Saturday, January 25, 2014

Location of Aratta and polemics of Witzel distorting Baudhāyana śrautasūtra

I think Aratta lies to west of river Indus and is known for horse breeding. Mahabharata talks about horses from  Aratta used in wars. Aratta was also known for irreligious and matriarchal culture. A place with a combination of all these -breeding of war-horses, matriarchy and unchaste women - to be the location of Aratta which was ruled by Sindhu kings in Mahabharata times. I am even tempted to connect it to Susa. I am giving below the Mahabharata sources.



Disregarding those arrows, the impetuous Vrikodara, with heart filled with rage, slew in that battle all the steeds, born in Aratta, of the king of the Sindhus.


And smiling the while, several warriors on thy side, with a large number of steeds consisting of the best of the Kamvoja breed as also of those born in the country of the Rivers, and of those belonging to Aratta and Mahi and Sindhu, and of those of Vanayu also that were white in hue, and lastly those of hilly countries, surrounded the Pandava army


Mighty steeds of gigantic size, of the Aratta breed, bore the mighty-armed Vrihanta of red eyes mounted on his golden car, that prince, viz, who, rejecting the opinions of all the Bharatas, hath singly, from his reverence for Yudhishthira.


Kritavarman, O king, also fled away, borne by his swift steeds, and surrounded by the remnant of his Bhoja, Kalinga, Aratta, and Valhika troops.


In former days a chaste woman was abducted by robbers hailing from Aratta.


 From Karna Parva

MB 8-45

Karna continued, Thou shouldst know all this, O Shalya. I shall however, again speak unto thee.

Listen with close attention to what I say. Once on a time a brahmana came to our house as a guest. Observing our practices he became highly gratified and said unto us, I dwelt for a long time on a peak of the Himavat quite alone. Since then I have seen diverse countries following diverse religions. Never, however, have I seen all the people of a country act unrighteously. All the races I have met will admit that to be true religion which has been declared by persons conversant with the Vedas. Travelling through various countries following various religions, I at last, O king, came among the Vahikas. There I heard that one at first becomes a brahmana and then he becomes a kshatriya. Indeed, a Vahika would, after that, become a Vaishya, and then a Shudra, and then a barber. Having become a barber, he would then again become a brahmana.

Returning to the status of a brahmana, he would again become a slave. One person in a family becomes a brahmana: all the others, falling off from virtue, act as they like. The Gandharas, the Madrakas, and the Vahikas of little understanding are even such. Having travelled through the whole world I heard of these practices, destructive of virtue, of these sinful irregularities amongst the Vahikas' Thou shouldst know all this, O Shalya. I shall, however, again speak to thee about those ugly words that another said unto me regarding the Vahikas. In former days a chaste woman was abducted by robbers hailing from Aratta. Sinfully was she violated by them, upon which she cursed them, saying, Since ye have sinfully violated a helpless girl who am not without a husband, therefore, the women of your families shall all become unchaste. Ye lowest of men, never shall ye escape from the consequences of this dreadful sin' It is for this, O Shalya, that the sisters' sons of the Arattas, and not their own sons, become their heirs.

- Jayasree


From: S. Kalyanaraman

Witzel's mistranslation or over-interpretation of Baudhāyana śrautasūtra

In a remarkable monograph, Vishal Agarwal, concludes that there is no Vedic evidence for an Aryan immigration into India. I will cite excerpts from this work of Vishal Agarwal. 

I will not digress on the arguments advanced to refute an Aryan Invasion Theory which was noted by an eminent linguist MB Emeneau as 'the linguistic doctrine' establishing the incursion of Indo-European in India. I will restrict the scope of this monograph to identifying and locating Aratta in Meluhha, in the context of my thesis on Meluhha -- a visible language.

In 1989, Witzel commented [Witzel, Michael. 'Tracing the Vedic Dialects'. In Dialectes dans les literatures indo-aryennes; Publications de l'Institute de Civilization Indienne, Serie in-8, Fascicule 55, ed. by C. Caillat, Diffusion de Boccard: Paris (1989)]: 

"In the case of ancient N. India, we do not know anything about the immigration of various tribes and clans, except for a few elusive remarks in the RV (= Rigveda), SB (= Shatapatha Brahmana) or BSS ( = Baudhāyana śrautasūtra). This text retains at 18.44 :397.9 sqq. the most pregnant memory, perhaps, of an immigration of the Indo-Aryans into Northern India and of their split into two groups: pran Ayuh pravavraja. Tasyaite Kuru-Pancalah Kasi-Videha ity. Etad Ayavam pravrajam. Pratyan amavasus. Tasyaite 
Gandharvarayas Parsavo 'ratta ity. Etad Amavasavam. "Ayu went eastwards. His (people) are the Kuru-Pancala and the Kasi Videha. This is the Ayava migration. (His other people) stayed at home in the West. His people are the Gandhari, Parsu and Aratta. This is the Amavasava (group)...the text makes a differentiation between the peoples of the Panjab and the territories West of it on one hand, and the "properly Vedic" tribes of Madhyadesa and the adjacent country East of it."  

