led (to be slaughtered)-it saw death befo re (it)'. AB. 2. 6. 8. Thls passage is clted by SpelJer. Ved. und Sanskllt Syntax, 287, P 91 among

ProleptIc NominatIve 11, Ex. 4-8.

for a lqng time expelled (from bIS realm),-be saw tbis Saindbu-k~ita (-Saman)' PB. 12. 12. 6.

For V rudh + apa = 'expel from one's kIngdom' cf AB 8 10. 7 rastrad apa-rudhyamanah; C;:B. 12 9. 3. 1 3 da~apurusamraJyad aparuddhah, AV. 3 3 4 anyaksetre aparuddham ca'antam. Cf. V bh + apa, 73, Ex. 13 and ef. for tha pbrase aparnddha; carat, Klelhorn's note to verse 14- of the Alhole Inscllption oLPulake"In 11 4I Eplgr. lnd VI, p.9 and hls note In JRAS 1907,11, P 1062f.

4. s'/,ndhuk~zd vm bharato raja jyog aparuddhar; caran so 'kamay-ata' 'va sva okasz gaccheyam tti, 'King Sllldhuk~lt Bharata wauder-ing for a long time expelled (from bis realm),-he desired' "May 1 return into my own borne".' JB. 3. 82 (Caland, AuswabI, p. 238).

For r gam + ava = 'leturn, be restlJred to one's real m' cf TS 2 3 1; K 11, {); MS. 2 2 1 where the verb m frequently used in connection wlth the nlual for the restoratlOn of an expelled (aparuddha) ruler. Further K.27 5 (p. 145,4) = 28. 1 (p. 152, 4); K 27. 5 (p 145, 8) = 28. 1 (p. 152, 7) K. 28. 1 (p 152, 9), AB. 8. 10. 7, PB. 2. 3. 7 In all these pa"sdges r gam + ava governs the aceusatIve:

v~{:am; anaparudhyam; rastram; ksatlam, devavifiim manusyavt~am; raJyam (ef. also PB 4. 3. 7 adhvanam,

11, Ex. 8E. ProleptIc Nommative.

8 B. so 'yam VI$nUr glanalJ I chandobhir abhttalJ pangrh~to 'gmk purastan na 'pakramanam asa sa tata evau '$adhinam mulany 'upamumloca 11 te ha deva uculJ: kva nu m$nur abhut kva nu yajno 'bhud tti, te ho ' cu~ chandobhtr abh~talJ parigrh'ito 'gnt'1J purastan na 'pakramanam asty atra'/, 'va 'nvtcchate 'tt, 'This same Vi~nu brmg ti red out,-he was enclosea on al! sides by the metres, Agni was in front, there was no escape-, het hereupon hid among the roots of the plants The gods said: "What hath become of

Vl~J:.lU? What hath become of the sacrifice?" They said: "He is enclosed on aU sides by the metres, Agni is in front, there is no escape, search (for him) right here".' QB. 1. 2. 5. 8-9.

Eggehng takes glanal,J, Delbrck, Altmd. Synt, p. 3fJ4, 37 -395, 1, both glanah and chandobhir abh~tab, pu,ngrh'itab, as eqUlvalent to a fimte narrabve tense. The sngulanty of such a use of the past partIclple was pomted out by Kelth, ZDMG 63 (1909) p.348-349 who correctly propose~ takmg sa 'yam Vtsnur glanar; chan-dobh~r abh~tal pang1hltah as prolepbc.subJecl lo upamumloca whlle he regards

agn~h .. apakramallam asa as a parenthesl!,. 1 dlfTer from hlm m taking only $0 'yam v~snul' gliinah as prolepllc subJect nommahve, whlle chandobhir abhltah pangrMtah wlth agnth put'ostiit and na 'pakramanam lisa forms a trmembral parenthesls after WhlCh the subJect IS resumed by sa (cf. 9, Ex 18 Rem. and 11, Ex. 1, A Rem). Thls seems to be demanded by paragraph 9 where the same three tems are grouped together (8 chandobhw abhttah p~ngrhto 'gnth purastan nli 'pakramallam asa :. 9 chandobhzr abhitah pangrhito 'gnth purastiin na 'pakramanam astl)

The Commentary takes apakramallam ntt 'sa = na Jagama as timle verb oC the firsl clause and appears to construe so 'yarh VISllur gllinah as ab-solute nommallve : glanasya tasya vlmor apagamanarlt nli 'sa

B. a. 9. deva ha va~ yajam tanvanas te 'surarak$asebhya asar/,-gad blbhayarh cakruh, 'The gods bemg engaged in performmg -the sacrifice,-they were afraid of the Asuras and Rak~ases lest they should chng to tbem'. QB. 1. 1. 2. 3; 1. 2. 1. 6 (cf Delbruck; Altmd. Syntax, 219, p. 394, 1); 1. 3. 1. 5; 1. 4. 4. 8, 1. 6. 1. 1~;1 3.3 3.16; 3.3.4.2; 3.5.3.15, 3.6 1. 27; 3.7.2 2,3.9.4. ti! 4. 1. 1. 6; 4.2.4. 19; 4 6.6. 1.

For lisaga cf. QB 1. 5 3 21 j 4. 1 1. 6 yadt no 'sU1'araksasliny asaJeyuh, 1 6 1. 12 su yady enan pUl'astlid asuraraksasany aSlsaiksanj 15 yady enam pura~tiid asllraraksasiny aSlsaksantt; 3.6. 1. 27, 4. 2. 4. 19 tan daksmato 'surarak~asliny iiseJuh.

The phrase asuraraksasebhya iisaglid bbhaylim cakruh m the fourleen QB' passages cIted ahoye appears lo be the only absolutely certam mstance m the Brah-

22

.

