Division .H33 Sect.on Digitized by the Internet Archive in 2016 https://archive.org/details/regvedarepetitio01bloo THE HARVARD ORIENTAL SERIES VOLUME TWENTY HARVARD ORIENTAL SERIES EDITED WITH THE COOPERATION OF VARIOUS SCHOLARS BY CHARLES ROCKWELL LANMAN PROFESSOR AT HARVARD UNIVERSITY; HONORARY MEMBER OF THE ASIATIC SOCIETY OF BENGAL, THE SOCIKTE ASIATIUUE, THE ROYAL ASIATIC SOCIETY (LONDON), AND THE DEUTSCHE MORGENI.ANDISCHE GESELLSCHAFT ; CORRESPONDING MEMBER OF THE ROYAL SOCIETY OF SCIENCES AT GOTTINGEN, THE IMPERIAL ACADEMY OF RUSSIA, AND THE INSTITUTE OF FRANCE IDolume XTvventv! CAMBRIDGE, MASSACHUSETTS Ibarvarb JUntvetstt'e press 1916 R i ^ vt A a. . c-T ion V e_ d a. s . RIG-VEDA REPETITIONS THE REPEATED VERSES AND DISTICHS AND STANZAS OF THE RIG-VEDA IN SYSTEMATIC PRESENTATION AND WITH CRITICAL DISCUSSION MAURICE BLOOMFIELD PROFESSOR OF SANSKRIT AND COMPARATIVE PHILOLOGY IN THE JOHNS HOPKINS UNIVERSITY BALTIMORE, MARYLAND PART 1: THE REPEATED PASSAGES OF THE RIG-VEDA, SYSTEMATICALLY PRESENTED IN THE ORDER OF THE RIG- VEDA, WITH CRITICAL COMMENTS AND NOTES CAMBRI DGE, MASSACHUSETTS Ibarvarb University press 1916 The volumes of this Series may be had, in America, by addressing Messrs. Ginn and Company, at New York or Chicago or San Francisco, or at the home-office, 29 Beacon Street, Boston, Mass. : in England, by addressing Messrs. Ginn & Co., 9 St. Martin ’8 Street, Leicester Square, London, W.C.; and in Continental Europe, by addressing Mr. Otto Harrassowitz, Leipzig.— For the titles and descriptions and prices, see the List at the end of volume 24. PRINTED FROM TYPE AT THE UNIVERSITY PRESS, OXFORD, ENGLAND BY FREDERICK HALL PRINTER TO THE UNIVERSITY First edition , 1916, One Thousand Copies CONTENTS PAGE Preface xvii Abbreviations xx Introduction 1 Parts and Sources and Purpose of the present work 1 The three principal parts or divisions of the work : Part I : The main body of the work (see page vi) 1 Text of repeated passages, in order of Rig-Veda, and with comments . . 1 Part 2 : Explanatory and analytic (see page vi) 1 Chapter 1 : Disposition of the repeated passages in ten classes (cf. p. vi) . . 1 Chapter 2 : Metrical variations by addition or subtraction or verbal change • 1 Chapter 3 : Lexical and grammatical variations of repeated padas (cf. p. vii) . 1 Chapter 4 : The themes of the repetitions (cf. p. viii) 1 Main subdivision A : Repetitions relating to the same god or group of divinities (p. ix) Main subdivision B: Repetitions relating to two different gods or groups (cf. p. xi) Chapter 5: Relative chronology of books and minor collections (cf. p. xv) . 1 Part 3 : Lists and Indexes (see page xvi) : Sources of the material for the present work 1 The published Vedic Concordance 1 The unpublished Reverse Concordance : present status of the same ... 2 Purpose of the present work 3 Character and scope and bearing of Rig-Veda repetitions .... 3 Most general statements as to the repetitions 3 Mass or amount of the repeated material 4 The nature of partial repetitions 4 The nature of concatenation or catenary structure 5 Illustrative examples of catenary structure 5 Concatenated lines which differ only in the order of their words .... 7 Repeated lines containing questions and answers 7 Concatenation of entire distichs 8 Boundary between repetitions and similarities an ill-defined one .... 8 Word-for-word repetitions distinguished from partial (less important) ones . . 10 Similarity of verses due to identical cadences 10 Illustrative examples of cadences 12 Hymns of like tenor which distinctly avoid verbal repetition 12 Imitative hymns : the Vaiakhilyas . 13 Other imitative hymns : 4.13 and 14 ; 9.104 and 105 13 Imitative strophes 14 VI Contents : Part 1 ; Part 2, Chapter 1 Juxtaposition of hymns with similar openings Consecutive imitative stanzas Imitative stanzas scattered through the RV Hymns parallel in structure (not wording) and with same final stanza . Similarity of obviously ritualistic hymns Similarities in mythic or legendary hymns Literary or historical repetitions On ‘ late hymns ’ and ‘ early hymns ’ Relative character of Rig-Veda chronology, and its criteria . Repetitions in their bearing on questions of exegesis .... On inconsistent renderings of repeated passages Part 1 : The repeated passages of the Rig-Veda PAGE 14 14 15 16 16 18 19 20 21 22 22 25 Systematically presented in the order of the Rig-Veda With the traditional statements of the Sarvanukramani As to their authorship and divinity And with critical comments and notes Especially as to the relative chronology of the passages repeated Explanations relating to Part 1 or the main body of this work Twelve notes as to the order of the repeated passages And as to their different kinds and varying importance, and so on The actual text of the repeated passages, with comments .... Repeated passages belonging to book 1 Repeated passages belonging to book 2 Repeated passages belonging to book 3 Repeated passages belonging to book 4 Repeated passages belonging to book 5 Repeated passages belonging to book 6 . Repeated passages belonging to book 7 Repeated passages belonging to book 8 Repeated passages belonging to book 9 Repeated passages belonging to book 10 Part 2 : Explanatory and Analytic Chapter 1 : Disposition of the repeated passages in ten classes . Classification according to extent and interrelations of the repeated passages Class 1. Groups of stanzas are repeated „ 2. Entire single stanzas unchanged as refrains at the end of hymns „ 3. Entire single stanzas, not refrains, repeated in any part of a hymn . „ 4. Substantially identical stanzas repeated with changes . „ 5. Similar stanzas „ 6. Distichs repeated unchanged „ 7. Distichs repeated with changes „ 8. Single pildas repeated with additional repetitive word or words . ,, 9. Two or more unconnected padas recurrent in the same pair of hymns or in a pair of adjacent hymns „ 10. Stanzas containing four or three or two pildas repeated in different places 27 29 29 162 180 211 242 274 306 336 402 463 489 491 492 493 494 495 498 501 505 508 511 514 Contents : Part 2 , Chapters 2 and ? vii PAGE Chapter 2 : Metrical variations as results of addition or subtraction or verbal change in repeated padas 523 General aspects of metrical variations 523 Expansion of one pada into two padas 523 Interrelation of tristubh and jagati and Interrelation of both with octosyllabic padas . 524 Metrical variation as criterion for relative chronology 525 Verbal changes as affecting minor matters of metric habit 526 Verses whose inferior metre indicates later date 527 Problematic cases of interchange between good and bad metre .... 528 Analytic grouping of the metrical variations 528 Class A : Variations as between several types of long (trimeter) lines . 529 A 1. Interchange between tristubh and jagati lines without change of meaning . 529 A 2. Interchange between the same with slight change of words and meaning . 530 A 3. Interchange between tristubh and jagati with grammatical change . . 531 A 4. Interchange between tristubh and jagati with change of meaning . . . 532 A 5. Interchange between tristubh and jagati as suggesting relative age . . 533 A 6. Interchange between tristubh and dvipada viraj 534 Class B : Variations as between short (dimeter) and long (trimeter) lines . 535 B 1. On ‘ false ’ jagati or tristubh 535 B 2. Padas of the Vimada-hymns which occur also without the refrain dipody . 536 B 3. Other refrain padas which occur also without the refrain dipody . . . 536 B 4. Padas with dipody appendage which is not refrain 537 B 5. Expansion in general of an octosyllabic pada into a tristubh or jagati . . 538 B 6. Expansion of an octosyllabic pada into a jagati 538 B 7. The same process with incidental changes 539 B 8. Expansion of an octosyllabic pada into a tristubh 540 B 9. Faulty verses of eight syllables interchanging with regular tristubh or jagati . 541 B 10. Cases where four syllables appear to be prefixed to an octosyllabic pada . 543 B 11. Cases where the expansion is by insertion 545 Chapter 3 : Lexical and grammatical variations of repeated padas . 548 Class A : Lexical variations 548 Class B : Grammatical variations 548 Matters preliminary to the subdividing of Class A : The terms ‘ synonymous ’ and ‘ non-synonymous ’ padas 548 Definition of synonymous padas 548 On uha-padas as indicated by change of theme in repeated padas .... 550 Downright uha-padas 551 Class A, Lexical variations : Six subdivisions 551 A 1. Synonymous padas with the same or closely similar words in changed order . 552 A 2. The same with interchanged synonymous words, but no change of metre . 553 A 3. Synonymous padas with interchanged synonymous words with change of metre 556 A 4. Synonymous padas with added or subtracted words ..... 557 Vlll Contents : Part 2 , Chapters 3 and 4 PAGE A 5. Synonymous piidas expressing or implying change of god or person or the like 558 A 6. Non-synonymous padas without or with change of metre .... 559 Matters preliminary to the subdividing of Class B : Character and scope of grammatical variations 561 Class B, Grammatical variations : Eleven subdivisions 562 B 1. Gratuitous and metrical variations in verbs or substantives .... 562 B 2. Variation of second and third persons 563 B 3. Variation of first and other persons 564 B 4. Variation of grammatical number in finite verbs and participles . . . 565 B 5. Variation of finite verbs and participles or gerunds or the like . . . 566 B 6. Sporadic and complex variations of verb 566 B 7. Variation of vocatives and other cases 567 B 8. Variation of nominatives and accusatives 567 B 9. Other variations of case 568 BIO. Variations of number and gender 569 B 11. Variation of pronouns 569 Supplementary statement as to suspension of the Nati 570 Chapter 4: The themes of the repetitions 571 Stability or flexibility of the verses according as they are applied to the same or to different themes 571 Critical significance of the use of the same line with different themes . . .571 Formulaic lines and their adaptation to different themes 572 Verses containing figures of speech adapted to different situations .... 574 Verses ascribing creative or cosmic act6 to the gods (Henotheism) .... 575 List of verses mentioning creative or cosmic acts 575 Cosmic acts connected with the sun and heaven and light 576 Control of the world and its creatures and its laws by the gods .... 576 Verses expressing more general ideas that befit a religious text .... 577 Piety and service of the gods : Pious men and households 577 Gods as source of inspiration 578 Barhis: spreading of the sacrificial straw as act of piety 578 Prayers and hymns : call upon the gods ......... 578 Soma-sacrifices and others 578 Expiatory formulas and the like 579 Rivalry for the favour and presence of the gods 579 Protection of the gods in misfortune, against enemies, &c. : Getting over misfortune 580 Protection and help in general 580 Against plot*, hostilities, and misfortune 580 Destruction of enemies 581 Contents : Part 2, Chapter 4 IX TAGE Prayers for long life, offspring, prosperity, and liberal patronage : Long life ' 581 Sons and servants 581 Goods and blessings in general 581 Wealth, especially in cattle and horses 581 Great or lasting fame 582 Liberal patronage 582 Figures of speech and Formulas 582 Various similes 583 Miscellaneous statements which have assumed a formulaic character . . . 583 Repetitions relating to the gods 584 Repetitions relating to one and the same god 584 Repetitions relating to different gods 585 Repetitions containing similes based on verses containing direct statements . . 586 Verses clearly transferred from one god to another 587 Three classes of repetitions relating to the gods 588 Class A : Repetitions relating to the same god or group of divinities (see below) . 588 Class B: Repetitions relating to two different gods or groups of divinities (p. xi) . 588 Class C : Repetitions relating to more than two divinities (p. xv) .... 588 Class A : Repetitions relating to the same god or group of divinities . . 589 Agni : General statement 589 Agni as burning or shining or consuming or pervading fire 589 Agni as mediator and messenger between men and gods 589 Agni as embodiment of the priesthood (Hotar, Rtvij, Purohita) .... 590 Agni as oblation-bearer and leader at the sacrifice 590 Agni in mythological and cosmic aspects 591 Agni as protector and enricher of men 591 Agni as recipient of praise and sacrifice 592 Indra : General statement 592 Indra as demiurge : Indra as slayer of Vrtra (Ahi) and releaser of the Waters 593 Indra as slayer of other demons and enemies 593 Indra’s other demiurgic or divine acts 593 Indra’s cosmic power and relation to other gods , 594 Indra’s warlike might . 594 Indra as chief consumer of Soma 595 Indra as protector and enricher of men 596 Indra as recipient of praise and sacrifice 596 Soma : General statement 597 Soma : ritual preparation of Soma : washing and cleaning 598 Soma: straining 598 Soma : pressing and flowing and clearing . 598 Soma and its admixtures 599 b [h.O.S. 20] X Contents : Part 2 , Chapter 4 A PAGE Soma and its vessels 600 Soma benefits Indra and other gods 600 Soma as protector and enricher of men ... 601 Soma’s divine and other qualities : Soma-worship 601 Acvins : General statement 602 A9vins : Their wonderful deeds . 602 Akins’ chariot 602 Afvins as protectors and enrichers of men 603 A^ins as recipients of praise and sacrifice 603 Usas 604 Maruts 604 Aditya-group : Mitra, Varuna, Aryaman, Aditi : General statement . . . 605 Adityas as upholders of the divine order, &c 605 Adityas as protectors and enrichers of men 605 Aditya-worship in general 606 Vicve Devah 606 Surya (Sura) and Savitar (Tvastar, Bhaga) 606 Rbhus 607 Vayu 607 Brhaspati 607 Rudra 607 Paijanya. 607 Visnu 607 Sarasvati 608 Vac 608 Tratar 608 Ahi Budhnya 608 Dadhikra 608 Devapatnyah 608 Pitarah 608 TJcijah 608 Oravan or Press-stones 608 Apri-divinities 608 Danastuti or praise of liberality to the priests 608 Dissimilar dual gods (Devatadvandvas) : General statement .... 609 Indra and Agni . 609 Indra and Viiyu 609 Indra and Varuna 609 Indra and Visnu 609 Indra and Brhaspati or Brahmanaspati 609 Dyava-PrthivI or Dyava-BhumI ... 610 Prthivi and Antariksa 610 Contents: Part 2, Chapter 4 B xi PAGE Class B : Repetitions relating to two different gods or groups of divinities 610 Agni with other divinities : Agni and Indra 611 Agni and Soma 612 Agni and Brhaspati or Brahmanaspati 613 Agni and Maruts 613 Agni and Vayu 613 Agni and A^ins 613 Agni and Surya or Savitar 613 Agni and Tvastar 613 Agni and Visnu 613 Agni and Pusan 613 Agni and Usas 614 Agni and Varuna 614 Agni and Yama 614 Agni and Aparii Napat 614 Agni and Manyu 614 Agni and Sarasvati 614 Agni and Ratri 614 Agni and Vijve Devah 614 Agni and dissimilar dual gods 614 Agni in miscellaneous relations 614 Indra with other divinities : Indra and Agni 615 Indra and Soma 615 Indra and Maruts 616 Indra and Afvins 616 Indra and Vayu 616 Indra and Rudra 617 Indra and Brhaspati or Brahmanaspati 617 Indra and Parjanya 617 Indra and Surya or Savitar 617 Indra and Tvastar 617 Indra and Visnu 617 Indra and Pusan 617 Indra and Usas 617 Indra and Varuna 617 Indra and Vena 617 Indra and Manyu 617 Indra and Sarasvati 617 Indra and Apva 618 Indra and Rodasi 618 Indra and Vifve Devah 618 Indra and dissimilar dual gods 618 Indra in miscellaneous relations . . 618 b 2 [h.O.S 20.] xii Contents : Part 2 , Chapter 4 B PAGE Soma with other divinities : Soma and Agni 618 Soma and Indra 618 Soma and Brahmanaspati 618 Soma and Vena 618 Soma and Savitar 618 Soma and Pusan 618 Soma and Usas 619 Soma and Sarasvant 619 Soma and Varuna 619 Soma and Sadasaspati 619 Soma and Anumati 619 Soma and Vifve Devah 619 Soma and dissimilar dual gods 619 Soma in miscellaneous relations 619 Acvins with other divinities : Afvins and Agni 619 Aijvins and Indra 619 A^ins and Usas 619 Ajvins and Surya 620 A9vins and Sarasvatl 620 A9vins and Adityas 620 A9vins and Maruts 620 A9vins and dissimilar dual gods 620 Adityas with other divinities : Varuna and other gods ... 621 Mitra and Vanina and other gods 621 Adityas and other gods 621 Maruts with other divinities : Maruts and Agni 622 Maruts and Indra 622 Maruts and A9vins 622 Maruts and Adityas 622 Maruts and V^ve Devah 622 Maruts and Rbhus 622 Maruts and Brahmanaspati 622 Maruts and Viiyu 622 Maruts and dissimilar dual gods 622 Maruts in miscellaneous relations 622 Usas with other divinities: Usas and Agni 622 Usas and Indra 622 Usas and Soma 622 Usas and A9vins 622 U?as and Surya or Savitar 622 Contents : Part 2, Chapter 4 B Usas and Saras vati Usas and Vac Usas in miscellaneous relations Vicve Devah with other divinities: Vi9ve Devah and Agni Vifve Devah and Indra Vicve Devah and Soma . Vifve Devah and Varuna Vifve Devah and Adityas Vi9ve Devah and Maruts Vi9ve Devah and Pitarah Vi9ve Devah and dissimilar dual gods . Surya or Savitar or Tvastar with other divinities : Surya and Savitar, and Agni Tvastar and Agni Surya and Savitar, and Indra Tvastar and Indra Savitar and Soma Surya and A9vins Surya and Savitar, and Usas Surya and Parjanya Surya and Savitar in miscellaneous relations . Rbhus with other divinities : Rbhus and Maruts Rbhus in miscellaneous relations . Vayu with other divinities : Vayu and Indra Vayu and Adityas Vayu and Maruts Vayu and Sindhu Vayu and Indra-Vayu Brhaspati or Brahmanaspati with other divinities : Brhaspati and Agni Brhaspati and Indra Brahmanaspati and Soma Brahmanaspati and Maruts Brhaspati and Rudra Brhaspati (Brahmanaspati) and Sarasvatl Brhaspati and Aponaptar Brahmanaspati and Indra-Agni .... Rudra with other divinities : Rudra and Indra Rudra and Brhaspati Parjanya with other divinities : Parjanya and Indra xiii PAGE 623 623 623 623 623 623 623 623 623 623 623 623 623 623 623 623 623 623 623 624 624 624 624 624 624 624 624 624 624 624 624 624 624 624 624 625 625 625 xiv Contents : Part 2, Chapter 4 B PAGE Parjanya and Surya 625 Parjanya and Vifvakarman 625 Visnu with other divinities : Visnu and Agni . 625 Visnu and Indra 625 Pusan with other divinities : Pusan and Agni 625 Pusan and Indra 625 Pusan and Soma 625 Pusan and Indra-Agni 625 Sarasvati (Sarasvant) with other divinities : Saras vati and Agni 625 Sarasvati and Indra 625 Sarasvant and Soma 625 Sarasvati and Afvins 625 Sarasvati and Usas 625 Sarasvati (Sindhu) and Vayu 625 Sarasvati and Brahmanaspati 625 Vac with other divinities : Vac and Usas 625 Vac and Vijvakarman 625 Vena with other divinities : Vena and Indra 626 Vena and Soma 626 Vi9vakarman with other divinities : Vifvakarman and Parjanya 626 Vifvakarman and Vac 626 Manyu with other divinities : Manyu and Agni 626 Manyu and Indra 626 Pitarah with other divinities : Pitarah and Vifve Devah 626 Pitarah and Indra-Agni 626 Gravanah or Gravanau with other divinities : Gravanah and Adityas 626 Gravanah and Rbhus 626 Gravanau and Usasanakta 626 Apr! divinities in miscellaneous relations 626 Danastuti in miscellaneous relations 627 Minor divinities in miscellaneous relations 627 Dual gods in relation to other dual gods and also to plural gods . . . 628 Afvins 629 Indra-Agni 629 Contents : Part 2, Chapters 4 and 5 xv PAGE Indra-Vayu .............. 629 Indra-Varuna 630 Indra Bfhaspati or Indra-Brahmanaspati 630 Indra-Soma 630 Indra-Visnu .............. 630 Indra-Pusan .............. 630 Indra's Harl .............. 630 Agni-Soma 630 Agni-Parjanya 630 Soma- Pusan 630 Mitra-Varuna 630 Usasa-Nakta 631 Dyava-PrthivI 631 Daivya Hotara 631 Gravanau 631 Class C : Repetitions relating to more than two divinities .... 631 General statement 631 List of correspondences 632 Chapter 5 : Relative chronology of books and minor collections . . 634 Untrustworthiness of Anukramanl-statements shown by the repetitions . . . 634 Critical value of author-names mentioned in the verses themselves .... 634 Intrinsic criteria of relative dates 635 How these criteria determine the relative dates of single hymns .... 635 Examination of such hymns for other indications of relative date .... 636 Massing of repetitions as a criterion of the relative date of mandalas or other collections 638 Massing of repetitions in the eighth book 639 Superior or inferior quality of repetitions in a given collection as a criterion of date 640 Application of this criterion to the Valakhilya hymns ...... 640 Application thereof to the eighth book as a whole shows its lateness . . . 641 Sporadic instances in which the eighth book shows superior verses .... 642 Quality of repetitions in the strophic collections of the first book (hymns 1-50) 643 The ninth or Pavamana Soma book ......... 644 Quality of the repetitions in the family-books ...... . 644 The second mandala 644 The third mandala 645 The fourth mandala 645 The fifth mandala 645 The sixth mandala 645 The seventh mandala 646 Conclusions as to the family books as a whole 646 On the relations of the third and seventh mandalas 646 The remaining groups of the first mandala (hymns 51-191) 647 The tenth mandala 649 XVI Contents : Part 3 PAGE Part 3 : Lists and Indexes 651 1. List of repeated cadences of Rig-Veda lines 653 Alphabetized reversely, that is, according to the sequence of the letters of each line taking those letters in a reversed order 654 2. List of lines repeated in one and the same hymn 675 3. List of refrain-lines 677 4. Index of Sanskrit words 681 5. Index of subjects 684 Additions and corrections 689 PREFACE The present work is a natural — one might say inevitable — outgrowth of my Vedic Concordance. I saw this early in the day when, soon after the publication of that work, I printed my article, ‘On Certain Work in continuance of the Yedic Concordance JAOS. xxix. 286 ff. In that article I outlined three principal tasks: 1. The treatment of the Rig-Veda Repetitions. 2. A Reverse Concordance. 3. The treatment of the Vedic Variants. Indeed, each of these three works is now well under way. The present work speaks for itself. The Reverse Con- cordance, though not ready for publication, exists in material form, and has played a very important part in supplying the materials for the Rig-Veda Repetitions. A brief account of its present status is printed on pp. 1-3 of this book. As regards the third work outlined in the above-mentioned article, namely the discussion of the Mantra-variants (some fifty thousand) from the point of view of grammar and lexicon and style, — I may refer, in the first place, to my two articles, ‘ On Instability in the use of Moods in earliest Sanskrit’, American Journal of Philology, xxxiii. 