Koenraad Elst took issue with the translation made by Witzel, of the Baudhāyana śrautasūtra passage (pages 164-165 of K. Elst, 1999. Update the Aryan Invasion Debate. Aditya Prakashan: New Delhi):

Amavasu is the subject of the second statement, but Witzel spirits the subject away, leaving the statement subject-less, and turns it into a verb, "amâ vasu", "stayed at home". In fact, the meaning of the sentence is really quite straightforward, and doesn't require supposing a lot of unexpressed subjects: "Ayu went east, his is the Yamuna-Ganga region", while Amavasu is the subject of the second statement, but Witzel spirits the subject away, leaving the statement subject-less, and turns it into a verb, "amâ vasu", "stayed at home". In fact, the meaning of the sentence is really quite straightforward, and doesn't require supposing a lot of unexpressed subjects: "Ayu went east, his is the Yamuna-Ganga region", while "Amavasu went west, his is Afghanistan, Parshu and West Panjab". Though the then location of "Parshu" (Persia?) is hard to decide, it is definitely a western country, along with the two others named, western from the viewpoint of a people settled near the Saraswati river in what is now Haryana. Far from attesting an eastward movement into India, this text actually speaks of a westward movement towards Central Asia, coupled with a symmetrical eastward movement from India's demographic centre around the Saraswati basin towards the Ganga basin..."The fact that a world-class specialist has to content himself with a late text like the BSS, and that he has to twist its meaning this much in order to get an invasionist story out of it, suggests that harvesting invasionist information in the oldest literature is very difficult indeed. Witzel claims (op.cit., p.320) that: "Taking a look at the data relating to the immigration of Indo-Aryans into South Asia, one is struck by a number of vague reminiscences of foreign localities and tribes in the Rgveda, in spite [of] repeated assertions to the contrary in the secondary literature." But after this promising start, he fails to quote even a single one of those "vague reminiscences"." 

Dr. S. Kalyanaraman, referred the matter to Dr. George Cardona- an international authority in Sanskrit grammar, and author of numerous definitive publications on Panini's grammar.Cardona clearly rejected Witzel's translation, and upheld the objections of Elst on the basis of rules of Sanskrit grammar. He stated (Message no. 3 (dated April 11, 2000) in the public archives of the Sarasvati Discussion list. The website of the discussion list was . The list is now defunct and messages are no longer available. In the original message, the word aayu was spelt incorrectly advertently as 'saayu'. This error was pointed out by Dr. Cardona himself, and has therefore been incorporated in the citation in the present article.):

"The passage (from Baudha_yana S'rautasu_tra), part of a version of the Puruuravas and Urva'sii legend concerns two children that Urva'sii bore and which were to attain their full life span, in contrast with the previous ones she had put away. On p. 397, line 8, the text says: saayu.m caamaavasu.m ca janayaa.m cakaara 'she bore Saayu and Amaavasu.' Clearly, the following text concerns these two sons, and not one of them along with some vague people. Grammatical points also speak against Witzel's interpretation.First, if 
amaavasus is taken as amaa 'at home' followed by a form of vas, this causes problems: the imperfect third plural of vas (present vasati vasata.h vasanti etc.) would be avasan; the third plural aorist would be avaatsu.h. I have not had the chance to check Witzel's article again directly, so I cannot say what he says about a purported verb form (a)vasu.h. It is possible, however, that Elst has misunderstood Witzel and that the latter did not mean vasu as a verb form per se. Instead, he may have taken amaa-vasu.h as the nominative singular of a compound amaavasu -meaning literally 'stay-at-home', with -vas-u- being a derivative in -u- from -vas. In this case, there is still what Elst points out: an abrupt elliptic syntax that is a mismatch with the earlier mention of Amaavasu along with Aayu. Further, tasya can only be genitive singular and, in accordance with usual Vedic (and later) syntax, should have as antecedent the closest earlier nominal: if we take the text as referring to Amaavasu, all is in order: tasya (sc. Amaavaso.h). Finally, the taddhitaanta derivates aayava and aamaavasava then are correctly parallels to the terms aayu and amaavasu. In sum, everything fits grammatically and thematically if we straightforwardly view the text as concerning the wanderings of two sons of Urva'sii and the people associated with them. There is certainly no good way of having this refer to a people that remained in the west." 

The noted archaeologist B. B. Lal (Lal, B. B. 1998. India 1947-1997, New Light on the Indus Civilization. Aryan Books International: New Delhi)  has also stated out that Witzel's translation is untenable and is a willful distortion of Vedic texts to prove the non-proven Aryan migration theory (AMT). Lal's criticism is along the same lines as that of Elst. 