Proleptlc Nomnatlve S 11, Ex. 9- 11

mana prose of V bh wh a double, paratacbcal ablabve of both the person and the acbon feared. The case form IS amblguous in agneh, TS 5 2. 5 2; 5. 2. 10 2, K. 20 3 (p 21,4); 20 9 (28, 8), tyam [prthiv'i] va agner atidahd abibhet, 'Thls (earth) was afrald of Agm, lest he should burn (her) excesslvely', where Delbruck, Altllld. Syntax, 69, P 111 takes agneh as ablabve, probably correctly, SIDce here al so Agm IS the person and attdaha the actlOn feared Contrast the ablatIve of tbe actlOn and the gemtJve of the person ID the foIlowing mstances: QB. 4. 3 3 11 apah'amad tf hm 'vat '~am etad btbhayarn caleara, K 29 9 (p 179, 6) te~al1t apalcramad ablbhet, 'He was afrald le~t they mlght go away', KB. 7 4 tasya lesalld mbhemt. '1 am afrald lest It pensh', MS 3 7. 10 (p 90, 1-4.) te va anyo 'nyasya 'bhtdro"fad abtbhayuh, 'They were afrald lest one mlghl decelve the other', AB 5 15. 9 prasrasad vii antasya bibhliya, 'He IS afrald (perfect = present) lest the end may become loose', AB 4.19 2, K.33 6 (p 31,11) tes,h vm det,li!t (K. deva Val) saptadal,(anlim pravlayad ablbhayuJ;, 'The gods were afrald lest these

Saptada~as should coIlapse', PB 8 9. 21 sa tayor [samnoh] aprayogad ablbhet,

11, Ex. 12-15: ProleptJc Nommatlve.

12. prajapaftk praJi},s sasrjanas sa vyasrasata, 'Prajapati having created offsprmg,-he became disjointed'. JB 2. 128 (Oaland, Aus-wahl, p 164); 409 (Oaland, Auswahl, p.164)

Contrast the normal wnJunct constructIon e. g PB. 15. 8. 2 = 16 5. 13 pm-Japatth proJa tesrata, 8a dugdho nNcano 'manyata, and compare below 12, Ex 3.

13. vag va~ yaJiiarn sasrJlna sa ~ 'r,;anayat (the mss. sarnanyat) sa 'p~pasat, 'Speech havmg created the sacrifice,-she was hungry, s11e was thlrsty'. JB. 3. 312.

The emendahon l~ based on JB. 1. 252 na ha va a9anyat~ na ptpasat., 2.6, 382 na 'I}anyat. na ptpasatt

14. agtraso vat svargam lokam yantas te 'J{iyam gharmam pra-sican, 'The Aglrases going to the heavenly world,-they poured forth the canldron upon the she-goat'. K. 21. 6 (p. 45, 2)

.The two parall"t texts have the normal conJunct constructlOll' TS 5 4. R 2-3 agirasah suvargam lokam !/antah I ajayilrh ghm'mam PI aSlcan; MS.3 3 4 (p. 37, 7\ agtraso vm 8var ynto (SchlOeder's text svarynto, see the note to Ex. 15) '7iiya1n gharmam prasiilcan

15 angtraso vat svargarn lokarn yantas te mekhalas saritnyaktrans, tatar,; r,;ara udatl~that, 'The Agirases going to the heavenly world,-they scattered (thelr) girdles about, thence arose the Qara-grass'. K. 23. 4 (p 78, 3).

The two palalIe[ passages have the normal, conJunct conslruction TS. 6. 1 3 3 agu'asah SIlVa gam loleam yanta rJarh vyabha)anta, t(lto yad atya~isyata te I}ara abha'Jan, MS 3. 6 7 (p 69, 5) agtraso vat svar ynto (Schroeder's text svarynto) yatra mekhalah samnyasans tatah 'iaro 'Jayata

24

Rere (MS 3. b 7, p. 69, 5) as well as at MS 1. 9. 5 (p. 135, 6); 3.3 4 (p 37,8), and 3. 9. (p 123,9) Schroeder's text, wJth the manuscrlpts, reads svarynto (so, and nol as Caland, WZKM, 23, p.55 prmts svryanto) as compaund. 8uth a compound would be wJthout parallel (cf. Wackernagel, Altmd. Gr 1I, 1, p. 193, 82, b) and an emendation to 8var ynto serlUS necessary m splte of the concurrent testJmony of the mss.

The entry aval'ynt III PW. col. 1455 and pw. p. 240, col. 3 fol' VS. 17 6~ is due to an overslght. At VS 17.68 the accent marks svr yantQ slgnify svar ynto, cf. Weber's IntroductlOD, p X, where y 'sau c = yo 'su ca IS clted, el. the paraIlel texts AV. 4. 14. 14, TA. 1. 27 5; QB. 9. 2 3. 27; TS. 4. 6.5 2, 5 4 7. 1; and svar yat VS 11.3,

PlOleptlC NommatIvE' 11, Ex. 16-18; 12, Ex 1-6.

16. addya val, svargarn lolvarn yantas te ' gtraSO ' smznl loke saha yaJfiena pratya?thali, 'The AdItyas gomg to the heavenIy worId, -they pushed (= kept) the AgIrases back m this worId together wIth the sacrIfice'. K. 28. 3 (p 166, U).

b) 17 deva va asurazr hanyarnanas te prajapattm upadhavan, 'The gods bemg on the point of bemg slain by the Asuras,-they had recourse to PrajapatI.

C. l8. pnyamedha ha vai bharadvaja yajamdo rnanyamtintis te ha sma na karn cana vedamdam upayantt, 'The Priyamedhas (and) BharadvaJa

12, EX. 7-13 ProJeptlc Nornmalve.