1 ff. ; and, ‘ On the variable Position of the Finite Verb in oldest Sanskrit’, Indogermanische Forschungen, xxxi. 156 ff. Sanskrit scholars will be even more interested in the following : I have associated myself in the interest of this last-mentioned work with my former pupil, Professor Franklin Edgerton of the University of Pennsylvania, and between the two of us we have now in hand a first draft of a work entitled Vedic Variants, a systematic presentation and critical discussion of the variant readings of the Vedic texts. We hope to begin to publish this soon, part after part, beginning with a first book on the Phonetic Variants, and continuing with parts on Noun-Formation; Noun-Inflexion; Verb-Inflexion; Variation in Pro- nouns and Particles; Order of Words; Lexical Interchange; Metrical Variations ; Interrelation of the Vedic Schools ; and so on. At all times students of the Rig-Veda have been aware of the existence in that text of verse, distich, and stanza repetitions. Aside from casual observations, Ludwig, Der Rig-Veda, iii. 95 ff. ; and Aufrecht, XV111 Preface Preface to his second edition of the text of the Rig -Veda, pp. xii if., have listed considerable hatches of correspondences. But probably neither of these scholars fully realized the extent of the repetitions (see p. 4, below). The real significance of these correspondences lies in their large number, and (on the whole) even distribution through the text. No theory as to the character and origin of the RV. can pass by these facts. They mark the entire Mantra-literature as, in a sense, epigonal, and they forbid pungent theories about profound differences between the family books, their authors, and their geographical prove- nience. E. g., the third book of the Vigvamitras and the seventh book of the Vasisthas, despite their traditional cleavage (p. 646), share not only the apri-stanzas 3.4.8-11 = 7.2.8—11, but will be found in general to participate in about as many repetitions as any two other family books. On the other hand text-critical and hermeneutic help is in proportion to the frequency of the repetitions. I believe that the Rig-Veda will be explained ultimately: every time a fish dies (dhiya-dhiya, TS. 2.6.6.1) some good point is made in the text, interpretation, grammar, or metre of the Veda. The kind and attentive reader will find that the under- standing of the RV. has been eased at many points through approach by the road of the repetitions. I might point out in particular that hitherto no treatise on Vedic metre has had the benefit of the consider- able mass of repeated passages which are varied as they are repeated ; see Part 2, chapter 2. I have endeavoured to extract from the repetitions their full signifi- cance. In this domain judgement is necessarily subjective ; there is room for difference of opinion, and scope for sharper eyes than mine. On the whole I have erred, I am sure, on the side of too little, rather than on the side of too much. Especially as regards the partial correspondences (p. 10), there are not a few passages which may in the future yield important information. What, e. g., is the full significance of the cosmo-mythic repetition: 7.33.7b, tisrah praja arya jyotiragrah: 7.101. la, tisro vacah pra vada jyotiragrah ; why this imitativeness in the words tisrah and jyotiragrah with themes otherwise so uncongenial? Or, let the reader judge for himself in just what way the meaning of the words mahas and tvacas is cleared up by their interchange in the item: 4.1. ll1', maho budhne rajaso asya yonau: 4.17.14, tvaco budhne rajaso asya yonau. Or, again, note the two brahmodya passages : 1.164.3®, sapta svasaro abhi sam navante: 10.71.3d, tarn sapta rebha abhi sam navante. It is scarcely necessary to recommend to the attention of serious Preface xix ' students of the Mantras the repetitions which are now so conveniently open to the eye. They are of interest not only for the direct explanation of many a given passage, but also for a critical comparison and estimate of the repeated matter in a given hymn as confronted with that of all the other hymns which are concerned in these repetitions. These are considerably more important than the variants in other Yedic texts, interesting as these are for the history of schools, the development of the language, and the later growth of Brahmanical ideas. As in the case of the Vedic Concordance I cannot conclude this Preface without grateful reference to the Editor and to the Founder of this Series. Professor Lanman has again brought to bear his great editorial talents and his sound scholarship on the production of this work. Its externals, or what may be called the mechanics, were un- usually intricate and difficult. If its form is convenient, its arrange- ment clear, if, in fact, the book is thoroughly usable, — all that is in proportion to his redactorial skill. Needless to say, he has also aided me much by his learning and critical acumen in many matters that concern the inner quality of the work. I can only regret that he could not help me systematically in the difficult and long-drawn task of reading the proofs ; hence, perhaps, the unduly large list of corrections at the end of the work. Once more it is my good fortune to express my high appreciation of the Founder, as he may be very properly called, of this Series, the late Mr. Henry C. Warren, of Cambridge, Massachusetts. Himself a scholar whose understanding of Buddhism and the Pali language is not excelled to this day, he has imparted to his interest in Indological Studies a life far beyond his all too short allotted time. The provision he left behind him has made it possible to publish in dignified style such a work as this, remote though it be from the beaten tracks of ordinary commercial enterprise and of average human interest. Maurice Bloomfield. Johns Hopkins University, Baltimore, May, 1916. ABBREVIATIONS The abbreviations for the names of Vedic texts are the same as those used in Bloomfield’s Vedic Concordance, and duly explained in the Introduction to that work, pages xvi-xxii. AJPh. American Journal of Philology. Arnold, VM. E. Vernon Arnold, Vedic Metre. Bergaigne. Abel Bergaigne, La Religion v6dique d’apres les Hymnes du Rigveda. Bezz. Beitr. Beitrage zur Kunde der indogermanischen Sprachen. Concordance. M. Bloomfield, A Vedic Concordance. Grassmann. Hermann Grassmann, Rig-Veda iibersetzt. GSAI. Giornale della Societa Asiatics Italians. Hillebrandt, Ved. Myth. Alfred Hillebrandt, Vedische Mythologie. IF. Indogermanische Forschungen. Ind. Stud. Albrecht Weber’s Indisclie Studien. JA. Journal Asiatique. JAOS. Journal of the American Oriental Society. KZ. Kuhn’s Zeitschrift fur vergleichende Sprachforschung. Ludwig. Alfred Ludwig, Der Rigveda oder die heiligen Hymnen der Brahmana. Ludwig, Die neuesten Arbeiten. A. Ludwig, Ueber die neuesten Arbeiten auf dem Gebiete der Rigveda-Forschung. Ludwig, Kritik. A. Ludwig, Uber die Kritik des Rigveda-Textes. Ludwig, Ueber Methode. A. Ludwig, Ueber Methode bei Interpretation des Rigveda. Muir, OST. J. Muir, Original Sanskrit Texts on the origin and history of the people of India. Oldenberg, Prol. Hermann Oldenberg, Die Hymnen des Rigveda. Metrische und text- geschiclitliche Prolegomena. Oldenberg, RV. Noten. Hermann Oldenberg, Rigveda. Textgeschichtliche und exegetische Noten. Pet. Lex. Sanskril-Worterbuch, herausgegeben von der Kaiserlichen Akademie der Wissen- schaften (St. Petersburg). SBAW. Sitzungsberichte der KOniglich-Preussischen Akademie der Wissenschaften. SBE. Sacred Books of the East. Ved. Stud. Richard Pischel und Karl F. Geldner, Vedische Studien. WZKM. Wiener Zeitschrift fur die Kunde des Morgenlandes. ZDMG. Zeitschrift der Deutschen Morgenlandischen Gesellschaft. INTRODUCTION Parts and sources and purpose of the present work The three main parts of the present work. — The bulk of this work naturally divides itself into three Grand Divisions, or Parts. Part i makes up what may be called the main body of the work and is occasionally so called. It presents in full quotation, in the order of the RV. text, the stanzas which are or contain repetitions. The repeated padas of each stanza are indicated by simple distinctions of type (see the Explanations for Part i, at p. 27). Each stanza is headed by the reports of Katyayana’s Sarvanukramanl, as to the author and divinity of a given stanza. And each item of repetition is accompanied by explanative, critical, and historical remarks, with special reference to the relative chronology of the repeated materials. Where it seemed profitable the stanzas are translated. Part 2 is explanatory and analytic. It is divided into five chapters. Chapter 1 disposes of the repeated passages in ten classes, according to their extent, their grouping, and their inter-relations (for details see the opening paragraph of that chapter). Chapter 2 deals with the metrical variations resulting from additions, subtractions, and verbal changes in repeated verse lines. Chapter 3 deals with the lexical and grammatical variations in repeated padas. Chapter 4 deals with the themes (divinities, objects, and ideas) of the repetitions. Chapter 5 contains a discussion, in the light of the repetitions, of the relative chronology of the books (mandalas) and miner collections, as assigned by tradition to particular authors or families of authors. Part 3, the concluding part, consists of three Appendixes. The first gives a list of repeated cadences (see p. xvi) ; the second one gives a list of the lines repeated in one and the same hymn (see p. xvi) ; and the third gives a list of the refrain lines. This is followed by an Index of Words and an Index of Subjects. Sources of the material for the present work : the Vedic Concordance and the Reverse Concordance. — The materials elaborated in this work are derived in the first place, and also in the main, from my Vedic Concordance, published in 1906, as volume X of the present series. The Concordance includes, of course, all RV. verses, arranged alphabetically from the beginning, so that it was no difficult task to extract from it all word-for-word repetitions, 1 [a.o.s. 20] 2 Introduction and also all partial repetitions whose opening syllables are identical. But it appeared very shortly that a work of this kind stands in need of a much broader basis. A great many partial Vedic repetitions are not brought out by alphabetic arrangement from the beginning, because the opening syllables of the verses concerned are changed more or less. Thus, e. g., no less than twenty -five octosyllabic (dimeter) padas reappear, with an increase of four syllables at the beginning, as dodeeasyllabic (trimeter) padas, e. g. : sasahyama prtanyatah 8.40.7 indratvotah sasahyama prtanyatah 1.132.1.1 Or, very frequently a single word 2 3 at the beginning is changed, for one reason or another, at times for reasons of the profoundest interest to Yedic criticism or interpretation, thus: a9atrur indra janusa sanad asi 1.102.8 anapir indra janusa sanad asi 8.21.13. There appeared to be but one way to reach these materials, and that promised to be, in some respects, of even greater interest than the word-for-word repetitions, namely, the compilation of a Reverse Concordance. The scheme of such a work had suggested itself to my mind on more general grounds,’ and I had announced the plan of it briefly in 1908. 4 The rough draft of a Reverse Concordance occupied a great deal of my time during recent years, and was completed in 19 11 up to the point where it could be relied upon to yield the information desired for the present purpose. About one-third of our material is derived from it : without it our work would have been very fragmentary indeed. These two Concordances may be relied upon to yield practically all the repeated verses in the broadest construction of that word. It will not often come to pass that a repeated verse will be disguised by changes both at the beginning and at the end. Such disguise is theoretically possible, but practically so rare as to be negligible. A repetition or two of this sort is incorporated in the present work, thus : agnl ratho na vedyali 8.19.8 agnim ratharh na vedyam 8.84.1. Negative assurance that such cases do not, after all, occur with considerable frequency could only be obtained by a word-for-word concordance, a task which lies beyond the scope of the present essay (see p. 3, bottom). The Reverse Concordance and its present status. — As just now men- tioned, I have in my possession a rough draft of a Reverse Concordance. The uncertainties and vicissitudes of human affairs may prevent me from elaborating it for publication. I desire therefore to describe this work, in order that 1 See p. vii, Class B 6. 4 ‘On certain work in continuance of the 8 Or even a single letter, as in the case of Vedic Concordance,’ JAOS. xxix, pp. 286 flf., verses beginning pra no and pra no. more particularly p. 288. 3 Vedic Concordance, pp. xb and xiv\ Introduction 3 Indologists may know both that it exists, and how far onward it has been carried. In its present state the work is altogether provisional. It includes precisely the materials incorporated in the published Concordance, no more and no less. Two copies of the published Concordance were cut up into the separate items contained therein ; one, so as to collect the items on the odd pages (i, 3, 5, &c.); the other, so as to collect the items on the even pages (2, 4, 6, &c.). The entire mass was then subjected to a reverse alphabetical arrangement, and pasted upon sheets in that arrangement. The work, so far, has not been elaborated beyond that stage. I will merely say that even in its provisional state it forms an inexhaustible mine of information on almost every imaginable question of Vedic language and literature. I am quite certain that, sooner or later, the work will commend itself for elaboration and publication either by myself, or some other scholar. In the meantime I shall be pleased to impart information derivable from it to any one who may desire. The purpose of the present work. — The aim of the present essay is to throw some light on the way in which the poets of the Rig-Veda exercised their art in the extant traditional collection, by studying the manner and extent to which they borrowed from one another, imitated one another, and, as it were, stood one upon the shoulders of another. There can be no doubt that they depend upon one another for many substantial units of verse-line (pada), distich, or stanza ; that such dependence cannot be imagined to have taken place without a considerable degree of consciousness ; and that it opei-ates to such an extent as to assimilate the entire body of hymns to a surprising degree. It will, I am confident, appear that the juxtaposition of these inter- dependent stanzas and parts of stanzas, when reinforced by pertinent comment, will not only put many questions as to the relative date of parts of the Yeda in a new light, but will also yield many a useful hint as to the exegesis of the Vedic texts. Character and scope and bearing of Rig-Veda repetitions Most general statements as to the repetitions. — Repetitions in the Rig-Veda range all the way from hymns which are made, intentionally, in the image of one another, as is the case in some of the so-called Valakhilya hymns, to mere collocations of two or more consecutive words. Between these two extremes lie repetitions of the same consecutive group of stanzas ; repetitions of single stanzas ; repetitions of three verses or padas of a stanza ; repetitions of distichs ; and repetitions of single verses or padas. With the class of repetitions involving merely consecutive words or set phrases, which do not result in the identity or close similarity of at least one single line, the pi’esent essay does not deal either systematically or fully. That would mean an entirely different work from the one here contemplated, namely, a word-for-word Concordance, 4 Introduction written out in full.1 The imitative moment in mere groups of words is, as a rule, faint, accidental, and more or less unconscious, because such collocations tend to assume the nature of set phrases. This phase of repetition is touched upon incidentally, provisionally, and yet perhaps sufficiently, in a paragraph or two, below, pp. 8 ff. As regards cadences, moreover, it is brought to light completely by the List of repeated cadences (Appendix I). It will be seen there that repetition of two or more consecutive words is an established feature of Rig-Vedic composition, as it is indeed of Yedic composition in general. Mass or amount of the repeated material.— Groups of stanzas, stanzas, parts of stanzas, distichs, and single verses amounting perhaps to a total of no less than 2,400 padas repeated entirely or partially, constitute the material with which the present treatise has to deal. These padas are repeated on the average nearly times, making a total of about 6,000 padas. This count does not include such as are repeated, for one reason or another, in the same hymn. Of these there are about 60, making a total of about 120, exclusive of the numerous rhetorical concatenations which often result in padas so much alike as to amount almost to identity ; see the next paragraph. Still more, a fortiori , this does not include refrain padas which abound in the Rig-Veda. Of these there are just about 150, repeated a total of about 1,000 times ; see p. xvi. Thus the total of repeated padas in the Rig-Veda, if we include close catenary imitation, is likely to concern not much less than 8,000 lines, that is to say , perhaps not less than one-fifth of the entire Rig-Veda collection.'1 The nature of partial repetitions. — These borrowings, as between different parts of the Rig-Veda, are not by any means restricted to mere mechanical word-for-word repetitions. A given verse unit may, indeed, appear in exactly the same form in two or more places. But quite as frequently it appears in a more or less changed form. Very frequently a line or stanza is changed to suit a different theme, especially a different divinity. The different metres in which the hymns of the Rig-Veda are composed may impose changes in repeated verse-lines. Especially the transfer of a line in the shorter (anustubh-gayatrl) metre to the longer (tristubh-jagatl) metre, or vice versa, involves extension or curtailment. Or, the same line may appear, with slight obligatory changes, both as tristubh and jagatl. Every imaginable form of change is theoretically possible when it comes to transferring verse or stanza from its original place to a new connexion. The poets rejoice in the utmost freedom in this respect. They curtail and extend, they vary and adapt previously existent verse units to suit their needs and their fancies. All 1 Cf. A. GuCrinot, Journal Asiatique (1907), 10. x. 585 ff. a The RV. is usually estimated at about 40,100 padas. According to the Caranavyuha, and the scholiast to 9®-, the number of stanzas in the RV. (Vaskala^akha) is 10,581, or (9&kala 9&kha) 10,417 ; see Weber, Indische Studien, iii. 256; x. 133, note. As to the number of words contained in the Rig-Veda, see the extract from the commentary to the Caranavyuha, given in Oldenberg’s Prolego- mena, pp. 5i4ff. In general see ibid., pp. 488ff. Introduction 5 these variations bring with them the opportunity for critical and historical study of the Rig-Veda texts and their inter-relations in the redaction. Especially the question of the relative date of the repeated materials comes to the fore constantly when the same metrical unit is found in two or more different forms or different connexions. The fruitfulness of this study will depend upon the degree of insight and sanity with which it is carried out. It is not likely that this will be done so as entirely to eliminate errors of judgement. The nature of concatenation or catenary structure. — I have stated above that the very large total of verse repetitions is due in part to the frequent catenary structure of the stanzas. Concatenation is a favourite rhetorical device of the Vedic authors from the beginning of our tradition. In the succession of the stanzas in a given hymn an expression, statement, or motif in one given stanza is taken up anew in the next stanza, in such a way as to modify, develop, or carry on further the events depicted, or the thought expressed in the first stanza. The practice at times runs through an entire hymn as in RV. 10.84, or is employed very artificially as in AY. 6.42.1, 2 ; 13. x. 46-48; ApQ. 14. 33.6. 1 I have treated briefly this feature of mantra composition as far as the AY. is concerned in my Prolegomena to that Veda.2 I would add here that this phase of rhetoric is known also in the Avesta ; 3 and that it is especially analogous to so-called parallelism in Hebrew poetry.4 Concatenation involves a very considerable amount of repetition, rarely word for word ; sometimes almost word for word ; and, very frequently, shading off to some sort of similarity in the general tenor of the two passages, accompanied by the verbatim repetition of one or two words.6 Illustrative examples of catenary structure. — These are in strictness beyond the proper scope of the present treatise, and are given only because they show in what manner concatenation contributes to the mass of repeated materials : tarn ma sam srja varcasa 1.23.23d sam magne varcasa srja 1.23.24“ rnor aksam na cakryoh 1.30. 14d rnor aksam na (jacibhih 1.30. 15d tvam na indra raya parlnasa 1.129.9“ tvam na indra raya tarusasa 1.129.10“ tasminn a tasthur bhuvanani vi^va 1.164. 13b tasminn arpita bhuvanani visva I.i64.i4d 1 Cf. also RV. 10.98.2, 3 and several of the stanzas that follow. Curiously RV. 10.1.7' concatenates with 10.2.1“. Since the theme (Agni) and the authorship of the two hymns are the same, the relation is, presumably, accidental. 2 See The Atharva-Veda (Indo-Aryan En- cyclopedia), § 40. In note 1 5 to that paragraph a list of illustrative Atharvan passages is cited. 3 E. g. Yasna 9.17,18; 45.1,2; Yasht 5.62,63; 10.82. 4 See David H. Muller, Die Propheten in ihrer urspriingliclien Form, e. g. pp. 180, 183, 186. 5 RV. 2.1 1 illustrates well this latter class of vaguer catenary structure. 6 Introduction tve deva havir adanty ahutam 2.1.13d asa deva havir adanty ahutam 2.1. 14b mandro vi^vani kavyani vidvan 3. 1. t 7b agnir vi§vani kavyani vidvan 3.1. i8d ya jagrvir vidathe ^asyamana 3.39.1° vi jagrvir vidathe 9asyamana 3.39. 2b ekam vicakra camasam caturdha 4-35.2d vy akrnota camasam caturdha 4.35.3* rayirii divo duhitaro vibhatih 4.51.10s tad vo divo duhitaro vibhatih 4.51.11s yad im somasah susuta amandan 5.3o.iod yad im soma babhrudhuta amandan 5.30.11* sa vy ucha sahiyasi 5.79.2® yo vy auchah sahiyasi 5.79.3° dhibliir viprah pramatim iehamanah 7.93. 3b girbliir viprah pramatim iehamanah 7.93.4s addhii deva mahaii asi 8. ioi.nd satra deva mahan asi 8.101.12 abhi tyam madyam madam 9.6.2s abhi tyam purvyam madam 9.6.3s yat te pavitram arcisi 9.67.23s yat te pavitram arcivat 9.67.24s tvam vipro abhavo ’ngirastamah 9.107.6® tvam kavir abhavo devavitamah 9.107.7® tebhih somabhi raksa nah 9.114.3d tena somabhi raksa nah 9. ii4-4b vi cid vrlieva rathyeva cakra 10. 10. 7d tena vi vrlia rathyeva cakra io.io.8d athem enam pra hinutat pitrbhyah 10. i6.id athem enam pari dattat pitrbhyah io.i6.2b yas te drapsa skandati yas te ai^uh 10.17.12s yas te drapsa skanno yas te ah^uh 10.17.13s vi9ved eta savana tutuma krse io.50-5d eta vi9va savana tutuma krse 10.50.6s atha deva dadliire havyaviiham 10.52.3d mam deva dadliire havyaviiham 10.52.4s te agneh pari jajnire 10.62. 5d ye agneli pari jajiiire 10.62.6s sarasvati saha dhibhih puramdhya 10.65.13d v^ve devah saha dhibhih puramdhya 10.65.14s dadhiimi te dyumatim vacam asan io-98.2d asme dliehi dyumatim vacam asan 10.98.3s utiiprnan marditaram na vindate 10.117. id uto cit sa marditaram na vindate io.ii7.2d apa9yam tva manasa cekitanam 10.183.1s apa9yam tvii manasa didliyanam 10.183.2s Introduction 7 Additional instances of this practice may be found in the following passages : 1. 11. 6, 7 3-32-9, 10 7-4*-4, 5 9.64.25, 26 1.22.16, 17 3-39-1, 2 7.104.4, 5 9.67.19, 20 1. 24.12, 13 3-5 *• 7, 8 7. 104.15, 16 9-67-3I, 32 1.32. 1, 2 (cf. 3-5) 3-55-6, 7 8. 1 1.8, 9 9-72.4, 5 i-73-6, 7 4.17.6,7 8.17.8, 9 10.28. 10, 1 1 i-85-4,5 . CO 1^. CO 8.19.22, 23 10.30.7, 8 1.108.9, 10 5-1-5, 6 8.52.7, 8 10.35.1, 2 1. 109.7, 8 5- 5 2- 1 3, 1 4 8-59-4, 5 10.65.13, 14 1.174* i» 2 6.27.4,5 8.86.2, 3 10.90.8, 9 2. 10. 1, 2 6.42.2, 3 8.94.10, 1 1 10.96.6, 7 2.14.6, 7 6.52.5, 6 9.10.1, 2 10-135-1, 2 2.18.5, 6 7-17-3, 4 9-5°-4, 5 10.164.1, 2 Concatenated lines which differ only in the order of their words. — At times the concatenating padas consist of the same words rearranged in different order. This kind of change, on account of its extreme simplicity, carries with it an extra touch of rhetorical liveliness, as compared with the more ordinary forms of concatenation : sam agnir idliyate vrsa 3.27.13' vrso agnih sam idhyate 3. 27.14“ ahus te trim divi bandhanani 1.163.3d trlni ta ahur divi bandhanani 1.163.4* v^ved eta savana tutuma krse jo. 50.3d eta v^va savana tutuma krse 10.50.6“ tena caklpra rsayo manusyah 10. 130.3d caklpre tena rsayo manusyah 10.130.6“ A few correspondences of this sort occur also in hymns widely apart : see the paragraph on padas which contain the same or similar words differently arranged (Part 2, chapter 3, Class A 1). Repeated lines containing questions and answers. — Allied to this theme are questions and answers, both stated in full. The effect is again rhetorical, either that of liveliness of diction, or mysterious solemnity. Cf. the brahmodya questions and answers at the a?vamedha sacrifice VS. 23.9 ff., et al. ; also RV. 5.44.14 and 15; 6.9.2 and 3; AV. 10.2.22,23. Thus: katham rasaya atarah payahsi io.io8.id tatha rasaya ataram payahsi io.io8.2d kas te jamir jananam 1.75.3® tvam jamir jananam 1.75.4® indrah kim asya sakhye cakara 6.27. ib indrah sad asya sakhye cakara 6.2 7.2b ko no mahya aditaye punar dat 1.24.1° sa no mahya aditaye punar dat 1.24.2° kam svid garbham prathamam dadhra apah 10.82.5° tarn id garbham prathamam dadhra apah 10.82. 6* 8 Introduction The two opening stanzas of 6.27 are made up entirely of a chain of question and answer lines whose obvious aim is to narrate in a sort of ballad style some particular events in which Indra figures as the hero : kim asya made kim v asya pltav indrah kim asya sakliye cakara, rana va ye nisadi kim te asya pura vividre kim u nutanasah. sad asya made sad v asya pitav indrah sad asya sakhye cakara, rana va ye nisadi sat te asya pura vividre sad u nutanasah. A similar ballad touch links the stanzas 5.44.14, 15 : yo jagara tarn rcah kamayante yo jagara tam u samani yanti, yo jagara tam ayam soma aha tavaham asmi sakhye nyokah. agnir jagara tam rcah kamayante agnir jagara tam u samani yanti, agnir jagara tam ayam soma aha tavaham asmi sakhye nyokah. Concatenation of entire distichs. — Concatenation may extend to an entire distich by carrying the parallelism beyond the limits of the single pada. The parallelism in such cases is, as a rule, less well sustained, presumably because the result would be too monotonous. The following instances are those of distichs more or less under the influence of this habit : yah pavamanir adhyety rsibhih sambhrtam rasam 9.67. 3iab pavamanir yo adhyety rsibhih sambhrtam rasam 9.67. 32ab amlva yas te garbham durnama yonim &9aye io.i62.icd yas te garbham amlva durnama yonim ajaye 10. 162. 2ab hiranyapanim utaye savitaram upa hvaye i.2 2.5*b apam napatam avase savitaram upa stuhi i.2 2.6ab a bharatam ^iksatam vajrabahu asman indragnl avatam 9aclbhih i.io9.7ab puramdara 9iksatam vajrahastasman indragnl avatam bharesu i.i09.Sab adveso no maruto gatum etana 9rota havam jaritur evayamarut 5-87.8ab ganta no yajnam yajiiiyah su9ami 9rota havam araksa evayamarut 5. 87-9ab a no gavyebhir a9vyaih sahasrair upa gachatam 8.73.i4ab ma no gavyebhir a9vyaih sahasrebhir ati khyatam 8. 73. 1 5ab esa divam vi dhavati tiro rajansi dharaya 9.3-7ab esa divam vy asarat tiro rajansy asprtah 9.3.8ab The phenomenon gradually fades out into such relation as appears in 4. 