Willem Caland's Dutch translation: It is he who first published the Baudhāyana śrautasūtra from manuscripts.(In three volumes, from 1903-13, by Bibliotheca Indica (Calcutta). In an obscure study15 of the Urvashi legend in Dutch, he focuses on the version found in Baudhāyana śrautasūtra 18.44-45 and translates the relevant sentences of text as (Caland, Willem. 1903. "Eene Nieuwe Versie van de Urvasi-Mythe". In Album-Kern, Opstellen Geschreven Ter Eere van Dr. H. Kern. E. J. Brill: Leiden, pp. 57-60).
"Naar het Oosten ging Ayus; van hem komen de Kuru's, Pancala's, Kasi's en Videha's. Dit zijn de volken, die ten gevolge van het voortgaan van Ayus ontstonden. Naar het Westen ging Amavasu; van hem komen de Gandhari's. de Sparsu's en de Aratta. Dit zijn de volken, die ten gevolge van Amavasu's voortgaan ontstonden." 
Translated into English (by Koenraad Elst.), this reads – 
"To the East went Ayus; from him descend the Kurus, Pancalas, Kasis and Videhas. These are the peoples which originated as a consequence of Ayus's going forth. To the West went Amavasu; from him descend the Gandharis, the Sparsus and the Arattas. These are the peoples which originated as a consequence of Amavasu's going forth." 
The text, as reconstituted by Caland (and also accepted by Kashikar – see below) reads 'Sparsus', which apparently stands for the peoples who are known as 'Parshus' elsewhere in the Vedic literature, and are often identified as the ancestors of Persians (or even of Pashtuns). Clearly, Caland interpreted the passage to mean that from a central region, the Arattas, Gandharis and Parsus migrated west, while the Kasi-Videhas and Kuru-Pancalas migrated east. Combined with the testimony of the Satapatha Brahmana (see below), the implication of this version in the Baudhāyana śrautasūtra, narrated in the context of the Agnyadheya rite is that that the two outward migrations took place from the central region watered by the Sarasvati. (Kashikar, Chintamani Ganesh. 2003. Baudhāyana śrautasūtra (Ed., with an English translation). 3 vols. New Delhi: Motilal Banarsidass/IGNCA).

In volume III of his translation, on p. 1235, Kashikar translates the relevant sentences of the text as follows- 
"Ayu moved towards the east. Kuru-Pancala and Kasi-Videha were his regions. This is the realm of Ayu. Amavasu proceeded towards the west. The Gandharis, Sparsus and Arattas were his regions. This is the realm of Amavasu." 

This is again a straightforward translation of the passage in accordance with the rules of Sanskrit grammar.
Toshifumi Goto's German Translation: 

In his recent study [Tushifumi Goto. 'Pururavas und Urvasi" aus dem neuntdecktem Vadhula-Anvakhyana (Ed. Y. Ikari)'. pp. 79-110 in Tichy, Eva and Hintze, Almut (eds.). Anusantatyai; J. H. Roll: Germany (2000)] of the parallel passages dealing with the Agnyadheya rite, Goto translates the Sutra passage in the following words (p. 101 sqq.) – 
""Nach Osten wanderte Ayu [von dort] fort. Ihm gehdie genannt werden: "kurus und pancalas, kazis und videhas."{87} Sie sind die von Ayu stammende Fortfuehrung. {88} Nach Westen gewandt [wanderte] amavasu [fort]. Ihm gehoeren diese: "gandharis, parzus, {88} arattas". Sie sind die von amAvasu stammende [Fortfuehrung]. {90} {87}iti kann hier kaum die die Aufzaehlung abschliessende Partikel (Faelle bei OERTEL Synt. of cases, 1926, 11) sein. In den beiden Komposita koennte der Type ajava'h' [die Gattung von] Ziegen und Schafen' vorliegen: pluralisches Dvandva fuer die Klassifikation, vgl. GOTO Compositiones Indigermanicae, Gs. Schindler (1999) 134 n. 26. {88} Gemeint ist hier wohl die Erbschaft seiner Kolonisation ("Fortwanderung"); mit 
bekannter Attraktion des Subj.-Pronomens in Genus und Numerus an das Pr 
{89} Mit WITZEL, Fs. Eggermont (1987) 202 n. 99, Persica 9 (1980) 120 n.126 als gandharayas parsavo statt -ya sparsavo aufgefasst, wofuer dann allerdings im rezenten BaudhSrSu die Schreibung gandharayah parsavo zu erwarten wals -SP- ausgesprochen wurde (wie z.B. in der MS, vgl. AiG I 342) und noch kein H (fÔr das erste s) eingefuehrt wurde. -yaspa- entging einer  (interpretatorischen) {90} Dahinter steckt wohl die Vorstellung von Ayu' als normales Adjektiv 'lebendig, beweglich' und entsprechend, wie KRICK 214 interpretiert, von amavasu-: "nach Westen [zog] A. (bzw.: er blieb im Westen in der Heimat, wie sein Name 'einer, der Gueter daheim hat' sagt."." 
Loosely translated23 into English, this reads – 
"From there, Ayu wandered Eastwards. To him belong (the groups called) 'Kurus and Panchalas, Kashis and Videhas' (note 87). They are the branches/leading away (note 88) originating from Ayu. From there, Amavasu turned westwards (wandered forth). To him belong (the groups called) 'Gandharis, Parsus (note 89) Arattas'. They are the branches/leading away originating from Amavasu. (note 90)." {90}: It appears that the notion of 'Ayu' as an normal adjectival sense 'living', 'agile' underlies this name. Correspondingly, Krick 214 interprets Amavasu as – "Westwards [travelled] A. (or: he stayed back in the west in his home, because his name says –'one who has his goods at home')". 