The parallel passage 1\18 4.6 8 (p 91, 15) has two paratactlCal clauses wdro vat vrtram ahan,. ,sa mahendro 'bhavat.

7. mdro vrtram hatva 'suran parabhavya so 'rnavasyam pratya-gaeehat, 'Indra baving slain Vrtra, havlUg defeated the Asuras,-he returned to the night of tbe new moon'. TB 1. 3. 10 1

8 ta [prevah] anyatra 'nnanz avdtvft tah prajapat~nz eval 'kadha 'nndyaya 'bJnsamavartanta, 'Tbese (creatUles) findmg food nowbere else,-they aU at once turned togetber to Prajapatl for food-eating'. K 37. 1 (p. 84, 15).

The parallel passage TB. 2 7 9 1 wJthoul prolepls' ta anyatra 'nnadyam aV1til a praJapatim praJ/i l~pavartanta

9. sa [prajapati[,] dar,;adha 'tmananz mdhaya mlthunarn krtva .sa ayatanarn mechat, He (PrajapatI) baving dlvided blmself in ten parts, bavmg copulated,-he desired a resting place'. K. 9. 11 (p. 112, 13).

The parallel passage M8 1 9 3 (p. 132, 9) wlthout prolepsis' sa da~adhii 'tma-nam vidhya mthunam krtva 'yatanam atOchat

] O tena devas samvatsaranz kalpaYitva svargarn lokam praJaya ta etad yajasya Jyottr upart~tad adadhu[" 'The gods having there-by ordered the year, havmg dlscerned (a way to) the heavenly world,-they placed tbis ligbt of the sacnfice on high'. K 28 2 (p. 154, 18).

11. sa [praJo,patl[,] praja asrJata, ta asya prajah sr#a(Q) stanav ev((, 'bltipadya tas tatal} sambabkavuk, 'He (Prajapatl) created crea-tUl'es, tbese creatures created by him, havmg recourse to (thelr) breasts,-tbey tbereby sustamed life'. QB. 2.5. 1. 3.

12 ahoratre deva abhijztya te vajram eva pandhzm akurvata, 'The goda havmg won day and mght,-they made the thunder-bolt the enclosure'. JB. 1. 209. 4 (Transactions of the Connecticut Academy of Arts and Sciences, 15, p. 163).

13. ahoratre dev abhijztya te 'mum adztyam savanar eva pratyaii-cam anayan, 'The gods having won day and night,-they led yOilder sun westward by meailS of the pressings'. JB. 1. 212. 3 (TransactlOns of the Connecticut Academy of Arts and Sciences, 15, p. 165). 26

Prolepbc Nominahve 13, Ex 1-6

13'. 4 The Proleptic Subject Nominative accompanied by both a Partwiple and a Gerund.

1. deva va asuran hatva va~radeyad i$amanas te dZQo 'mohayan,

14, Ex. 1-2, 15, Ex. 1. Prolephc NominalIve.

In both passages Caland construes ayanta7J, as eqUlvalent to a filllte verb (see aboye 10), rendermg JB. 1. 226 (Ex.5) by: 'Als dIe Nachkommen des Kanva eine rnehrtaglge OpfersIlzung beendlgt hatten, kamen sle zuruck, sich Ilnschlrrend (?). Da sbeoen sle ... auf eme mlt Kurhlssen dlCht hedeckte .. .', and JB 3 203

tEx. 6) by' 'Als uie RS1S eme mehr\aglge Opfelsltzung beenulgt hallen, kamen Sle zurck, slch anschirrend (?). Sle sagten' .. '

1 have adopted m both passages the readID~ of Whitney's transcrlpt at JB. 3. 203 (Ex.6) utthaya ylintall, preferrmg yantah \0 yantah because \he former is

u~ed e g. AB. 5. 30. 1 (yatho 'bhayata;cakrena yan) wIth reference to travel by a car, which IS Imp]ed III OUI' two passages by liyuJanlih. The compounds V ya + a OI' V ~ + Oto come' do not seem to tit the context

B. The Proleptic Subject Nominative is not resumed by a De-monstrative Pronoun but its Severance from the Rest of the

Clause is indicated by atha (cf aboye 3, 1, bj 7,1, b). 14. 1. The Prolaptic Subjact Nominativa accompaniad by

a Present Participle. Both of the following mstances, Ex. 1-2, are noted by Delbruck, Altind. Syn,

tax, p 539, 36 Fol' atha sea aboye 7, b. 1. catu~pad't va~ gam, tasmac catur [l{lam] upahvayate 11 sa L'ai

catttr upahvayamano 'tha nane 'vo 'pahvayate 'Jam~taya~ 'Four-footed is the cow; therefore he calls four times upon (tbe 1q.a); venly, he, calling four tImes upon (ber),-yet he calls upon (her) in dlfferent ways, as it were, in order to avoid repetition'. QB. 1. 8.1.24-25.

2 brahmavadino vadanti : sa tvai ya jeta yo ' gm~tomena yajamalw 'tha sarvastomena yajete 'ti, 'The theologians say: "He, In sooth, would sacrifice who (while) sacrificing with the Agnil?toma, -yet should sacrifice with the Sarvastoma'. TS. 7. 1. 3. 1.

Tbis passage shghtly dffers from the preceding one (Ex. 1) ID that the pIOleptIc constructlOn occurs in a relativa clause

13. Tha Proleptic Subject Nominative accompanied by a Gerund.

1. tato deval) sarvam yajam samvrjya 'tha yat pap~~tharh yajnasya bhagadheyam asit tena~ 'nan [asuran] nirabhajan, 'Tbereupon the gods having seized the whole sacrifice (fol tbemselves),-then dls-28

Pendent Nommative. 16, 1 ..

possessed these (Asuras) by means of that which was the worst portion of the sacrifice'. QB. 1. 9 2. 25.