20. iab, 2ab ; or 8.26.21, 22. I need hardly say that my treatment here of this theme of concatenation is a mere sketch, which, I hope, may point the way for some younger scholar to a thorough investigation of this feature of the Rig-Veda. I am sure that it will prove valuable not only for the rhetoric, but also for the criticism and interpretation of that Veda. Boundary between repetitions and similarities an ill-defined one. — In another way also, the boundary line which separates the repeated verses of the Veda from the rest of the mass is not absolute, and cannot easily be drawn, even for practical purposes. Since repeated padas, in the sense which is given to the phrase in this book, are not always perfectly identical in their Introduction ft wording, it follows that the differences in the wording of two similar verses may outweigh their similarities. Or, put in another way, the question may arise whether the similarities in wording or structure of certain verses entitle them to be treated as repeated verses. For such similarities may, on the one hand, be due merely to the homogeneous character of a closely related body of semi-technical literary products, such as make up the Samhita of the RV. On the other hand, they may fade to a point where dissimilarity overrides similarity. It has therefore not always been easy to decide what to include or what to keep out. I have been guided, to some extent, by the intrinsic importance of the similarities in deciding what to regard as repeated padas, in distinction from mere accidental agglomerations of similar words. Nevertheless the theme becomes elusive at certain points: it frays, so to speak, at the edge. Thus we have the frequent expression covering the space of a pada, yajamanaya sunvate 5.26.5 ; 8.14.3; 17-IO> 10. 175.4. Unimportant, formulaic, and hap-hazard as is this expression, it is entitled in our plan to the full dignity of a repeated pada, if for no other reason, because it is a metrical unit of the sort we engaged to collect and discuss. But the same expression occurs at the end of several heterogeneous lines, as a more or less accidental cadence, to wit : bhadra 9aktir yajamanaya sunvate 1.83.3 viijved aha yajamanaya sunvate 1.92.3 rjuyate yajamanaya sunvate to. 100. 3 supravye yajamanaya sunvate 10. 125. 2. With these I have not dealt as repeated padas, content to state, once for all, under 5.26.5, that the expression yajamanaya sunvate is cadence in the above- mentioned four padas. Again the pada, yajamanasya sunvatah 6.54.6 ; 60.15, is not treated directly as a repetition of yajamanaya sunvate (5.26.5). A cross- reference from 5.26.5 to 6.54.6 is thought sufficient to secure the proper attention to this unimportant stylistic or metrical accident. Another set of examples, which illustrates well the instability of our criteria in this matter, brings up the question whether the following large group of padas is to be treated entirely, or in part, as repeated, or merely as similar material. It will be noted that all padas refer to Agni : agnih ijukrena 9ocisa 8.56(Val. 8). 5 agnim 9ukrena 90cisa 1.45.4 agne 9ukrena 90cisa 1.12.12 ; 8.44.14; 10.21.8 agnis tigmena 9 see under that item. I am inclined to surmise that the first hymns in these pairs were the patterns, the second the imitations. On the evidence of the opening stanzas of the first two pairs I am almost tempted to speak of 1 and 3 as India hymns, of 2 and 4 as £akra hymns ; $akra seems to me to be a clumsy refinement. Other imitative hymns: 4.13 and 14; 9.104 and 105. — Aside from the Valakhilyas, I know of but two pairs of hymns in the KV. which pattern one after the other, namely 4.13 and 14; and 9.104 and 105. The first pair is ascribed to Vamadeva Gautama, and addressed to Agni in the morning. The fifth and last stanza of each hymn is the same ; pada 2a is repeated with a variant. As for the rest there is considerable variation, but also marked similarity in corresponding padas, e.g. ia, ic, 4a. Without doubt the hymns are intentionally imitative, but I cannot say which was the model. Hymns 9.104 and 105, to Pavamana Soma, ascribed to Parvata and Narada, 1 See Max Muller, History of Ancient San- p. 35, note 21 ; Oldenberg, Prol. p. 494 if. skrit Literature, p. 220; SBE. xxxii, p. xlviff. ; 508. Theological explanations of the term Roth, ZurLitteratur und Geschichte des Veda, valakhilya in KB. 30.S ; £B. 8. 3. 4.1. P- 35 i Weber, Indische Literaturgeschichte®, 2 Cf. p. xv, line 13 from bottom. 14 Introduction are both in usnih metre. Each of the two hymns has six stanzas aggregating 1 8 padas ; each pada in one hymn is a mere variation of the corresponding pada in the other. It is as though the poet of the second hand had made a deliberate effort to change the wording of the first hand, without, however, really changing the sense. The parallelism of the two hymns is even more marked than that of the Valakhilya pairs. In PB. 13.x 1.3, 4 ; 14.5.4 both these hymns, too, are designated as Valakhilya. In the view of the Brahmana both the words khila and khilya have the meaning apparently of ‘ repetition Imitative strophes. — Similar to these doublet hymns are the two pragatlia strophes 8.87.1, 2 and 8.87.3,4, two small hymns in which the Priyamedhas address the Alvins in lines that differ but little from one another. In the first pair they offer hot milk (gharma), in the second soma. The repetition is either rhetorical or ritualistic, in any case intentional.1 Similarly there are two successive strophes of three stanzas each at the beginning of 6.15, i.e. 6.15.1-3 and 6.15.4-6, whose parallelism in metre, wording, and sense shows that they are two redactions of the same theme. The first stanzas of the two trcas are more particularly similar. Juxtaposition of hymns with similar openings. — This leads me to observe that successive hymns in the Rig -Veda occasionally betray parallelism, because the redactors inclined to place hymns with similar openings together. So the three hymns 9.31-33 each begin with pra somasah; 9.31. 1 and 9.32.1 have pra somasah . . . akramuh ; and 9.32.5 and 9.33.5 exhibit the intentional parallelism, abhi gavo anusata = abhi brahmlr anusata. Similarly 9.29. 1 opens with the words prasya dhara aksaran, which are repeated in 9.30.1 as pra dhara asya . . . aksaran. Again, 4.39.1 begins: i^urii dadhikram tam u nu stavama divas prthivya uta carkirama, uchantlr mam usasah sudayantu. This opening connects the hymn definitely with 4.40.1 whose first hemistich x-eads : dadhikravna id u nu carkirama vi9va in mam usasah sudayantu. In my Prolegomena to the Atharva-Veda I pointed out long ago that similar verbal resemblances, treated much more mechanically or stupidly, explain why certain Atliarvan hymns, whether related in theme or not, follow one after the other in the benighted arrangement of the 9aunaka school of that Veda.2 Consecutive imitative stanzas. — The Valakhilya mood, as we may call it, betrays itself occasionally in two successive stanzas of the same hymn. So the doublet stanzas 8.40.10 and 11 make a tolerably elaborate statement in 1 Cf. Oldenberg, Prol. p. 217. 2 See, The Atharva-Veda, p. 39. Introduction 15 almost the same words ; 1 the changes are rung in such a way as to alter the sense of some words without much changing their outer form or sound. The words in thick type are significant: 8.40.10 : tain 9i9lta suvrktibhis tvesam satvinam rgmiyam, uto uu cid ya ojasa 9usnasyandani bhedati jesat svarvatlr apo nabhantam anyake same. 8.40. 1 1 : tarii 9 i 9 1 1 a svadhvaram satyam satvanam rtviyam, uto nu cid ya ohata anda 9usnasya bhedaty ajaih svarvatlr apo nabhantam anyake same. Oldenberg in the second part of his KV. Noten, p. 108, remarks aptly that the two stanzas, belonging to an Indragnl hymn, yet addressed each to one god, refer respectively to Indra (st. 10) and Agni (st. u). But the real point of the repetition is in the heightened rhetorical effect of the ‘ prophetic ’ aorist ajaih in 11, as compared with the milder modal jesat in 10. 2 Similarly 2.12. 14 and 15 are little more than rhetorical uhas of the same theme.3 See also the following chains of stanzas all of which, more or less, are under the in- fluence of concatenation : 1. 8. 8-10 ; 15.7-9; 2I*I-4> 25. 7—9 ; 189.1-4; 2.18.4-7; 4.2.6-95 32.19-21; 7.63.1-4; 8.3.11,12; 35.4-6; 96.13-15; 19-21; 9.4. 1-3; 49.1-4; 10.17. 1 1— 1 3 ; 18.11,12; 19. 1-3; 94.1-4; 107.8-11; and cf. also the looser correspondences of certain stanzas of 10. 101 (Grassmann’s Translation, ii. 491). As a specimen of Valakhilya variation of single stanzas in the AV., see e.g. AV. 3.1. 1 and 3.2.1. Imitative stanzas scattered through the RV. — In all these cases there is some sort of juxtaposition of the parallel materials, showing that the redactors wTere awai-e of the fact, and put some sort of appraisal upon it. But there are quite a few pairs of single stanzas scattered through the collection in places far apart which exhibit the same sort of likeness. The original, which- ever that is, is not actually or entirely repeated, but it hovers before the mental eye of the later poet who is, possibly, hardly aware that he is reproducing rather than producing. E. g. : 1. 24. 1 : kasya nunam katamasyamrtanam manamahe caru devanam nama, ko no mahya aditaye punar dat pitaram ca dpjeyarii mataram ca. 10.64.1 : katha devanam katamasya yamani sumantu nama 9rnvatam manamahe, ko mrlati katamo no mayas karat katama utl abhy a vavarti. Similarly the following pair : 1.114.9 : upa te stoman pa9upa ivakaram rasva pitar marutarh sumnam asme, bliadra hi te sumatir mrlayattamatha vayam ava it te vrnlmahe. 10.127.8 : upa te ga ivakaraiii vrnlsva duhitar divah, ratri stomam na jigyuse. 1 Cf. Grassmann, i. 457 ; Hillebrandt, Ved. 2 Cf. the author, JAOS. xxix. 295. Myth. iii. 64, 300, note 3 ; Geldner, Ved. Stud. * Cf. Weber, Proc. Berlin Academy, 1900, iii. 64. p. 606. 16 Introduction The following little list is more or less of the same sort : 1.64.4 : 5*54- 11 1.114.2 : 2.33.13 1.143.8: 6.8.7 2. 11. 4, 5 : 10.148.2 2.18.7 : 7*29.2 3.19.2: 4.6.3 7.1 1.2: 10.70.3 8.100.2: 10.83.7 8.45.4, 5: 8.77.1, 2. 3.41.7: 7.31.4 3.62.10: 5.82.1 4.7.8: 4.8.4; 8.39.1 The resemblances in these pairs are, for the most part, only of the general order, and it is not my purpose to exhaust them. On the one hand they fade out into mere verbal correspondences ; on the other hand they are likely to repeat some one pada, so that they figure in the body of this work. Hymns parallel in structure (not wording) and with same final stanza. — There are, further, hymns constructed intentionally on parallel principles, in which the wording scarcely, or not at all, suggests the parallelism. The similarity of the hymns is rather in the number of the stanzas ; the metrical structure ; or the general theme. Above all they share the same final stanza. Thus in 8.36 and 37, two hymns of seven stanzas each, addressed to Indra1 by Qyava and 10.63 and 64. For these and other, fainter, indications of parallelism see Oldenberg, Prol. pp. 129, note 2 ; 145 ; 205 ; 218, note 3 ; 236, note 3. Similarity of obviously ritualistic hymns. — The ritualistic aprl-hymns, 1 According to A9-7-I2-9, 16 ; 99- 10.6.9, *6 at the Niskevalya. the first at the Marutvatlya^stra ; the second 2 Cf. Oldenberg, Piol. p. 266. Introduction 17 1. 13, 142,188; 2.3; 3.4; 5.5; 7.2; 9.5; 10.70, no, show many identical passages, from the equation 3.4.8-11 = 7.2.8-11, down to the identity or similarity of single padas. Hymn 1.13 shares no less than six padas with 1. 1 42. These old blessings presumably contain prehistoric stock which passed on from ancient times to the Rishis of the RV. Nevertheless, there is a good deal of difference in the style and the age of the aprl-sQktas. Some are purely formulaic; others, like 7.2, approach the diction and style of ordinary sQktas. Doubtless the nearer an aprl-stanza is to the ordinary style, the later it is. In one apram, namely 9.5, God Soma Pavamana is qualified successively for the functions of each of the divinities and potencies of the aprl-list. Oldenberg, Prol. pp. 28, note, 194, has shown that this inane application to Soma is accompanied by criteria of language and metre which show that some later poet, having in view the diaskeuasis of the ninth book, composed this hymn, apparently because he was bound, at all hazards, not to let that book go without an apram. It will be observed that the presence of Pavamana in this hymn disturbs its similarity to the other aprl hymns, so that only two padas of the usual stock appear in that hymn (9.5. 4*: 1.188.4® ; 9.5.8°: 5.5.7°). A second class of ritual stanzas correspond to the rtupraisas of the ?rauta- ritual. They are a class of formulas in which the various kinds of priests are correlated with certain definite divinities. These formulas are employed at the ceremony of choosing priests as well as at the so-called rtuyaja, or rtugraha, a class of offerings in which these priests and their divinities seem to symbolize the seasons of the year (rtu).1 The hymns containing these stanzas are i. 15, ii.36, and ii.37 (cf. also ii.5).2 3 The correspondences between the twelve stanzas of 1.15 with the twelve stanzas in 2.36 and 37 — these are in reality but a single hymn — are correspondences of theme with occasional verbal parallelism ; they do not rise to the repetition of entire stanzas or padas. The two sets of hymns, 1.2 and 3 confronted with 2.41 ; and, again, 1.23 confronted with 1.135 and 136, contain invitations to drink soma addressed to divinities which appear in fixed order: Vayu, India -Vayu, Mitra-Varuna, &c. This order reflects a definite ritualistic arrangement of the soma ritual (praiiga- 9astra). In general the resemblance between the corresponding stanzas is restricted to looser verbal similarity (cf. e.g. 1.2.7 with 1.23.5, and again with 2.41.3), but in one instance this ritual parallelism is supported by the identity of 1.23.8 with 2.41.5. See Bergaigne, JA. xiii (1888), 123 ff. ; Hillebrandt, Ved. Myth. i. 259 ff. Sundry touches of parallelism obviously connect the two ritual hymns 3.28 and 3.52. The former accompanies purodafa-offerings to Agni ; the 1 See Hillebrandt, Ritual-Litteratur, p. 131 ; Ved. Myth. i. 260 ff. ; iii. 147 ff . ; Oldenberg, Religion des Veda, p. 455, note ; Prol. p. 193 ; Caland-Henry, Agnistoma, pp. 224®. ; Olden - 3 | H.O.S. 20] berg, RV. Noten, p. 1. 2 Cf. also the rtupraisa AV. 20.20 ; and the khilas, adhyaya 7, in Scheftelowitz, Die Apokryplien des Rig-Veda, p. 148. 18 Introduction latter to Indra and companion gods. The offerings are arranged according to the three daily savanas. The wording is similar throughout ; e. g. purola agne pacatas in 3.28.2; purola9am pacatyam in 3.52.2; cf. Bergaigne, JA. xiii (1888), p. 20 ff. ; Hillebrandt, Yed. Myth. i. 229 ; Oldenberg, Rig-Veda Noten, I, p. 236. Even more marked are the special ritualistic theme and structure which connect 5.40 and 5.78, as described by Bergaigne, ibid., p. 60 ff., and Oldenberg, ibid., p. 335. We are not in the position to determine to what extent and for what purpose the more external parallelism in metre, order of gods, &c., harbours ritual identity or similarity, because the Sutras are, as a rule, silent on the subject. So, e. g., in the cases of 8.36 and 37, or 7.34 and 56, above (p. 16). Hillebrandt, in the course of his work on Yedic Mythology, has drawn attention more than once to the difference in worship of the gods and practice as carried on by the traditional Vedic Rishis, on the evidence of their respective books (mandalas).1 The aprl-hymns are the classical example. The Yi?ve Devah hymns are scarcely less ritualistic and formulaic. These differences must have been accompanied by a good deal of sameness, so that parallelism of all sorts reflects without doubt to some extent occupation with the same theme, much in the same manner as in the later schools (fakhas) of the ritual. If we had before us the ritual practices which accompanied the Rig-Vedic hymns at the time of their composition, the Rig-Veda would lose much of its obscurity. For ritualistic correspondences in general see Bergaigne’s posthumous studies in JA. xiii (1888). Similarities in mythic or legendary hymns. — A somewhat different order of parallelisms manifests itself at times in hymns addressed to certain divinities of marked physiognomy and more or less definitely limited legendary apparatus. The wondrous deeds of the A9vins are liable to be stated in the same formulaic language anywhere in the A9vin hymns. Yet a special tie connects the two A9vin hymns 1.116 and 1.117, both ascribed by tradition to Kakslvat Dairgha- tamasa. Each contains 25 stanzas ; the last stanzas are similar ; and there is close verbal correspondence between 1.1 16.7*: 1 . 1 1 7. 7a ; and i.n6.7d: 1.1 i7.6d; and i.n6.i6n: i.n7.i7a. Again, 1.117 and 1.118 have points of contact with 10.39 (authoress, Ghosa Kakslvatl) ; see i.n7.20d: 10.39. 7b J and 1.118.9“: 10.39.10“. The two Rbhu hymns, 4.34 and 4.35, are connected by constant expressions that contain the stem ratna (ratnadheya, vajaratna, ratnadha, ratnam dha). Again, 4.33 and 4.36 are connected by the padas 4.33.8“ ratharn ye cakruh suvrtarii narestham, and 4.36.2“, rathaxii ye cakruh suvrtam sucetasah ; or, 4.33 and 4.35 are connected by the padas 4.33.3“, punar ye cakruh pitara yuvana, and 4-35.5a> 9acyakarta pitara yuvana; or, 4.35 and 4.36 are connected by the padas, 4-35.2d, ekam vicakra camasam caturdha, and 4.36.4“, ekam vi 1 See especially iii. 394, and i, Index, p. 540*, under mandala ; iii, Index, p. 456b, under Ritual. Introduction 19 cakra camasaih caturvayam. In an analogous manner many points of contact between 2.12 and 10.121 show that the typical Indra of the so- called saianlva hymn has been made to serve as pattern for the hymn to the God Ka.1 In an even more general way we may expect to find more or less striking similarities between hymns addressed to the same divinity, because the later author is coaxed into a state of reminiscence by the exhaustive habits of his predecessors. We may say bluntly that it is almost impossible for a later author to compose a hymn to Agni or Indra or Soma without imitating his predecessors. The earlier poets have exploited these themes so thoroughly that there is nothing left for him to do but to follow their habits ; it is a mere question of degree how closely he will follow them. So, e.g., it is almost impossible to define exactly the nature of the similarities that connect the first hymn of the RV. with hymn 3.10. Both are hymns by Vi^vamitrid poets, addressed to Agni, and it is certain, furthermore, that their similarities go beyond the limits of accident. The first stanza of the first hymn seems to depend upon motifs that reappear in 3.10.1, 2 (note particularly 3.io.2“b, tvarii yajnesv rtvijam agne hotaram ijate). The pada 3.10.2°, gopa rtasya dldihi sve dame is reproduced in i.i.8b, gopam rtasya dldivim ; 3. 10. 4b, agnir devebhir a gamat is practically identical with 1.1.5°, d©vo devebhir a gamat; 3.10.4“, sa ketur adhvaranam is not very far in sense from i.i.8“, rajantam adhvaranam ; and 3.10.2 as a whole may be compared with 1.1.6. I think it likely enough that 1. 1 is really patterned after 3.10, but this is not certain: either hymn, or rather both hymns, may have been composed independently enough as regards everything except the settled and coercive habits of dealing with Agni ‘ Fire ’, the ritual god, for an indefinitely long time and for the same purposes. Literary or historical repetitions. — In all these cases of similarity the imitative element is, as it were, incidental or corollary. They are produced each by the natural circumstances of the case. But the most of the repetitions of stanzas, distichs, and padas in the Rig-Veda are, as we might say, literary or historical. The Hindus seem even at this early time to have been afflicted by an imperfect sense of literary proprietorship. What we stigmatize as plagiarism is to them the healthy exercise of utilitarian pragmatism. So at a much later time, and indeed at all times. E.g. the recent Bengali edition of Kalidasa’s Qakuntala shares two of its strophes with Bhartrhari’s Centuries ; see Hillebrandt, Uber das Kautillyafastra (Breslau, 1908), p. 28 ; Gottingische Gelehrte Anzeigen, 1909, p. 931. Kautsavya’s Nirukta is but an extract with scanty additions from Yaska’s work of the same name ; 2 * 4 see Bloomfield, JAOS. 1 Cf. Oldenberg, Prol. p. 315 ; Deussen, Geschichte der Philosophic, vol. i, part 1, p. 1 28 If. ; Bloomfield, Religion of theVeda, p. 240. 4 These additions are unfortunately often disguised by the evil state of the text which no editorial ingenuity may hope to cure entirely. In § 115 (p. 315) read dhruvarksam for dhruvadraksam (MSS. also dhruva iksam), that is dhruva + rksam (dvandva) ‘ the polar star and the Pleiades ’. 20 Introduction xv, pp. xlviiiff. ; The Atharva-Veda, p. 17 ; Bolling and von Negelein, Parifistas of the Atharva-Veda, i, p. 315. The many Brhats and Laghus testify to the same freedom. Professor Franke is at the present time engaged in elaborating a series of Concordances between the Buddhist texts, which shows that any good thing was regarded by the monks as common property.1 The numberless floating didactic stanzas in Sanskrit literature (Bohtlingk’s Indische Spriiche) are evidence of the same habit. In didactic or gnomic literature this freedom is not strange and scarcely reprehensible. A late Tamil writer, Pavanandi, has given a sort of canonical sanction to unrestrained borrowing : ‘ On what matters, with what words, in what way high men have spoken — so to speak is the convenience of style ’ ; see Vinson, Revue de Linguistique, vol. xlii, p. 155. It is entirely likely that this characteristic statement itself is but a repetition of some classical Sanskrit apothegm, though I have not been able to trace it. An occasional protest against plagiarism is not wanting ; see Kavyaprakafa 7 5 ; Rajataranginl, ed. Troyer 5.159; ed. Calc. 164; ed. Stein 164. To a small extent even RV. repetitions represent floating verses which have become common property, such as the padas, devo na yah savita satyamanma, 1.73.2 ; 9.97.48; or, deva iva savita satyadharma, 10.34.8; 139.3. But in the main RV. repetitions are the result of the freedom or licence with which later authors borrowed the products of their predecessors. On ‘ late hymns ’ and ‘ early hymns ’. — The Rig -Veda makes a distinction between old and new hymns, an interesting theme that scholars in the past have referred to quite frequently.2 The hymns themselves allude in clearest language to songs of old that were composed by the Rishis of the past. The later poets undertake to compare, more or less boastfully or complacently, then- own compositions with those of the ancient masters. ‘ A new song for a right old god ’, as I have explained in this book the puzzling RV. expression, navyam sanyase ; see under 8.8.12. Such reports are significant because they show that the Vedic poets were aware of the fact that Rig-Vedic composition stretched over a long period, preceding their own time. The suggestion has also been made that hymns which refer to themselves as ‘ new ’, as ‘ having been patterned after old ’, as ‘ having been made in the manner in which Atri, Kanva. Jama- dagni, and other worthies made their hymns’, are of recent origin.3 In the light of the materials which are worked up in this book, I have grown more sceptical as to our judgement in these matters. The mass of hieratic hymns seem to be in an advanced state of reciprocal assimilation. The hymns which refer to themselves as ‘ new ’ unquestionably presuppose antecedent hymns, 1 See recent issues of ZDMG. and WZKM. 2 See Muir, Original Sanskrit Texts, iii. 224ft. ; Ludwig, Der Rig-Veda, iii. 180; Hillebrandt, Ved. Myth. i. 123 ; Weber, SBAW. June 14, 1890, p. 605 ; Bloomfield, Religion of the Veda, p. 203. 3 Here figure traditionally important Rishis : atrivat, vasisthavat, or kanvavat, as well as less important or dubious names: jamadagnivat, vya9vavat, or nabhakavat : see Grassmann’s Lexicon, under these words. Introduction 21 but which are these antecedent hymns ? The absence in many hymns of the word ‘ new ’ does not prove that these hymns are pristine. Where, in brief, in the Rig-Veda are these pattern hymns that the poets refer to ? It seems to me safer to say, as I have said recently,* that ‘ the hymns of the Rig-Veda as a body are largely epigonal, or born after a long period of hymn production which must have, once upon a time, been much freer from conventional thought and mechanical utterance ’. The Rig-Veda is rather the final expression of its own type of composition, presupposing at all points a long period of antecedent activity. Consequently it is very often easy to point out signs of relative lateness, but I have yet to find any hymns in the collection which show positive signs of coming from the archetype period, that is to say, from the period when hymns of this sort were first composed. The authors, we know, are traditional, rather than historical. Their names are echoes from an olden semi-mythic time. Therefore, too, it seems to me very doubtful that the body of hymns assigned to any one of the more prominent of these traditional authors is, as a whole, superior or anterior to those of another author. In any case they one and all abound in repetitions. Many hymns of the pragfitha collection of Kanva and the numerous Kanvids are most certainly late clap-trap, but the important role which these hymns play in the Sama-Veda canon should warn us from condemning the rather banal compositions of the eighth book, because this involves the condemnation of the Sama-Veda to a late date. So, e.g., Indra’s epithet rclsama, ‘he for whom the saman is made upon therk’, belongs strictly speaking to this book alone, and yet conveys in grammatical form and content a very archaic conception.1 2 A fortiori the other family books, more checkered in their contents, must not, it seems to me, be judged chronologically as units : old and new, or, shall we rather say, older and newer, is contained in each of them, gathered together without reference to either absolute or relative chronology. The tenth book certainly contains a great deal of late matter.3 E.g., the Prajapati-hymn, io. 121, is certainly later than its relative, the sajanlya-hymn to Indra, 2.12, though the latter, in its turn, belongs to the later Indra compositions. But I would not venture, in reason, to condemn the funeral stanzas as a whole, as a class, either because they affect a more popular diction, or because they happen to treat a theme apart from the stock themes of the Rig-Veda. I cannot, for my part, imagine any Vedic time without just such or nearly such funeral stanzas. Relative character of Rig-Veda chronology, audits criteria. — Rig-Veda chronology is relative chronology in a very narrow sense. Because the mass shows unmistakable and innumerable points of likeness, every hymn or stanza is open to comparison with every other hymn or stanza. These comparisons are not in vain. Many a time we can say positively that such and such a hymn 1 JAOS. xxix, p. 287. family books, below, Part 3, chapters. 