A very strong piece of evidence for deciding the correct translation of Baudhāyana śrautasūtra 18.44 is the passage that occurs right after it, i.e., Baudhāyana śrautasūtra 18.45...From this text, it is clear that Urvasi, Pururava and their two sons were present in Kurukshetra in their very lifetimes. There is no evidence that they traveled all the way from Afghanistan to Haryana (where Kurukshetra is located), nor is there any evidence that she took her sons from Kurukshetra to Afghanistan after disposing off the pitcher. The passage rather only to indicate that the family lived in the vicinity of Kurukshetra region. Therefore, the possibility that Amavasu, one of the two sons of Pururava and Urvasi lived in Afghanistan from where Ayu, the other son, migrated to India is totally negated by this passage. Rather, BSS 18.45 would imply that the descendants of Amavasu, i.e., Arattas, Parsus and Gandharis migrated westwards from the Kurushetra region. (It may be pointed out that in Taittiriya Aranyaka 5.1.1, the Kurukshetra region is said to be bounded by Turghna (=Srughna or the modern village of Sugh in the Sirhind district of Punjab) in the north, by Khandava in the south (corresponding roughly to Delhi and Mewat regions), Maru (= desert, noting that the Thar has advanced eastward into Haryana only in recent centuries) in the west, and 'Parin' (?) in the east. This roughly corresponds to the modern state of Haryana in India)...

According to Witzel, Hertha Krick and Asko Parpola, BSS 18.44 designates the 
homeland of Gandharis, Parsus and Arattas as 'here' ('ama' in 'amavasu'). Prima facie, this suggestion is illogical, because the territory inhabited by these three groups of people is a vast swathe of land comprising a major portion of modern-day NWFP/Baluchistan provinces of Pakistan, and much of Afghanistan. To denote such a vast territory by 'here', while contrasting it with supposed migrations of Kurus and other Indian peoples from 'here' to 'there' (= northern India) is somewhat of a stretch. Baudhāyana (or whoever wrote BSS 18.44) was definitely a resident of northern India, and for him,  Afghanistan and northwestern Pakistan would be 'there', and not 'here' or 'home'...

The Location of 'Aratta' of Baudhāyana  Kalpasūtra

In an online paper, Witzel tries to minimize the important he placed earlier on BSS 18.44 as the only important direct evidence for an Indo-Aryan immigration. He also argues (Michael Witzel. 2001. 'Autochthonous Aryans? The Evidence from Old Indian and Iranian Texts." In Electronic Journal of Vedic Studies, vol. 7, issue 3. See footnote 45 on page 16 at online paper available at

"…However, the passage plays, in the usual Brahmana style, with these names and their Nirukta-like interpretations and etymologies. They are based (apart from Ayu: ayus 'full life span'), on the names of the two sons of Pururavas, Amavasi: ama vas 'to dwell at home', as opposed to Ayu: ay/i 'to go', contrasting the 'stay home' peoples in the west (Amavasyavah: Gandhara, Parsu, Aratta) with those (Ayavah: Kuru-Pancala, Kasi-Videha) who went/ went forth (ay/i + pra vraj) eastwards, as the text clearly says. A note of caution may be added: The missing verb in the collocation pratyan Amavasus allows, of course, suppletion of pravavraja. If one follows that line of argument, one group (the Ayavah) 'went east', the other one (the Amavasyavah) 'went west', both from an unknown central area, to the west of the Kuru lands. The Kuruksetra area is excluded as the Kurus went eastwards (i.e., toward it!), apparent from somewhere in the Punjab, (e.g., from the Parusni, the place of the Ten Kings' Battle, RV 7.10)…..The passage in question is just one point in the whole scheme of immigration and acculturation… The Gandhari clearly are located in E. Afghanistan/N. Palistan, the Parsu in Afghanistan and the Aratta seem to represent the Arachosians (cf. Witzel 1980); note the Mesopot. Aratta, the land of Lapis Lazuli (cf. Possehl 1996b, Steinkeller 1998)."
We may easily dismiss Witzel's attempt to impose his Nirukta like etymologies in this Sūtra passage for the simple reasons that they are opposed to the rules of Sanskrit grammar (as elaborated by George Cardona cited by me above), and because the parallel passages from Baudhāyana śrautasūtra 18.55,  Śathapathabrāhmaṇa XI.1.5 and Vādhula Anvakhyāna 1.1.1-2 clearly pre-suppose the Kuruksetra region as the scene of action involving Pururava and Uruvasi. Witzel refers to his publication 'Witzel (1980)' as proof that Arattas were 'Arachosians' (= residents of Helmand valley in S W Afghanistan), but when that publication was checked (See footnote 3 in Witzel, Michael. 1980, 'Early Eastern Iran and the Atharvaveda', in Persica, vol. IX, pp. 86-128.), it was found to place the Arattas in the Badakhshan area in extreme N E Afghanistan! 