For atha see aboye 7, b.

H. The Pendent Nominative. The disjunct nominative does not form the snbject of the clause

but anticipates an oblique case SpelJer, Sansknt Syntax, 499, P 387-388 cItes a f>w mstances of the Pen-

dent NommalIve III classical Sansknt as 'a kmd of anacolouthon'

A. The Pendent Nominative Resumed by a Demonstrati:ve Pronoun in its Proper Oblique Case.

16. 1. The Pendent Nominative anticipates a Genitive A) The Pendent Nominative consists of a simple Noun. 1. In the Pariplava Akhyana, a legend to be recited at the

AQvamedha-sacrifice and to be repeated at stated intervals' during the space of ayear, occur the following ten clauses'

The ;B. Inserto; an ity aha after raja and before tasya, the parallel texts AQ. 10. 7 and QQ. 16.2 omIt thls Insertron, WhlCh do es not belong to the text of the l~gend, but they also omrt the word raja.

a) manur vawasvato raja tasya manu~ya vz()al,t, 'King Manu Vai vasvata,-his people are men'. QB. 13. 4. 3. 3.

b) yamo vaivasvato raja tasya pitaro vi~af, (King Yama Vaivas-vata,-his people are the Fatbers'. QB. 13. 4. 3. 6.

e) varuna addyo raja tasya gandharva vi~alJ, 'King VaruJ;la Adi-tya,-his people are the Gandharvas' QB. 13. 4. 3. 7.

d) somo val~navo raja tasya 'psaraso vz~af, (King Soma Vail!J;la-va,-his people are the Apsarases'. QB. 13.4.3. 8,

e) arbudaf kadraveyo raja tasya sarpa vifah, (Kmg Arbuda Kadraveya,- his people are the, Serpents' QB. 13. 4 3. 9.

f) kubero vaz~avano raja tasya rak~aS1 vifaf, (King Kubera VaiQravaJ;la,-his people are the Rak~ases'. QB. 13. 4. 3. 10.

g) aszto dhanvo raja tasya 'sura vi~al,t, 'King Asita Dhanva,-bis people are the Asuras'. QB 13. 4. 3. 11.

b) matsyah sammado raja tasyo :clakecara vi~ah, (King Matsya Sammada,-his people are the Water-dwellers'. QB. 13. 4. 3. 12.

29

16, Ex. 2-5, 17, Ex 1 Pendent Nommahve,

~~. reads IIdakacariih for udakecarah, tf Wackernagel, Altlnd Gl n, 1, 84, e, 1', p. 199, 14 and 88, b, P ~10, ~3. . ,

1) tark$Yo vaipat;yato raja tasya vayas~ vlt;ah, 'Kmg Tark~ya Vaipa~yata,-his people are the Birds'. QB. 13.4.3. 13

j) dharma mdro raja tasya lleva vtfah, 'King Dharma Indra,-his people are the gods'. QB. Id. 4. &. 14

2. yad rdhvarh ca~alad dvyaftgulam va tryangularn va sadhya ~ti devas tena te{lam lokam jayatz, 'What (space), two or three fingers Wlde, there ]s aboye the top-ring (of the sacrlfiClal stake),-the gads called "Sadbyas",-their world he wins by that'. QB. 3. 7. 1. 25:

The parallel passages TS b 3 4 8-9, MS 3 9 4 (p 1 Hl, 3 ff.) and K 26 6 (p 129, 7 ff.) have notbmg 5)ntachcalIy eomparable.

3. vaSavo mdra adltyas te{lililn da) (the mS. et'/,) jyotw agram yad asavadityaQ" 'The Vasus, the Rudras, (and) the Adityas,-their leader is this light, (viz) yonder sun'. JB. 2 240 (JAOS. 18, p. 34).

The fragment of the Qatyayana Brahmana eIted by Sayana on RV 7 33.7 shows the same construetlOn

B) The Pendent Nominatlve consists of a Noun accompamed by a Present PartIciple.

4 deva vaz svat'gam lokam yantas te{lam yam chandarisy ani~ '1'tektini svargyany asas tazs saha svargam lokam ayan, 'The gods going to the heavenly world,-what their undefined heavenly metres were with them they went to the heavenly world'. K 21. 2 (p. 68, 14).

5. tau [a~vmalt] devan upavartamanau tayor ya bhl{lajya tan,r ast tam tredha vinyadadhatam: agnau trtiyam brahma't}e trtyam apstt trtiyam, 'These two (A~vins) turning toward the gods,-what their healing form was that they deposited in three differeut place s : one third in Agni, oue third in the Brahmana, (and) one third in the water'. K. 27.4 (p. 143, 14)

The two parallel passages l\1S 4.6.2 (p. 80, 8) and TS 6 4.9 1-2 have nothmg syntactically comparable

17. 2. The Pendent Nominative, consistmg of a Series of Simple Nouns, anticipates an Instrumental.

1. pa~avo vag ~ndnyam pranpanau tair va mdro 'kmayata sayujyam gaccheyam iti, '''Cattle, speech, vigour, m-breathing (and). 30

Pendent Nommatlve 17, Ex 2

out-breathmg,- wh them", Iudra desired, "may I be untted"'. MS. 3. 10 6 (p. 137, 13).

2. rathantaram pr~tham rathantaram t;astram agnz~tomo yaJi/as tene 'marn lokarn rdhnott, 'The Rathantara Pr;;tha (Stotra), the Qastra containing the Rathantara, (and) the Aglllr;;toma sacnfice,-by means of that he prospers in regard to this world'. QB 13, 5. 1. 2. .