2 See the more concrete results of the 3 See the same Part and chapter, comparison of the eighth book with the other 22 Introduction was composed after such and such a hymn, and, still more frequently, that such and such a stanza was composed after such and such a stanza. In these detailed comparisons the repetitions must figure as a criterion of the first importance. The double or triple appearance, in different connexions, of the same larger metrical unit is liable to betray the place where it was originally composed. I need hardly state that it does not do so always, because a great many of the repeated passages consist of commonplaces, or are mere formulas. The repeated passages are to a very large extent modulated in the manner of the uhas or vikaras of the ritual stanzas and prayers. In such cases the variation in the form (metre) or contents of the passage may teach with yet greater certainty which is the model, and which the imitation. E.g., in two stanzas, 3.32.7 ; 6.19.2, Indra is addressed fitly as brhantam rsvam ajaram yuvanam. In 6.49. 10 Rudra is addressed as brhantam rsvam ajaram susumnam. So superior is the word sequence ajaram yuvanam, and so obvious is the uha nature of susumnam in reference to Rudra’s character (Rudra is mldhvas and $iva), that it follows with almost mathematical certainty that the Rudra passage is modelled after, is an uha of, the Indra passage; see under 3.32.7. A con- siderable number of times the same line occurs in unexceptionable surroundings in one place, but as an obvious parenthesis in another ; see, e.g., under 1.10.7 ; 12.1 ; 15.9 ; 24.10 ; 124.3 5 Index under Parenthesis ; and below, Part 2, chap. 4. A refrain which is original with a certain book, or with a traditional Rishi, may betray an outside hymn as late and imitative. Anacoluthon, tautology, ellipsis, solecism, dilution, addition and subtraction, imperfect metre, and occasional other criteria, too subtle for classification, can often he handled securely so as to establish relative dates, when a passage affected by these faults occurs elsewhere without these faults. Repetitions in their hearing on questions of exegesis. — It is almost needless to say that repetitions are an important element in interpretation : a given passage which is obscure in one connexion may be successfully inter- preted, because its recurrence in another connexion helps to determine its meaning; see, e.g., under 1.47.7; 2.13.19; 5-87-5 J 6.23.3, and many times more. On inconsistent renderings of repeated passages. — Conversely, and as a corollary to the last statement, it may be added that existing translations of the RV. betray their defects and provisional character in no particular more than in the way in which they deal with repeated passages. Many a time a repeated passage is rendered ad hoc, regardless of its appearance a second or third time.1 Repetitions are often mutually interpretative, and so it may happen that a repeated passage forbids a translation which seems to fit fairly in the first instance. The body of this work offers many cases of this sort : the subject by itself would yield an interesting and profitable dissertation. By way of illustration, Grassmann translates the pada, carur rtaya pltaye, 1 Khande-khande piindityam : Lagliucanakyam, ed. Teza, p. 8. Introduction 23 in 1. 137. 2, by ‘bereit zu regelrechtem Trunk’; the same pada at 9.17.8, by ‘dem Heiligen beliebt zum Trunk’. Ludwig (95), at 1. 137.2, ‘schoner zum opfer, zum tranke’; (807) at 9.17.8, ‘ schdn dem opfer als trunk zu dienen’. Similarly it was possible for Grassmann to render the pada, (ma) ni rlraman yajamanaso anye, 2.18.3; 3-35- 5» ‘ nicht mogen jetzt andre Priester dort zur Einkehr lenken ’ (vol. i, p. 25 ; here correctly), and ‘nicht mdgen andre Opfrer ergdtzen (vol. i, p. 83; here incorrectly). See under 2.18.3. Other cases of diverging translations are brought out under 1.8.5 » 1-10.7 ; 1.23.12 ; 1.37.4 ; 1.47.7; 1.50.2; 1.73.3; 1.82.2; 1.176.1; 2.5.8; 2.17.4; 2.26.2; 3.1. 15; 3-48.4; 4.3.10; 4.9.5; 4-29.1; 6.45.33; 7.15.8; 7.20.3; 7.35.14; 7-84.5; 8.2.32; 8.22.2; 8.24.8; 8.31.17; 9.13.9; 9.23.1; 9.33.2; 9.63.8, &c. Pischel, Ved. Stud. ii. 58, has undertaken to show, unsuccessfully, I think, that the pada, upa srakvesu bapsatah, has a different meaning in 7.55.2 from that which belongs to it in 8.72.15 ; see under 7.55.2. It is a fundamental fact that a given verse-unit has the same meaning everywhere, except in so far as it is altered verbally to suit a different theme or a different connexion. I hope that the present work will go far to eliminate, at least, this kind of fault from future RV. exegesis, and that it will also help positively our understanding of the oldest book of India. PART THE FIRST THE REPEATED PASSAGES OF THE RIG-VEDA SYSTEMATICALLY PRESENTED IN THE ORDER OF THE RIG-VEDA WITH THE TRADITIONAL STATEMENTS OF THE SARVANUKRAMANI AS TO THEIR AUTHORSHIP AND DIVINITY AND WITH CRITICAL COMMENTS AND NOTES ESPECIALLY AS TO THE RELATIVE CHRONOLOGY OF THE PASSAGES REPEATED 4 [h.o.s. so] Explanations relating to Part 1 or the main body of the work Explanation 1. — The repetitions of the RV. are treated in the following pages in the current order of that Veda as divided into books, hymns, and stanzas. Thus, beginning, e.g., with the thirteenth hymn of the first book, the verse-correspondences are as follows : i.i3.2a : 1.142.2b [1.13.2c : 6.53.10°] I*I3*3b> 7b [1.13.4° : 1. i4-“a; 6.16.9“; 8.34.8“] 1. i3*5a : 3-4i-2b; 8.45. ib 1.13.6“ : 1. 142.6“ 1.13.6b : 1.142.611 1.13.7“ : 1.142.7b 1.13.7° : 8.65.6° ; 10.188.1° 1, i3.8bc : 1. 142. 8bc ; i.i88.7bc 1.13*9 : 5-5-8 1.13.10° : 1.7.10°. With this arrangement the main body of this work serves of itself also as an index of RV. repetitions. This makes superfluous the printing of an index, in the above manner, at the end of the work, in the place usually assigned to an Index of Citations. Explanation 2 . — A distinction is made between either word-for-word repeti- tions, or important repetitions, on the one hand, and partial, less important repetitions, on the other hand. This distinction is discussed on pp. 8 tf. of the Introduction. Explanation 3. — Stanzas containing word-for-word repetitions, or important repetitions, are written out in full in their first listing in the order of the RV. ; in their second or third listing the repeated padas alone are quoted. Thus the stanzas 1.1.2 and 4.8.2, containing the completely repeated pada, sa devan eha vaksati, are printed entire in the order of 1.1.2. But in the order of 4.8.2 the repeated pada alone is quoted. It is understood, of course, that the discussion, if any, of such repetitions is, eveiy time, under the head of the earliest citation. Explanation 4. — Partial, less important repetitions are stated in each order of listing without writing out in full the stanzas within which these repetitions occur. They are, moreover, placed within square brackets. Thus, e.g., [1.91.6°, priyastotro vanaspatih : 9.12.7“, nityastotro vanaspatih.] Explanation 5. — The number-citation at the head of each item is printed in Clarendon (black-faced type), so as to facilitate the survey of the extent and order of the entire mass of repetitions. 28 Explanations relating to Part 1 Explanation 6. — The repeated padas of the sort described in Explanation 3, in their first occurrence, are printed in Clarendon (black-faced type) in all the stanzas involved. Explanation 7. — Additional repeated words (see p. vi, line 4 from below) are also printed in Clarendon, so as to help make clear to the eye, at a glance, the full scope of the repetition. Thus, e. g. : 1.3.10b (Madhuchandas Vaifvamitra ; to Sarasvatl) pavaka nah. sarasvatl vajebhir vajinivati, yajnam vastu dhiyavasuh. 6.61.4b (Bharadvaja; to Sarasvatl) pra no devl sarasvatl vajebhir vajinivati, dhlnam avitry avatu. Explanation 8. — Partly repeated padas, i.e. those printed within square brackets (see Explanation 4), are printed in ordinary type. Explanation 9. — In a great many of the stanzas written out in full in the interest of a given pada or given padas, there occur also other padas which are repetitions of padas pertaining to other stanzas in other places. Such padas are enclosed between two angles or el-brackets, and the place of their first occurrence and primary treatment in the order of the RV. is given in the margin, preceded by a little hand (®r). Thus, e.g. : 1.15.1b (Medhatithi Kanva ; to Indra) indra somam piba rtuna tva viqantv indavah, matsarasas tadokasah. 8.9 2. 2 2a (Qrutakaksa Angirasa, &c. ; to Indra) a tva vigantv indavah Lsamudram iva sindhavahj 8.6.35b na tvam indrati ricyate. This means that 8.92.22 shares its first pada with 1.15. 1 ; and, further, that its second pada also is repeated in a group stated first under 8.6. 35b. Explanation 10. — Frequently stanzas written out in full involve, in the same way, additional repeated padas of the partial and less important sort (see Explanation 2). They also are enclosed between two angles or el-brackets, and their first occurrence in the order of the RV. is marked on the margin with a little hand (44?), and with ‘ cf.’ (compare) before the citation. Thus, jw cf. 6. 16.7* Explanation 11. — Padas repeated within a single hymn only are listed only the first time, in ordinary print. Thus, under 1.13.3b, but not again under 1.13.7b : 1.13.3b, 7b> asmin yajna upa hvaye. Explanation 12. — In the same way refrain padas are listed and printed in ordinary type only at the place of their first occurrence. Thus, at 1.97.1, and not again: 1.97. ia, ic-8c, apa nah yofucad agham. REPEATED PASSAGES BELONGING TO BOOK I Group 1. Hymns 1-11, ascribed to Madhuchandas Vai9vamitra 1.1. 2e (Madhuchandas Vaigvamitra ; to Agni) agrn'h purvebliir fsibhir idyo nutanair uta, sa devan 6ha vaksati. 4.8.2° (Vamadeva Gautama ; to Agni) sa hi v6da vasudhitim mahan arodhanaih divali, sa devan 6ha vaksati. The metrical sequence of vasudhitim (never vasu°) is indeed bad at 4.8.2 ; but in my opinion it is one that has become stereotyped in the RV., and we should therefore not follow Arnold (YM. p. 124) in emending it. In this, Oldenberg (ZDMGr. lx. 160) agrees with me. Accordingly this fault may not be used as a criterion for the posteriority of the Vamadeva stanza. How this stereotyping could happen appears pretty clearly below, under 3. 31. 17. 1.1.4b (Madhuchandas Vaigvamitra ; to Agni) agne yam yajnam adhvaram vigvatah paribhur asi, sa id devesu gachati. i.97.6b (Kutsa Angirasa ; to Agni or Agni Quci) tvam hi vigvatomukha vigvatah paribhur asi, Lapa nah gofucad agham.j refrain, 1.97.1“, ic-8c Variations upon paribhu-phrases at 2.2.5 (with adhvaram) and at 5.13.6 (with devan). [1.1.5°, devo devebhir a gamat: 3.io.4b, agnir devebhir a gamat.] Hymn 3.10 is ascribed to Vigvamitra. The two hymns show much general resemblance. Its significance is discussed on p. 19. 1.1.8“ (Madhuchandas Vaigvamitra ; to Agni) rajantam adhvaranam Lgopam rtasya dldivim,j cf. 3.10.2° vardhamanaiii sve dame. 1.27.1° (^unahgepa Ajlgarti, called Devarata; to Agni) agvaih na tva varavantam vandadhya agnim namobhih, samrajantam adhvaranam. 1. 1.8 — ] Part 1: Repeated Passages belonging to Rook I [30 1.45.4° (Praskanva Kanva ; to Agni) mahikerava utaye Lpriyamedha ahusata, j 1.45.4b rajantam adhvaranam agmm gukrena gocisa. 8.8. 1 8° (Sadhvansa Kanva ; to Agvins) La vain vigvabhir utibhihj Lpriyamedha ahusata, j a: 7.24.4a ; b: i.45.4b rajantav adhvaranam agvina yamahutisu. Here the original is rajantam adhvaranam ; it is primarily an Agni-motif, as patir hy adhvaranam agne at 1.44.9, s:i ketur adhvaranam at 3.10.4, and netaram adhvaranam at 10.46.4 clearly show. As applied to the A9vins at 8.8.18, it is obviously secondary,1 — an observation made long ago by Oldenberg (Prol. p. 262) and backed by his opinion that the Praskanva hymns (1.44-50) are related to the ‘ Vatsa-group ’ (8.6-1 1) and prior to them. The secondariness of 1.2 7.1' is equally obvious ; and the trick (the prefixion of sam-) by which the trochaic cadence is secured to match that of a and b, is equally transparent. The word samrajantam, although clearly made ad hoc, we must not (with Pet. Lex., Grassmann, and the Concordance2) take as a compound of raj with sam; but rather (considering the absence of the paripanna samdhi : RPr. 4.7) as a denominative of samraj (Sayana : samrat- svarupam svaminam), ‘ playing the role of over-lord of’. The variation gopa rtasya didihi (at 3.10.2® ; 10.118.7°) might properly be called a ‘ phrase- inflection ’ (nom.-voc.) of the original (acc.) at i.i.8b. — For I.45.4"*, see under 1. 12.12, and cf. p. 9. 1.2.7ab (Madhuchandas Vaigvamitra ; to Mitra and Varuna) mitram huve putadaksam varunam ca rigadasam, dhiyam ghrtaclm sadhanta. 7.65.1b (Vasistha ; to Mitra and Varuna) [««* 7.63.5° Lprati vam sura udite suktairj mitram huve varunam putadaksam, yayor asuryam aksitam jyestham vigvasya yamann acita jigatmi. 5. 64. ia (Arcananas Atreya; to Mitra and Varuna) varunam vo rigadasam rca mitram havamahe, pari vrajeva bahvor jaganvansa svarnaram. The poor trochaic pada 1.2.7® may be a reminiscence of the faultless tristubh 7.65. ib : see Part 2, chapter 2, Class B ir. 1.2.8a (Madhuchandas Vaigvamitra ; to Mitra and Varuna) rt6na mitravarunav rtavrdhav rtasprga, kratum brhantam agathe. 1.152.1^ (Dlrghatamas Aucathya ; to Mitra and Varuna) yuvam vastrani plvasa vasathe yuvor dchidra mantavo ha sarg&h, avatiratam anrtani vigva rtdna mitravaruna saeethe. As between 1.2.8® and i.i52.id no very clear relation is apparent. It is, however, noteworthy that the enclisis of rtavrdhau (which the Pratigakhya expressly prescribes at 982) suggests that the redactor of 1.2.8 vaguely felt that -varunav was not the end of pada a, and marks rtavrdhau as what may very aptly be termed a 4 metrical vox media ’, or word used airti koivov so far as the metre is concerned, and counting now as part of pada a and then Analogous secondary applications under 1. 44.11 ; 7.11.1. Correct this by transferring the pada from 947 a to 986 b. Hymns ascribed to Madliucliandas Vaigvdmitra [ — 1.4.1 31] again as part of pada b. Just so the short u of vasudhitl marks yeniate of 4.48.3 as a metrical vox media : see under 3. 31. 1 7. That the text does not read mitravarunk rtavrdhav (Lanman, Noun-Inflexion, p. 575) points also to the ‘ Verquickung beider Pad as ’, as Oldenberg observes at RV. Noten, p. 2. 1.3.0b (Madhuchandas Vai9vamitra ; to Indra) indra yahi tutujana upa brahmani harivah, sutd dadhisva naf canah. 10. 104.6s (Astaka Vaifvamitra ; to Indra) upa br&hmani harivo haribhyam somasya yahi pltaye sutasya, indra tva yajnah ksamamanam anad Lda9van asy adhvarasya praketah.j 7.n.ia It is most tempting to regard i.3.6b as a fragment taken over from pada a of the faultless tristubh 10.104.6 ; the more so, inasmuch as the three indra yahi invocations waver clumsily between iambic (1.3.5) and trochaic (1.3.4, 6) cadences, and because our fragment is metrically so characterless as to fit with neither. 1.3.7l> (Madhuchandas Va^vamitra ; to Vi9ve Devah) omasa9 carsanldhrto vi(jve devasa a gata, da9vanso da9tisah sutam. 2.41. 1 3a (Grtsamada ; to Y^ve Devah) = 6.52.7s (Rji^van Bharadvaja ; to Vi9ve Devah) vigve devasah a gata L9rnuta ma imarii havam,j Ledam barhir ni sldata.j Oldenberg, Noten, p. 3, takes omasas as a umasas, with BR. 1.3.10b (Madhuchandas Va^vamitra ; to Sarasvatl) pavaka nah sarasvatl vajebhir vajinivati, yajnam vastu dhiyavasuh. 6.6i.4b (Bharadvaja; to Sarasvatl) pra no devi sarasvatl vajebhir vajinivati dhinam avitry avatu. One is obviously patterned after the other : but which ? 1.4.1b (Madhuchandas Va^vamitra ; to Indra) surupakrtnum utaye sudugham iva goduhe, juhumdsi dyavi-dyavi. 8.52(Val.4).4c (Ayu Kanva ; to Indra) yasya tvam indra stomesu cakano vaje vajin chatakrato, tarn tva vayam sudugham iva goduho juhumasi 9ravasyavah. The word goduh occurs thrice. At 1.164.26 and 8.52.4 it must mean ‘milker’. At 1.4.1 Sayana and Ludwig take it as ‘ milker ’, and Grassmann as ‘ a milking ’. The rendering as nomen actionis is supported by the obviously intentional correspondence between utaye and goduhe in the proportion surupakrtnum : sudugham = utaye : goduhe. Commonplace as the stanza is, it may yet have served as a suggestion to the versifex of 8.52, considering the admittedly late character of the Valakhilya. As to 8.52. 4d, see under 6.45.10. 8»*cf. 2.41. 13b 3.4i.i3c 1.4.3 — ] Part 1 : Repeated Passages belonging to Book 1 1.4.3b (Madhuchandas Yaigvamitra ; to Indra) atha te antamanam vidyama sumatlnam, ma no ati khya a gahi. 10.89. 17b (Renu Vaigvamitra ; to Indra) eva te vayam indra bhunjatlnam vidyama sumatlnam navanam, Lvidyama vastor avasa grnantoj Lvi9vamitra uta ta indra nunam. j 0:1.177.5°; d:6.25.9d Pada i-4.3b, though catalectic, is faultless (resolution not necessary : JAOS. xi, p. xxviii). Resolution and the (despite 8.51.5° or 10.160.5°) very otiose navanam stretch it indeed to a tristubh, of which, however, the secondary character is glaringly revealed by its almost intolerable cesura. This judgement tallies well with that of Oldenberg (Prol. p. 267) who calls the Vigvamitrid of 10.89 one of the ‘modern Epigones’. — The latter half of 10.89.17 moreover is merely an uha (or ‘ borrowing, mutatis mutandis ’) from the Bharadvaja stanza 6.25.9 (°f- under 6.25.9). Thus rudely does the critic show up Renu’s stanza as (to 75 percent.) a thing of 1 shreds and patches ’. [l.4.4c, yas te sakhibhya a varam: 9.45.2°, devan sakhibliya a varam] 1.4. 6e (Madhuchandas Vai^vamitra ; to Indra) uta nah subhagah arir voceyur dasma krstayah, syamdd indrasya garmani. 8.47.5° (Trita Aptya ; to Adityas) pari no vrnajann agha durgani rathyo yatha, syamdd indrasya garmany adityanam utavasy Lanehaso va utayah suutayo va Qtayah.j (W refrain, 8.47. ief-i8ef Arnold, YM., p. 45, notes evidence of late date for 8.47. The banality of the whole hymn and its heavy refrains attest a low degree of artistic skill in its author. The solitary Indra- pada, 8.47.5°, in a l°ng Aditya-hymn seems at first blush a palpable intrusion ; but pada d (on account of its uta and its need of the complementary syamaof c) cannot be disjoined from c : they make a unit. Accordingly we may suppose that the Adityu-couplet was worked out upon a reminiscence of 1.4.6°, by an author who was not staggered by its partial impertinence. 1.4.8° (Madhuchandas Yaigvamitra ; to Indra) asya pitva gatakrato ghano vrtranam abhavah, pravo vajesu vajinam. i.i76.5d (Agastya ; to Indra) avo yasya dvibarhaso ’rkesu sanusag asat, ajav indrasyendo pravo vajesu vajinam. The grave difficulties which beset the rest of 1.176.5 do not concern its final pada. This is loosely appended and may well have been borrowed from 1.4.8. 1.4.10ab+° (Madhuchandas Vaigvamitra ; to Indra) y6 rayd ’vanir mahan suparah sunvatah sakha, tasma indraya gayata. 33] [ — 1-5-4 Hymns ascribed to Madhuchandas Vaigvamitra 8.32.i3nb+c (Medhatithi Kanva; to Indra) y6 rayd ’vanir mahan suparah sunvatah sakha, tdm indram abhi gayata. i.5.4c (Madhuchandas Vaigvamitra ; to Indra) yasya saxhsthe na vrnvate harl samatsu gatravah, tasma indraya gayata. 1.5.1b (Madhuchandas Vaigvamitra ; to Indra) a tv eta ni sldat6ndram abhi pra gayata, sakhaya stomavahasah. 8.92. i b (Qrutakaksa Ahgirasa, or Sukaksa Ahgirasa ; to Indra) pantam a vo andhasa indram abhi pra gayata, vifvasiiharii ^atakratum manhistham carsanlnam. 1.5.2ft+b+c (Madhuchandas Vaigvamitra; to Indra) purutamam purunam iganarh varyanam, indram s6me saca sut6. 6. 45.29“ (^arnyu Barhaspatya ; to Indra) purutamam purunam stotfnarh vivaci, vajebhir vajayatam. i.24.3b (^unahgepa Ajlgarti, called Devarata ; to Savitar) abhi tva deva savitar igaham varyanam, sadavan bhagam imahe. 8.7i.i3b (Sudlti Angirasa, or Purumldha Ahgirasa; to Agni) agnir isarh sakhye dadatu na ige y6 varyanam, agnirii toke tanaye gagvad imahe vasum santaiii tanupdm. x°.9-5a (Trigiras Tvastra, or Sindhudvipa Ambarlsa ; to Waters) igana varyanam ksayantl? cai-sanlnam, apo yacami bhesajam. 8.45.29° (Trigoka Kanva ; to Indra) rbhuksanam na vartava ukthesu tugryavrdham, indram some saca sut6. Noteworthy is the discordance of rendering to which the repeated purutamam purunam gives occasion : Ihn der der Reichen reichster ist, or Den Reichsten unter Vielen (Grassmann) ; dem vollsten der vollen, or dem reichsten der reichen (Ludwig). Pischel, Ved. Stud. i. 36, renders 6.45. 29*** thus : ‘ der du am meisten (angerufen) wirst im Wettstreit vieler Lobsanger ’ ; but the parallel at 1.5.2 makes against taking purunam out of its own pada ; and if, as is natural, i.5.2b means about the same as 1.5.2% then Grassmann’s first version and Ludwig’s second are to be deemed good. In ige yo varyanam we have again a case of ‘ phrase-inflection ’, the nominative to fganam varyanam. 1.5.4° : 1.4.10°, tasma indraya gayata : 8.32.13°, tarn indram abhi gayata. 5 [h.o.s. 20] 1.5.5 — ] Part 1: Repeated Passages belonging to Book 1 [34 1.5.5b+c (Madhuchandas Vaigvamitra ; to Indra) sutapavne suta imd gucayo yanti vltaye, sdmaso dadhyagirah. 8.93. 2 2b (Sukaksa Angirasa ; to Indra) patnlvantah suta ima uganto yanti vitaye, apam jagmir nicumpunah. 1 . 1 3 7* 2b (Parucchepa Daivodasi ; to Mitra and Varuna) ima a yatam indavah sdmaso dadhyagirah sutaso dadhyagirah, uta yam usaso budhi' Lsakam suryasya ragmibhih, j cs* 1.47. 7*1 suto mitraya varunaya pltaye Lcarur rtaya pltaye. j 1.137.28 5.51.713 (Svastyatreya Atreya ; to Vigve Devah) Lsuta indraya vayavej sdmaso dadhyagirah, 6s* 5- 5 1 . 7a nimnarii na yanti sindhavo ’bhi prayah. 7-32.4b (Vasistha; to Indra) ima indraya sunvire sdmaso dadhyagirah, tan a madaya vajrahasta pltaye haribhyam yahy oka a. 9. 22.3b (Asita Kagyapa, or Devala Kagyapa ; to Soma Pavamana) Lete puta vipagcitahj sdmaso dadhyagirah, 6s* 9.22.3s vipa vy anagur dhiyah. 9.63. i5b (Nidhruvi Kagyapa; to Soma Pavamana) suta indraya vajrine sdmaso dadhyagirah, pavitram aty aksaran. 9.ioi.i2b (Manu Samvarana ; to Soma Pavamana) Lete puta vipagcitahj sdmaso dadhyagirah, 9.22.3s siiryaso na dargataso jigatnavo dhruva ghrte. The streams of Soma at 5.51.7 (like Southey’s water that ‘comes down at Lodore’) come fairly tumbling on their way to Indra ; and so, elsewhere, the songs of praise (see under 8.6.34). 1.6. 8C (Madhuchandas Vaigvamitra; to Indra) tvam stoma avlvrdhan tvam uktha gatakrato, tvam vardhantu no girah. 8.44.19° (VirQpa Angirasa ; to Agni) Ltvam agne manlsinasj tvam hinvanti cittibhih, 6S* 3.10.1s tvam vardhantu no girah. To begin successive padas, or even stanzas (5.8. 1-7 ; 5.9.1), with the same word is so common a procedure as to have no critical significance. Nor is the epanaleptic tv&m of 8.44.19’’ censurable: see Oldenberg, Noten, 427a (dasselbe Wort pleonastisch mehrfacli gesetzt). [1.6.10°, igano yavaya vadham : 10.152. gd, varlyo yavaya vadham] [1.0. 9b, divo va rocanad adhi : 1.49.1b : 5.56.1*1 • 8.8.7s, divag cid rocanad adhi] 35] Hymns ascribed to Madhuchandas Vaigvamitra [—1.7.9 1.7.3b (Madhuchandas Vaifvamitra; to India) indro dlrghaya caksasa a suryam rohayad divi, vi gobhir adrim airayat. 8.89. 7b (Nrmedha Aiigirasa and Purumedha Augirasa ; to Indra) amusu pakvam airaya a suryam rohayo divi, gharmam na saman tapata suvrktibhir justaiii girvanase bill at. 9. 107.711 (Sapta Rsayah ; to Pavamana Soma) somo mldlivan pavate gatuvittama rsir vipro vicaksanah, tvarii kavir abhavo devavitama a suryam rohayo divi. 10.156.4° (Ketu Agneya ; to Agni) agne naksatram ajaram a suryam rohayo divi, dadhaj jyotir janebhyah. 1.7.4° (Madhuchandas Vaifvamitra ; to Indra) indra vajesu no ’va sah&srapradhanesu ca, ugra ugrabhir utibhih. 1.129.5° (Parucchepa Daivodasi ; to Indra) ni su namatimatirii kayasya cit tejisthabhir aranibhir notibhir, ugrabhir ugrotibhih, [ityadi]. 1.7.8° (Madhuchandas Vaifvamitra ; to Indra) vrsa yutheva vansagah krstir iyarty ojasa, igano apratiskutah. 1.84.7° (Gotama Rahugana ; to Indra) ya eka id vidayate Lvasu martaya dafuse,j (T 1.84. 7b igano apratiskuta indro anga. The stanzas 1.84.7,8,9 read each (apart from the last two words) like a brahmodya, — of course not necessarily in interrogative form. The last two words, indro anga, form the ‘answer’ and are simple prose, — quite out of the metre, as is intrinsically clear, without any reference to the parallel 1.7.8°, which is however a faultless metrical unit. 1.7. 9a (Madhuchandas Vaifvamitra ; to Indra) ya 6kag carsanmam vasunam irajyati, indrah panca ksitlnam. i.i76.2b (Agastya ; to Indra) tasminn a vefaya giro ya ekag earsanlnam, anu svadha yam upyate yavam na carkrsad vrsa. For the desperate 1.176.2, see citations in Oldenberg’s Noten, and especially Ludwig, 5.48 end, and Grierson’s Bihar Peasant Life, p. 182. The parallel, alas, injects no decisive new factor into the ample discussion, beyond this, that ya ekaf earsanlnam in 1.176 seems to be a fragment and a dislocated one. i.7.i o — ] Part 1 : Repeated Passages belonging to Book I [36 1.7.10c (Madhuchandas Vaigvamitra ; to Indra) l'ndram vo vigvatas pari havamahe janebhyah, asmakam aatu kdvalah. 