...Witzel's interpretations are valid only if Aratta can be removed from W. Panjab (which is where the entire length and breadth of Indian literature places it) and transplanted in Arachosia (S W Afghanistan), as Witzel has done above, without any proof ( A Czech scholar Václav Blažek relies on the mistranslation of the passage in Witzel [Witzel, Michael. 'Rgvedic History: Poets, Chieftains and Politics'. in The Indo-Aryans of Ancient South Asia ed. by Erdosy, George Walter de Gruyter, Berlin: 1995: 320-321] to reinforce his conclusion that the Arattas were localized in the Helmand basin. See Blažek, Václav. 2002. 'Elamo-Arica'. In The Journal of Indo-European Studies, Vol. XXX, Nos. 3-4 (Fall/Winter 2002): pp. 215-242 (see page 216). Interestingly, in the 'Acknowledgements' section on page 235 of the paper, the Blažek says – "I wish to thank Michael Witzel for providing an opportunity to present the first version of this paper at the conference held at the Department of Indic Languages at Harvard University in May 2002…."). This alone would leave W. Punjab as a 'central area' from which some people move east and some move west. However, we may reject this possibility because as a natural corollary, it would imply that W Punjab itself did not receive any progeny of Pururavas and Uruvasi, even though regions to the east and west of it did so. 

Secondly, and more important, the other occurrences of the word 'Aratta' in the Vedic texts indicate that these people were residents of W. Punjab (north of Multan, just as in the historical period) and not of Helmand valley as proposed by Witzel and others. (The word 'Aratta' is conspicuous by its absence in Vedic literature proper, i.e., in the Samhitas, Brahmanas, Aranyakas and in the older Upanishads. The oldest text where it occurs for the first time is Baudhāyana śrautasūtra. In all later Sanskritic literature, the word denotes western and central Punjab. 'Aratta' is also mentioned as a source of Lapis Lazuli in a Mesopotamian text. Since this mineral was obtained from extreme northern regions of Afghanistan, as well as from regions just north of Quetta, some scholars have often assumed that it denoted the Helmand valley. See for instance – Hansman, J. F. 1978. 'The Question of Aratta'. In The Journal of Near Eastern Studies, vol. 37, no. 4, pp. 331-336 However, even if this identification in Mesopotanian texts is correct, we need not assume that the Aratta in Baudhāyana Kalpasūtra also meant the same region because the Mespotamian text and this Kalpasūtra are separated from each other by great distance and time. In my opinion, it is more appropriate to interpret this Kalpasūtra using data from successor Hindu traditions, rather than data from distant Mesopotamian traditions! ) The Baudhāyana śrautasūtra is a not stand-alone text of its particular Sakha of Krshna Yajurveda. It is in fact a (major) part of a larger text – the Baudhāyana Kalpasūtra. The various parts of the Kalpasūtra are the śrautasūtra, theHautrasūtra, Grhyasūtra, Sulbasūtra and the Dharmasūtra. Hindu tradition attributes all the portions of the Sutra to the same person, viz. Muni Baudhāyana. To modern scholarship however, the Kalpasūtraappears to be a stratified text. (A major portion of the Kalpasūtra is attributed to Baudhāyana himself, the Dvaidhasūtra is said to be an addition by his direct disciples (or near immediate disciples), portions ofGrhyasutra are attributed to him with the Grhyaparisesha being a late addition. The Dharmasūtra  is considered a very late addition. Patrick Olivelle [2000. Dharmasūtras, annotated text and translation. New Delhi: Motilal Banarsidass, p. 10] assigns a date of 150 B.C.E. to Baudhāyana dharmasūtra, a date which is ridiculously recent in my opinion.) It is not relevant here to discuss the merits of these various views related to the authorship of Baudhāyana  Kalpasūtra here. Even if the entire Kalpasūtra is not from the same author, the later parts nevertheless reflect the understanding of the older tradition by the later-day Baudhayaniyas. 

Even if we assume that 'Brahmana-like' portion BSS 18.44 is an older part of the text, it is worthwhile investigating what the words Aratta, Parshu and Gandhara mean in other portions of the Kalpasūtra text. Parsu does not appear to occur elsewhere in Baudhāyana Kalpasūtra. Aratta and Gandhara however are found mentioned in BSS 18.13 and in Baudhāyana Dharmasūtra (= BDS). 
Let us consider BDS (I have used the edition by Umesh Chandra Pandey. 1971. Baudhāyana Dharmasūtra (with Govindswami's commentary, and a gloss by Chinnaswami Shastri). Chaukhamba Sanskrit Series: Varanasi) First Sūtra defines Aryavarta as the land west of Kalakavana (roughly modern Allahabad), east of 'adarsana' (the spot where Sarasvati disappears in the desert), south of Himalayas and north of the Vindhyas. An alternate definition of Aryavarta in sūtra restricts Aryavarta to the Ganga-Yamuna doab. The text then enumerates the following peoples who are of 'mixed' origins, and therefore whose traditions are not worthy of emulation by the residents of Aryavarta – 
"Avanti (-Ujjain), Anga (= area around modern Bhagalpur in Bihar), Magadha, Surashtra (= modern Kathiawar), Upavrta, Sindhu (= modern Sindh), Sauvira (= modern Bahawalpur, and Pakistani Panjab south of Multan) are (i.e., the residents of these regions are) of mixed origin."  Baudhāyana  Dharmasūtra "Aratta, Karaskara (=Narmada valley?), Pundra (=northern Bengal), Sauvira, Vanga (= southern Bengal), Kalinga – whosoever visits these areas should perform Punastoma or Sarvaprshthi sacrifices as an expiation." Baudhāyana Dharmasūtra 
Clearly, all these regions lie outside the Aryavarta where Vedic orthopraxy prevailed. What needs to be noted here is that all these 'impure' regions lie on the periphery of Aryavarta. Distant regions such as Gedrosia, Arachosia (Helmand valley), Kashmir, Kabul Valley etc., are so far that they are not even mentioned. Again, Baudhāyana śrautasūtra 18.13 also mentions that whosoever visits Sauvira, Aratta, Kalinga, Karaskara and Gandhara, should perform ritual expiations. 
In both these cases from the Baudhāyana Kalpasūtra, the word 'Aratta' denotes a region or peoples who live on the periphery of Aryavarta, or close to Aryavarta but outside it. Witzel's interpretation (Witzel 1995: 320-321) of Baudhāyana śrautasūtra 18.44 however requires Arattas to 
be in the Helmand valley in Arachosia (south-west Afghanistan), from where they then migrated into South Asia. Much of ancient Sanskrit and Vedic literature considers the people and region of central and western Punjab as impure and outsiders. But equating Aratta with Arachosia would mean that there is no region or people corresponding to central and western Punjab that is considered polluting in the Baudhāyana Kalpasūtra