Excursus: The Nature o the Accusabve wlth Y'l'dh, y'l'adh, and yplt~. A. The aeeusatJve wJth y pus IS, m the Brahmana proge, always to be regarded

as an aceusatJve of eontent or referenee (Delbluck, Altmd Syntax, 122, p. 177, 3:3-178,7) rather than as aeeusahve of the dlreet obJeeL Thus, PE. 25 16 3 te sahasramsahasrarh put1an aplt~yan and te sahasram-sahas/"am putran pu~yantz. 'They throve (thnvel wlth thausandfold male Ilrogeny' (Delbruek: 'Slchel helfi,t das mclJt .Sle la,sen tausend Sohne gedeIhen" , sondern dab Gedelhen haftet an dem SubJect von pusyatt'), cf RV. 1 64. 14 tokam pusyema tanayam r,;atwh htmah, 'May we thflve a hundred years wlth progeny (and) offspnng'; PB 10. 1 6 aro dhamasar,;O ht praJah par;ava oJo balam pusyantt, 'For every half-month offsprmg (and) cattle merease m strength (and) vIgour', AE. 2 1 7 pusyati praJarh ca pa~ir ca, 'He prospers wlth regard to offsprmg and caUle'. In these examples the aceusatIve IS semantrcally eqUlvalent to an Illstrumental, cf. TS 2 1. 9. 4 2.4.6.2 pusyatt pra}aya pap,bhth (synonymous wh y vrdh TS. 7. 4. 3. 2 Val dhante prajaya pa~ubhth) i TB. 2 :3 3 2 sau [prthwI] 'sadhbhtr vanaspattbhtr apusyat, sa [vayuh) marlcfbhw apusyat; tad [antartksam] vayobhtr apusyat; sa [ad,tyah] raemron,r apu~yat, sa [dyauhl naksatratr apusyat, so [candramah] 'ho) atratr ardhamsatr masotr rtubhlh samvatsarena 'pusyat. In ihe end, AB. 2. 1 7 pu~yati prajam ca pa~il1i~ ca and TS. 2. 1. 9. 4 pusy(ltt praJaya par,;ubluh have very nearly tbe same meamng as TS 5. 7. 8. 3 praJC!m pa:n . aVal unddhe

Wlth a slIghtly dlfferent shade of meamng V pus lS \lsed at \iB 13 'l ;1 8', TB. 3 9. 7 2 tasrmid raja pagn na pusyatl WhICh IS rendered by Delbruck ~ 'Deshalb hat em Adhger seme Starke meht m Heerden'.

Eggelmg renders' 'Wbence the kmg do es not lear eattle'. But whlle .. pus wrth causatIve meanmg oecurs m Mantra (e. g RV 3 45. 3 kl atum pusyasi ga fva 'thau mcreasest thy power hke kme') the Brahmana prose uses the causatIve in thIS sense, e g TB. 1 6. 2 5 praja eva tad yaJamanah-po.~ayatt 'Thus the sacnficer causes (rus) offsprmg to mcrease'

Fmally K. and J8. use the accusatives varnam and rilpam dependent on y pus: thus K 9. 11 (p. 113, 1) ahna devan asrJata, te fukluth varnam apusyan ('were resplendent m whlte colour') ratrya 'sura1i8, te krsna abhavan; JB 3 303 sal [prthiui] 'kamayata: prcmm varnal/! pusyeyam itt, sat 'tam mantram apavyad : a 'yam gauh pr~nir akramd (RV 10. 189. 1) ttt, tato va iyam prth1vi pr~mm varnam apusyat ('attallled vUllegated appearance') pr~nim ha val varnaril _praJaya

31

17, Ex. 2. Pendent Nommatlve

pa~"bhih pusyatl ya evam veda; K. 11. 2 (p. 146, 2) bahu va efad anyad anya) Juhott, sa ifval'o 'nyad rupam posto~.

cr, in Mantra, VS.4 2 bhadmm varllam pusyan, RV 8 41 5 sa kat'h kavya pUI'U rpam dyaur ,va pu~yatz. AV. 7. 60 7 vl"va rpam pusyata 13. 2. 10 vl~va rupani pusyasi.

Tbese accusalIves of content or reference are easy developments of tbe frequent cognate accusative wllh V pus, thus wlh man eosan MS 1. i.2 (p 111,6); K. 9. 1 {p. 104, 14-15); MS. 1. 7. 3 (p 112.2), K 8 15 (p. 99. lU). PB. 21. 10. 5; with eMn posan TS 7. 1. 9. 1, PB. 8 4. 4.; wIlh tan posan TB 2 3. 3. 2; wilh para-mait posam PB 21 10. 9; wth bahn posan PR. 19 5. 10, wIth sarvan poslJn PE 21. 10. 7; 22 7.2, \VIlh bhyiisalh posam ~B. 3.7 14, 16, wilh sahasrapo~am K. 24. 6 (p. 95, 18); wlth sahasraposlJlt JB. 1 21 (sahasraposiin ptt$yatl ya evam fJdvan agnihotram juhotr; wilh pltstm PB. 19. 14. 1; simlldry MS. 1. 5. 9 (p. 78, t)

8ahasrapo~asye '!fe, pusyatl sahasram; TB. 2. 3. 3. 1 8a [agmh] sahasram (Blbl. Ind,' sahasram) apu~at, 'He lhnves (throve] a thousandfold'.

er. the YaJus ~B. 14. 9. 4.23 (= BAH 6. 4. 23 Madhy. = 24 K!tnv.) asmin sahasram pusylJsam; Kau~ 89. 13 asmm sr:..hasram pusyasma; ApMB. 2 11. 32 asmlnn aham sahasmm pusyami.