1.13.10° (Medhatithi Kanva ; to Tvastar) iha tvastaram agriyam vifvarupam upa hvaye, asmakam astu kdvalah. The form havamahe is hieratic; hvaye is popular. The latter occurs no less than five times in 1.13 and marks this Apri-hymn as late. 1.8.4° (Madhuchandas Vaifvamitra ; to Indra) vayam furebhir astrbhir indra tvaya yuja vayam, sasahyama prtanyatah. 8.40.7(1 (Nabhaka Kanva; to Indra and Agni) yad indragm jana ime vihvayante tana gira, asmakebhir nrbhir vayam sasahyama prtanyatd Lvanuyama vanusyatdj Lnabhantam anyake samOj e : 1.132.1° ; f : refrain, 8.39. if ft. 9.61.29° (Amahlyu Angirasa ; to Soma Pavamana) Lasya te sakhye vayanij tavendo dyumna uttame, 9.61. 29“ sasahyama prtanyatah. Prefixion of the four syllables indratvotah expands 1.8.4° to a full jagatl at i.i32.ib, q. v. Under 2.8.6 it appears that -yama prtanyatah is a Vedic cadence. Reminiscence of 1.8.4 ’n tvaya ha svid yuja vayam, 8.2i.na ; 102. 3a. 1.8.5° (Madhuchandas Vaifvamitra ; to Indra) mahan indrah paraf ca mi mahitvam astu vajrine, dyaur na prathina gavah. 8.56(Val.8).i° (Prsadhra Kanva ; Danastuti of Praskanva) Lprati te dasyave vrka radhoj adar9y ahrayam, c®*ab: 8.55.1° dyaur na prathina gavah. ‘Great is Indra, aye, more than great: greatness be his that wields the bolt; might, wide as the heaven.’ Here the ascription of might or 9&vas to Indra is an entirely natural sequel to the ascription of greatness. ‘ Might wide as the heaven ’ is not much different from the might of Indra which at 8.24.9 called aparltam (Sayana : 9atrubhir aparigatam avyaptam). But as a nearly synonymous sequel to r&dho ahrayam, ‘ a gift that brings no shame to the giver’ (Sayana on 5.79.5 : alajjavaham), ‘no shabby gift’, pada c is plainly not fit. It is not fit, even if we assign to 9iivas the unwarrantable meaning of fiille (Ludwig) or Rulim (Grassmann). This unfitness adds one more (unneeded) item to the cumulative evidence touching the lateness of the Valakhilya. 1.8.7’’ (Madhuchandas Vaifvamitra ; to Indra), followed by 1.8.8 yah kuksfh somapiltamah samudra iva pinvate, urvir apo na kakudah, — 37] Hymns ascribed to MadhucJiandas Vdigvdmitra [ — 1.9.8 eviv hy iisya sQnfta virapgi gomatl main, pakvil gakhiT, na daguse. 8. i2.5b (Parvata Kfinva ; to Indra) imam jusasva girvanah samudra iva pinvate, Lindra vigvabhir Qtibhir vavaksitha. j ] Hymns ascribed to Medhatithi Kanva [ — 1.23.7 8.i2.27b (Parvata Kanva: to Indra) yada te visnur ojasa trini pada vicakramd, Lad ft te haryata harl vavaksatuh.j refrain, 8.i2.25°-27° Cf. 8.53(Val. 4).3C, yasm&i visnus tnni pada vi cakramd. 1.22.21ab (Medhatithi Kanva ; to Visnu) tdd vipraso vipanyavo jagrvansah sam indhate, vfsnor yat paramam padam. 3. io.9ab (Vifvamitra Gathina ; to Agni) tam tva vlpra vipanydvo jagrvansah sam indhate, Lhavyavaham amartyarii sahovfdham.j 6*' 3.9.10° The repeated disticli is primary in 3.10.9 : * The bards, skilled in song, on waking, have kindled thee (Agni, fire).' The application of the same idea in 1.22. 21 is mystic: the bards kindle the highest stepping-place of Visnu, the sun-fire at its zenith, the abode of the blessed. Cf. 1.22.20; 1. 154.5; 10.1.3, &c., and Hillebrandt, Ved. Myth. i. 354. We may admire the ingenuity which enables the epigonal poet to express the thought that the inspired song of the poets kindles the light of the heavens, just as it accompanies the rubbing of the sacrificial fire. But the fact remains that he has adapted an ordinary sense motive effectively, yet mechanically, to his high idea. Without the former we should hardly have had the latter. Cf. also Oldenberg, RV. Noten, p. 1 7. 1.23.1a (Medhatithi Kanva ; to Vayu) tlvrah sdmasa a gahy a^irvantah suta ime, vayo tan prasthitan piba. 8.82.2“ (Kusldin Kanva ; to Indra) tivrah sdmasa a gahi sutaso madayisnavah, piba dadhfg yathocise. 1.23.2a: 1.22. 2b, ubha deva divispf9a. 1.23.2°: 1.22.1°; 4.49.5°; 5.71.3°; 6.59.iod: 8.76.6°; 8.94.10°, 11°, 12°; asy£ somasya pltaye. [1.23.6°, karatam nah suradhasah: 3.53.13°, karad fn nah suradhasaL] 1.23.7a (Medhatithi Kanva ; to Indra Marutvant) marutvantam havamaha indram a sdmapitaye, sajur ganena trmpatu. 8.76.6b (Kurusuti Kanva; to Indra) indram pratnena manmana marutvantam havamahe, Lasya somasya pltaye. S3" 1.22.1° i.23.8 — ] Part 1 : Repeated Passages belonging to Book I [56 1.23.8 (Medhatithi Kanva; to Indra Marutvant, better Vigve Devah) = 2. 4 1. 1 5 (Grtsamada ; to Vigve Devah) indrajyestha marudgana ddvasah pusaratayah, vigve mama gruta havam. See Bergaigne, ii. 371, 383, 390, 428 ; Oldenberg, RV. Noten, p. 18. Cf. Weber, Proceedings of the Berlin Academy, June 14, 1900, p. 603, note 1 ; our introd. p. 17. Ludwig’s (244) emendation of pusaratayah to gusaratayah is intrinsically unnecessary. 1.23.9° (Medhatithi Kanva ; to Indra Marutvant) hata vrtram sudanava indrena sahasa yuja, ma no duhgahsa igata. 2.23.10° (Grtsamada ; to Brhaspati) tvaya vayam uttamam dhlmahe vayo brhaspate paprina sasnina yuja, ma no duhganso abhidipsur igata pra sugansa matibhis tarislmahi. 7.94.7° (Yasistha ; to Indra and Agni) indragnl avasa gatam Lasmabhyam carsanlsaha,j 5.35.1° ma no duhgahsa igata. 10. 25.7d (Vimada Aindra, or others ; to Soma) Ltvam nah soma vigvatoj gopa adabhyo bhava, 1.91.8s sedha rajann apa sridho vl vo made ma no duhgahsa igata vivaksase. Cf. raksa makir no aghagansa Igata, under 6.71.3, and ma na (and, va) stena Igata maghd- gansah, under 2.42.3. — The pada 10.25. 7d with its tetrasyllable refrain (vivaksase) is certainly secondary ; and abhidipsuh in 2.23.10 looks very much like a gloss. 1.23.10b (Medhatithi Kanva ; to Vigve Devah) vigvan devan havamahe marutah somapitaye, ugra hi prgnimatarah. 8.94.3° (Bindu Angirasa, or Putadaksa Angirasa ; to Maruts) Ltat su no vigve arya a sada grnanti karavah,j ts? 6.45.33ab marutah sdmapitaye. 8.94.9° (The same) a y 6 vigva parthivani paprathan rocana divah, marutah sdmapitaye. [1.23.15°, gobhir yavam na carkrsat : i.i76.2d, yavam na carkrsad vfsa.] 1.23.20ftb° (Medhatithi Kanva ; to Waters) apsu me s6mo abravid antar vigvani bhesaja, agnim ca vigvdgarhbhuvam apag ca vigvabhesajlh. io.9.6ab° (Trigiras Tvastra, or Sindhudvlpa Ambarlsa; to Waters) apsu me sdmo abravid antdr vigvani bhesaja, agnim ca vigvdgambhuvam. 57J Hymns ascribed to Medhatithi Kanva [ — 1.24.9 The two stanzas are identical, except that 10.9.6 lacks the fourth pada of 1.23.20. The latter is certainly surplusage, as Grassmann observes, ii, p. 504. The entire passage 1.23.20-23 is repeated at 10.9.6-9. At 1.23 it is part of an appendix of six stanzas (19-24) which follows upon six trcas, each addressed to a different divinity. Four of these stanzas (20-23) are taken bodily from the well-knit hymn 10.9 (6-9). This excerpt is preceded in 1.23 by the metrically irregular (AnukramanI, puraiisnih) stanza 19, and followed by the Agni-stanza 24, which continues and expands in a concatenary way the theme of the preceding distich. Cf. Oldenberg, Prol. pp. 225, 234, and for further reference, his RV. Noten, p. 17. 1.23.210 (The same) = 10.9.70 (The same) apah prnlta bhesajam varutham tanvd mama, jy6k ca suryam dr 6. 1.23.22 and 23 (The same) = 10.9.8 and 9 (The same) idam apah pra vahata yat kim ca duritam mayi, yad vaham abhidudroha yad va gepa utanrtam. apo adyanv acarisam rasena sam agasmahi, payasvan agna a gahi tarn ma sam srja varcasa. Group 3. Hymns 24-30, ascribed to i=d . 1.25.10° (^!unah9epa Ajlgarti, &c. ; to Varuna) ni sasada dhrtavrato varunah pastyasv a, samrajyaya sukratuh. 8.25.8b (V^vamanas Vaiya9va ; to Mitra and Varuna) rtavana ni sedatuh samrajyaya sukratu, dhrtavrata ksatriya ksatram a9atuh. Pischel, Ved. Stud. ii. 212, starts with 1.25. 10 to show that pastyS, feminine, means ‘ river’, or ‘ water ’, in distinction from pastyii, neuter, ‘ home ’. The passage does not seem to me well chosen, as is shown by the imitative stanza 8.25.8, where no water is mentioned. A further parallel, the nivid stanza in the form of an unanswered riddle, 8.29.9, dva cakrate upama divi samraja sarpirasutl, shows that the seat of Varuna and Mitra is high in heaven. This does not, of course, make it impossible, here and there, when a god’s home happens intrinsi- cally to be water, that pastyii may refer to water; cf. Mahidliara to VS. 10.7 (TS. 1.8.12.1 ; MS. 2.6.8; KS. 15.6). A poet may refer to Varuna’s, Agni’s, or Trita’s domicile, and mean ‘ watery domicile ’ ; then next a Hindu commentator may remain well within the bounds of his 1 Ludwig, 200, tries, ineffectively, to smooth out the roughness : ‘ Aryaman [Mitra und Varuna], die opferwiirdigen, sind uns Aditi,’ &c. [ 1.25.20 59] Hymns ascribed to Qunahgepa Ajigarti reprehensible habits, and translate pastya by ‘ river’. For the present it would seem to me that pastyasu (plurale tantum) in 1.25. 10 had best be rendered ‘ seat’, or ‘ dwelling In the next stanza (next item) Varuna is described as surveying from there (atah), Odin-like, the wondrous events of the cosmos which have been and shall be. This is scarcely Varuna in the waters. Of the two stanzas 8.25.8 is obviously secondarily ‘ breitgetreten ’. So also judged, many years ago, Aufrecht in his second edition of the text of the Rig-Veda, vol. ii, p. xxvi. 1.25.11b (Qunahgepa Ajigarti, &c. ; to Varuna) dto vlgvany adbhuta cikitvan abhi pagyati, Lkrtani ya ca kartva.j «wcf. 1.25.11c 8.6. 29*’ (Vatsa Kanva; to Indra) atah samudram udvatag cikitvan ava pagyati, yato vipana ejati. For 8.6.29 see Geldner, Ved. Stud. iii. 56. [1.26.11°, krtani ya ca kartva: 8.63.6b, krtani kartvani ca.] 1.26. 15b ((^unahgepa Ajigarti, &c. ; to Varuna) uta y6 manusesv a yagag cakrd asamy a, asmakam udaresv a. io.22.2J (Vimada Aindra, or somebody else ; to Indra) iha gruta indro asme adya stave vajry fclsamah, mitro na y6 janesv a yagag cakrd asamy a. The banality of 1.25.15 leads Grassmann to misrender the stanza : ‘ Und der den Menschen Herrlichkeit verleiht, die ganz vollkommen ist, und selbst an unsern Leibern auch.’ But 10.22.2 shows that yagag cakre means ‘obtained glory’, rather than ‘conferred glory’. So Ludwig, 82, but he, in his turn, resorts to emending udaresv to duryesv: * Der sich unter den menschen vollkomne herlichkeit geschaffen, in unsern eignen hiiusern.’ The rough and insipid pada 1.25.15° shows that the stanza is late imitative manufacture, patterned, doubt- less, after 10.22.2, but it makes sense as it stands. After stating in 1.25. 14 that Varuna is unassailable and so on, the present stanza says : * Who, moreover, of men exacts undivided respect, and of our bodies (bellies) ’, that is to say, by punishing, when we sin, our bodies with his disease, the dropsy. — Cf. Geldner, Rigveda-Kommentar, p. 5. 1.25. 20b (Qunahgepa Ajigarti, &c. ; to Varuna) tvam vigvasya medhira divag ca gmag ca rajasi, sa yamani prati grudhi. 5. 38.3d (Atri Bhauma ; to Indra) gusmaso ye te adrivo mehana ketasapab, ubha devav abhistaye divag ca gmag ca rajathah. Grassmann, to 5.38.3 (following Sayana): ‘ Die Helden, die, 0 Schleuderer, dir reichlich zu Gebote stehn, ihr Gotter beid’ beherrscht zum Heil den Himmel und die Erde auch.’ The heroes, according to Sayana, are the Maruts, a very sensible suggestion as far as the second distich is concerned, though gusmaso cannot, of course, mean heroes. Ludwig, 539, also refers the dual to gusmaso and Indra. In ZDMG. xlviii. 571, I took gusmaso in the sense of light- nings, and referred the two gods to the lightnings and Indra, or to the press-stones (adrivah) and the lightning. I now consider this no more probable than does Oldenberg, RV. Noten, P- 333- who remarks : ‘ Welcher zweite Gott neben Indra gemeint ist konnen wir nicht wissen.’ Judging from 1.25.20, we might now guess Varuna, if it were not for the very vague and commonplace quality of the formula divag ca, &c. Therefore, perhaps better, Indra and Soma ; cf. 9.95.5, indrag ca yat ksayathah siubhagaya. [60 i.26.i — ] Part 1: Repeated Passages belonging to Book I 1.26.1C: 1.14.11°, semam no adhvaram yaja. 1.26. 4b+c ((^unahfepa Ajigarti, &c. ; to Agni) a no barhi ri^adaso varuno mitrd aryama, sidantu manuso yatha. i.4i.ib (Kanva Ghaura; to Varuna, Mitra Aryaman) yam raksanti pracetaso varuno mitro aryama, nu cit sa dabhyate janah. 4.55.10’’ (Vamadeva ; to Vifve Devah) Ltat su nah savita bhagOj varuno mitr6 aryama, l'ndro no radhasa gamat. 5. 67.3^ (Yajata Atreya ; to Mitra and Varuna) vifve hi vi9vavedaso varuno mitr6 aryama, vrata padeva sa?cire Lpanti martyam risah.j 8.i8.3b (Irimbithi Kanva ; to Adityah) Ltat su nah savita bhagOj varuno mitrd aryama, L9arma yachantu sapratho yad Tmahe. j 8. 28. 2a (Manu Vaivasvata ; to Vi9ve Devah) varuno mitrd aryama smadratisaco agnayah, patnlvanto vasatkrtah. 8.83.2^ (Kusldin Kanva ; to Vi9ve Devah) te nah santu yujah sad a varuno mitr6 aryama, vrdhasa9 ca pracetasah. 9.64. 29a (Ka9yapa Marlca; to Pavamana Soma) hinvano hetfbhir yata a vajam vajy akramlt, sidanto vamiso yatha. Ludwig, 251, and Grassmann render 1.26.4 : ‘May Varuna, &c., sit upon our barhis like men ’ ; Bergaigne, La Religion V^dique, i. 67 ; Melanges Renier, p. 78 ; and Oldenberg, SBE. xlvi. 13 : 1 May Varuna, &c., sit down on our barhis as they did on Manu’s.’ The latter trans- lation, which Ludwig also suggests in his commentary, is in its general sense eminently satisfactory ; it would call for no comment but for the curious parallel in 9.64.29. Ludwig, 854, renders that stanza, ‘ ausgeschiittet gelenkt von denen, die es laufen lassen, ist zur krafttat das kraftross geschritten, wie kampfer die ihren platz einnehmen.’ Grassmann, almost the same, except that he renders pada c by ‘wie Krieger sitzend nach dem streit’. Both are preceded by the Pet. Lex. under i.vanus, ‘die beim soma sitzen wie Kampfbereite ’. All three renderings of the repeated pada are more or less whitewashed : as regards Ludwig, sidanto means ‘ sitting ’, which is very different from a warrior’s taking his place, presum- ably, in the ranks ; as regards Grassmann, vanuso does not mean ‘ nach dem streit ’, though it may mean ‘striving’; as regards Pet. Lex., ‘ Kampfbereite ’ is open to similar criticism. I believe that vanuso means ‘ desiring’ : the soma steed attains his prize, and so do the sacri- ficing priests who desire the soma ; cf. 10.96.1, pra te (sc. indrasya) vanve vanuso haryat£m madam, ‘I desire the golden drink of thee who (also) desirest it.’ Considering, now, the facile interchange between m and v (cf. JAOS. xxix. 290 ff.), one of the two padas 1.26.4° and 9.64.29° is pretty certainly patterned after the other. I incline to think that 1.26.4° i9 model, 9.64.29° the imitation. Be this as it may, the construction of manuso in r.26.4, as nominative plural, rather than genitive ts- 4.55.ioa 6S* i.4i.2b 6S-4.55.109 6S- 8. 18. 3° 61] Hymns ascribed to Qunahgepa Ajlgarti [ — 1.28.9 singular, rises in the scale in spite of a certain prima facie insipidity : the gods Varuna, &c., are compared with sacrificing men. Both sit upon the barliis, equally interested in the progress of the sacrifice ; soma and dafcsina are doubtless in the mind of the poet. Rather curiously, we have much the same variant as between RY. 1.44.11° and TB. 2.7.12.6°, manus- vdd (TB. vanusvad) deva dhimahi priicetasam. Here the commentary to TB., vanusvat paricaranavat, has in mind the same idea as ours in reference to vanusali in RV. 9.64.29°. — The pfida, varuno mitrd aryama, also as refrain in io.i26.3b-7b ; cf. varuna mitraryaman, under 5.67.1° ; and see p. 11. 1.26.5C ((^unahgepa Ajlgarti, &c. ; to Agni) purvya hotar asya no imindasva sakhyasya ca, ima u su grudhl girah. r*45-5b (Praskanva Kanva ; to Agni) ghftahavana santyema u su grudhl girah, yabhih kanvasya sQnavo havante ’vase tva. 2.6.1° (Somahuti Bhargava ; to Agni) imam me agne samidham imam upasadam vaneh, ima u su grudhl girah. 1.26.10b (^Junahgepa Ajlgarti, &c. ; to Agni) vi'gvebhir agne agnibhir imam yajnam idam vacah, cano dhah sahaso yaho. 1.9 1. ioa (Gotama Rahugana ; to Soma) imam yajnam idam vaco Ljujusana upagahi, i.9i.iob soma tvam no vrdhe bhava. 10. 150.211 (Mrlika Vasistha; to Agni) imam yajnam idam vaco Ljujusana upagahi, 1.91. iob martasas tva samidhana havamahe mrllkaya havamahe. Antecedently it is probable that 1.26. iob is borrowed from the compact distich of the two others. 1.27.1°, samrajantam adhvaranam : i.i.8a; 45.4°, rajantam, &c. ; 8.8.18° ra- jantav, &c. 1.28.1cd-4°d, ulukhalasutanam aved v indra jalgulah. 1.28. 9b (^unahgepa Ajlgarti, &c. ; to Prajapati Harigcandra, or [Adhisavana-] carmapragansa) uc chistam camvor bhara s6mam pavitra a srja, ni dhehi gor adhi tvaci. 9.i6.3b (Asita Kagyapa, or Devala Kagyapa ; to Soma Pavamana) anaptam apsu dustaram somam pavitra a srja, Lpunihmdraya patave. j 9. 16.3° 1.28.9 — ] Part 1 : Repeated Passages belonging to Book I [62 9.51. 1 b (Ucathya Angirasa; to Soma Pavamana) adhvaryo adribhih sutam s6mam pavitra a srja, Lpunlhindraya patave.j 6®* 9.16.3° For stanza 1.28.9 °f- Hillebrandt, Ved. Myth. i. 170; Geldner, Rigveda Kommentar, p. 5 ; for the entire hymn, Oldenberg, RV. Noten, p. 24. Cf. also Ludwig, 784. — The cadence, gor &dhi tvaci, also at 9.65.25 ; 79.4 ; 101.11. 1.29.1b (Qunahgepa Ajlgarti, &c. ; to Indra) yac cid hi satya somapa anagasta iva smasi, a tu na indra gansaya gosv agvesu gublirisu sahasresu tuvimagha. 2.41.16° (Grtsamada ; to Sarasvatl) ambitame nadltame devitame sarasvati, apragasta iva smasi pragastim amba nas krdhi. The two stanzas show subtle relationship of structure which may be expressed in propor- tional form : 1.29.1, an^astah : a 9ahsaya = 2.41.16, apracastah : pr^astim krdhi. [l.29.2a, giprin vajanam pate: 6.45. iob, indra vajanam pate.] 1.30.7° (Qunahgepa Ajlgarti, &c. ; to Indra) yoge-yoge tavastaram vaje-vaje havamahe, sakhaya indram utaye. 8.21.9° (Sobhari Kanva; to Indra) yo na idam-idam pura pra vasya aninaya tarn u va stuse, sakhaya indram utaye. The stanza 8.21.9 d°es not sound so well as 1.30.7, as regards either sense or metre. 1.30. 8b (Qunahgepa Ajlgarti, &c. ; to Indra) a gha gamad yadi gravat sahasrinibhir utibhih, vajebhir iipa no havam. io.i34.4d (Mandhatr Yauvanagva ; to Indra) ava yat tvam gatakratav indra vigvani dhanus6, rayirii na sunvate saca sahasrinibhir utibhir Ldevi janitry ajljanad bhadra janitry ajljanatj <5®* refrain, 10. i34.ief-6ef 1.30.9a ((^unahgepa Ajlgarti, &c. ; to Indra) dnu pratnasyaukaso huve tuvipratim naram, yam te purvam pita huve. 8.69. i8a (Priyamedha Angirasa ; to Indra) dnu pratnasyaukasah priyamedhasa esam, purvam anu prdyatim vrktabarhiso hitaprayasa agata. 1.30.10° (^unahgepa Ajlgarti, &c. ; to Indra) tarn tva vaydm vigvavara gasmahe puruhata, sdkhe vaso jaritfbhyah. 63] Hymns ascribed to Qunahgepa Ajigarti [ — i. 30.21 3.5i.6d (Vi^vamitra; to Indra) tubhyam bnihmani gira indra tubhyam satra dadhire harivo jusasva, bodhy hpir avaso nutanasya sdkhe vaso jartrbhyo vayo dhah. 8.7 1. 90 (Sudlti Afigirasa and Purumllha Aiigirasa ; to Agni) sa no vasva upa masy urjo napan nnihinasya, sakho vaso jaritrbhyah. It would seem that metre and sense both justify us in assuming that siikhe vaso jaritfbhyo v£yo dhah is the mother pada. See Part 2, chapter 2, class B 8. 1.30.18b ((^unalifepa Ajigarti, &c. ; to Afvins) samanayojano hi vam ratho dasrav amartyah, samudre afvineyate. 5* 75*9li (Avasyu Atreya ; to Afvins) tibhad usa riifatpafur agnir adhayy rtviyah, ayoji vam vrsanvasQ ratho dasrav amartyo Lmadhvl mama 9rutam havam.j 6ft* refrain, 5-75- ie— 90 1.30. 19b (Qunah^epa Ajigarti, &c. ; to A9vins) ny aghnyasya murdhani cakram rathasya yemathuh, pari dydm anyad iyate. 5. 73-3b (Paura Atreya ; to A9vins) Irmanyad vapuse vapu9 cakram rathasya yemathuh, parjr anya nahusa yuga mahna rajahsi dlyathah. For these difficult cosmic-mythological stanzas cf. the recent discussions of Pischel, Ved. Stud. i. 212 ff . ; Ludwig, Ueber Methode, p. 30; Hillebrandt, Ved. Myth. iii. 384, note; Oldenberg, RV. Noten, p. 361 (who cites the parallels from RV. and other sources). Olden- berg’s remarks on anya in 5.73.3® are pertinent. It seems to me also that the word there, not too aptly, agrees with yuga, that is to say, that it has a different meaning than in 1. 30.19. Possibly, therefore, 5.73.3 is reminiscent of earlier treatments of the idea of the 1 other wheel * such as appears in 1.30. 19 or 8.22.4. 1.30.21c ((^unal^epa Ajigarti, &c. ; to Usas) vayarn hi te amanmahy antad A parakat, agve na citre arusi. 4.52.2® (Vamadeva ; to Usas) agveva eitrarusl mata gavam rtavarl, sakhabhud a9vinor usah. Bergaigne, La Syntaxe des comparaisons v6diques (Melanges Renier, p. 75 ff., especially pp. 77, note 1, 88); Pischel, Ved. Stud. i. 91 ff.,have treated the phenomenon of case attraction in comparisons. They show that the primary word in a comparison attracts to its own case- form the secondary, or simile word. On page 92 Pischel remarks that he has found scarcely more than one case of attraction to the vocative, namely, this very pada 1.30. 21°. But he has failed to note the parallel, 4.52.2% which stamps 1.30.21° as imitative. I do not wish to say that the vocative attraction in 1.30. 21 violates any habit, notwithstanding its rareness, especially as Bergaigne, 1. c., p. 80, and Delbriick, Altindische Syntax, p. 106, cite one i. 3o.2i — ] Part 1 : Repeated Passages belonging to Book I [64 more case from the first book, 1.57.3. But of the two repeated padas above one must be the model, and that is 4.52.2% making it likely, after all, that the construction in 1. 30.21 is for the nonce. We must remember here the frequent cases in which the secondary or simile word is in the nominative while the primary word is in the vocative, e.g. 1.16.5 ; 1.36.13 ; 7.13.3, &c. More precisely, therefore, a9ve na in 1. 30.21 imitates a9veva in 4.52.2 ; the inter- dependence of the two is not to be doubted, especially as the cadence of both lines is irregular (y w ii), and it is not to be supposed that two poets would happen upon the same metrical irregularity. 1.30.22c (Qunahfepa Ajlgarti, &c. ; to Usas) tvarii tyebhir a gahi vajebhir duhitar divah, asm6 rayim ni dharaya. 10.24. ic (Vimada Aindra, or others ; to Indra) Lindra somam imam pibaj madhumantam camu sutam, €«*cf. 8.17.1b asm6 rayim ni dharaya vi vo made sahasri'nam puruvaso vivaksase. Cf. the pada, 9uddho rayim ni dharaya, 8.95.8% also octosyllabic, which helps to show that 10.24.1° with refrain is secondary. Group 4. Hymns 31-35, ascribed to Hiranyastupa Angirasa 1.31. 8d (Hiranyastupa Angirasa ; to Agni) tvam no agne sanaye dhananam ya^sam karum krnuhi stavanah, rdhyama karmapasa navena devair dyavaprthivi pravatam nah. 9.69. iod (Hiranyastupa Angirasa ; to Pavamana Soma) l'ndav indraya brhate pavasva sumrllko anavadyb rifadah, bhara candrani grnat6 vasuni devair dyavaprthivi pravatam nah. 10.67. 1 2*^ (Ayasya Angirasa; to Brliaspati) Lindro mahna mahato arnavasyaj vi mQrdhanam abhinad arbuddsya, 8»* 10.67. 1211 Lahann ahim arinat sapta sindhunj devair dyavaprthivi pravatam nah. C*s* 4.28.1° [1.32.1% indrasya nu vliykni pra vocam : 2.21.3d, indrasya vocarh pra krtani viryL] 1.32.3b (Hiranyastupa Angirasa ; to Indra) vrsayamfino ’vrnlta somam trikadrukesv apibat sutasya, a sayakarii maghavadatta vajram ahann enam prathamajam ahinam. 2.15.1° (Grtsamada ; to Indra) pra gha nv asya rnahat6 mahani satya satyasya karanfini vocam, trikadrukesv apibat sutasyasya mfide ahim indro jaghana. 65] Hymns ascribed to Hiranyastupa Angirasa [ — 1.33. 12 [1.32.4°, at suryarii janayan dyixm usiisam : 6.30.5'!, sakam suryarii, &c.] [1.32.5J, ah ih fayata upaprk prthivyah : 10.89. 1 4'!, prthivya apfg amuya 91'iyante.] 1.32.12'1 (Hiranyastupa Angirasa : to Indra) a^vyo varo abhavas tad indra srke yat tva pratyahan deva ekah, ajayo ga ajayah 90ra somam avasrjah sartave sapta sindhun. 2.12.12!' (Grtsamada ; to Indra) Lyah saptara^mir vrsabhas tuvismanj avasrjat sartave sapta sindhun, fris* cf. 2. 1 2. 1 2a yo rauliinam asphurad vajrabahur dyam arohantam sa janasa indrah. Cf. Pischel, Ved. Stud. ii. 91 (improbable suggestion) ; Geldner, ibid. 183. — For echoes of 1.32.12 see perhaps AV. 2.29.7; TS. 6.5. 5. 2; TB. 1.1.8. 3. [1.32.15'', aran na nemih pari til babhuva: 1. 14 1.9'*, ariin na nemih paribhur ajayathah.] Cf. 5.13.6- [1.33.5°, pra yad divo hariva sthatar ugra: 6.41.3°, etarii piba hariva, &c.] 1.33.12° (Hiranyastupa Angirasa ; to Indra) ny avidhyad illbifasya drlha vi frnginam abhinac chusnam indrah, yavat taro maghavan yavad ojo vajrena fatrum avadhih prtanytim. 7-9i.4a (Vasistha ; to Indra and Vayu) yavat taras tanvo yavad ojo yavan naraf caksaslt didhyanah, 9ucim somam 9ucipa patam asme indravayu sadatam barhir edam. Both Ludwig and Grassmann translate the second distich of 1.33. 12 with a diplomatic touch that disguises its plainest sense. The former, 965, ‘ wie gross seine schnelligkeit, Maghavan, wie gross seine gewalt, mit dem keile totetest du den kampfeslustigen feind.’ But the vocative Maghavan shows clearly that it is not a question of the enemy’s, but of Indra’s swiftness and strength. In that regard Grassmann is quite correct : 'Nadi deiner Kraft und Schnelle, macht’ger Indra. erschlugst den Feind, den Kampfer, mit dem Blitz du.’ This rendering, however, in its turn, veils an intrinsic insipidity. What the distich really says, damning Indra with faint praise, is this : ‘ As far as held out thy alertness, 0 Maghavan, and thy strength, thou hast slain the fighting enemy with thy bolt.’ Now 7.91.4 shows that this is indeed the meaning of the passage just discussed. Ludwig, 715 : ‘ Sovil die eigene riirig- keit, die [eigene] starke so vil manner mit einsicht schauend [vermogen] ; trinkt den reinen soma bei uns, o trinker von reinem, Indra und Vayu, sitzt nider auf unserm barhis.’ Grassmann : ‘ Soweit des Leibes Riistigkeit und Kraft reicht, soviel die Manner schau’n mit ihren Blicken, trinkt reinen Soma, ihr des Reinen Trinker, und setzt auf diese Streu euch, Indra-Vayu.’ The stanza has its own difficulties, because it states in turgid style what might have been stated in plain language : the poet invites Indra and Vayu to a protracted drinking- bout which shall last as long as the sacrificers’ bodily endurance holds out, and as long as they can keep their mental faculties unimpaired. I make little doubt that the repeated pada is primary in this connexion, and that it is secondary and weak in 1. 33.12. This weakness the Western translators, in their several ways, instinctively disguise in their versions. 9 [h.o.s. 20] 1.33.14 — ] Part 1 ■' Repeated Passages belonging to Book I [66 1.33.14a+b (Hiranyastupa Angirasa ; to Indra) avah kutsam indra yasmin eak&n pravo yiidhyantam vrsabham dagadyum, gaphacyuto renur naksata dyam lie chvaitreyo nrsahyaya tasthau. i.i74.5a (Agastya ; to Indra) vaha kutsam indra yasmin cakan syumanyu rjrd vatasydgva, Lpra surag cakram vrhatad abliikej ’bhi sprdho yasisad vajrabahuli. i.i74.5c 6.2 6.4b (Bharadvaja ; to Indra) tvam ratham pra bharo yodham rsvam avo yiidhyantam vrsabham dagadyum, tvam tugrarii vetasave sacahan tvam tujim grnantam indra tutoh. See Geldner, Ved. Stud. ii. 171 ; Rigveda-Kommentar, p. 7 ; and cf. under 1.174.5°. 1.34.10b (HiranyastOpa Angirasa ; to Agvins) a nasatya gachatarh huyate havir madhvah pibatam madhup6bhir asabhih, yuvor hi purvam savitosaso ratham rtaya citram ghrtavantam isyati. 4.45. 3a (Vamadeva ; to Agvins) madhvah pibatam madhup^bir asabhir uta priyarn madhune yuhja- tham ratham, a vartanim madhuna jinvathas patho drtim vahethe madhumantam agvina. One may imagine the repeated pada in 1.34. 10 to have been borrowed from a madhu- stanza and a madhu-hymn like 4.45, and equally well one may imagine the same pada expanded gloatingly into the theme of the four padas of 4.45.3. For the connexion between Agvins and madhu see Hillebrandt, Ved. Myth. i. 239 ff. [l.34.11n, a nasatya tribhir ekadagair iha : 8.35.3“, vigvair devais tribhir, &c.] 1.34.11cd (Hiranyastupa Angirasa ; to Agvins) La nasatya tribhir ekadagair ihuj devebhir yatarii madhupeyam agvina, S«“cf. 1.34. na prayus taristam ni rapansi mrksatam s6dhatam dv^so bhavatam sacabhuva. x* 1 57* 4cd (Dlrghatamas Aucathya ; to Agvins) La na urjam vahatam agvina yuvaihj madhumatya nali kagaya mimik- satam, Cw* 1.92.17° prayus taristam ni rapansi mrksatam sedhatam dvdso bhavatam sacabhuva. 1.34.12'1 (Hiranyastupa Angirasa ; to Agvins) a no agvina trivfta rathenarvahcam rayiiii vahataih suviram, grnvanta vam avase johavimi vrdh6 ca no bhavatam vajasatau. i.ii2.24d (Kutsa ; to Agvins) apnasvatlm agvina vacam asme krtarii no dasra vrsana manlsam, adyQtye ’vase ni hvaye vam vrdhd ca no bhavatam vajasatau. The word adyutya in i.ua.24d seems to me to mean ‘ darkness or ‘ trouble rather than 1 unlucky gambling as the Pet. Lexicons and the translators assume. 67] Hymns ascribed to Hiranyastupa Angirasa [ — 1.36.5 [l.35.2c. hirany¥a savita rathena: 4.44.5%hiranyayenasuvftilr&thena; 8.5. 35a, hiranyayena rdthena.] [1.36.8°, hiranyaksuh savita deva agat: 2.38.4% aramatih savita, &c.] [1.35.8% dadhad ratnR daguse vitryani: see under 1.47. ib.] [1.35. 9b, ubho dyavaprthivi antar lyate : 1. 160. i°, sujanmanl dhisane an tar lyate.] See the context of each stanza. 1.36.10*5 (HiranyastOpa Angirasa ; to Savitar) hiranyahasto asurah sunlthah sumrlikah svavan yatv arvan, apasedhan raksaso yatudhanan asthad devah pratidosam grnanah. 1. 1 1 8.i b (KaksTvat Dairghatamasa, son of U?ig ; to Afvins) a vam ratho afvina ^yendpatva sumrlikah svdvan yatv arvan, yo nnirtyasya manaso javlyan Ltrivandhuro vrsana vataranhuh.j 1. 1 i8.i‘> The epithet ‘tenderly merciful’ (sumrllkii) is applied to Savitar in 1.35. 10, to the Akins’ chariot in 1.118.1. There can be no doubt that the repeated pada is primary in the former. Cf. the relation of 1.108.1 to 7.61.1 (under 1.108.1). On the other hand the fourth pada of 1.118.1 has a parallel in 1.183.1, to wit : tarn yunjatham manaso yo javlyan trivandhuro vrsana yas tricakrah, yenopayathah sukrto duronam tridhatuna patatho vir na parnfiih. From the point of style, or expression, 1.183.1, especially its first distich, seems, in turn, decidedly inferior and afterborn in relation to 1.118.1; the three stanzas therefore may involve a case of double relative age : 1.35.10; 1.118.1 ; 1.183.1. — For pratidosam in 1.35. 10 see Ludwig, 131 (who suggests pr£ti dosam); Bartholomae, Bezz. Beitr. xv. 198, note. [l.35.11d, raksa ca no adhi ca brQhi deva ; 1.1 14.10°, mrla ca, &c.] Group 5. Hymns 36-43, ascribed to Kanva Ghaura [1.36.3% pra tva diitam vrnimahe : 1.12.1®, agnim dutam vrnlmahe ; 1.44.3% adya dutam vrnl mahe.] Cf. 8.io2.i8b. 1.36.3b: i.i2.ib; 1.44.7% botararh vi^vavedasam. [1.36.4% devdsas tva varuno mitro aryama : 1.40.5°, yasminn indro varuno, &c. ; 7.66.12°, yad ohate varuno, &c. ; 7.82. ioa ; 83.10% asme Indro varuno, &c. ; 8.19.16°, yena caste varuno, &c. ; 8.26.11% sajosasa varuno, &c. ; x 0.36.1% dyavaksama varuno, &e. ; 10.65.1% agnir indro varuno, &c. ; 10.65.9% indravayu varuno, &c. ; 10.92.6°, tebhif caste varuno, &c.] 1.36.5b (Kanva Ghaura ; to Agni) mandro hota grhapatir agne dut6 vigam asi, tve vigva samgatani vrata dhruva yani deva akravata. 1 .36.5 — ] Part 1 : Repeated Passages belonging to Book 1 [68 1.44.9’’ (Praskanva Kanva ; to Agni) patir liy adhvaranam agne duto vigam asi, usarbudha a vaha somapltaye devan adya svardr^ali. For 1.44.9* cf. the padas, rajantam adhvaranam, &c., under 1.1.8. 1.36.7ab (Kanva Ghaura ; to Agni) tam ghem ittha namasvina upa svarajam asate, hotrabhir agnim. manusah sam indhate titirvSnso ati sn'dhah. 8.69.17^ (Priyamedha Angirasa ; to Indra) tam ghem ittha namasvina upa svarajam asate, artham cid asya sudhitam yad etava avartayanti davane. For 1.36.7° cf. 2.2.8°; xo.n.5b, hotrabhir agne manusah svadhvarah. 1.36.8b (Kanva Ghaura ; to Agni) ghnanto vrtram ataran rodasl apa uru ksayaya cakrire, bhuvat kanve vrsa dyumny ahutah krandad a?vo gavistisu. 7.60. 1 id (Yasistha ; to Mitra and Varuna) yo brahmane sumatim ayajate Lvajasya satdu paramasya rayah,j 6®* cf. 4.i2.3b siksanta manyurii maghavano arya uru ksayaya cakrire sudhatu. Cf. 6.50.3 ; 8.68.12. 1.36.10’’ (Kanva Gham’a ; to Agni) yam tva devaso manave dadhiir iha yajistham havyavahana, yam kanvo medhyatithir dhanasprtam yam vrsa yam upastutah. i.44.5d (Praskanva Kanva : to Agni) stavisyami tvam aham vifvasyamrta bhojana, agne trataram amftarh miyedhya yajistham havyavahana. 7.15.6° (Vasistha Maitravaruni ; to Agni) semam vetu vasatkrtim agrn'r jusata no girah, yajistho havyavahanah. 8.19.21° (Sobhari Kanva; to Agni) lie gira manurhitaih yarn deva dutam arati'm nyerire, yajisthaih havyavahanam. [l.36.12d, sa no mrla mahan asi: 4.9.1“, agne mrla malian asi.] 1.36.14° (Kanva Ghaura ; to Agni) urdhvo nah pahy anhaso nf ketiina vi'fvam sam atrinam daha, krdhi na urdhvan carathaya jivase vida devesu no duvah. 1.172.3° (Agastya ; to Maruts) trnaskandasya mi vifah pari vrhkta sudanavah, urdhvan nah karta jivase. [ — i-37-it 69] Hymns ascribed to Kanva Ghaura 1.36. 15ab (Kanva Ghaura ; to Agni) pahi no agne raksasah pahi dhurtdr aravnah, pahi rfsata uta va jighahsato bfhadbhano yavisthya. 7.i.i3ab (Vasistha Maitravaruni ; to Agni) pahi no agne raksaso ajustat pahi dhurtdr araruso aghayoh, tva yujd prtanayunr abhi syam. It seems pretty clear that the fuller form of the distich, 7. 1.13*’’, whose author is said to be Vasistha, is the primary form. For the cadence of 1.36. 15* ( oii) is a severe infringement, of course not unparalleled, of metrical law, whereas 7.1.13“ is unexceptionable. In 1.36. 15b aravnah (catalectic dipody) cleverly takes the place of araruso (o ^ — ). In 8.60. io4, pahi vi^asmiid raksaso aravnah, we seem to have a tertiary descendant from this distich, namely a contraction of 1.36. i3,b. 1.37.4C (Kanva Ghaura ; to Maruts) pra vah pardhaya ghrsvaye tvesadyumnuya $usmine, devattam brahma gayata. 8.32.27° (Medhatithi Kanva ; to Indra) pra va ugniya nistiire ’salhaya prasaksine, devattam brahma gayata. We render 1.37.4/ Sing for yourselves a god-given song to the tiery host (of the Maruts), the brilliantly luminous, the mighty ! ’ The word vah in the second place is that immensely common vah in just that position (the second word of the stanza, e. g. 5.52.4 ; 6.10. 1 ; 16.22 ; 8.19.7; 62.16; 71.12), an enclitic dative of interest, quasi German, ‘singt eucli ’ (cf. Bezz. Beitr. xxvii. 268). None of the translations do justice to this subtle idiom : see Grassmann, ii. 40 ; Ludwig, 673 ; Max Muller, SBE. xxxii. 63 (where older renderings are quoted). Ludwig is enticed by this use of vah to take gayata as passive : ‘ Eurer kttnen schar, von blendender herlichkeit, der kraftvollen, soli ein von den gOttern eingegebenes brahma gesungen werden.’ In his commentary he retracts the lapsus, and follows the other translators. The other stanza involves a remarkable type of repetition : ‘Sing for yourselves a god- given song to the strong, conquering, unconquered, overwhelming (Indra) ! ’ Ludwig, 598, not having in mind his rendering of the repeated pada in 673, translated gayata as active, ‘ singt das den gSttern entnommene brahma eurem gewaltigen &c. Grassmann: ‘ Auf eurem starken . . . singt das gottverliehene Gebet.’ Now Aufreclit in the Preface of the second edition, p. xxv, notes the repetition of the pada, and remarks : 1 Das gottgegebene brahma hat der Nachahmer (the author of 8.32.7) geraubt.’ Aufrecht thus judges, because he presumably construes vah in 1.37.4 as referring to the Maruts ; then, finding the same plural in a parallel stanza to Indra, he condemns the latter as an imitation. But with our con- struction of vah the supposed reference to the Maruts is cancelled, and, as far as I can see, one stanza is as good as another. 1.37.1a, 5b, krllarii vah £ardho (sb, krihirii yac chardho) mdrutam. [1.37. 8C, bhiya yiSmesu rejate (sc. prtliivi) : 8.20.5°, bhumir yamesu rejate.] I.37.11c (Kanva Ghaura ; to Maruts) tyam cid gha dlrgham prthum miho napatam amrdhram, pra cyavayanti yamabhih. 1.37-h — ] Part 1: Repeated Passages belonging to Book I [70 5.56.44 (Qyavafva Atreya; to Maruts) ni ye rinanty ojasa vftha gavo na durdhurah, afmanarii cit svaryam parvatam girim pra cyavayanti yamabhih. We may render 1. 37.11, ‘Verily, even that long and broad child of the cloud (the rain) that does no injury, they cause to fall in their course ’. All translators agree on some such sense: Ludwig, 673 ; Grassmann, i. 41; Max Muller, SBE. xxxii. 64. For miho napat cp. Bergaigne, ii. 18, 46, 256, note 2. For the first distich of 5.56.4 cf. Geldner, Yed. Stud. i. 117. The second distich is perhaps, though by no means certainly, interpreted by 1. 37.11 : a^manam cit svaryam (also 5.30.8°), ‘ the heavenly stone ’ may be lightning ; parvatam girim, 1 cloud-mountain ’. I render : ‘ Even the heavenly stone (lightning) (and) the (cloud) moun- tain they cause to fall (as rain-storm) in their course.’ Therefore again rain-storm (for parvatarii girim cf. Bergaigne, i. 258). This, if, indeed, it be correct, does not appear quite clearly enough in Ludwig, 690 : ‘ sogar den himlischen keil, den fels, den berg, auf ihren ziigen stiirzen sie.’ Grassmann, i. 208 : 1 Des Himmels Felsen auch und den gewalt’gen Berg erschiittern sie durch ihren Gang.’ Max Muller, ibid. p. 337 : ‘ they by their marches make the heavenly stone, the rocky mountain (cloud) to shake.’ For both stanzas cf. 8.7.4, vapanti maruto miharn pra vepayanti parvatan, yad yamam yanti vayubhih. Though the parallels seem to call in both stanzas for cloud mountains, we must not forget that the Maruts shake also real mountains, trees, and so on; e.g. 1.37. 12; 1.39.5; 1.85.4. In that case Grassmann’s translation comes closest to the sense of the original. 1.37.12a (Kanva Ghaura ; to Maruts) maruto yad dha vo balam janan acucyavltana, girmr acucyavltana. 8.7. 1 ia (Punarvatsa Kanva ; to Maruts) maruto yad dha vo divah sumnayanto havamahe, a tu na lipa gantana. The anacoluthic quality of 1.37.12° suggests the question whether its similarity to 8.7.11* is accidental, especially as several stanzas of 1.37; 1.38; and 1.39 have padas repeated in 8.7 (1.38.1“ : 8.7.31* ; 1.39.5* : 8.7-4b; i.39.6b : 8.7. 28b). Ludwig, 673, renders 1.37. 1 2, ‘0 Marut, so wie eure kraft ist, warft ihr die leute nider, warft ihr die berge nider ’. Oldenberg, SBE. xlvi. 172 : ‘ O Maruts, with such strength as yours, you have caused men to tremble.’ Other renderings in Max Muller, SBE. xxxii. 79. Oldenberg illustrates the anacoluthon by com- paring 1. 147. 3. If the correspondence between 1.37. 12* and 8.7.1 1* is not, after all, fortuitous, the former, of course, is the epigonal piida. Note the enclisis of acucyavltana after the relative pronoun yad, which heightens the anacoluthic effect. 1.38.1n (Kanva Ghaura ; to Maruts) kad dha nunaih kadhapriyah pita put ram na hastayoh, dadhidhve vrktabarhisah. 8.7.3ia (Punarvatsa Kiinva ; to Maruts) kad dha nunam kadhapriyo yad indram ajahatana, ko vah sakhitva ohate. Recent discussions of kiidliapri, and the like, by Piscliel, ZDMG. xxxv. 714 ; Geldner, Ved. Stud. iii. 64 ; Oldenberg, RV. Noten, p. 26. In 8.8.4 the Akins’ epithet adhapriyfi (dual) seems to mean ‘then-friends’, something like ‘reliable friends’; therefore kadhapriyah means ‘ when-friends ’, i.e. ‘uncertain, fickle, or capricious friends’; in 1.30.20 kadhapriye 71] Hymns ascribed to Kanva Ghaura [ — 1.39.6 (enclitic) seems to be vocative feminine singular of a transition form kadhapriya, derived from kadlmprl. In the stanzas above the repeated pada fits equally well in both cases (see the sequel in 1.38). The hymns 1.38 and 8.7 are otherwise related as to authorship (see the preceding item) ; I am unable to discover any indication as to priority. Perhaps we may render 1.38.1 : ‘ What is up with you now, ye tickle friends ? As a father his son in his arms so have ye been placed (accommodated), O ye (gods) for whom the barhis is prepared.’ For the middle of root dhii in passive sense see, e. g., 1.24.4. The stanza would then seem to express surprise or disgust because the Maruts do not respond to kind treatment. The idea is continued effectively in the next five stanzas. If, however, dadhidhve is to be taken as active we may render: ‘What now, ye fickle friends, did you, like a father his son in his arms, place (us) ? &c.’ Again complaint, expressed rhetorically in question form, at the neglect of the Maruts. The implication would be that the Maruts did not cherish their worshippers, as might properly be expected of them. 1.39.6a+‘1 (Kanva Ghaura ; to Maruts) pra vepayanti parvatan vi vihcanti vanaspatln, pro arata maruto durmada iva d6vasah sarvaya viga. 5.26.9c (Vasuyava Atreyah ; to Vigve Devah) edaih maruto agvina mitrah sldantu varunah, devasah sarvaya viga. 8.7.4b (Punarvatsa Kanva ; to Maruts) vapanti maruto mihaiii pra vepayanti parvatan, yad yumam yanti vayubhih. Note that 1.39 and 8.7 share another pada ; see under i.39.6b. For 8.7.4 see under l.37-”°- [l.39.6a, upo rathesu prsatlr ayugdhvam : 1.85.5®, pra yad rathesu prsatlr ayugdhvam.] 1.39.61’ (Kanva Ghaura; to Maruts) Lupo rathesu prsatlr ayugdhvam j prastir vahati rohitah, tar cf. 1.39.6® a vo yamaya prthivi cid agrod ablbhayanta manusah. 8.7.28 (Punarvatsa Kanva; to Maruts) yad esam prsati rathe prastir vahati rohitah, yanti ubhra rinann apah. We render 1.39.6, ‘ And ye have hitched the spotted mares to your chariot ; a red stallion draws as leader. Even the earth hath listened at your approach, and men were frightened ’. Cf. Ludwig, 675 ; Grassmann, ii. 43 ; Max Muller, SBE. xxxii. 97. The word prsatlr which the translators render by ‘antelopes ’ means in fact ‘spotted mares’, because the Maruts have the epithet prsadagva. See Bergaigne, ii. 378, and, very explicitly, Naighantuka 1 . 1 5 ; Brhad- devata 4.144 (catalogue of the spans of the gods), where we have the express statement, prsatyo ’gvas tu marutam. The word prasti (pra + sti, like abhisti, lipasti, and paristi) means literally ■ being in front ’, ‘ leading horse ’. It is the analogue of purogava, and TrpioPvs, ‘ leading steer ’. Both refer to what is known as a ‘ spike-team ’, or ‘ unicorn ’. To a team of two animals a third is hitched in front for better control. See the author in American Journal of Philology, xxix. 78 ff. The second stanza may be rendered, ‘ When the red stallion guides as a leading horse their speckled mares at the chariot, then the bright chariots approach and let the waters i.39-6 — ] Part 1 : Repeated Passages belonging to Book I [72 flow Subtly, and yet in a peculiarly certain way, this stanza is secondary, directly patterned after 1.39.6. The entire characteristic and imaginative description of the span of the Maruts in 8.7.28 is crowded incidentally, as it were, into a subordinate clause (note orthotone v£hati in 8.7.28 ; enclitic vahati in 1.39.6), whereas in 1.39.6 the description is the set theme of the first distich. I cannot doubt that this important bit of mythography was first stated in the explicit terms of 1.39.6, before it could be referred to incidentally, yet in the very same words, in 8.7.28. The same conclusion, namely priority of 1.39.6% applies to the partial relation of 1.39.6* to 1.85.5* (in neither of which padas, by the way, we should read ayugdhuam with Oldenberg, RV. Noten, pp. 41, 85). The original description was categorical, and not subordinate ; cf. also 5.57.3d. For the general character of 8.7 see p. xv, line 15 from below. — Note that 1.39 shares another pada with 8.7 ; see under 1.39.5. [1.39.7b, riidra avo vrnlmahe : 1.42.56, pusann avo, &c.] 1.40.21’ (Kanva Ghaura ; to Brahmanaspati) tvam id dhi sahasas putra martya upabrutd dhand hitd, suviryam maruta a svagvyam dadhlta yo va acake. 6.6 1. gb (Bharadvaja; to Sarasvatl) yas tva devi sarasvaty upabrut6 dhan6 hit6, indram na vrtraturye. 1.40.4a+b+<1 (Kanva Ghaura ; to Brahmanaspati) yd vaghate dadati sunaram vasu sa dhatte aksiti gravah, tasmai ilam suvlram a yajamahe supraturtim anehasam. 5.34. 7b (Sarhvarana Prajapatya ; to Indra) sam irii panel- ajati bhojanam muse vi dacjuse bhajati sunararia vasu, durge cana dhriyate vifva a puru jano yo asya tavislm acukrudhat. 8. ic>3.5b (Sobhari Kanva ; to Agni) ssi drdhe cid abhi trnatti vajam arvata sa dhatte aksiti .59-9d (Bharadvaja ; to Indra and Agni) IndrSgnl yuvor api vasu divyani parthiva, a na iha pra yachatam rayim vigvayuposasam. 1.79.12b (Gotama RahOgana ; to Agni) sahasrakso vicarsanir agni raksansi sedhati, hota grnlta ukthyhh. 7.i5.ion (Vasistha Maitravaruni ; to Agni) agni raksansi sedhati ^ukrafocir amartyah, L9\icih pavaka ldyah.j €«* 2.7.4“ Note that 1.79.4s = 7.15.11'. 1.80.ie-18e, arcann anu svarajyam. 1.80. 6b (Gotama Rahugana; to Indra) adhi sanau ni jighnate vajrena gataparvana, mandana Indro andhasah sakhibhyo gatum ichaty Larcann anu svarajyam. j iff refrain, 1.80. xe-i6e 8.6.6b (Vatsa Kanva ; to Indra) vi cid vrtrasya dodhato vajrena gataparvana, giro bibheda vrsnina. 8.76.2° (Kurusuti Kanva ; to Indra) ayam Indro marutsakha vi vrtrasyabhinac ehiralji, vajrena gataparvana. 8.89.3d (Nrmedha Ahgirasa and Purumedha Angirasa ; to Indra) pra va Indraya brhate maruto brahmarcata, vrtram hanati vrtraha 9atakratur vajrena gataparvana. Note the correspondence of 8.6.38* with 8.76. 1 1*, and also the occurrence of the expression vrtrasya dodhatah in 1.80.5 as well as in 8.6.6, [1.80.8°, mahat ta indra \Tryam: 8.55(Val. 7). ic, bhund Indrasya vxryam.] 1 3 [h.o.s. 20] 1.80.9 — J Port 1: Repeated Passages belonging to Book I [98 1.80.9d (Gotama Rahbgana; to Indra) sahasram sakam areata pari stobhata vinfatfh, ^atdinam anv anonavur indraya brahmddyatam Larcann anu svarajyam.j refrain, i.8o.ie-i6e 8.69. 9d (Priyamedha Ahgirasa ; to Indra) ava svarati gargaro godha pari sanisvanat, pi'nga pari caniskadad indraya brahmddyatam. For 1.80.9 see Neisser, Bezz. Beitr. xx. 67 ; for 8.69.9, Hillebrandt, Yed. Myth. ii. 237, [1.80.10c, mahat tad asya paunsyam : 8.63.3°, stuse tad, &c.] 1.80.10d (Gotama Rahugana ; to Indra) l'ndro vrtrasya tavislm m'r ahan sahasa sahah, Lmahat tad asya pauhsyaihj vrtram jaghanvan asrjad Larcann dnu svarajyam.j c : cf. 1.80.10°; e : refrain, 1.80. ie-i6e 4. i8.7d (Samvada Indraditivamadevanam) kim u svid asmai nivido bhanantendrasyavadyam didhisanta apah, mamaitan putro mahata vadhena vrtram jagbanvan asrjad vi sindhun. 4. 19.8b (Vamadeva ; to Indra) purvfr usasah faradaf ca gurta vrtram jaghanvan asrjad vi sindbun, paristhita atrnad badbadhanah slra l'ndrah sravitave prthivya. This case is remarkable, because it is both definite and simple. The pada i.8o.iod fails to end in an iambic dipody, and its verb has no object. Ludwig, 460, translates diplomatically ‘ als er den Vrtragetotet liess er fliessen’ ; Grassmann, more freely, 1 schlug Vrtra und ergoss die Fluth’. But the parallels show that the pada is the truncated torso of another pada, regular in its final cadence and the preceding anapaest, and duly furnished with that object which every reader of this Veda would supply anyhow, namely sindhun ; from these a later poet over-familiarly has extracted the short form to suit his metre. Cf. also Oldenberg, RV. Noten, p. 83, to RY. 1.82.2. See Part 2, chapter 2, class B 9. (1.81.5a, a paprau parthivam rajah: 6.61. nab, apapriisl parthivany urii rajo antariksam.] 1.81.5cd (Gotama Rahugana ; to Indra) La paprau parthivam rajoj badbadhe rocana divi, 6S* cf. i.8i.5a na tvavan indra ka• 359- 1.92. 16c (Gotama Ralitigana ; to A9vins) a^vina vartir asmad a gbmad dasra hiranyavat, arvag ratham samanasa ni yachatam. 7.74.2° (Vasistha ; to Alvins) yuvarii citram dadathur bhojanam nara codetham sQnftavate, arvag ratham samanasa ni yachatam Lpibatarii somyam miidhu.j fro* 6.60. 1 5J 8.35. 22ft (^yavafva Atreya; to Afvins) arvag ratham ni yachatam jnbatarii somyam m£dhu,j 8»* 6.60. 1511 a yatam afvina gatam avasyur vam aham huve dhattam ratnani dafiise. The rigmarole of 8.35.22, repeating, as it does, two padas of 7.74.2, seems late and imitative. 1.92.17° (Gotama Rahugana ; to A9vins) yav ittha 9lokam a divo jyotir janaya cakrathuh, a na urjam vahatam agvina yuvam. 1. 157.4“ (Dlrghatamas Aucathya ; to A9vins) a na urjam vahatam agvina yuvam madhumatya nah ka9aya mimik- satam, Lprayus taristaih ni rapansi mrksatam sedhatam dveso bhavatam saca- bhuva.j fro* 1.34. n°d For i.i57-4b cf. the entire stanza 1.22.3. 1.92.18b (Gotama Rahugana ; to A9vins) eha deva mayobhuva dasra hiranyavartani, uusarbudho vahantu somapltaye.j So* cf. 1.92.18° 5.75.2° (Avasyu Atreya ; to A9vins) atyayatam a9vina tiro vi'9va aham sana, dasra hiranyavartani susumna sindhuvahasa Lmadhvl mama 9rutam havam.j ear refrain, 5.75. ie— 9e 8.5. 1 ib (Brahmatithi Kanva; to A9vins) vavrdhana Lgubhas patij dasra hiranyavartani, Lpibatam somyam madhu.j 8a* 6.6o.i5d 1.92.18 — ] Part 1 : Bepeated Passages belonging to Book I [108 8.8. ic (Sadhvahsa Kanva ; to Agvins) La no vigvabhir utibhirj Lagvina gaehatam yuvam,j 6s*a: 7.24.4a; b: 5-75-3b dasra hiranyavartani Lpibatam somyam madhu.j 6.60. i5d 8*87.5° (Dyumnlka Vasistha, ox1 others ; to Agvins) La nunam yatam agvinagvebhih prusitapsubhih,_, tsr a : 8.8.2a ; b : 8.i3.nb dasra hiranyavartani gubhas patl Lpatam somam rtavrdha. j 1.47.38 Cf. rudra hiranyavartani 5.75.3°. There can be no doubt that the composite pada 8.87.5° marks the stanza as late. Note the enclisis of 9ubhas patl, in connexion with the orthotone vocatives preceding it (contrary to 1.3.1), due to numerous passages in which this expression occurs without other vocatives in the final iambic dipody of dodecasyllabic padas. Cf. Olden- berg, RV. Noten, Index, p. 427 (Vokativbetonung). [1.92.18°, usarbudho vahantu somapltaye: 8.i.24d, vahantu soxnapltaye.] 1.93.2d (Gotama RahQgana ; to Agni and Soma) agnlsoma yo adya vam idam vacah saparyati, tasmai dhattarh suviryam gavam pdsaria svagvyam. 9.65. 1 7b (Bhrgu Yaruni, or Jamadagni Bhargava ; to Soma Pavamana) a na indo gatagvi'nam gavam p<5sam svagvyam, vaha bhagattim utaye. 1.93.3d (Gotama Rahugana ; to Agni and Soma) agnlsoma ya ahutim yo vam dagad dhavi'skrtim, sa prajaya suviryam vigvam ayur vy agnavat. 8.31. 8b (Manu Vaivasvata; Daxiipatyor agisah) putrina ta kumarina vigvam ayur vy agnutah, ubha hi'ranyapegasa. 10.85.42h (SQrya Savitrl ; to Surya) l'haiva stam ma vi yaustam vigvam ayur vy agnutam, krilantau putrair naptrbhir modamanau sve grhe. These and similar formulas carry on a lively existence in AV. and the Yajus-texts ; see my Vedic Concordance under vigvam ayur, &c. [l.93.4a, agnlsoma ceti tad vlryarii vam: 3.12.9c tad vam ceti prd vlry&rn. ] 1.93.6d (Gotama RalxQgana ; to Agni and Soma) anyarii divo matarigva jabhaiumathnad anyam pari gyeno adreh, agnlsoma brahmana vavrdhandrum yajnaya cakrathur u lokam. 7.99.4“ (Vasistha; to Indra and Visnu) uriim yajfiaya cakrathur u lokam janayanta suryam usasam agnfm. dasasya cid vrsagiprasya mfiya jaghnathur nara prtaniijyesu. L — 1 *95*5 109] Hymns ascribed to Gotama Bdhugana 1.93. 8^ (Gotama KiihQgana ; to Agni and Soma) y6 agnisoma havisil saparyad devadncfi manasa y<3 ghrtena, tasya vratam raksataih patam anhaso vigd janaya mahi garma yachatam. 7.82. ib (Vasistha ; to India and Varuna) tndravaruna yuvam adhvaraya no vigd janaya mahi garma yachatam, dlrghaprayajyum ati yo vanusyati vayarh jayema pftanasu dndhykh. Group 11. Hymns 94-115, ascribed to Kutsa Angirasa 1.94.1'*-14d, agne sakhye ma riskma vayiim tava. 1.94.3*' (Kutsa Angirasa ; to Agni) gakema tva samidharii sadhaya dhiyas tv6 deva havir adanty ahutam, tvam adityan a vaha tan hy ugmasy Lagne sakhy6 ma risama vayarii tava.j Crw” refrain, 1.94. id-i4d 2. 