The Mitanni HOCK'S ARGUMENTS AGAINST THE OIT H. H. Hock presents various arguments against the OIT in two papers included in the volume "Aryan and non-Aryan in South Asia" (Ann Arbor, 1997, edited by Madhav M. Deshpande and Johannes Bronkhorst): "Out of India? The Linguistic Evidence" (p. 1-17), and "Through a glass darkly: Modern 'racial' interpretations vs. textual and general prehistoric evidence on arya and dasa/dasyu in Vedic society" (p. 145-?). The papers are interesting, and afford scope for some fundamental studies on certain points, which produce strong evidence on matters pertaining to the Indo-Aryan problem, though, as we shall see, not in the direction indicated by Hock. In the first paper, Hock touches on the "Vedic-Sanskrit=Proto-Indo- European" theory and the alternate "Proto-Indo-European-in-India" theory, and argues strongly against both. The first of these theories is, of course, untenable. But, in the course of his arguments, Hock deals with the issue of the Mitanni language in a debatable manner. In discussing the second theory, he takes up two issues, both of which invite debate: the "equine evidence", and the evidence of ancient IE dialectology. In the second paper, he discusses the AIT arguments about the racial differences between the so-called invading (or immigrating or "acculturating"?) Indo-Aryans, as derived from textual analyses, and, fortunately, dismisses them as baseless. However, in the course of his paper, he presents other arguments from the AIT side on two issues, which, again, invite debate: the identity of aryas and dasas/dasyus in the RV, and the evidence of river names with particular reference to the identity of the Sarasvati in the RV. We will, therefore discuss the following issues here: The Mitanni evidence. The Equine evidence. The evidence of ancient IE dialectology. The identity of aryas and dasas/dasyus in the RV. The Rivers, especially the Sarasvati, in the RV. A. The Mitanni Evidence. Hock argues against the arguments of S.S. Misra, "that the Mitanni form of Indo-Aryan must be later than Vedic Sanskrit and must have been imported to the Near East from India", and concludes, to the contrary, that the Mitanni language is in fact not a "form of Indo-Aryan" at all, but a form of Indo-Iranian, and that this "near Eastern variety of Indo-Iranian appears to predate the earliest attested stages of both Indo-Aryan (which has changed (d)zh to h) and Iranian (with s>h)" [HOCK 1997:3]. His sole argument, on the basis of which he reaches this conclusion, is that the Mitanni word "wasanasaya' of the chariot', appears to reflect a stage prior to the change of pre-Indo-Aryan voiced aspirate (d)zh>h, assuming that the word corresponds to Skt, vahanasya (see MAYRHOFER 1986, s.v. vah-)" [HOCK 1997:2]. Witzel, in the present volume under review here, modifies this to suggest that the language is indeed Indo-Aryan, but "an early pre-Rgvedic stage of IA, seen in the preservation of IIr ? zdh-> Ved. ?edh-, Iir ai> Ved. e, as well as in the absence of retroflexion .... there is no retroflexion in mani-nnu, or the Southwest Iranian, Elam. O.P. bara-mani and in the East Iranian dialect, Avest.: ma ini (in spite of the very specific phonetic alphabet used by the Zoroastrians!) .... Mit. IA also does not have typical South Asian loan words such as ani'lynch pin'." (p. 361-2). He amplifies this in his footnote: "Note ?zd- in Priyamazdha (Bi-ir-ia-ma-as-da, Mayrhofer 1979:47 in Palestine, cf. Priya-asva: bi-ir-ia-as-su-va): Ved. Priyamedha: Avest. –mazda. Or, note retention of Iir ai> Ved. e (aika: eka in aikavartana), and retention of j'h> Ved. h in vasana(s)saya of 'the race track' = [vazhanasya] cf. Ved. vahana- (EWA II 536, Diakonoff 1971: 80; Hock 1999: 2). Mit. IA also shares the Rgvedic and Avest. Preference for r (pinkara for pingala, parita for palita)" (p. 389). The evidence for the language being Indo-Aryan rather than Indo- Iranian is overwhelming ?every single Mitannian "Aryan" word is Indo-Aryan, and an overwhelming majority of the words are absent or unknown in Iranian. Hock has to indulge in special pleading [HOCK 1997:2-3 footnotes] to explain away the absence of Vedic/Mitanni deities like Varuna/uruwana, or the Vedic/Mitanni numeral word eka/aika, in Iranian; but the evidence is much more wide-based: as Witzel puts it, the words cover "the semantic fields of horses, their colors, horse racing and chariots, some important 'Vedic' gods, and a large array of personal names adopted by the ruling class" (p. 361) And all these words point towards Indo-Aryan, and away from Iranian. Witzel, therefore, only concentrates on showing that the "Mit. IA" words "belong to an early, pre-Rgvedic stage of IA" (p. 361). And his evidence to this effect consists only of the absence of retroflexion (eg. in mani-nnu), the absence of what he calls "typical South Asian loan words" (the word ani, "lynch pin"), and the ai in aika, zd in Priyamazdha, and zh in vazhanasya. The evidence is clearly flimsy and argumentative: the absence of retroflexion in Iranian is a separate matter. The absence of retroflexion in the Mitanni words is perfectly natural: Indo-Aryan languages migrating from India often tend to lose their retroflexes. It is possible that the Mitanni, like the Iranians before them (if they had retroflex sounds) and the Romanies or Gypsies after them (who definitely did), lost the retroflexes after emigration. In any case, the languages which borrowed and used the Mitanni words, in the records, did not use alphabets which had letters for retroflexes (and when, even today, millions of Indians write Indian words in the Roman alphabet without seeing any need to indicate the retroflex sounds, it would be too much to expect the non-IE languages which borrowed some Mitanni words to invent special alphabets to represent retroflex sounds if found in those words. Modern Arabic words used in Hindi, also, do not indicate the exact Arab sounds in the words). "Typical South Asian loan words" is an insolent phrase: How does Witzel decide that the word ani, "lynch pin", is a "typical South Asian loan word", and how does he decide that the word is absent in the Mitanni language? The sound ai instead of e in aika is too flimsy to be of any value as an indicator of its pre-Rigvedic vintage. It is definitely not my claim that the Rigveda was composed in 5000 BCE or completed in 3100 BCE, or that the Mitanni language is a form of Prakrit. But it is my claim that the Mitanni were emigrants from India in the Late Period of the Rigveda, which I have always roughly placed between 2300 BCE or so and 1500 BCE. And the evidence of the Mitanni words in West Asia proves this beyond the shadow of any doubt. It is the same story, of ara ("spokes") or of the "Bactria-Margiana words", all over again: A large number of Mitannni names end with the suffix –atti. Parpola lists the following from the Mitanni records: Biratti, Mittaratti, Asuratti, Mariatti, Suriatti, Intaratti, Paratti and Suatti [= Vedic Sanskrit Priyatithi, Mitratithi, Asuratithi, Maryatithi, Suryatithi, Indratithi, Pratithi and Suatithi]. Other names end with the suffix ?medha such as Biiriamasdha/Priyamazdha [=Vedic Sanskrit Priyamedha], the suffix ?asvasuch as Biiriaassuva [=Vedic Sanskrit Priyasva], the suffix ?sena such as Biiriasena [=Vedic Priyasena], the suffix ?ratha such as Tusratta/Tuiseratta [=Vedic Tvesaratha], or start with the prefix rta- such as Artaassumara and Artataama [Vedic Rtasmara and Rtadhaman]. Excluding the names Vadhryasva and Vrsanasva, which have a different grammatical form, and with the sole exception of one name (which is in fact an exception that actually proves the rule, as we shall see) names with the above suffixes and prefix are absent in the Mandalas of the Early Period and the Middle Period of the Rigveda, and are found only in the Mandalas of the Late Period (the non-family Mandalas I, VIII, IX and X, and in the