B. The accusattve wllb v"adh, where it occurs m the Brahmana prose, must he Interpreted In lhe same way as accusalive of contenl or reference. Thus,

Pendenl Nommatlve. 17, Ex. 2.

tad yam eva praJapatzr rddh!m ardhnot tlim eva rddhtm rdhnoti . ya evam veda; JB. 2 98, 2. 132; 3.186 samlin'im rddhlm rdhnuyliva,. ., samlinim rddhtm ardhnutam: JB. 2.213 (bIs) saman'im rddhlm rdhnuyat, ... , sai 'tam eva rddhim {irdhnot; JB 2. 215 (bIs) samlinim rddhtm rdhnuylim, .. , film eva rddhtm rdhnoti; JUB 3.19.5.

Contrast with the~e cognate accusahves the cognate mstrumental al 1\'[8 4. 1 2 (p. 3,16) sli 'rdhnQd rddhyli, GB 2. 1 12 rddhYli rdhnotz.

Flom thls cognate accusabve other accusatlves of contenl or ref~rence develop. Thus, as wlth v' radh, the accusalIves kamam and kamlin- KA. 26. 1 etan and .tan kamlin rdhnuvantt; 25. 2; 26 15 sart'ltn kltman rdhnuvantt; 1\'[S 1. 6. 8 (p 98,3) tam kamam a"dhnot and rdhnotl; K. 7. 5 (p. 67,5) klimam asya [agneh] rdhnotl, manusasye 'n nu yah klimam rdhnott sa msiylin bhavatz, JB.~. {24 (bis); 330 tam kamam rdhnavani yo da;aratre, JB 2.330 rdhnott hai 'nenai 'tlin kamlin ya .etasmm [atlrlitre] klimah, gA 1 2 sarran kaman rdhnuvantt.

kamam, kliman rdhnotz, 'He succeeds m regard lo his wIsh(es)' is thus nearly equlValent to kamam, kn/n pnoh, 'He attalnS hls wlsb(es), gB. 2.1. 4. 3; 4; 5, 30, 2.2 1. 5; 9,3.1. 4. 2~, 4, 5.8.14 (bIS); 6.2 1. 17, 7 2. 2 21, 9 2. 1. 18, 9.3 3. 2 (bis); 8 (bis), 9 4 4- 10 (hls); !l 5.2 9; 10 2.6.15, 13.4.1. 1 (biS), 12, 13; 13 5.1. 9,14; 13 5.2. 9 22, 13 8. 1,3, AB. 3. 48 5, 8 14 4, 8 19 2; KB. 2. 3, 4.4 (bIS), 6 1" (bls~; 7.1; 11 6,13.1; 21l. 1 (ter), 29.6; GB 1. 1. 2; 2.1 2b; 2. 2.18; PB. 20. 3. 4 (bIS), ~B. 3.7.10, 1'8. 3.7 14.2; K. 2~ 8 (bl~, p.65 1, 4), 1\'[S,3 6 4 (p 64, 6); 4 6. 8 (p 91 14), and to kamam, kliman avarunddhe, 'He oblams hls wish(es)', PB. 7. 7 8, 11. 8. 8, 12.5.20, 13 3 12; 13.6.9; 13.12. 5; 14.4 7; 14.5 15: 14.6 6; 8: 10: 14. 12. 5; 23.3.4,' TS. :~. 3.5 !!!; 5.1. 8. 2; 5.4 8 1; AB 6.2 4; 2. 17. lO, 2.20.25, K 23.2 (p 76,1)

Wlth the accusatives devam, devatam, or the name of a speClfic dlVlnity, y' rdh has the sense of 'to succeed wlth the gods' = 'to wm thelr fdvour', 'to WIll tbem over'. Tbus PB 25 14.4 devatli rdhnott, AB 1 1. 5 devn rdhnuvanti; TS.5 7.4.3 devan rdhnott, JB.2 16 (quatel) tlim [devatam] eva tena ,dhnott, 17 tam [devatlim] eva rdhnuylit, PB. 17 83 ogmm rdhnott, K.8. 13 (p 97, lb); G8. 2. 1. 7 praJlipattm rdhnotl, K 22 8 (p. 64 23) enam [pra1lipatm] rdhnQtt, TS 1 7. 3. 2 ~liksad eva pro)lipattm rdhnoti, 5.4.12.2; 5 7 1. 3 tam [praJllpatlm] eva siiksad rdhnott, A B 1. 30. 31 rdhnoty enam [proJapattm J; JB. 2 34i; 356, ::J57; ~58 pra1iipatun rdhnalama; 2 236 krlsllam pl'aJapattm rdhnavlima; M8 4 8 10. {p. 119, 10) lam [iiddyam] rdhnott; K 30. 5 (p 187,10) tam [adttyam} rdhnuvant~.

1 cr v' p wllh deviin gB 7 2.2.21; K.30 7 (p 188,20); 33.2 (p 28,10); PB 17.1 1, wIlh deta/ah, K 10,1 (p. 125,20); PB 17 11. 3,22, 7 4, TS. 1. 8 7 1, AB. 4 13 7; GB. 2 1 11; wlth pra)lipatlm, AB 4. 22 9; GB.2 5 4; 2 5 8, PB.4 9.14, 16 4.12,13; 16 16 4, 17.94, 17.11. 3, 18.6 4; 5, 8; 18 7.5, 19 7.6; 21. 14 4, 22 5 5.8; 22 7.5. 8; ~B 3.7 2, TS 7.4.4.1 (biS), TB 1 8.7.1; 3.8 16.1 (biS), K.8 3 (p. 86, 14); 8 11 (p 95,1); 8.13 (blS,'p 97,7,16), 12.9 (p 172,3); 14.7

() ertel, The Syntax of Cases 8 33

17, Ex. 2 Pendent Nommabve.