1 . 1 3d (Grtsamada Bhargava (^aunaka, formerly Angirasa Qaunahotra ; to Agni) tvam agna adityasa asyarii tvam jihvam gucayas cakrire kave, tvam ratisdco adhvaresu sagcire tv6 deva havir adanty ahutam. The repeated pada seems well enough in both stanzas. In 2.1. 14 the idea is taken up a second time concatenatingly, and with vast expansion : tve (agne vigve amrtaso adruha iisa) deva havir adanty ahutam. Most of the words which I have bracketed are little more than empty glosses; cf. vigve devSso adruhah, 1.19.3; 9.102.5. We may assume with some con- fidence that this stanza at least is the work of an after-poet. This does not of itself establish the secondary origin of 2.1. 13 (Aufrecht, Preface to Rig-Veda, Second Edition, p. xxiv), because the concatenating stanza may have been added by a later hand, as a kind of gloss ; see Bloomfield, The Atharva-Veda, p. 43, bottom. Still 2.1 is a litany every pada of whose first twelve stanzas begins with a case-form of stem tva 1 thou ’ ; such a composition is, to say the least, not primary. I am therefore inclined to think that Aufrecht’s conception of the relative age of our repeated pada is correct. [1.94.13°, garman syama tava saprathastame : 5.65. 5b, syama saprathastame.] 1.94.16°'*; 95.1 i°d ; 96.9°'*; 98.3c11; ioo.i9°d; io2.n°d; 103. 8^ ; 105. 19cd; 106.7^*; 107. 3cd; 108.13^; 109. 8°d; 110.9°'*; 112.25^; ii3.20°d; 114.11°'*; U5.6°d; 4.97.58°d, tan no mitro varuno mamahantam aditih sindhuh prthivi uta dyauh. [l.95.5b, jihmanam urdhvah svayaga upasthe; 2.35.9b, jihmanam urdhvo vidyutam vasanah.] i.95-8 — ] Part 1: Repeated Passages belonging to Book I [110 1.95.8a (Kutsa Angirasa ; to Agni) tvesam rupam. krnuta uttaram yat samprncanah sadane gdbhir adbhih, kavir budhnam pari marmrjyate dhih sa devatata samitir babhuva. 9.71.8 (Rsabha Vaifvamitra ; to Pavamana Soma) tvesam rupam krnute varno asya sa yatragayat samrta sedhati sridhah, apsa yati svadhaya daivyam janam sam sustuti nasate sarh goagraya. For 1.95.8 see Bergaigne, i. 73, and especially ii. 67 ; Oldenberg, SBE.xlvi. x 1 5, 118 ; RV. Noten, 95; for 9.71.8, Bergaigne, i. 162, 176, 189; iii. 172, and especially ii. 67. Notwith- standing the bizarre mysticism of both stanzas, we can see that they are related beyond the verbal similarity of their opening padas. In 1.95.8° ‘sage prayer’ (kavir . . . dhih, hendia- dyoin) purifies Agni’s foundation, after his highest part has assumed brilliant colour (pada a), so that it becomes the meeting-place among the gods (devatata). In 9.7I.81 Soma assumes brilliant colour, goes to the divine folk (pada b) and there associates himself with praise accompanied by kine (sustuti . . . goagraya, pendant to kavir . . . dhih). We may never find out what is the exact value of these mystic utterances of the Rishis, but in any case, these two stanzas which assimilate Agni and Soma, must be considered in their mutual relation, before explanation of either is possible. 1.95.11 = 1.96.9 (Kutsa Angirasa ; to Agni) eva no agne samidha vrdhano revat pavaka gravase vi bhahi, tin no mitrd varuno mamahantam aditih sindbuh prthivi uta dyauh. The second hemistich is refrain in i. 94.16011 ff. 1.96.1(l-7d, deva agm'm dharayan dravinodam. 1.96.6a (Kutsa Angirasa ; to Agni) ray 6 budhnah samgamano vasunam yajnasya ketur manmasadhano veh, amrtatvam raksamanasa enarn Ldeva agnim dharayan dravinodam. j refrain, 1.96. id-7<1 10.139.3“ (Vigvavasu Devagandharva ; to Surya) rayd budhnah samgamano vasunam vfgva rupabhi caste gaclbhih, Ldeva iva savita satyadharmejndro na tasthau samare dhananam. 1 0.34.8b The entire stanza 10.139. 3 is rubbish without real sequence, certainly secondary to 1.96.6. Cf. under 1.73.2°. 1.96. 8n, dravinoda dravinasas turasya: 1. 1 5-7a, dravinoda dravinasah. 1.96.9 = 1.95.H. 1.97.1a, ic-8°, apa nah fofucad agham. [1.97.3b, prasmakasa? ca surayah : 5.10.6°, asmakasag ca sOrayah.] 1.97.6b; 1.1.4b, vifvatah paribhur asi. Ill] Hymns ascribed to Kutsa Angirasa [ — i. 100.15 1.98.2'l+<1 (Kutsa Angirasa ; to Agni, or Agni ViUgvanara) prstb divi prstb agnih prthivyam prsto vigva osadhlr a vivega, vaifvanarah sahasa prsto agnih sa no diva sa risah patu naktam. 7.5.2“ (Vasistha ; to Vfiigvanara) prstb divi dhayy agnih prthivyam Lneta sindhunfuii vrsabha stiyanam,j C«* 6.44.2ib sa manusTr abhi vigo vi bhati vaigvanaro vavrdhano varena. 10.87. id (P5yu Bharadvaja ; to Agni Raksohan) raksohanaiii vajinam a jigharmi mitram prathistham upa yami ^arma, ^i^ano agnih kratubhih samiddhah sa no diva sa risah patu naktam. Note that a variant of 7-5.ab, vrsa sindhunam vrsabha stiyanam, is addressed in 6.44.2 ib, to Indra (more fittingly?). For the entire item see Oldenberg, RV. Noten, p. 95 ; and note AV. 2.2.2*, divi sprstd yajatah sdryatvak. [1.99.1°, sa nah parsad ati durgani vi'9va : 1.89. 2b ; 10.56. 7J, svastibhir ati, &c.] 1.100.1d-16d, manitvan no bhavatv indra Qti. 1.100.11° (Rjra9va, or others ; to Indra) sa jamibhir yat samajati milhe ’jamibhir va puruhQta evaih, apam tokasya tanayasya jesd Lmarutvan no bhavatv indra txtf.j Cw'i’efrain, i.ioo.id-i5d 6.44.18° (^amyu Barhaspatya ; to Indra) asu sma no maghavann indra prtsv Lasmabhyarii mahi varivah sugam kah;J 6^*1.102.4° apam tokasya tanayasya jesa indra surln krnuhi sma no ardham. For the meaning of the repeated pada see Bergaigne, ii. 177, note, 185, note ; for 6.44.18, Noisser, Bezz. Beitr. vii. 233. 1.100.12b (Rjra9va Varsagira, and others ; to Indra) sa vajrabhfd dasyuha bhrma ugrah sahasracetah gatanitha rbhva, camrlso na 9avasa pancajanyo Lmanitvan no bhavatv indra uti.j 6s= refrain, i.ioo.id-xs11 10.69. 7b (Sumitra Badhrya9va ; to Agni) dlrghatantur brhaduksayam agnih sahasrastarih gatanitha rbhva, dyuman dyumatsu nrbhir mfjyamanah sumitresu dldayo devayatsu. If we compare i.ioo.i2b with the general drift of 3.60.7 and 8.96.18 it will seem quite clear that the locution sahasracetah 9atamthah in 1.100.12 is superior and prior to the insipid sequence sahasrastarih catanlthah in 10.69.7. Tl1® change in the latter stanza is due to the suggestion of brhaduksa in pada a. The case is one of the clearest in relative chronology. [1.100.15b, apag cana gavaso antam apuh : 1.167.9^ arattac cic chavaso, &c.] 1. 1 00.19 — ] Part 1: Repeated Passages belonging to Book I [112 1.100.19 (Rjrafva ; to Indra) = 1. 102. ii (Kutsa ; to Indra) vigvah.6ndro adhivakta no astv aparihvrtah sanuyama vajam, Ltan no mitro varuno mamahantam aditih slndhur prthivi uta dyauh.j 6»* refrain, i.94.i6c fg : refrain, 10.133.1fe Grassmann renders 8.21.13 : ‘Denn du bist ja von Hause aus ganz ohne Vettern, Indra und Verwandtschaft auch ; durch Kampf begehrst Vorwandte du.’ Ludwig, 596, ‘oline nebenbuhler nilmlich, ohne genossen, Indra, bist du von jeher ; im kampfe suchst du den gefiihrten’ ; cf. his commentary, vol. v, p. 147. The paradox in the repeated padas is only apparent : linapih, as well as a9atriih, expresses Indra’s solitary greatness as a warrior god from his birth on ; he requires no ally and no enemy dares him. Ludwig conceives 8.21.13° as irony. This is quite unnecessary ; the stanza continues the thought very aptly and effectively : ‘ with battle alone dost thou seek friendship that is to say, ‘ all you care for is fight’. Yet it is a good guess that 8.21.13 states in paradoxical and heightened rhetorical form, therefore in later form, what the other versions state in simple and primary form. I have for my part no hesitation in believing that 8.21.13 imitates 1.102.8. — For i.io2.8°dseo under 1 .81 -5cd ; for the repeated pada cf. also 8.15. iob, manhistha indra jajiiise. 113] Hymns ascribed to Kutsa Angirasa [ — 1.105.1-18 1.102.11 = 1. 100. 19. 1.103.2a (Kutsa ; to Indra) sa dharayat prthivim papr&thac ca vajrena hatva nfr apah sasarja, ahann ahim abhinad rauhinarii vy ahan vyahsam maghava ?aclbhih. 2.15.2° (Grtsamada ; to India) avan?6 dyam astabhayad brhantam a rodasl aprnad antiiriksam, sa dharayat prthivim paprathac ca somasya ta miida indra? cakara. I render 1.103.2, ‘He sustained the earth and spread it out. Having slain (Vrtra) with his bolt he released the waters. He slew the dragon, cleft Rauhina ; Maghavan slew Vyahsa with might.’ The theme of the first pada is none too sympathetic with the rest of the stanza. Of course, it contains ono of the stock ideas about Indra (2.17.5 ; 3-44-3), and loose linkage of motifs is one of the standard failings of the RV. But the other stanza shows the samo pada so closely knit with the rest of the stanza that it seems hard, in this instance, to doubt direct borrowing on the part of 1. 103.2 : ‘On no timbers (resting it) he supported the high heaven ; he filled the two (cosmic) hemispheres and the mid-air ; he sustained the earth and spread it out. Inspired by soma Indra hath done these deeds.’ The stanza is perfect, and the sequence dyam, antiiriksam, prthivim intentional and original. 1.103. 7 (Vllmadeva ; to Vifve Devah) devair no devy dditir ni patu devas trata trayatam aprayuchan, Lnahi mitrasya varunasya dhasimj arhamasi pramiyam sanv agneh. fr^cf. 4-55*7c [1.107. 2“, upa no deya avasa gamantu : 10.35.13°, vifve no devil, &c.] Cf. 1.89. 7d. 1.107.21' (Kutsa ; to Vi?ve Devah) Lupa no deva avasa gamantvj afigirasiliii samabhi stuyamanah, Wcf. 1.107.2“ indra indriyuir manito marudbhir adityair no aditih (jarma yansat. 4.54. 6d (Vamadeva; to Savitar) ye te trir ahan savitah savaso dive-dive saubhagam asuvanti, Indro dyavaprthivf sfndhur adbhir adityair no aditih garma yansat. io.66.3b (Vasukarna Vasukra ; to Vifve Devah) indro vasubhih pari patu no gayam adityair no aditih 5.31. 1. 1.118.4d (Kakslvat Dairghatamasa ; to A9vins) a vam 9yenaso a9vina vahantu rathe yuktasa a9avah patamgah, ye apturo divyaso na gfdhra abhi prayo nasatya vahanti. 6.63. 7b (Bharadvaja ; to A9vins) a vam vayo ’9vaso vahistha abhi prayo nasatya vahantu, Lpra vam ratho manojava asarjljsah prksa isidho anu pQrvih. 6.63.7° For the difficult pada 6.63. 7a cf. 8.23.3, and Oldenberg, RV. Noten, p. 408 ; for prksa Pischel, Ved. Stud. i. 96. 1.118. 6a, ud vandanam airatam dahsanabliih : i.ii2.5b, ud vandanam airayatam svkr dr9e. 1.118. 9a (Kakslvat Dairghatamasa; to A9vins) Sruvam gvetaih pedava indrajutam ahihanam agvinadhattam agvam, johutram aryo abhibhutim ugram sahasrasam vrsanam vldvangam. 1 0.39. 1 oa (Ghosa Kakslvatl ; to A9vins) 5ruvam gvetam pedave ’gvinagvam navabhir vajair navati ca vajfnam, carkrtyam dadhathur dravayatsakham bhagam na nrbhyo havyarn mayobhiivam. Tlie problem of interpretation is johutram. The Pet. Lex., followed by Grassmann, and Hillebrandt in the vocabulary of his Chrestomathy, renders ‘ laut wiehernd \ Suyana had previously indicated the same translation, and accounted for it by ati9ayena sam- gramesv ahvataram, ‘ the caller to battles par excellence ’. Bergaigne, ii. 452, * invoqu6 125] Hymns ascribed to Kaksivat Dairghatamasa [ — 1.122.6 par les prtUres ’, which is nearly correct. Ludwig, 30, hits the nail on the head with * laut zu riihmen The suffix -tra makes the noun one of instrument, with incidental passive value, e.g. patra ‘instrument of drinking’ ; joliutra means ‘subject to fervent invo- cation \ This is probable grammatically and intrinsically ; it is made certain by the closely parallel carkrtyam ‘worthy of ardent praise' in 10.39.10 (cf. also 1.119.10). The expression johutram aryah is paralleled even more closely by carkrtyam aryah in 4.38.2 ; cf. also htivyo aryah in 1.116.6. I cannot agree with Geldner, Ved. Stud. iii. 77, that carkrtyam arydh means ‘der zu riihmen ist noch melir als ein Reicher or, by the same terms, that luivyo aryah means ‘to be invoked more than a rich man ’. Nor can I believe that Bergaigne, Lexique du Rig- Veda, p. 170, and Oldenberg, ZDMG. liv. 178, are right in translating johdtram aryah, and carkrtyam ary&h by ‘ he is to be praised or called by the poor ’. ari is in these passages synonymous with suri and maghiivan : carkrtyam aryah ‘ fit to bo praised by the rich (sacrificer)’ ; johutram aryah ‘to be fervently invoked by the rich (sacrificer)’. — Noto also that i.ii7.2od = io.39.7b. — For the repeated pada cf. i.ii7.9b. 1.121.5cd (Kaksivat Dairghatamasa ; to Indra, or Vi^ve Devah) tubhyaiii payo yat pitarav anltam radhah surdtas turane bhuranyu, ^tici yat te rekna ayajanta sabardughayah paya usriyayah. 10.61. 1 icd (Nabhfinedistha Manava ; to Vi?ve Devah) maksu kanayah sakliyam naviyo radho na rdta rtam it turanyan, $uci yat te rekna ayajanta sabardughayah paya usriyayah. For these stanzas, both of which come pretty close to intentional bralimodya, see Ludwig, 47°, 997 (with notes) ; Grassmann, ii. 448, 475 ; Bergaigne, ii. 1 10, ill, 309 ; iii. 233 ; Olden- berg, RV. Noten, p. 118. Stanza 10.61.10 begins with a pada almost identical with 10.61. ii* maksu kanayah sakhyarii navagviih. 1.121.13b (Kaksivat Dairghatamasa ; to Indra, or Vi$ve Devah) tvaiii suro harito ramayo nfn bharac cakram 6ta 7.2.4 ; 9.5.4 ; 10.70.4) introduce, of course, the notion that the barhis is wide : in 1.188.4 it has room even for a thousand heroes. And yet we can see unerringly that the repeated pada is borrowed directly from the Usas imagery and diction. It may have been, as hinted above, suggested by pada b, vastor asya vrjyate dgre ahnam, which introduces Usas in person. See vastor usasah, or usasam 1.79.6; 7.10.2 ; and agre ahnam in 5.1.4; 5.80.2. The ritualistic poet as he spreads the sacrificial straw in the morning when Usas rises does not miss the opportunity to make this bold comparison between his ‘ God Barhis' (devabarhis, TS. 1.1.2.1, and many other times) and the Goddess Usas. — For i.i24.5ab cf. 1.92. i*b. 1.124.7® (Kakslvat Dairghatamasa ; to Usas) abhrateva punsa eti pratlci gartarug iva sanaye dhananam, jaydva patya ugati suvasa Lusa hasreva ni rinlte apsah.j cf. i.i24.7d 4.3.2b (Vamadeva Gautama ; to Agni) ayam yonig cakrma yam vayam te jaydva patyd ugati suvasah, arvaclnah parivlto ni sldema u te svapaka pratlcih. 10.71. 4d (Brhaspati Angirasa ; to Juana) uta tvah pagyan na dadarga vacam uta tvah grnvan na grnoty enam, uto tvasmai tanv&m vi sasre jaydva patya ugati suvasah. 10.91.13d (Aruna Yaitahavya ; to Agni) imam pratnaya sustutim navlyasim voceyam asma ugate grnotu nah, bhuya antara hrdy asya nisprge jaydva patya ugati suvasah. The repeated pada offers an interesting illustration of the art of simile as handled by the Vedic poets, and at the same time contributes to the higher criticism of the Veda. We know that the idea of the repeated pada is as staple with these poets as, e.g., that of the * cow lick- ing the calf’, the standard expression for mother’s love. The four repetitions show that the verse was in what we may call a state of flotation — any poet’s fair game. Yet I venture to assume that it originated in the Usas stanza, 1. 124.7 (f°r which see Bartholomae, Bezz. Beitr. xv. 2 ; Pischel, Ved. Stud. i. 308). The poet of 10.71.4 uses high art in braiding the notion with his theme vac, ‘ the holy word precursor of brahma : ‘ There are some who are able to see, yet do not see Vac ; yea there are some who are able to hear Vac, but do not hear her. But to some she unfolds her person as a finely robod, loving wife to her spouse.’ In 10.91.13 another poet desires that his recent clever song of praise shall insinuate itself into Agni’s heart as the same kind of a wife is pleasing to her husband. It will be observed that the construction of the repeated pada begins here to loosen somewhat. In 4.3.2 it is very loose indeed. Oldenberg, SBE. xlvi. 325 renders, ‘ This is the home which we have prepared for thee (sc. Agni, meaning, of course, Agni’s hearth, yoni) as a well-dressed, loving wife (prepares the marriage-bed) for her husband’. I doubt that the poet had any such com- parison in mind ; he wishes to say, it seems, that Agni’s hearth shall please him as an attractive wife pleases her husband. The metaphor limps decidedly, though we cannot say definitely whether the repeated pada is borrowed directly from 1.124.7, or from the floating mass. Still one is tempted to put the relative chronology of the stanzas in the order of the Hymns ascribed to Kaksivat Dairgliatamasa [ — 1. 127.2 129] applicability of the repeated pflda : 1.124.7; 10.71.4 ; 10.91. 13; 4.3.2, and this order, in any case, remains valid when we appraise the relative stylistic merits of the stanzas. — The four padas of 1. 124.7 each contain a simile whose interpretations engage the native commentators ; see Geldner, Rigveda Kommentar, p. 22. For 1.124.7% al80 tbe author, SBE. xlii. 258. [1.124.7d, usit hasreva nt rinlte apsah : 5.8o.6b, yoseva bhadra ni rinlte dpsah.] 1.124. 10b (Kaksivat Diiirghatamasa ; to Usas) pra bodhayosah prnatd maghony dbudhyamanah panayah sasantu, revad ucha maghavadbhyo maghoni revat stotre sonrte jarayantl. 4.51.30 (Vamadeva ; to Usas) uchantir adya citayanta bhojan radhoddyayosdso maghdnih, acitr6 antah panayah sasantv abudhyamanas tamaso vimadhye. The obscure word jarayantl (Sayana, sarvapriininah ksapayantl !) in i.i24.iod seems to mo to be intelligible best in the light of such an expression as, usasam . . . prati vipraso matibhir jarante, 5.80.1. Since the bards sing to Usas habitually, Usas, in her turn, may be said ‘to cause songs to be sung which result in wealth for the singer ’, rev.lt stotre jarayantl. This is, in fact, what happens on the morning of each (sacrificial) day when Usas appears. In x-IJ3-5 > 7.76.6 Usas is herself said to be the first singer (in the morning). We are thus saved the assumption of a stem jaraya in the sense of ‘ awaken ’, as suggests Foy, KZ. xxxiv. 251. The root gar (jagar) never shows initial j. For other suggestions, none of them alluring, see Oldenberg, RV. Noten, p. 128. — Note that the first padas of the two stanzas transfuse the same idea, and that the two stanzas are, in fact, imitative throughout, in spirit even more than in words. 1.124.12 (Kaksivat Dilirghatamasa ; to Usas) = 6.64.6 (Bharadvsja ; to Usas) ut te vayag cid vasat6r apaptan narag ca y6 pitubhajo vyustau, ama sate vahasi bhuri vamam uso devi daguse martyaya. Geldner und Kaegi, Siebenzig Lieder, p. 37, regard stanzas 1.124.11-13 as appendix; Grassmann, ii. 449, impugns stanzas 11, 12. The present stanza is particularly well joined in 6.64.6, but there is no real indication as to where it originated. — For pada c cf. 10.42.8% ni sun vat 6 vahati bhuri vamam. Group 13. Hymns 127-139, ascribed to Parucchepa Daivodasi [I.127.1b, vasum sunum sahaso jatavedasam : 8.71.11% agni'rh sunum, &c.] 1.127. 2c+e (Parucchepa Daivodasi ; to Agni) yajistham tva yajamana huvema jyestham angirasam vipra manmabhir vipre- bhih gukra manmabhih, parijmanam iva dyam hdtaram earsaninam, goci'skegam vrsanam yam imd vlgah pravantu jutaye vigah. 17 [h.o.s. 20] [130 1.127.2 — ] Part 1: Repeated Passages belonging to Book I 8.60.3d (Bharga Pragatha ; to Agni) agne kavlr vedha asi hota pavaka yaksyah, Lmandro yajistho adhvaresv Idyoj viprebhih gukra manmabhih. 5^4. 7. ib 8.23.7b (Vifvamanas VaiyaQva ; to Agni) agnim vah parvyam huve hdtaram carsanlnam, tam aya vaca grne tam u va stuse. 8.60. 1 7d (Bharga Pragatha; to Agni) agnim-agnim vo adkrigmh huvdma vrktabarhisah, agnim hitaprayasah QaQvatisv a hdtaram carsanlnam. Oldenberg, SBE. xlvi. 129, translates the first tristich of 1.127.2 : ‘May we, the sacrificers, call thee hither, the best of sacrificers, the first of the Angiras, 0 priest, with our prayers, with priestly prayers, 0 bright one.’ Ludwig, 281, more diplomatically, holds to the ordinary meaning of the words of the third pada, ‘ mit den heiligen siingern,0 heller, mit gedenkenden liedern Grassmann, like Oldenberg, ‘ mit weisen liedern, reiner Oldenberg in a note points out the recurrence of the pada, viprebhih $ukra manmabhih, in 8.60.3, without discuss- ing the circumstances under which it appears. But they cannot be passed by lightly ; the stanza in question is translated most naturally: ‘O Agni, thou art an ordering sage, a worshipful priest, 0 Purifier ; lovely, best sacrificer, fit to be revered at the offerings by the sages with their prayers, O bright god.’ For idyo with the instrumental of person performing the reverence, cf. 1.1.2 ; 3.29.2. The same statement in the active at 8.23.25, vipra agnim . . . Ilate. There is no reason for denying the author of 8.60.3 the primary and real authorship of the pada, nor need we fear to say that Parucchepa adapted it loosely, especially as it happened to fit in with the needs of his atyasti rhyme ; cf. under 1.82.2. We may note that the other repeated pada in 1.127.2, namely hotaram carsanlnam, recurs in the same hymn, 8.60.17 (also in 8.23.7), and that the next item shows connexion between 1.127.8 and 8.23.25. Hymns 8.23 and 8.60 correspond in three padas, to wit : 7b = i7d ; 22b = 2d ; 27* = I4d. 1.127. 8d (Parucchepa Daivodasi ; to Agni) vifvasam tva vigam patim havamahe sarvasam samanarh dampatim bkuje satya- girvahasarh bhuje, atithim manusanam pitur na yasyasaya, ami ca vi'fve amrtasa a vayo havya devesv a vayah. 8.23. 25a (Vifvamanas Vaiyafva ; to Agni) dtithim manusanam sunum vanaspatlnam, vipra agnim avase pratnam ilate. Cf. at the end of the preceding item. — For the repeated pada cf. 4.1.20”, vi<;vesam dtithir manusanam. 1.127.9do (Parucchepa Daivodasi ; to Agni) tvam agne sahasa sahantamah fusmintamo jayase devatataye rayir na devatataye, Qusmintamo hi te mado dyumnintama uta kratuh, adha sma te pari caranty ajara frustivano najara. 1 75-5ab (Agastya ; to Indra) Qusmintamo hi te mado dyumnintama uta kratuh, vrtraghna varivovida manslstha afvasatamah. It would seem clear that the connexion of the repeated couplet in 1. 175.5 is more original. The combination of mada and kritu is common in Indra stanzas : 5.43.5 ; 6.40.2. — On the metre of 1.127.9* c^- Oldenberg, Prol., p. 69. 131] Hymns ascribed to Parucchepa DCiivodasi [ — 1.128.6 1.127.10e (Parucchepa Daivodasi ; to Agni) pra vo maho sahasft sahasvata usarbudhe pa^uso nagnaye stomo babhutv agnaye, prati ydd irh havisman vigvasu ksasu j6guve, agre rebho na jarata rsQnaiii jurnir hota rsQnam. 5.64. 2d (Arcananas Atreya ; to Mitra and Varuna) ta bahava sucetiina pra yantam asma arcate, fevam hi jaryhm vam vi9va.su ksasu j6guve. The repeated piida is used in slightly different constructions. The passage r.i 27.1c159 is to be rendered, ‘ when (the worshipper) who gives offerings has praised him in all places 5.64.2°*, ‘for your praiseworthy kindness has been praised in all places’. The word sucetuna in the latter stanza occurs also in 1.137.11. — For i.i27.io‘b cf. Pischol, Ved. Stud. i. 91 ; its metre, Oldenberg, RV. Noten, p. 132. 1.128.2’’ (Parucchepa Daivodasi ; to Agni) tarn yajnasadham api vatayamasy rtasya patha ndmasa havismata devatata havismata, sa na Qrjam upabhrty aya krpd na jQryati, yam matarifva manave paravato devarii bhah paravatah. 10.70.2° (Sumitra Badhrya^va; Apra, here to Narafansa) a devanam agrayaveha yatu nara^anso vifvarupebhir a9vaih, rtasya patha namasa miyddho devebhyo devatamah susudat. io.3i.2b (Kavasa Ailusa ; to Vifve Devah) pari cin marto dravinam mamanyad rtasya patha namasa vivaset, uta svena kratuna Siim vadeta 9reyansarii daksam manasa jagrbhyat. For 1.128.2 see Oldenberg, SBE. xlvi. 137 ; RV. Noten, p. 132 ; Max Muller, SBE. xxxii. 202, 437 ; for 10.70.2, Hillebrandt, Ved. Myth. ii. 104 ; ii. 448. — The cadence namasa vivaset also in 6.i6.46d. 1.128. 6e+e (Parucchepa Daivodasi ; to Agni) V19V0 vihaya aratir vasur dadhe haste daksine taranir na 9i9rathac chravasyaya na 9i9rathat, vi9vasma id isudhyate devatra havyam 6hise, vi9vasma it sukrte vSram rnvaty agnir dvara vy rnvati. 8.19.1° (Sobhari Kanva; to Agni) tarn gurdhaya svarnaram devaso devam aratim dadhanvire, devatra havyam ohire. 8.39.6d (Nabhaka Kanva; to Agni) agnir jata devdnam agnir veda martanam aplcyam, agnih sa dravinodd agnir dvara vy ftrnute svahuto navTyasa Lnabhantam anyake same.j refrain, 8.39. ib ff. Stanza 1.128.6 has obscure spots. The change from third to second person in ohise (Pada- patha, 1+ uhise) leads Bartholomae, Bezz. Beitr. xv. 230, to suggest the infinitive a+uhise to wit : ‘ fur jeden flehenden ist von ihm (namlich Agni) das opfer gotterwarts zu faren.’ The parallel ohire (Padapatha, a4-uhire) does not go to support that view ; cf. also Neisser, 1.128.6 — ] Part 1: Repeated Passages belonging to Book I [132 ibid, xxvii. 265 ; Oldenberg, RV. Noten, p. 133. As regards the difficult first pada, Oldenberg, SBE. xlvi. 140, proposes the radical change to vi^va vihaya aratir v£su dadhe, which makes easy sense: ‘the far-reaching steward has taken all goods in his right hand’ (cf. 9.1S.4). But in RV. Noten, p. 132, he is assailed by doubt : there is, indeed, no compelling reason why the nominatives VI9V0 vihaya aratir vasur should be severally doubted as Agni’s epithets : 1 The universal, far-reaching steward, the Vasu, has put into his right hand (sc. goods, vasu, which is to be supplied with punning allusion to the nominative vasur). So Sayana ; differently Madhava to TB. 2. 5. 4.4. For isudhyatd see Pischel, Ved. Stud. i. 141 ; Ludwig, Uber Methode, p. 63. That 1.128.6 and 8.19.1 are directly imitative of one another is shown not only by the repeated pada but also by the parallel aratir and aratim. — For the interchange between rnvati and urnute cf. in my Vedic Concordance : tvesas te dhuma rnvati (urnotu). 1.128. 8a+b (Parucchepa Daivodasi ; to Agni) agnim hbtaram ilate vasudhitim priyam c^tistham aratim ny drire havya- vaham ny erire, vifvdyum v^vavedasam hotaram yajatarii kavi'm, devdso ranvam avase vasuyavo glrbhi ranvam vasuyavah. 5.i.7b (Budha Atreya, and Gavistkira Atreya ; to Agni) pra nu tyam vi'pram adkvaresu sadkum agnim hotaram ilate namobhih, d yas tatana rodasl rtena nityam mrjanti vajinarh gkrtena. 