Conclusion: Rather than insisting on seeing evidence for 'movement' or 'migration' in the word 'Ayu', and correspondingly 'remaining in their home' in the word Amavasu, it is perhaps less tortuous to interpret this passage figuratively in a different manner that is more consistent with the Indian tradition. Indian tradition holds that the Kuru-Panchalas, and later Kashi-Videhas followed Vedic orthoproxy (i.e., they performed fire sacrifices to the Devas) and therefore were 'alive', whereas the progeny of Amavasu did not sacrifice to the Devas and hoarded their wealth in their homes.
Whatever be the interpretation, there is no convincing way to uphold Witzel's mistranslation or over-interpretation of Baudhāyana śrautasūtra 18.44. One must be extremely wary of using at least the Vedic versions of this legend to construct real history of human migrations, otherwise we would have to deduce an emigration from India in the direction of Central Asia. There is absolutely no need to read modern and colonial Aryan invasion and migration theories into ancient ritual texts. 
Therefore, we may conclude there still exists no Vedic evidence for an Aryan immigration into India. 

Araṭa: locus and identification

The detailed background provided in the previous paragraphs, on the importance of the Baudhāyana śrautasūtra text is central to the identification of Aratta (transliterated more precisely as Araṭa.

Let me summarise the key evidences provided in the ancient text of Baudhāyana śrautasūtra:

The text makes a differentiation between the peoples of the Panjab and the territories West of it on one hand, and the "properly Vedic" tribes of Madhyadesa and the adjacent country East of it.

Moving to the east are Kuru, Panala, Kasi and Videha, together called the Ayava group.

Moving to the west are Gandhari, Parsu and Araṭa people together called the Amavasa group. The region occupied by the goup is a vast swathe of land comprising a major portion of modern-day W. Punjab, North-west Frontier Province/Baluchistan provinces and much of Afghanistan. Araṭa were residents of W. Punjab (north of Multan). Baudhāyana śrautasūtra 18.33 and Baudhāyana Dharmasūtra refer to Araṭa and Gandhara. 

Baudhāyana Dharmasūtra clubs together many regions with people of 'mixed origins' such as: "Avanti (Ujjain), Anga (region around modern Bhagalpur, Bihar), Magadha, Surashtra (modern Kathiawar), Upavrta, Sindhu (modern Sindh), Sauvira (modern Bahawalpur and Pakistani Punjab south of Multan). Extending the regions with people of 'mixed origins',  the list extends in Baudhāyana Dharmasūtra to: Araṭa, Karaskara (Narmada valley), Pundra (northern Bengal), Sauvira, Vanga (southern Bengal), Kalinga (Orissa)" -- visit to which regions, requires performance of punastoma or sarvaprshthi as expiation. 

All these regions listed in the two lists of Baudhāyana Dharmasūtra and , as 'mixed persons' and including Aratta (which may be a region around Santal Paraganas, Bastar and eastern Uttar Pradesh) can be explained as Meluhha (mleccha) area, since Meluhha (mleccha) are characterised by their use of apa-śabdas while performing yajña-s.

I entirely agree with Vishal Agarwal's explanation which notes that these as 'impure' regions lie on the periphery of Aryavarta (Ganga-Yamuna doab) where Vedic orthopraxy prevailed". If one (from Aryavarta, governed by the sacred Vedic tradition) visits these 'impure' regions detailed prescribed procedures require (punastoma or sarvaprshthi) yajña-s to be performed as expiation. I submit that these regions are the Meluhha (mleccha) speech areas. The prescribed expiation procedures are also outlined in Baudhāyana śrautasūtra 18.31 which lists 'impure regions' as: Sauvira, Araṭa, Kalinga, Karaskara and Gandhara. Visiting these 'impure regions' required performance of prescribed expiation procedured.