(p. 206,5), 14.9 (p 208,20),14.10 (p. 209, 3), 18.19 (p. 280,7), 232 (p 75,18),29.6 (biS, P 174,11,12), 34. ~J (p. 43, 19),37 11 (p 92,18); MS. 1. 6. 7 (p 96, 20), 1. 6. 12 (p. 105, 16), 1. 11 6 (p 168, 9); 1 11. 9 (p. 171,3), 218 (p 10,5); 33.4 (p.36,1), 441 (p. 51, 7); 44.7 (p 58,15); 4 4.10 (p 61,14), 4 5 3 (p 67,8), 4 7.5 (p. 99,15); 4 7 7. (p. 102, 21); wllh ltd!tyam KB 25.3, TB 3. 12.5.8, wlth jjdtyltn PB. 16.14.2 (biS); wlth mdragnt KB 3.2; and v' rudh + ava wIth devatlih TS.3.5.4.4, 54.1.2,1) 4.8.3; 6;tJ.51.4-5; 5 7.4.1, 6.3.7 3; 7 4.1. 2; TE 1. 2. 2 5; 1. 2 5 2; 1 6.1.6.; 2.7.1. 3, 4, MS. 3.7.1 (p 75,15), K. 8.12 (p. 96,6), 10.1 (biS, p. 125, 3,4), 23.2 (p. 75, 12), 31. 11 (biS, p. 13, 11 and 14, 12); 37 7 (p. 88, 6), wJth alJmm, TS 5. 3 11. 2; TE 2 1 2 1 (biS), MS 1. 8.9 (p 129,8), K 8 15 (p 99,3), wh praJltpattm GB 2 2. 13, MS 1. 11. 6 (p. 168,10), wlth somam MS. 4.1. 1 (p 1,7)

2. JB. 2 262 etas,e cat 't'a devatltsv rdhnoty etanr; ca lokan ltpnott, 285 tltsv [devatasu] evat 'tad rdhnott tasu prattstMya 8vargam lokam eti show the locatlve devatasu mstead of the accusatIve devatah, allhuugh at JB. 2. 285 the accusative dtr;ah wIlh v' rdh Irnmediately precedl\s. dtfa eval 'tena rdhnott dksu prattsthaya svargarn lokam eti.

3 Delbruck, Altmd Syntax, p. 176, 30 renders v' rdh in this connectlOn by 'befnedlgen', PW. s ardh, no 3. by 'genugen, befnedlgen'; pw. s al'dh, no 2 by 'fordern'. Kelth in bIS note to AB 1 1. 5 (ef also hl5 note to. TS 1 7. 3. 2) regards the aecusah ve as one of referenee; hls rendermgs vary: AB. 1 1 5 'They prosper as regards the gods'; TS. 1. 7. 32m the-text 'He enJoys PraJapah' m the note 'He prospers m relatlOn to P.'; TS 5 4 12 2 and 5 7.4 3 'He wms hlm'; 5.7 4 3 'He prospers the-gods', takmg appdrently ),/ rdh, hke Haug, wJth causatlve force.

v' I dh m the sense of 'to be successful m regard to somethmg' = 'to succeed in wmning or attalmng somethmg' lS also. found III the followmg passages'

K. 11. [) (p. 149,21) ena [osadhh] Idhnotl; Cf. TB. 3. 8. 17. 4; K. 20 3 (p 21, 12) ta evo 'bhayIr [osadhIh] avarunddhe;

KB. 20 1 apnuvantl. osadhlh. MS 3 2. 1 (p 16, 2) sapta . .. saptany agner rdhnot~, 'He wins Agm's seven

sevens'.

The 'seven sevens' refers to VS 17. 79 sapta te agne 8amtdha~ sapta Jthvah.

The paIallel texts TS.5 4 7.5 and K 19.11 (p. 13, 19) read prllllUt fQr rdhnotl.

JB. 2. 237; 323, 324 (biS) atho yan lman ekaikan stoman upa,gama tan sar-dham rddhva tesu pratlsthayo 'drcam a:navamaha ltl, 'May we attalll the goal wmnmg togelher each one of these Stomas wruch we have smgly unuertaken (and) takmg our Etand in them', JB. 2. 285 d~a eval 'tena 'dhnott dtksu pi atls-thiiya svargam lokam eit, "He thereby WIllS the quarters, takmg hls stand m the quarters he goe~ to lhe !JeJ.venly world',

34

Pendenl NommatJve 17, Ex 2

Contrast with the accusalive di;a rdhnoti the locative dtksv rdhnoti al TS.3 5.10.2, MS.4. 7. 7 t(bIS, P 103, 1); 3.6.1 (biS, p. 60, 16), K 8.1 (bIS, P 84, 1-2), 29. 6 (p. 174, 9) and cf. GB. 2. 4. 4, etah paca dl;a apnott; TS. 6.4.4. 3 duta eva 'sma avarunddhe

JB.2 129 brahmana Velva sa tad brahmavarcasam ardhnod, brahmanai 'va brahmavarcasam rdhnott ya eva/n veda, 2.131 brahmana brahmavarcasam,rdh-nattima, 2. 175 yado val brahmavarcasam rdhnoty atha sa tasya purodham gacchatt; 2. 288 jyotir iva ha vai sa bhtiti yo brahmavarcasam rdhnoti; 2 345 tad brahmavarcasam rdhnuvantt tad u brahmavarcasino bhavant.; JUB. 1 37.6;