6.14.2° (Bkaradvaja Barkaspatya ; to Agni) agnir id dki praceta agnir vedkastama rsik, agnim hdtaram ilate yajiiesu manuso vifah. 7.16.1° (Vasistka Maitravaruni ; to Agni) ena vo agnim namasLorjo napatam a kuve,j 7.16. ib priyam c6tistham aratirix svadhvaram vigvasya dutam amftam. It is obvious that i.i28.8a is composite and secondary in the light 6.14. 2C and 5.1. 7b; cf. also 3.io.2b, agne hotaram Ilate. It does not seem necessary with Arnold, VM., p. 124, to read vasudhitim; cf. under 1.1.2°. But the pada points to the secondary workmanship of 1.128.8. [1.129.26, prksam atyaiii na vajinam: 1.135.5°, 3fum atyaiii, &c.] 1.129.3fg (Parucckepa Daivodasi ; to Indra) dasmo ki sma vrsanam pinvasi tvacam kiirii cid yavlr araram 9ura martyam parivrnaksi martyam, indrota tubkyam tad dive tad rudraya svaya9ase, mitraya vocam varunaya saprathah sumrbkaya saprdtkah. i.i36.6b° (Parucchepa Daivodasi ; Lingoktadevatah) namo dive brhat6 rodaslbhyarii mitraya vocam varunaya mllhuse sumrlikaya milhuse, Lindram agnim lipa stuhij dyuksam aiyamanarii bhagam, frw" cf. 1.12.7° jyog jivantah prajaya sacemaki somasyoti sacemahi. For i.i29.3°cf. Max Muller, SBE. xxxii. 142 ; Pischel, Ved. Stud. i. 109. For i.i29.3ab°, Oldenberg, RV. Noten, pp. 133. 133] Hymns ascribed to Parucchepa Daivodasi [ — 1.130.6 1.129.5°, ugrabhir ugrotibhih: 1.7.4°, ugra ugrabhir Qtibhih. 1.129. 0ll+K (Parucchepa Daivodasi ; to Indra) tvam na indra raya parinasa yahi pathan anehasa puro yahi araksasa, sacasva nah paraka a sacasvastamlka a, pahi no dQrad arild abhistibhih sada pahy abhistibhih. 4.31. i2b (Yamadeva ; to Indra) asman aviddhi viyvah^ndra raya parinasa, asman vifvabhir utibhih. 8.97. 6d (Rebha Kajyapa ; to Indra) sa nah somesu somapah sutesu favasas pate, madayasva rfulhasa sQnftavat6ndra raya parinasa. 10.93.11° (Tanva Partha ; to Vi9ve Devah, here Indra) etarii fansam indrasmayiis tvam kucit santam sahasavann abhistaye sada pahy abhistaye, medataiii vedata vaso. The obscure stanza 10.93. 11 (Ludwig, 240) with its irregular metre (prastarapankti) approaching the asti type, invites the belief that it was composed under the influence of x.i 29.9. The expression . . . abhistaye sada pahy abhistaye seems to be an odd and gratuitous variation of . . . abhistibhih sida pahy abhistibhih ; see Part 2, chapter 3, class B 9. — Cf. 5.10.1°, pra no raya parinasa. 1.130.1s (Parucchepa Daivodasi ; to Indra) endra yahy lipa nah paravato nayam acha vidathamva satpatir astam rajeva satpatih, havamahe tva vayarii prayasvantah sute saca, putraso na pitaram vajasataye mahhisthaih vajasataye. 8.4. i8d (Devatithi Kanva ; to Indra or Pusan) para gavo yavasarh kac cid aghrne nityam rekno amartya, asmakam pusann avita fivo bhava manhistho vajasataye. 8.88. 6d (Nodhas Gautama ; to Indra) nakih paristir maghavan maghasya te yad da^use da9asyasi, asmakam bodhy ucathasya coditti manhistho vajasataye. To the treatments of the difficult expression nayam dcha, cited by Oldenberg, RV. Noten, p. 1 2 1, add Ludwig, Kritik, p. 37 ; Uber Methode, p. 23. 1.130.6b (Parucchepa Daivodasi ; to Indra) im5m te vacam vasuyanta ayavo ratham na dhirah svapa ataksisuh sumnaya tvam ataksisuh, jumbhanto jenyam yatha vajesu vipra vajinam, atyam iva 9a vase sataye dhana vi'9va dhanani sataye. 5.2.nb (Kumara Atreya, or Vrsa Jana; to Agni) etam te st6mam tuvijata vipro ratham na dhirah svapa ataksam, yadid ague prati tvam deva haryah svarvatlr apa ena jayema. 1.130.6 — ] Part 1: Repeated Passages belonging to Book I [134 5.29.13d (Gaurivlti (^aktya ; to Indra) rndra brahma kriyamana jusasva ya te gavistha navya akarma, vastreva bhadra sukrta vasuyu ratham na dhirah svapa ataksam. The repetition of the word ataksisuh in i.I30.6c, belonging as the word does to the formu- laic repeated pada b, marks tho composition of this rhyme pada, as well as the stanza which contains it, as secondary. We may consider as quite certain that this sentiment was first uttered in the first person singular. [1.130. 7d, atithigvaya gambaram : i.56.6b, arandhayo ’tithigvaya gambaram ; cf. 9.6i.2b.] 1.130.8? (Parucchepa Daivodasi ; to Indra) indrah samatsu yajamanam aryam pravad vlgvesu gatamutir aji'su svkrmllhesv aji'su, manave gasad avratan tvacam krsnam arandhayat, daksan na vi'fvam tatrsanam osati ny argasanam osati. 8. 12.9^ (Parvata Kanva; to Indra) indrah suryasya ragmibhir ny argasanam osati, agnir vaneva sasahih pra vavrdhe. Cf. Muir, OST. i. 174 ; Oldenberg, KV. Noten, p. 135. 1.130. 9d (Parucchepa Daivodasi ; to Indra) sura? cakram pra vrhaj jata ojasa prapitve vacam aruno musayatlfana a musayati, ugana yat paravato ’jagann utaye kave, sumnani vigva manuseva turvanir aha vigveva turvanih. 8.7.26° (Punarvatsa Kanva ; to Maruts) ugana yat paravata uksno randhram ayatana, dyaur na cakvadad bhiya. The appraisal of the repeated pada depends upon the two mythic snatches told in the two stanzas. Of these the second, 8.7.26, seems to say distinctly enough : ‘When, (O Maruts) ye came with Uijana from a distance to Uksno Randhra, he bellowed from fright, as the sky (thunders).’ So Max Muller, SBE. xxxii. 392, 397 ; Geldner, Ved. Stud. ii. 169 (differently, Ludwig, 701). Though we know nothing further about this legend, the context fixes u^ina as instrumental. U5ana (later U9anas) Kavya is an ancient priest-ally of the gods (Bergaigne, ii. 338 ff.). And so he figures in 1. 130.9 : When, O seer, thou didst come with Ujana from a distance to help.’ The allusions otherwise, mythical or legendary, in 1. 130.9 are veiled from our ken ; see Bloomfield, JAOS. xvi. 34 ff. ; Hillebrandt, Ved. Myth. iii. 290, note 2 ; Geldner, Ved. Stud. ii. 175 ; Ludwig, Die neuesten Arbeiten, p. 174 ; Oldenberg, RV. Noten, p.135. Later stories throw no light on the matter ; see Spiegel, Die arische Periode, 284 ff. Connexion with Avestan Kava Usa(Shah Nameh, Kai Kaus : Spiegel, ibid. 285) is doubted, perhaps over- scoptically, by Bartholomae, Altiranisclies WOrterbuch, s.v. 2. usant. [1.131.1f; 8.12. 22b, devaso dadhiro purih: 5.i6.id, martaso dadkir6 pur&h: 8.i2.2 5b, dev&s tva dadhiro purah.] 135] Hymns ascribed to Parucchepa Daivodasi [ — 1. 133.7 [1.131.4h, puro yad indra 9a rad I r avatirah : 1.174.2'’; 6.20. ioc, sapta yat purah 9&rma 9iiradlr dart.] 1.132. l’,c (Parucchepa Daivodasi ; to Indra) tviiya vayarii maghavan purvye dhana indratvotah sasahyama prtanyatd vanu- yama vanusyatah, n6dhisthe asminn ahany adhi voca nu sunvatS, asmin yajne vi cayema bhare krtarii vajayanto bhare krtam. 8.40.7 9-61.39°; the cadence vanavad vanusyatah at 2.35.1*, 2* ; 26.1*. [1.132. 4b, yad angirobhyo ’vrnor apa vrajam: 1.51.3°, tvam gotram ahgirobhyo ’vrnor apa.] 1.132. 5? (Parucchepa Daivodilsi ; to Indra) sam yaj janan kratubhih 9ura iksayad dhane hite tarusanta 9ravasyavah pra yaksanta 9ravasyavah, tasma ayuh prajavad id bddhe arcanty ojasa, indra okyarii didhisanta dhitayo devan acha na dhitayah. 1.139.1s (Parucchepa Daivodasi ; to Vi9ve Devah) astu 9rausat puro agnirn dhiya dadha a nu tac chardho divj7am vrnlmaha indravayii vrnlmahe, yad dha krana vivasvati nabha samdayi navyasi, adha pra su na upa yantu dhitayo devan acha na dhitayah. Cf. for 1. 132. 5 Oldenberg, RV. Noten, p. 137; for 1.139.1, Pischel, Ved. Stud. i. 69, 70; Hillebrandt, Ved. Myth. i. 488 ; Ludwig, Kritik, pp. 12, 19 ; Uber Methode, p. 34 ; Oldenberg, RV. Noten, p. 141. 1.133. 7e (Parucchepa Daivodasi ; to Indra) vanoti hi sunvan ksayarn parlnasah sunvano hi sma yajaty ava dviso devanam ava dvisah, sunvana it sisasati sahasra vajy avrtah, sunvanayendro dadaty abhuvam rayim dadaty abhuvam. 8.32. i8b (Medhatithi Kanva; to Indra) panya a dardirac chata sahasra vajy avrtah, indro yo yaj van 0 vrdhah. Cf. Neisser, Bezz. Beitr. xix. 148. 1.134-2 — ] Part 1 : Repeated Passages belonging to Book I [186 1.134.2a+e (Parucchepa Daivodasi ; to Vayu) mandantu tva mandino vayav indavo ’smat kranasah siikrta abhidj’avo gobhih krana abhi'dyavah, yad dha krana iradhyai daksam sacanta utayah, sadhrlcina niyiito davane dhi'ya upa bruvata im dhiyah. 2.n.nb (Grtsamada ; to Indra) Lpiba-pibed indra fura somarhj mandantu tva mandinah sutasah, 2. 1 1. 1 ia prnantas te kuksi vardhayantv ittha sutah paura l'ndram ava. 3.13.2^ (Rsabha Vaigvamitra; to Agni) rtava yasya rodasl daksam sacanta utayah, havfsmantas tam llate tam sanisyanto Vase. We may render 1.134.2 as follows : ‘ May the delightful drops of Soma delight thee, they that have been mixed by us, the well prepared, that tend to heaven ; they that are mixed with milk, and tend to heaven. When indeed the mixed (Soma drops) are for well-being, when the helps (of the gods) attach themselves to solid piety, then do our prayers engage Vayu’s span together to bestow gifts.’ For translations differing more or less, see Ludwig, 711 ; Grassmann, ii. 137 ; Pischel, Ved. Stud. i. 68. Pischel here defends the translation of krana by ‘ mixed ’. This suggestion, as well as the comparison with Ktpaw, dates back to Roth, as early as 1852 ; see Yaska’s Nirukta, Erlauterungen, p. 46, bottom. Cf. also Ludwig, Kritik, p. 12; Uber Methode, p. 24; Oldenberg, RV. Noten, p. 58; Geldner, Rig-Veda Kommentar, p. 26. The matter that concerns us here is the recurring pada 3.i3.2b. Ludwig, 312 : ‘der ordnungsmassige den die beiden welthiilften, mit des tiichtigkeit hilfe verbunden, den flehen an die havis bereitet haben, die gewinnen wollen zur gnade.’ Grassmann, L 67 : ‘ Den Heil’gen dessen Kriifte starkt das Weltenpaar, das Opferwerk, ihn flehn die opferreichen an, um Hiilfe die verlangenden.’ Ludwig’s translation is desperately obscure ; Grassmann is very hazardous in co-ordinating daksam with rddasi. Oldenberg, SBE. xlvi. 266, more recently renders the first distich : ‘ The righteous one to whose skill the two worlds (Heaven and Earth), and (all) blessings cling.’ The doubtful point in this rendering is the rather bizarre grammatical co-ordination of rodasl and utayah, with asyndeton, as the author assumes. I wonder whether Oldenberg, if he had happened to note the recurring pada, daksam sacanta utayah, in 1.134.2, would have adhered to his construction. It seems to me that the pada in question means ‘ the helps (of the gods) attach themselves to solid piety (or, pious solidity)’, and that the pada forms a parenthesis in 3.13.2. I paraphrase explicitly what the stanza seems to me to declare : ‘ The righteous (Agni) whose are the two worlds (Heaven and Earth) — (whose) helps attach themselves to solid pious work — him do men with havis revere, him they who desire gain, that they may obtain his blessing.’ It is another question whether we should accept the consequence of this construction and say that the author of 3.13.2 has borrowed pada b from 1.134.2. The pada may have been afloat as a sort of proverb. Cf. also Ludwig, Neuesto Arbeiten, p. 59. 1.134. 3bc (Parucchepa Daivodasi ; to VSyu) vayur yunkte rbhita vayur aruna vayu rathe ajira dhuri vblhave vahistha dhuri vdlhave, pra bodhaya puramdhim jara a sasatim iva, pra caksaya rodasl vasayosasah ?ravase vasayosasah. 137] [ — 1. 135-2 Hymns ascribed to Parucchepa Didvodasi 5.56.6cd (QyAvRgva Atreya; to Maruts) Lyungdhvarii hy 6rusT rathej yuhgdhviiih nithesu rohitah, (~nr 1.14.12a yungdhvam hari ajira dhuri vblhave v&histha dhuri vblhave. For the relation of the repeated p&das see under 1.14.12*. 1.134.0c+8 (Parucchepa Daivodilsi ; to Vayu) tvarii no vayav esam fipQrvyah s6manam prathamah pitim arhasi sutanam pitim arhasi, uto vihutmatlnarii vigam vavarjuslnam, vi$va it te dhenavo duhra iigirarii ghrtam duhrata agiram. 4.47.2b (ViUnadeva; to India and Vayu) Li'ndraf ca vayav esaxhj sbmanam pitim arhathah, yuvarii hi vantlndavo Lnimnam apo nil sadhryhk.j 5.5i.6b (Svastyatreya Atreya ; to Vifve Devilh) Ltndrag ca vayav esaihj sutanam pitim arhathah, tan jusetham arepasRv abhi prayah. 8.6. 1 9b (Vatsa Kanva ; to Indra) imas ta indra prgnayo ghrtam duhata agiram, enam rtasya pipyiislh. Tlie difficult word vavarjuslnam, 1.134.6, in the light of vihutmatlnan suggests the common use of root varj in connexion with barliis ; vigath vavarjuslnam would then mean, ‘ of people that have prepared (the barhis)’. In AV. 7.50.2 avarjuslnam looks like an artificial negative of the same word, perhaps haplologically a(va)varjuslnam, something like ‘ impious ’ (cf. asunvant, and the like). Cf. Geldner, Ved. Stud. i. 144 ; Ludwig, Ueber Methods, p. 28 ; Oldenberg, RV. Noten, p. 138. In 8.6.19 the pada, ghrtam duhata agiram, is apparently a modernized and metrically less fit version of ghrtirh duhrata agiram in 1.134.6. However, Aufrecht, in the Preface to the second edition of the Rig-Veda, p. xix, note, remarks whim- sically and pertinently anent i.i34.64b0: ‘Was hat der gute Parucchepa dabei gedacht als er die beiden Adjectiva (meaning apurvyah and prathamdh) setzte ? Der Vers musste ausgefullt werden.’ The secondary manufacture of 1.134.6 is unmistakable. — Cf. 2.14.2. 1.135. 2a+f (Parucchepa Daivodasi ; to Vayu) tubhyayam sbmah pariputo adribhi sparha vasanah pari kogam arsati gukra vasano arsati, tavayam bhaga ayusu somo devesu buyate, vaha vayo niyuto yahy asmayur jusano yahy asmayuh. 8.82. 5a (Kusldin Kanva ; to Indra) tubhyayam adribhih suto gobhih grlto madaya kam, pra soma indra huyate. 7.90. 1 c (Vasistha ; to Vayu) pra vlraya gucayo dadrire vam adhvaryubhir madhumantah sutasah, vaha vayo niyuto yahy acha Lpiba sutasyandhaso madaya. j i-a' 5.5 1.5° Oldenberg, ZDMG. Ixi. 825, is struck by the unfitness of the combination pariputo adribhih in 1. 135. 2*: ‘mit den steinen wird der soma ja gepresst, nicht gereinigt’ (cf. under 5.86.6). 18 [h.o.s. 20] 4-47-2a C»- 4.47. 2d 4-47-2tt 1. 1 3 5. 2 — ] Part 1: Repeated Passages belonging to Boole I [138 The repeated pada 8.82.5s illustrates his misgivings, and points to the later, mere jingly, manufacture of Parucchepa, 1.135.2s. For other points in the same stanza see the same author, RV. Noten, p. 139. — The correspondence between I.I35.21 and 7.90.1° suggests the praiigasastra ; see Bergaigne, JA. xiii. (1888) 127. 1.135.3ab+c (Parucchepa Daivodasi ; to Vayu) a no niyudbhih gatinlbhir adhvaram sahasrinibhir upa yahi vitaye vayo havyani vitaye, tavayam bhaga rtviyah saragmih surye saca, Ladhvaryubliir bharamana ayansata j vayo gukra ayansata. I-I35-3f 7.92. 5ab (Vasistha ; to Vayu) a no niyudbhir gatinlbhir adhvaram sahasrinibhir upa yahi yajfiam, Lvayo asmin savane madayasva j Lyuyam pata svastibhih sada nah.j &w-c: cf. 7.23.5d ; d: refrain, 7.i.20d ff. The pada 1.1 35.3s is repeated in the next stanza 1.135.4®. Ludwig, Der Rig-Veda, iii. 97 : ‘die stelle des vii. mandala scheint die wiederholung zu sein.’ He does not say why, but it seems to me this view is borne out by the metre. Arnold, VM., p. 310, remarks that 7.92.5s is ‘ extended tristubh ’. We see, of course, that it is not exactly extended, but a jagatl line repeated in exactly the same form, in 1.135.3s. Pada c is a tristubh of established form in the seventh mandala, e. g. asmin chura savane madayasva, 7-23.5d; asminn u su savane madayasva, 7.29.2®. The fourth pada is refrain. It looks for all the world as though 7.92.5 were a latter appendage in broken metre to the four stanzas which originally made up the hymn. Pada b is shortened from a jagatl to a tristubh in deference to the prevailing type. In any case the correspondence between the two stanzas suggests the praiiga9astra ; see Bergaigne, JA. xiii. (1888) 127. 1.135.3C, 4C, vayo havyani vitaye. 1.135. 31, 6b, adhvaryubhir bharamana ayansata. 1.135.4b+c (Parucchepa Daivodasi ; to Vayu) a vam ratho niyutvan vaksad avase ’bhi prdyansi sudhitani vitaye vayo havyani vitaye, pibatam madhvo andhasah purvapeyaiii hi vain hitam, Lvayav a candrena radhasa gatamj l'ndrag ca radhasii gatam. ewef. i.i35.4f 6.i6.44b (Bharadvaja ; to Agni) acha no yahy a vahabhi prayansi vitaye, La devan somapltaye.j 1.14.6° Pada 1. 135. 4® is identical with 1.135.3® Oldenberg, SBE. xlvi. 133; RV. Noten, p. 139, make various suggestions regarding the penultimate pada of 1.135.4, which just fall short of carrying conviction. For the padas repeated in this item see also the closely similar pftdas treated under 6.15.15s. [I.135.4f, vayav a candrena radhasa gatam: 4.48. i°-4°, viiyav a candrena rathena.] [1.135.5°, agtim atyaiii na vdjinam : 1.129.2P, prksam atyaiii, &c.] Hymns ascribed to Parucchepa Daivodasi [ — 1.136.2 189] 1.136. 6C (Parucchepa Daivodasi ; to Vayu) im6 vam soma apsy a suta ihLadhvaryubhir bharamana ayansataj vayo gukra ayansata, W*i-i35-3b et6 vam abhy hsrksata tirah pavitram agavah, yuvayavo ’ti x-omany avyaya somaso aty avyaya. 9.62. ib (Jamadagni Bhargava ; to Soma Pavamana) ete asrgram indavas tirah pavitram agavah, vigvhny ablii saubhaga. 9.67. 7b (Gotama ; to Soma Pavamana) Lp;ivamanasa indavasj tirah pavitram agavah, 9.24. ib indram yamebhir agata. It seems natural to suppose that the repeated pfula, tirah pavitram agavah, in 1.135.6, is borrowed from the sphere of Soma Pavamana in the ninth book. 1.136. 7C (Parucchepa Daivodasi ; to Vayu and Indra) ati vayo sasato yahi gagvato yatra grava vadati tatra gachatam grham indrag ca gachatam, vi sQnrta dadrge nyate ghrtam il purnaya niyuta yatho adhvaram indrag ca yatho adhvaram. 4.49. 3b (Vamadeva ; to Indra and Brhaspati) a na indrabrhaspatl grham indrag ca gachatam, Lsomapa s6mapltaye.j ew* 1.23.3° 8.69.7b (Priyamedha Angirasa ; to Indra) ud yad bradhnasya vistapam grham indrag ca ganvahi, madhvah pltva sacevahi trih sapta sakhyuh pade. It seems to me that the repetition of Indra’s name in 4.49.3ib shows that pada b is employed here formulaically and secondarily. Cf. 6.36.6; 7.88.3 ; 8.25.2 ; 10.86.22, and for the entire phenomenon, Edgerton, KZ. xliii. no ff. 1.136. ltl (Parucchepa Daivodasi ; to Mitra and Varuna) pra sii jyestham nicirabhyarii brhan namo havyarn matim bharata mrlayadbhyam svadistham mrlayadbhyam, ta samraja ghrtasuti yajne-yajna upastuta, athainoh ksatram na kutag canddhrse devatvam nu cid adhrse. 2.4 1. 6a (Grtsamada ; to Mitra and Varuna) ta samraja ghrtasuti Laditya ddnunas patl,j 1.136.3! sacete anavahvaram. Cf. 8.29.9b, samraja sarpirasuti ; and 8.8.i6d, vasuyad danunas patl. Note that 2«4i.6b = 1.136.3'- 1.136.2e (Pai-ucchepa Daivodasi ; to Mitra and Varuna) adargi gatur urave variyasl pantha rtasya samayansta ragmibhig caksur bhagasya ragmibhih, dyuksam mitrasya sadanam aryamno varunasya ca, atha dadhate brhad ukthyam vaya upastutyam brhad vayah. r. 136.2 — ] Part 1: Repeated Passages belonging to Book I [140 8.47.9d (Ti'ita Aptya ; to Adityas) aditir na urusyatv Laditih garma yachatu,j £5” 6.75.13d mata mitrasya revato aryamno varunasya caLnehaso va utayah suutayo va utayah. j «s* refrain. 8. 4 7. 1 ef- 1 8ef 1.136. 3e (Parucchepa Daivodasi ; to Mitra and Varuna) jyotismatlm aditirii dharayatksitim svarvatlm a sacete dive-dive jagrvansa dive- dive, jyotismat ksatram agate aditya danunas pati, miti’as tayor varu no yatayajjano ’ryama yatayajjanah. 2.4 1. 6b (Grtsamada ; to Mitra and Varuna) Lta samraja glirtasutlj aditya danunas pati, £»* 1.136. id sacete anavahvardm. For the repeated pada cf. 8.8. i6d, vasuyad danunas pati. Note that 2.41.6*= i.i36.id. [I.136.4a, ayaxh mitraya varunaya gamtamah : 9.104.3°, yatha mitraya, &c.] 1.136. 6bc, mitraya vocam varunaya mllhuse sumrllkaya mllhuse : 1. 129.3ft, mitraya vocam varundya saprathah sumrllkaya saprathah. 1.137.1e, 3d, asmatra gantam upa nah. 1.137.1S (Parucchepa Daivodasi ; to Mitra and Varuna) susuma yatam adribhir gogrlta matsara ime somaso matsara ime, a rajana divisprgLasmatra gantam upa nah,j gs* 1.137.1® ime vam mitravaruna gavagirah sdmah gukra gavagirah. 9.64.28° (Kagyapa Marlca: to Soma Pavamana) davidyutatya ruca paristobhantya krpa, sdmah gukra gavagirah. It seems as though 9.64.28 treated the repeated pada loosely and secondarily, as compared with 1. 1 37. 1. Grassmann, ‘die milchgemiscliten Soma’s sind erliellt von lichtem Stralilen- glanz, versehn mit rauschender Gestalt’. Ludwig, 854, not very differently. Note, however, that 1.137.1, 2 are really not much more than Soma Pavamana stanzas, done over for Mitra and Varuna. Therefore 1.137.1 is likely to be later than 9.64.28. 1.137. 2b: 1.5.5°; 5-5Mb; 7-32-4b> 9. 2 2.3b ; 63. 1 5b ; ioi.i2b, somaso dadhy- agirah. 1.137.2°: 1.4 7*7d ; 5.79.8°; 8. ioi.2d, sakarh suryasya ragmibhih. 1.137.28 (Parucchepa Daivodasi ; to Mitra and Varuna) ima a yatam indavah Lsdmaso dddhyagirahj sutaso dadhySgirah, uta vam usaso budhf Lsiikam suryasya*ragmlbhih,j sut6 mitraya varunaya pltaye carur rtaya pltaye. i-5-5° C»* 1 .4 7-7d 141] Hymns ascribed to Parucchepa Daivodasi [ — 1. 142.3 9.17.8° (Asita Kil^yapa, or Devala Ka$yapa ; to Soma Pavamilna) madhor dharfim anu ksara tlvrah sadhastham asadah, carur rtaya pltaye. For the repeated pad a see the Introduction, p. 22 ft. [1.137. 3b°, ah?um duhanty adribhih somaiii duhanty adribhih: 9.65. i5b, tlvram duhanty adribhih.] 1.139. 1^: 1.132.5s, devan acha na dhltayah. [1.139.3d, yuvor vifva adhi yrtyah: 8.92. 20a, yasmin vifva, &c.] [1.139. 0K, sumrllko na a gahi : 1.91.11°, sumrllko na a vi?a.] Group 14. Hymns 140-164, ascribed to Dirghatamas Aucathya 1.140.10a (Dirghatamas Aucathya ; to Agni) asmakam agne maghavatsu dldihy adha cv a si van vrsabho damunfih, avasya fi^umatlr adlder varmeva yutsu parijarbhuranah. 6.8.6a (Bharadvaja Barhaspatya ; to Vaiyvanara) asmakam agne maghavatsu dharayanami ksatram ajaram suvliyam, vayarh jayema ^atinaiii sahasrinam vai5vftnara vajam agne tavotibhih. [1.141. 9d, aran na nemih paiibhur ajayathah : 1.32.13d, aran na nemih pari ta babhuva.] Cf. 5.13.6. 1.142.1° (Dirghatamas Aucathya ; Apra, here Agni) samiddho agna d vaha devdn adya yatasruce, tantum tanusva purvyam sutasomaya da£use. 8.13.14° (Gosuktin Kanvayana, and Afvasuktin Kanvayana; to Indra) d tu gahi pra tu drava Lmatsva sutasya gomatah, j 8. 1 3. 1 4b tantum tanusva purvyam yatha vid6. I feel quite certain that the repeated idea fits less well with Indra and the otherwise banal statement about him, in 8.13.14. Moreover the refrain appendage yatha vide (tetra- syllabic pada throughout the hymn) betrays late workmanship for 8.13. See Part 2, chapter 2, class B 3. 1.142. 2b: 1.13.2a, madhumantam tanunapat. [1.142.2°, yajnam viprasya mdvatah: i.i7.2b, havam viprasya. &c.] 1.142.3a (Dirghatamas Aucathya ; Apra, here Narajansa) Qiicih pavako adbhuto madhva yajnam mimiksati, nara9ansah trir d divo devo devesu yajniyah. [142 1.142.3 — ] Part 1: Repeated Passages belonging to Booh I 8.13.19c (Narada Kanva ; to India) stota yat te anuvrata ukthany rtutha dadhe, (Trita Aptya : to Soma Pavamana) abhi brahmlr anQsata yahvir rtasya matarah, marmrjyante divah fifum. 143] Hymns ascribed to Dirghatamas Aucathya [ — 1.142.11 9.102.7b (Trita Aptya ; to Soma Pavamfma) samlclne abhi tmana yahvi rtasya matara, tanv&na yajnam finusag yad anjate. 10.59.8b (Bandhu Gopayana, or others ; to Dyavaprthivyau) 9am rodasl subandhave yahvi rtasya matara, bharatam apa yad rapo dyauli prthivi ksama. rapo Lmo sii te kirn cana- mamat.j W refrain, 10.59.8° if. 8.87.4b (Dynmnlka Vfisistha, or others; to Alvins) pibatam somarii madhumantam agvina barhih sidatam sumat, ta v&vrdh&na upa sustutlm divo gantarii gaurav iverinam. The dual form, yahvi rtasya matara, to Day and Night, 1.142.7 ; 5.5.6 ; to Heaven and Earth, 10.59.8, and probably also 9.102.7 (cf. 9.74.2 ; 10.44.8), is original. The plural form, 9.33.5, to the Prayer Cows (Ludwig, 823), in the Rishi’s best style of untrammelled fancy, is secondary. — For i.i42.7d cf. a barhih sidatam narii, 8.87.2b. — For the correspondence of 1.142 and 5.5 see also under i.i42.4lb. 1.142. 8bc: 1.13.8^ ; 1.188.7^, hotarfi daivya kavf, yajnam no yaksatam imam. 1.142.8‘' (Dirghatamas Aucathya ; Apra, here Divine Hotars) mandrajihvS jugui’vanl Lhotara daivya kavl,j €«* 1.13.8b Lyajharii no yaksatam imaihj sidhram adya divisprgam. 1.13.8° 2.41.20b (Grtsamada ; to Dyavaprthivyau, or Havirdhane) dyava nah prthivi imam sidhram adya divisprgam, yajniim devesu yachatam. 5.13.2b (Sutambhara Atreya ; to Agni) agne stomam manamahe sidhram adya divisprgah, devasya dravinasyavah. The question of interpretation involved is this : Is divisprgah, in 5. 1 3. 2b, genitive singular, agreeing with Agni, or is it nominative plural, agreeing with the subject of manamahe ? The translators, Ludwig, 351 ; Grassmann ; Oldenberg, SBE. xlvi. 395, take the first view. The last mentioned scholar, in a note, definitely : ‘ divispf9ah, no doubt, is genitive sing, referring to Agni, not nominative plural, referring to the worshippers.’ I cannot say whether Oldenberg, at the time of his writing, had in mind the parallels, but they seem to me rather to point to the opposite view, namely that the worshippers attain to heaven by means of their song of praise (stoma), just as they accomplish the same end in the parallel stanzas by means of their sacrifice (yajna). For examples of the frequent juxtaposition of stoma and yajna see RV. 1.156.1 ; 2.5.7 > 5-52-4 ; 6.16.22 ; 8.6.3 ! 99- 10.9.17. We may accept this conclusion notwithstanding that divispr 10.99.6 ; 105.2); and, to match, the sense of the pada fits more primarily the yajamana in 10.93.6. 1.151.4b (Dlrghatamas Aucathya ; to Mitra and Varuna) pra sa ksitfr asura ya mahi priya tftavanav rtam a ghosatho brhat, yuvam divo brhato daksam abhuvam gam na dhury upa yuhjathe apah. 8.25.4° (Vifvamanas Vaiyafva; to Mitra and Varuna) mahanta mitravaruna samraja devav asura, rtavanav rtam a ghosato brhat. For 1.151.4 see Oldenberg, SBE. xlvi. 224 ; RV. Noten, p. 149. I.152.1d, rtena mitravaruna sacethe; i.2.8a, rtena mitravarunau. [1.152.4d, priyam mitrasya varunasya dhama; 7.6 1.4% ^ansa mitrasya, &c. • 10.10.6°, brhan mitrasya, &c. ; 10.89.8°, pra ye mitrasya, &c. Cf. also under 2.27.7° and 4-5- 4C*] 1.162. 5a (Dlrghatamas Aucathya ; to Mitra and Varuna) ana