Thus, it is clear that in --Baudhāyana śrautasūtra 18.13 and Baudhāyana Dharmasūtra, both constituent parts of Baudhāyana Kalpasūtra -- Aratta is identified as a region on the periphery of Aryavarta (Ganga-Yamuna doab) but close to it. Such a region was peopled by Meluhha (mleccha) speakers who can be distinguished from Arya vācas, speech of residents of Aryavarta. With such a distinction, it is possible to postulate Meluhha (mleccha) as proto-Indo-Aryan or precursor versions ofPrākts or deśiSuch Mleccha vācas of 'impure regions' detailed in both the texts identified the Meluhha region  and Meluhha artisans/traders had their sea-faring merchandise and donkey caravans along the Tin road of the bronze age extending from Meluhha into the Fertile Crescent. See: Proto-Indian Meluhha, a precursor of Prākṛts and deśya

Reference to acquisition of lapis lazuli in a Mesopotamian text has to be explained. Lapis lazuli was obtained from northern regions of Afghanistan and also from regions just north of Quetta. While the regions might have lapis lazuli mines, the stones might have been routed through Araṭa (Meluhha speakers and Meluhha merchants), the same way as tin and tin-bronzes were routed along the tin road which extended from Meluhha across the Persian Gulf into Sumer/Elam/Mesopotamia.

The role of Mleccha speakers during the bronze age in an extended contact area is elaborated in the following monographs:

See: Notes on spread of lost-wax casting from Meluhha and tin trade from Meluhha
Meluhha hieroglyphs: 1. Dhokra kamar lost-wax metal caster; 2. Dance-step of Mohenjo-daro metal cast
Meluhha and tracking the Tin Road. After all, what is a Bronze Age without bronze?

The identification of Araṭṭa as a region outside of Aryavarta, but within the framework of Indus-Sarasvati Civilization core region of a sacred Vedic Sarasvati river basin, is consistent with the archaeological sites associated with the bronze age extending from Sohri-Sokhta to Rakhigarhi, from Shahdad to Dholavira, Lothal and Daimabad with most of about 2000 (or 80% of 2600) archaeological sites of the civilization located on the Sarasvati River Basin. The civilization sustained the bronze age initiatives in trade with the use of Meluhha hieroglyphs to communicate catalogs of metalware and stoneware.

Just as Paul Thieme and Satyaswarup Misra had  traced proto-Indo-aryan words through Mesopotamia-Anatolia-Mitanni using Mitanni treaties, Koenraad Elst presents the following possibilities in the context of the ongoing search for urheimat of IE speakers: 

"Their (Mitanni treaties) language was mature Indo-Aryan, not proto-Indo-Iranian.  Satya Swarup Misra argues that the Mitannic languages already showed early Middle-Indo-Aryan traits, e.g. the assimilation of dissimilar plosives (sapta > satta), and the break-up of consonant clusters by interpolation of vowels (anaptyxis, Indra > Indara). This would imply that Middle-Indo-Aryan had developed a full millennium earlier than hitherto assumed, which in turn has implications for the chronology of the extant literature written in Middle-Indo-Aryan. In the centuries before the Mitanni texts, there was a Kassite dynasty in Mesopotamia, from the 18th to the 16th century BC.  Linguistically assimilated, they preserved some purely Vedic names: Shuriash, Maruttash, Inda-Bugash, i.e. Surya, Marut, Indra-Bhaga (Bhaga meaning effectively 'god', cfr. Bhag-wAn, Slavic Bog). The Kassite and Mitanni peoples were definitely considered as foreign invaders.  They are latecomers in the history of the IE dispersal, appearing at a time when, leaving India out of the argument, at least the area from Iran to France was already IE.  They have little bearing on the Urheimat question, but they have all the more relevance for mapping the history of the Indo-Iranian group. Probably the Kassite and Mitannic tribes were part of the same migration, with the latter settling in a peripheral area and thereby retaining their identity a few centuries longer than the Kassites in the metropolitan area of Babylon.  According to Babylonian sources, the Kassites came from the swampy area in what is now southern Iraq: unlike the Iranians, who migrated from India through Afghanistan, the Kassites must have come by sea from Sindh to southern Mesopotamia.  While the Iranians migrated slowly, taking generations to take control gradually of the fertile areas to the south of the Aral Lake and of the Caspian Sea, the Kassites seem to have been a warrior group moving directly from India to Mesopotamia to carry out a planned invasion which immediately gave them control of the delta area, a bridgehead for further conquests of the Babylonian heartland.  They were a conquering aristocracy, and having to marry native women, they lost their language within a few generations, just like the Vikings after their conquest of Normandy. If the earlier Kassite and the later Mitanni people were indeed part of the same migration, their sudden appearance falls neatly into place if we connect them with the migration wave caused by the dessiccation of the Saraswati area in ca. 2000 BCE. Indian-Mesopotamian connections relevant to the Urheimat question have to be sought in a much earlier period.  Whether the country Araṭṭa of the Sumerian sources is really to be identified with a part of the Harappan area, is uncertain; the Sumerian legend Enmerkar and the Lord of Araṭṭa (late 3rd millennium BCE) mentions that Araṭṭa was the source of silver, gold and lapis lazuli, in exchange for grain which was transported not by ship but over land by donkeys; this would rather point to the mining centres in mountainous Afghanistan, arguably Harappan colonies but not the Harappan area itself.  However, if this Araṭṭa is the same as the Indian Araṭṭa (in West Panjab) after all, it has far-reaching implications.  Araṭṭa is Prakrit for A-rASTra, 'without kingdom'.  The point here is not its meaning, but its almost Middle-Indo-Aryan shape.  Like sapta becoming satta in the Mitannic text, it suggest that this stage of Indo-Aryan is much older than hitherto assumed, viz. earlier than 2000 BCE."