only Family Mandala which falls in the Late Period, Mandala V): Atithi: Medhatithi VIII.8.20 Medhyatithi I.36.10,11,17; VIII.1.30; 2.40; 33.4; 49.9; 50.9; 51.1; IX.43.3 Nipatithi VIII.49.9; 51.1 Mitratithi X.33.7 Medha: Asvamedha V.27.4,5,6; VIII.68.15,16 Priyamedha I.39.9; 45.3,4; VIII.2.37; 3.16; 4.20; 5.25; 6.45; 8.18; 32.30; 69.8,18; 87.3; X.73.11 Nrmedha X.80.3; 132.7 Sumedha X.132.7 Asva: Aghasva I.116.6 Istasva I.122.13 Rjrasva I.100.16-18; 116.16; 117.17 Ninditasva VIII.1.30 Marutasva V.33.9 Vyasva I.112.15; VIII.9.10; 23.16,23,24; 24.22,23,28,29; 26.9,11; IX.65.7 Vidadasva V.61.10 Syavasva V.52.21; 61.5; 81.5; VIII.35.19-21; 36.7; 37.7; 38.8 Sena: Rstisena X.98.5,6,8 [Tvesaratha is found as a phrase, though not a name, in V.61.13] Ratha: Priyaratha I.122.7 Brhadratha I.49.6; X.49.6 Srutaratha V.36.6 Svanadratha VIII.1.32 [possibly also Dasaratha I.126.4 and Aristaratha X.6.3] Rta: Rtastup I.112.20 [Rtadhaman itself, as a name or phrase, is found in post-Rigvedic Samhitas] The word mani, referred to by Witzel, is another example. The word is very common in the post-Rigvedic texts, and in all later periods, but, in the Rigveda, it is found only in the Mandalas of the Late Period, namely, in I.33.8 and 122.14. In addition, it may be noted, about the Mitanni and late Rigvedic names beginning with Priya- above, that it is not just names, but allcompound words with priya- as the first element are restricted only to the Mandalas of the Late Period, and are very common later on, but completely missing in the Mandalas of the Early and Middle Periods of the Rigveda. The only exception referred to by us above, ie the only name found in Mandalas earlier than those of the Late Period, is the name Citraratha, found in a Mandala of the Middle Period, in IV.30.18 (and again, later, in X.1.5). This, far from disproving the rest of the evidence, actually confirms it. The only two other names which, although occurring in Mandala I, are found in the upa-mandalas of the Middle Period, are the names Istasva and Rjrasva. All these three names, the three earliest occurrences in the Rigveda of the categories of names listed above, together provide us with the period and area of the provenance of these names: they all refer to the great battle "beyond the Sarayu" between the forces of Rjrasva (Arjaspa) and Istasva (Vistaspa) in Afghanistan in the early part of the Middle Period of the Rigveda, in which Citraratha (a Puru or Vedic Aryan, who fought on the side of the Iranians) was killed [see TALAGERI 2000:214-224].This battle took place after the events of the Early Period which took place in Haryana, and then in the Punjab, and the subsequent westward expansion [see TALAGERI 2000:210-14]. In my earlier writings, both in my books as well as in my debates with Witzel-etc., I have always expressed my unwillingness to postulate "hard dates" for the events in the Rigveda without "hard evidence" like dateable inscriptions and documents, etc. Nevertheless, Witzel-etc. compelled me to express my precise views on the subject, which I did (roughly): Early Period ?3400-2600 BCE; Middle Period ?2600-2200 BCE; Late Period ?2200-1400 BCE. Witzel-etc. introduced the subject of spokes (ara) and "Bactria- Margiana words", both of which confirmed my dates, at least for the Late Period. Now, the subject of the Mitanni words, again reintroduced by Witzel, has led to an examination of the Mitanni evidence, which clearly provides irrefutable evidence for my dating for the Late Period once more, this time on the basis of actual dateable inscriptions and documents ?if not in India, then in West Asia. The Mitanni are clearly emigrants from India in the Late Period of the Rigveda. All this overwhelming evidence cannot be ignored or refuted, and it is my hope that at least scholars like Hock, if not evangelical crusaders like Witzel, will care to weigh the evidence and reconsider their positions. Source: Note from Shrikant Talageri, Dec. 16, 2005