Cf. brahmavarcasam wJth v' ap, GB 2 5.8: WJlh V rudh + ava, TS 1. 5. 8 4: 2 1. 7. 3, 7; 2 2. 10. 2; 7 2 3 1, 7. 4. 1. 3, 7. 4. 2 5; TB. 1. 1. 311,27.11 (ter); 2; 27.3.3; 38 18.5; 3.9.5.4; 5,3.9.3.3, MS.l 5. 11 (p 79, 12); 1. b 8 (p 99, 16); 3 3 2 (p 33,8), K. 10 2 (p 126, 17); 20. 13 (p. 34, 1); 21. 2 (p. 38,8): 32. 1 (p 19, 20), 34 9 (p. 43, 11), 34. 10 (p. 44,4); 37.7 (p. 88,4), PB. 8 10 2; 16.14 5; 16.16 6: 18 7.2; 19.5.8,19.77,19.105; 19.11.6; 19.17.6, 19.18.3; 20.10.1. 25. 6. 3; ~B. 3. 7. 20, GB. 1. 2. 2

MS 4.3.8 (p 47, 7-8) relstram rdhnotl, The parallel passage TB. 1 7 3 1 has ta eva 'smai riistram prayacchanti,

and ef. rastram wJth v' rudh + ava TB 1. 7. 3. 6; TS. 3. 4. 8 1; 5 4 9 3; and rii.Jyam wlth V rudh + ava M~. l' 6 11' (p. 104, 2-3).

JUB. 1. 37. 5 avyaslktam elcastham ;nyam rdhnott 'He successfulIy altains good forlune not poured asunder (but) closely umted',

For avytiskta cf. KB 11. 8 and probably 27.6 below. JUB. 1. 37. 7 pranam rdhnoti, PB 4. 1 6 yaJnnmukham too rdhnuvanti;

Here the Commenl glosses~ yaJnamukharh yalniinij,h mukhabhutam agm .. stomam p,'upyii "'dhnuvantt sattrlllah samrddhu bhavanti. K 24. 8 (p. 99, 18) uses the loeatlve yaJnamukha evai 'tena rdhnoti. And ef. MS. 3. 4 1 (p. 45, 17) yaJiiamukham ella 'varunddhe; K. 8 10 (p 94,3) yaJiiamukham evO. 'labhya ..

KB. 11..8, 27. 6 ap~nm vacam avyaslktam (at 27. 6 LIndner's lext, wlthout fiSS. varlant, reads abhyaSlkliim WhlCh should probably be emended to avya-slMam, ef aboye JUB. 1. 37. 5) prathamata (27 6 antata) rdhnavuma, 'May we In the begmnmg (at the end) suecessfully attam Speech faUened, not pon red nsunder'.

Dvandvam al PB. 24 12. 4 adlty dt'andvam ardhnuvan, 'The AdJtyas prospered m couples' S, of course adverb; the Comment. glosses It by dvau-dvau mllltva; cf AB 3 50 4; KB. 26 7 (p. 122, 21); 29.8 (p. 141, 20), QB. 1. 1. 1 22 (bis); 14. 1 3. 1 (bis), TS 5. 4. 2. 1 (ter), 5. 6. 2. 5, K. 27 1 (p. 138, 20); 29. 9 (p 179, 3)

In vlew of all the preeedmg examples Jt seems advisable to regard the accu-sahve tmam lokam at

18-20A. Pendent Nominabve

tlivantai svlit'Jyarn lokam rdhnavlima as or1ginalIy quasicognate accusabves of content or referencej ~mam lokam rdhnoti combInes the meamngs of asml1il loka rdhnot~ and imam lokam ava1"ltnddhe or iipnot~: 'He lS successful in winmng this world and thnves having attamed It'

" lS. 3. The Pendent Nominative consisting of a Simple Noun anticipates a Locative.

1. athai 'tac charrarn tas1tJ,m na rllso 'sti, 'Now tbis body,-there is no sap in it'. QB. 4 4. 5. 1.

19. 4. The Pendent Nominative consisting of a Simple Noun accompanied by Ad.Jectival Qualifiers anticipates an Accusative.

1. athe 'tara! [apah} JJ('ya~ svadvya~ f}antas tas tat'rai 'va 'bhya-~ramyad abhyatapat samatapad, 'But the other (waters) WhlCh were drinkable, palatable, appeased,-these he then toiled over, heated, thoroughly haated'. GB. 1. l. 3.

The Proleptic Nommabve 1mmedmtely precedmg tbls passage is hsted aboye 9, Ex. 4.

20. 5. The Pendent Nominative consisting Of a Noun ac-compamed by a Gerund anticipates an Accusative.

1. atithyena va~ deva i~tva tant samad avzndat, 'The gods having sacrificed with the guestoffering, -them discord befell'. QE. 3. 4. 2. 1; 3. 4. 3: 1.

1. SpelJer, Ved and Sansknt Syntax, 287, p.91 cItes tbIS passage as an example of 'anacoloutblC' cOllslructlOns, and Delbruck, Altmd. Synt. p 409 notes it as an mstance of tbe absolute use of the jerund w~lb the remark tbat ttvii IS nearly eqUlvalent to a fimte verh. Cf KeJtb, ZDMG &3, 1 g09, p 346

2. At ~B 8 1. 1. 3 prafiipater visras/lit prana udakrliman devata bhtvli, tan abravU' .. , 'From PralapatI bemg d1slomted the breaths (= orgam of sense), ha Vlllg beco me dl VlmtJeS, departed; to tbem he sald' ',the punctuatlOn must be placed after devata bhtvli, not, as Eggehng does, before 1t. '

r Addition from the Aranyakas and Upall1$ads. 20 A. 5 A. The Pendent Nominative consisting of a Noun ac-

companied by a Present Participle which governs an Object Accu-sative anticipates an Accusative.

1 A. tam va~ nu tva pan~vaja i