This brilliant linguistic analysis points to the word Araṭṭa itself as a Proto-Indo-Aryan gloss with the semantics 'without kingdom' treating the word as a Mleccha (Meluhha) or Proto-Prākṛt tadbhava from a-rāṣṭra (Arya vācas or Sanskrit of the Aryavarta, Ganga-Yamuna doab).

Since there are significant indicators in the Baudhāyana kalpasūtra texts pointing to the locus of Araṭṭa in West Punjab, the implications detailed by Koenraad Elst deserve to be taken note of and the identification of Araṭṭa in the extended region attempted by scholars like DT Potts and Steinkeller may have to be re-visited. In this note, I have argued that Marhashi and Meluhha are the same. 

Potts and Steinkeller argue that Araṭṭa was Marhashi.

Attached two pdf documents.

Potts also discusses: J. F. Hansman, 1978 The Question of Araṭṭa, JNES 37, 331-336 (who locates Aratta at Shahr-i-Sokhta).

In my view, the evaluation of JF Hansman is close to the mark of identifying Araṭṭa as a region close to Shahr-i-Sokhta consistent with his earlier article in Iran 10 (1972): 118, n.97. Shahr-i-Sokhta is loated near the western border of Afghanistan.  In his 1978 paper, Hansman refers to Majidzadeh suggesting the possibility of locating Araṭṭa in Kerman province, a district which ies to the west of Iranian Sistan. Majidzadeh locates Araṭṭa between the present city of Kerman and the town of Shahdad. Elsewhere, the Shahdad standard has been explained as a set of Meluhha hieroglyphs. It is possible that Meluhha speakers were settled in Shahdad.

In the light of the textual evidence from ancient Indian texts discussed in this monograph, the suggestions and critical comments made by JF Hansman have to be taken into account. He suggests that a system of direct trade existed between Araṭṭa and the Sumerian states of southern Mesopotamia. It is certainly not far-fetched to postulate donkey caravans moving from Meluhha into these Sumerian states. If Araṭṭa was in W. Punjab as the Indian texts seem to indicate, the trade from Araṭṭa could have been part of the trade with Meluhha using sea-faring means and using caravans on the Tin Road as mentioned in cuneiform texts

There are lots of arguments about Araṭṭa . See: One comment says (2011) about a hypothesis of Afghanistan as Araṭṭa : "it has been universally agreed in recent years that the most famous Old World lapis lazuli mines, those on the upper reaches of the Kokcha river, a tributary of the Oxus (Amu Darya), in the Badakhshan district of Modern Afghanistan, described by Marco Polo (Yule 1929; i. 157), were the primary source for the ancient Near East and Egypt. Evidence for exploitation of these mines in the third millenium BC has been strengthened by the discovery of raw lapis lazuli and evidence of bead manufacture of Shortughai on the river Oxus (Francfort and Pottier 1978; Francfort 1987) in a settlement where the material culture is described as largely 'Harappan'".

I find it amazing that these views of Elst are NOT taken into account by scholars who discuss the location and identification of Aratta. The Wikpedia polemics (URL given above) make a mention of this inexplicable approach in the search for truth.

As the conclusive identification of Araṭṭa can await further detail re-evaluations of many conjectures made, I submit that the Indian texts referred to in this note should also be taken into account within the context of use of Meluhha hieroglyphs on many metalware and stoneware trade transactions and I posit the hypothesis that Araṭṭa referred to a region in the present-day Western Punjab where Meluhha speakers were artisans/traders working in stoneware and metalware, with particular reference to tin trade along Tin Road and spread of lost-wax casting technologies exemplified by the Dancing Girl statue ofMohenjo-daro which finds an echo on a potsherd from Bhirrana on the banks of River Sarasvati. As the secular desiccation of River Sarasvati progressed with recurrent tectonic events resulting in river migrations depriving Sarasvati River from glacial sources, there could have been movements of Meluhha speaker artisans, say, from Dholavira to Bhirrana and eastwards upto Rakhigarhi.

A simple conclusion is that Araṭṭa was in Indus-Sarasvati civilization area referred to in the overall context of cuneiform text references as Meluhha and that Araṭṭa artisans/traders were Meluhha (mleccha) speakers.

Embedded are the following for ready reference:

  • Vishal Agarwal's Vedic evidence for Aryan Migration Theories
  • DT Pott's Exit Aratta
  • Piotr Steinkeller's New light on Marhashi
  • JF Hansman' Question of Araṭṭa

The Question of Aratta Author(s): J. F. HansmanSource: Journal of Near Eastern Studies, Vol. 37, No. 4 (Oct., 1978), pp. 331-336

S. Kalyanaraman

No comments: