Language ‘X’ is mleccha in Sarasvati linguistic area
mlech ‘to utter indistinctly’ (Skt.S’Br.)
mlecchayati id. (Dha_tup. xxxii , 120) (cf. Tibetan ‘lalo’)
Early references to mleccha (meluhha) do indicate it as a dialect and NOT as a term referring to speakers or groups of people. The distinction between arya vaacas and mleccha vaacas is only in reference to, respectively, the grammatical or non-grammatical forms of the lingua franca. That a term should have been coined to represent the writing system of mleccha language is also significant. That it was called mlecchita vikalpa and that a study of this cryptography was a prescribed art by Vatsyayana should make us pause and rethink the early 'meaning' of mleccha. The famous Mesopotamian cylinder seal (showing the meluhhan merchant carrying the antelope (read: ranku, tin; ranku, antelope) also refers to meluhha as a language (requiring an interpreter). Kumarila: "Then the mlechchhas who have got no qualms of conscience in the doing of any action, could never be said to be incurring any sin, if your theory [of dharma depending upon one's conscience] ... were true." Kumarila refers to Tamil as a_ndhradra_vid.a; he also calls it ‘mleccha bhaashaa’. Bhaashaa refers to the spoken language. The categorization of aaryavaacas and mlecchavaacas is comparable to the Tamil classification of centamir.. and koduntamir.. (good and broken Tamil, respectively). (See Harold Schiffman’s notes at http://ccat.sas.upenn.edu/~haroldfs/540/handouts/indiapol/node5.html ) In the Vedic tradition, particular emphasis was laid on ‘purity’ of utterance, that is, grammatically correct rendering, as distinct from indistinct forms of ungrammatical speech referred to as mleccha. The distinction between aaryavaacas and mlecchavaacas in Manu is clearly in terms of the grammatical or non-grammatical forms of the spoken language or parole. Barbaric, that is, ungrammatical, use of language extends to treating the speakers of such dialect themselves as barbaric (that is, violators of the purity in tradition or conventions related to karmakaand.a): on an epigraph, the first independent Reddy king, Vema, "restored all the agraharas of Brahmanas, which had been taken away by the wicked Mleccha kings". (cf. Ramayya, J. (1981). "Vanapalli plates of Anna-Vema" Epigraphica Indica, vol.VIII. India: Archaelogical Surver of India, 9; v.9-12.) Deviation from the codes of conduct in performance of yajna is the context for Baudhayana (~800 BCE) defining a Mleccha as someone "who eats meat or indulges in self-contradictory statements or is devoid of righteousness and purity of conduct." It is not far-fetched to expand the classifications of aarya-speakers and mleccha-speakers (aaryavaacas and mlecchavaacas) to the classification of deva and asura, respectively. Deva-asura divergence may explain the divergence of proto-iranian from proto-prakrit – both tracing their roots to mleccha as the spoken idiom of the language (Language ‘X’) and to Samskrtam as the grammatically correct version of the same language – mleccha, meluhha. Arvind Sharma refers to the dilemma posed by George Coedes, “How can we explain the maritime drive of a people who regarded crossing the black water and contact with the mleccha barbarians as bringing defilement and pollution?” (Arvind Sharma, 1992, Ancient Hinduism as a missionary religion, Numen, Vol. 39, No. 2 (Dec., 1992), pp. 175-192 http://links.jstor.org/sici?sici=0029-5973(199212)39%3A2%3C175%3AAHAAMR%3E2.0.CO%3B2-L Is Kalki avata_ra a mleccha? http://www.svabhinava.org/abhinava/abhIntro-superframe.php śābarôtsava m. a partic. festival of the Mlecchas KālP. (Skt. Lex.)
“The Padma Purana (6.71.279-282) relates that Lord Kalki will end the age of Kali and will kill all the wicked mlecchas and, thus, destroy the bad condition of the world. He will gather all of the distinguished brahmanas and will propound the highest truth. He will know all the ways of life that have perished and will remove the prolonged hunger of the genuine brahmanas and the pious. He will be the only ruler of the world that cannot be controlled, and will be the banner of victory and adorable to the world.” http://www.stephen-knapp.com/kalki_the_next_avatar_of_God.htm A variant is a reference to mleccha in the context of Buddha avatara, as daitya-daanava mohine: [quote] In Chapters 17-18 of the 3rd Section of Vishnu Purana, Buddha has been designated as ‘Mayamoha’. Once, while bathing in the waters of the Yamuna, Akrura was astonished to see Krishna-Balarama within the river. Coming out, he saw Them seated in a chariot as They had been before appearing in the water. Again he immersed himself in the water, and saw the yellow-clad four-handed Vasudeva Sri Krishna along with His associates, graciously seated on the lap of the thousand-hooded Sri Anantadeva while being worshiped by Brahma and other demigods. At that time, he prayed to the Lord in the following manner: namo buddhaya suddhaya daitya-danava-mohine mleccha-praya-kshatra-hantre namas te kalki-rupine (Srimad-Bhagavatam 10.40.22) [unquote] http://www.bvml.org/SBBTM/buddha.html
http://download.yousendit.com/CA1BAE8B67ECB5FB Formation and Evolution of Bharatiya languages (ebook for download)
My hypothesis is that language ‘X’ is mleccha of Sarasvati linguistic area. I may further add that Munda in Lahurdewa metallurgical area is a spark from this integral anvil. So are Nahali the language integrate of Tapati basin, Tamil of Sangam period, Gautama’s Prakrit and Mahavira’s Pali, rendering the entire nation of Bharatam as a linguistic area tracing its roots to mleccha.
Bronze age sites and Mleccha-speaker regions
I suggest that it is not mere coincidence that regions of austro-asiatic speakers correlate with the bronze age and neolithic metal-bearing sites.
The frames of slides used by Prof. Shivaji Singh in his lecture are presented in text form (below), as we wait for the final paper with references and bibliography. I deem it a privilege to circulate these insights of Prof. Shivaji Singh for comments and suggestions.
The thoughts expressed by Prof. Shivaji Singh open up a new paradigm, a new field of archaeolanguage studies which should occupy many archaeologists and language scholars for generations to come. It will be a great day when every archaeological artefact found can be ‘named’ using the concerned community’s lingua franca. We have miles to go. We have to agree on a method to carry this archaeolanguage enquiry further.
I find that Nahali is omitted from the lists of North and South Munda subgroups based on David Stampe. It may be useful to Stampe’s views on Nahali. (Kuiper has done some work on Nahali and IE linguists seem to think that Nahali is a language isolate since 50% glosses are Indo-Aryan, 25% Dravidian and 25% Munda).
Bharatam being the homeland of Austroasiatic is a thesis that should be pursued further. Has David Stampe countered Robert Blust’s arguments?
There will be lots of counter views by IE linguists who will point to libraries of their books showing inheritance of characteristics in languages. Are the views of Subhash on adequate to counter the IE linguists’ objections?
The powerful argument is on a framewith Emeneau commending the linguistic area as a way of studying language families. Unfortunately, IE linguists dismiss Emeneau-type approach as ‘areal linguistics’.
The key frame is that which refers to a language community and NOT a language family.
The language community is offered as an alternative paradigm in language evolution studies. When revolutions such as the organized cultivation of rice or alloying of two or more metals occurs, surely, the language community would ‘invent’ new glosses to cover these revolutionary surges in community-living. Is there a way to isolate these ‘inventions’ in the Munda community which is coterminus geographically with the metals and minerals zones in Bharatam and southeast asia?
Explaining how Bellwood is ignoring crop genetic researches, a problem will occur with the statements made about migrations of people. The key issue will be the directions of migrations of ancestors of munda-speakers of today. A possibility of migration from Sarasvati basin eastward cannot be ruled out. Maybe there is a grain of truth in the possibility of munda speakers or mleccha community in sarasvati civilization times (say 5000 BCE onwards, with particular reference to metals working) attested in this community’s words in Rigveda.
The last frame tantalizingly leaves the question of chronology unanswered. We just do not have enough language analysis tools to date glosses.
The question of Language X is moot. Language X is, in fact, mleccha (meluhha) as shown by the glosses related to metallurgy and the hieroglyphs of Sarasvati hieroglyphs representing these glosses rebus through images (both signs and pictorial motifs). This mleccha converges into the proto-Prakrit languages including Dravidian community languages to create the mosaic of Bharatiya language community.
Language is a social contract as detailed at the following monograph: http://docs.google.com/View?docid=ajhwbkz2nkfv_620hs8zfc&pli=1 Indo-European Linguistics (IEL), a belief system; reclaiming history of bharatiya languages
mleccha, bharatiya who employ Prakrit form of speech Ø Manu notes (10.45): mukhaba_hu_rupajja_na_m ya_ loke ja_tayo bahihØ mlecchava_cas’ ca_ryava_cas te sarve dasyuvah smr.ta_h Ø ‘All those people in this world who are excluded from those born from the mouth, the arms, the thighs and the feet (of Brahma) are called Dasyus, whether they speak the language of the mleccha-s or that of the a_rya-s.‘ Ø S’Br. 188.8.131.52: upajigya_sya_m sa mlecchas tasma_n na bra_hman.o mlecched asurya_ hais.a_ va_g evam trans. ‘he [who speaks thus] is a mleccha, hence let no bra_hman.a speak ungrammatical – mlecched — language since such is the speech of Asuras’ Ø A Pali text, Uttara_dhyayana Sutra 10.16 notes: ladhdhan.a vi maanusattan.am aariattam pun.raavi dullaham bahave dasyoo milakkhuyaa; trans. ‘though one be born as a man, it is rare chance to be an aarya, for many are the dasyu and milakkhu’. Milakkhu and dasyu constitute the majority, they are the many. Dasyu are milakkhu (mleccha speakers). Dasyu are also aarya vaacas (Manu 10.45), that is, speakers of Sanskrit. Both aarya vaacas and mleccha vaacas are bharatiya, dasyu, people. Mahabharata alludes to ‘thousands of mlecchas’, a numerical superiority equaled by their valour and courage in battle which enhances the invincibility of Pandava (MBh. 7.69.30; 95.36). Ø Na_t.yas’a_stra XVII.29-30): dvividha_ ja_tibha_s.a_ ca prayoge samuda_hr.ta_ mlecchas’abdopaca_ra_ ca bha_ratam vars.am as’rita_ ‘The ja_tibha_s.a_ (common language), prescribed for use (on the stage) has various forms. It contains words of mleccha origin and is spoken in Bha_ratavars.a only… I have attempted a synthesis in my work, Bharatiya Languages – formation and evolution (in press). We have miles to go in delineating the diffusion of language communities and movement of mleccha (bharatiya) language-speakers in space and time.
Archaeology and Dispersal of Munda Languages (in special context of the beginnings of rice cultivation in Middle Ganga Plain)
— Lecture by Professor Shivaji Singh, Gorakhpur on 6 November 2007 at the meet organized by Director of Archaeology, Min. of Culture, Govt. of UP
“Admittedly, archaeological research can be done, and is done, in the absence of any notion of the subject population’s linguistic repertoire. Yet language is the glue that holds speech communities together, and when linguistic information is available and can be made to fit with the archaeological evidence, the result can be a richer and more detailed picture of prehistory.” – Franklin C. Southworth (in his 2005 publication: Linguistic Archaeology of South Asia, page 334)
Who can deny the genuineness of the above observation of Southworth? But, alas, the unfortunate ground reality is that linguistics is found to be often at odds with archaeological findings in South Asia.
Instances are not far to seek.
- The old and well-known linguistic-archaeological divide in Aryan invasion/ migration debate is still not bridged, and
- Now we have before us the recent case of linguistics’ incompatibility with archaeo-logical data from Lahuradewa suggesting independent domestication of rice in Ganga Plain.
Despite the vast spatial expanse of India (3.29 million sq. km.), and in spite of the richness in variety of her physical geography, geology and climate, and the resulting multitude of natural flora and fauna, its inhabitants have failed to be creative enough to start any agriculture of their own.
This is what the linguistic researches have shown.
According to linguists, the art of cultivation of all major crops being grown in this vast country today has been introduced here from outside: wheat and barley by Aryans coming from the northwest and rice by Austro-Asiatics arriving from the east.
On the face of it, one fails to understand this ‘barrenness ’ of India that linguistics portrays.
A parody of politics in academia
It all started with a parody of politics acted in the theatre of academia in the late 18th century.
A paradigm of historical linguistics was generated with its own specific concept of ‘Family of Languages’ which, as we shall see, was highly incredible. By mid-19th century, this paradigm became enchanting enough to overwhelm even the honest and non-political academicians.
As a result of studies in Indo-European, Elamo-Dravidian, and Austric/Austroasiatic language families, the earliest speakers respectively of Sanskritic, Dravidian, and Munda languages in India were proved to be outsiders.
Archaeology does not support these ‘Arrival in India’ theories propounded by linguists. Then, how to tackle with these archaeological-linguistic divides?
We must remember that various scientific disciplines are not equally scientific. Positive sciences (as physics and chemistry) are more scientific than natural sciences (like Botany and Zoology). And natural sciences are more scientific than social sciences (economics, political science, archaeology, linguistics, etc.). In this grading on the scale of scienticism, archaeology is definitely more scientific than linguistics.
This is because archaeology is ‘pratyaksh pramaana’ while linguistics is ‘anumaana’.
With these preliminary remarks, now let us come to today’s lecture proper.
Two main issues under consideration in this lecture are:
First, whether the beginning of rice cultivation in Mid-Ganga Plain is basically an independent indigenous achievement or else it has been introduced in this region by an outside inspiration from East or Southeast Asia as commonly claimed by linguists and even by a few archaeologists?
Second, is Michael Witzel’s new, rather startling, hypothesis, which claims that the Harappans spoke a form of Para-Munda language, credible?
Munda languages: Extent and affinities
Munda (known also as Kol) languages are divided into two groups:
North Munda and South Munda.
North Munda languages are spoken mostly in the Chhota Nagpur plateau of Jharkhand, in parts of Orissa, West Bengal and Bangladesh, and on the Assam-Nepal border.
South Munda, on the other hand, is spoken mainly in central Orissa and along the border between Orissa and Andhra Pradesh.
On linguistic considerations, the Munda languages are thought to be most closely related to the Mon-Khmer languages, spread intermittently across an area that ranges from Assam (in India) to Vietnam, and from Yunnan to Malaysia and the Nicobar Islands in the Bay of Bengal; and the two together are taken to constitute what is designated as the Austro-Asiatic ‘family’ of languages.
Linguistic affinities of Austro-Asiatic languages have been further found with Austronesian languages spoken mainly in the islands of Southeast Asia and the Pacific (from Taiwan to Timor and from Madagascar to Easter Island), and an Austric hypothesis has been formulated according to which Austro-Asiatic and Austronesian language ‘families’ belong to a presumed single ‘super-family’ of languages known as Austric.
Austric ‘Super-Family’ of Languages
Munda Mon-Khmer Nicobarese
Position of Munda in the Austric ‘super-family’ of languages
The Austric Hypothesis
The Austric hypothesis was proposed in 1906 by Pater W. Schmidt, a German missionary working in Southeast Asia.
He hypothesized that Austroasiatic (AA) and Austronesian (AN) language families have a common origin and that they should, therefore, be grouped together in a single super-phylum. He gave the name ‘Austric’ to it.
He presented phonological, morphological and lexical evidence in support of his hypothesis.
The general perception of linguists is that while phonological and morphological pieces of evidence are quite convincing, the lexical evidence does not stand the test of linguistic scrutiny.
Various versions of the Austric hypothesis
Not every linguist accepts the Austric hypothesis. And those who accept it differ in their views about the languages included in the Austric ‘super family’. This has resulted into several versions of the original hypothesis.
The most well-known among these is the one proposed by Paul K. Benedict in 1996. This takes Austric to include, besides AA and AN, Daic, Hmong-Mien, and Japanese. It is known as the ‘Greater Austric Hypothesis’.
Other ones (with their specific language sets) go by the name of their propounders albeit without much support from comparative linguists in general.
Historical linguistics started in the heyday of racial studies. Down to mid-20th century and later still, ‘racial culture’ and ‘language culture’ concepts were not fully disentangled.
The Austric-speakers were thought to be Proto-Australoids who reached and settled in India in very ancient times from the east Mediterranean area (Palestine).
Later, they partly moved on towards east and southeast peopling several areas en route like Myanmar, Cambodia, Laos, Vietnam, Malay Peninsula, Indonesia, the Philippines, Taiwan, and the islands of Melanesia, Micronesia and Polynesia as far as up to Easter Island. There were some migrations of these people back to India in due course.
Linguistic paradigm has since changed
Race concept has been discredited, and linguists are now using archaeological data to locate language homelands.
However, like speakers of other so-called ‘Proto’ languages, the Proto-Austric-speakers too could not be definitely correlated with any specific cluster of archaeological assemblages.
The Hoabinhian-Austric equation, suggested by some, is not accepted by most of the scholars for Hoabinhian is considered to be merely a sort of stone tool industry and not an ‘archaeological culture’.
Search for Austric, Austroasiatic and Austronesian homelands continues
It is worth noting that Pater W. Schmidt, the propounder of the Austric hypothesis, was of the opinion that the Austroasiatic homeland was in India. However, his view was opposed by scholars such as Robert von Heine-Geldern who believed that it lay in Southeast Asia.
This now over a century old debate still continues unabated.
The possibility of India being the homeland of Austroasiatic is quite strong
At the 2004 annual meeting of the Southeast Asian Linguistic Society held in Bangkok, David Stampe reiterated the view that “India may be the homeland of Austroasiatic, and Mon-Khmer reflects an offshoot that migrated eastward”.
And, more recently on Monday, 16th July 2007, Gerard Diffloth has in a lecture presented some fresh arguments suggesting that “the origins of the Austroasiatic family are more likely to be found somewhere on the shores of the Bay of Bengal
Enters Robert Blust with his ‘rice argument’
‘Austroasiatic homeland in India’ hypothesis received a jolt with Blust’s researches published in 1966.
He based his arguments on: (A) the ‘linguistic theory
of least moves’ and (B) archaeology of rice.
As Blust’s is the dominant voice in the field at present, let us note his conclusions and examine the bases of his arguments.
Blust’s main conclusions in nutshell are:
(a) the Austric homeland was located in the general region where the Salween, Mekong and Yangzi rivers run parallel on the Burma-Yunnan frontier, that is, roughly in northwestern Yunnan,
(b) Proto-Austric was spoken in this area in c. 7000-6500 BCE,
(c) the dispersal of various branches of Proto-Austric seems to have followed different river courses,
(d) Proto-Austronesian dispersal followed down the Yangzi to its mouth and thence down the Fujian coast to Taiwan, and Taiwan became the Austronesian homeland,
(e) Proto-Austroasiatic had bifurcated into its western (Munda) and eastern (Mon-Khmer) branches from its homeland in upper Burma by 7000-7500 BP,
(f) Munda probably moved down the Brahmaputra Valley into Assam and Bangladesh, whence they gradually spread westward, and Mon-Khmer down the Salween and Mekong valleys into mainland Southeast Asia, and
(g) Khasi, a Mon-Khmer language spoken in the region around Khasi and Jaintia Hills of Meghalaya, was separated from the other Mon-Khmer languages by subsequent spread of Tibeto-Burmese-speaking people.
Blust’s taking Austric hypothesis as prehistory is mistaken because:
* his arguments are based on an untenable ‘linguistic theory of least moves’,
* he accepts the flawed concept of ‘language family’ current in linguistics, and
* he pushes the rice argument to its extreme, taking rice cultivation to be
almost an ethnic feature of Austric-speakers and their descendents.
Let us take up these points one by one.
The linguistic ‘theory of least moves’
Also known as ‘the theory of minimum displacement’, it was first clearly enunciated by Sapir in 1916, and formalized by Dyen in 1956.
According to this theory, the area of greatest language diversity is the most likely center of origin. That means, languages follow the rule of minimum displacement.
This theory is used to prove external origin of Sanskrit too.
How unreliable this theory could be is indicated by a simple analogy given by Arvind Sharma of McGill University, Montreal.
Consider a scenario
A thousand years from now the linguist is trying to identify the original land where English was spoken, which is now found in various forms flourishing in Australia, New Zealand, North America, U.K. and the Indian subcontinent.
Kindly remember that the actual facts, as known to you and me, have dissolved in the mist of antiquity.
How very plausible it would be to conclude that North America is the homeland of the English language, whence it spread eastward to Europe and Asia, and westward to Hawaii, Australia and New Zealand.
‘Language family’ concept is scientifically untenable
Language family concept was developed in 1860s by German philologist August Schleicher and others. The concept has been so repeatedly used since then that it is taken for granted, something not to be doubted or disbelieved.
Those were the heyday of mechanistic physics, before the laws of genetics and quantum mechanics had come to be known. Since the discovery of these laws, no successful attempt has been made to establish a rational basis for inheritance of characteristics in languages.
According to scientist Subhash Kak, “Insights from the theory of non-linear dynamics indicate that the multitude of interactions among speakers would lead to the formation of just a few languages. Strongly interacting systems of very many components, like assemblies of neurons or human speakers, have only a few stable interaction states, called attractors, associated with their behaviour, and these, for speakers are the various languages”.
This deeper understanding of the process of language formation, made possible by modernday science, goes against the language family concept.
Voices against language family concept have been raised by linguists too
The concept of ‘Linguistic Area’ given by H. V. Velton in 1943 (under its German name ‘sprachbund’, meaning ‘language space’), is one such voice.
According to M.B. Emeneau, who accepts this concept, Indo-Aryan, Munda and Dravidian languages have so much typological commonality that they should be taken to constitute a single super-phylum, and India considered a linguistic area.
Another voice of descent comes from the side of areal linguists
Although ‘research on areal linguistics’, to use the words of Martin Haspelmath, ‘is currently still in the hunting and gathering stage’, it has nevertheless posed a serious challenge to genetic linguistics.
It is a growing forceful voice as indicated by a 2001 publication (Aikhenvald and Dixon edited Areal Diffusion and Genetic Inheritance) which contains as many as twelve papers on inherited versus diffusional origins of linguistic features.
The ‘punctuated equilibrium’ idea in linguistics
According to it, relationships between languages usually expressed by the tree model are not the norm, but arise only as a consequence of ‘punctuations’ (occasions of stirring historical events such as the introduction of agriculture).
It has been introduced in linguistics from biological sciences where it was initially proposed in 1972 by Stephen Jay Gould and Niles Eldredge.
A sort of ‘revolution in evolution’, it appears to have been inspired by Thomas Kuhn’s conception of ‘paradigm change’ in his book ‘The Structure of Scientific Revolutions’ (1962).
In reality, there are language communities, not language families
Languages are, indeed, found to be related, but this relatedness is not ‘genetic’ in nature.
Languages have normally developed by a process of interactive convergence, not through divergence or bifurcations as taken in language family concept.
Language process is basically evolutionary, punctuated only rarely by revolutions (such as beginnings of agriculture).
Tagging rice cultivation and Munda
From the very beginning of historical linguistics rice cultivation was taken to be a contribution of Austric-speakers.
In Indian context, rice and Munda are tagged together.
Over half a century ago, the renowned Indian linguist Suniti Kumar Chatterjee wrote that the word ‘chaawala’ for rice would appear to be based on Middle Indo-Aryan ‘chaamala’ (OIA ‘chaama’, ‘aa-chaama’) and this ‘chaama-la’, in the original sense of ‘food’ may be connected with Munda root ‘jom’ to eat.
This derivation of ‘chaawala’ from ‘jom’ is now considered to be extremely farfetched.
Blust’s reconstruction cannot be considered credible for it accepts the old racial notion of rice cultivation being an ethnic feature of Austric-speakers (the erstwhile Proto-Australoids), besides believing in the untenable linguistic ‘theory of least moves’ and flawed ‘language family’ concept.
It is a different matter, however, that this reconstruction of Austric dispersal is still supported or at least tacitly accepted by many scholars like Paul Jen-kuei Li (2001) and Franklin C. Southworth (2005).
Michael Witzel’s Para-Munda Hypothesis
Munda linguistic archaeology has acquired a new, rather startling, dimension with Witzel’s Para-Munda Hypothesis.
In Aryan invasion/migration models, it is taken that the so-called Aryans met the Dravidians when they entered India from the northwest. According to Witzel’s Para-Munda hypothesis, the Aryans met the Munda-speakers in the Indus valley, not the Dravidians whom they encountered only later on.
According to Witzel, the language of Harappa was in all probability a sort of Munda. In his opinion, all attempts at deciphering the Harappan script have failed because it was presumed to be an early form of either Dravidian or Sanskrit.
Munda-speakers are not found in Ganga-Yamuna doab and other western areas; but Witzel traces them in Panjab!
How does Witzel do that?
He does it on the basis of: (a) some 380 ‘foreign’ words found in the Rigveda, and (b) his stratification of the Rigveda in Early, Middle and Late portions.
Shifting the ‘foreign’ (loan) words chronologically, he shows that the earliest portions of the text have Munda loans but no Dravidian. That means that in the beginning the Rigvedic people came in contact with Munda-speakers, not with speakers of Dravidian.
The so-called ‘foreign’ words in the Rigveda are the words that do not appear to be OIA.
On linguistic considerations, some of these words are identified as Para-Munda (Proto-Munda according to Kuiper), that is, belonging to an early (reconstructed) form of Munda language.
Others are linguistically distinguished as Proto-Dravidian, Proto-Tibeto-Burman, etc. or belonging to some unknown/extinct language.
These languages are considered to be substrate languages in the Rigveda.
Let me illustrate the method by an example.
The Rigveda (3.53.14) refers to a person named Pramaganda, who is chieftain of the Kikatas. This word ‘Pramaganda’ is taken to be a Para-Munda term.
It is said that this word has double prefixes (pra and ma) which a characteristic of Munda languages.
Pra (Munda per) means ‘son of’, and ma indicates possession. Ganda is Munda *gad/gand meaning ‘water’.
Thus, the word ‘Pramaganda’ means ‘son of the river/water/canal possessor’.
It is surmised that the same *gad/gand appears in the river name ‘Gandaki’ and ‘Ganga’.
Para-Munda hypothesis is incredible.
Identifying loan words is highly conjectural. For, why can’t prefix pra in Pramaganda be taken to mean ‘great’ as in pra-pautra. Maganda is a Sanskrit word meaning money-lender, a term equivalent to kusiidin (a usurer).
Witzel’s stratification of the Rigveda is different from that of Talageri (2000). Witzel does not provide any reason why his stratification should be preferred, not Talageri’s.
In fact, Munda-speakers are not found in any region to the west of Bihar, Jharkhand, Orissa, southern Madhya Pradesh and northeastern Maharashtra; and there is no solid evidence for their having been more widespread in earlier times.
Lahuradewans: Rice farmers, not wild-rice reapers
The discovery of cultivated rice in 7th millennium BCE context at Lahuradewa has attracted international attention.
Peter Bellwood (Australia) and Dorian Q. Fuller (UK) particularly have discussed this discovery in several of their recent papers.
I will base my comments on these discussions in the light mainly of linguistics and genetic studies.
The culture sequence at Lahuradewa is well defined and securely dated.
Carbonized rice grains and rice impressions in the core of pottery have been recovered from the lowest levels of Lahuradewa IA.
Dr. K.S. Saraswat, an eminent palaeobotanist of the country, has examined them and identified domesticated as well as wild varieties of rice.
A husk clot of domesticated rice is directly AMS dated to circa 6400 BCE, which is in tune with conventional 14-C dates ranging between 6000 and 3000 BCE.
So, there is hardly any doubt that Lashuradewans were cultivating domesticated rice in 7th millennium BCE.
This finding is very significant for it suggests that Middle Ganga Plain was one of the areas where rice was domesticated indigenously.
Indeed, at such an early time, domesticated rice grains could not have reached Lahuradewa from any far off place. At least there is no evidence to show that.
The debate on this finding is generated neither by any discrepancy in the excavations done nor by any misreading of the archaeological data by the excavators, but by certain models and perceptions shared by some scholars influenced by linguistic paradigm.
Peter Bellwood is one of those archaeologists who uncritically accepts the linguistic inference that “Munda-speakers with rice cultivation spread into eastern India from Southeast Asia around, or perhaps before, 3000 BC” (Bellwood 2006:58).
His reading of archaeological data at Lahura-dewa is coloured by his above perception, and so he concludes: “Thus, I suggest that Lahuradewa IA provides us with a window into the Munda past in the Ganga Basin, prior to the arrival of Indo-Aryan speaking peoples in this region.” (Bellwood, in forthcoming publication UP State Arch. Deptt.).
It would have been better had Bellwood stopped at this point. We would have overlooked him thinking he is more a linguist than an archaeologist.
But, no. He is vain enough to doubt the domestic identity of rice grains under reference and he continues: “It is difficult to argue seriously for a Munda arrival in this region much before 3000 BC, so the identity of rice dated to 6400 in Lahuradewa IA remains a little uncertain. Does it represent Mesolithic rice exploitation that left traces in the topsoil prior to the foundation of Lahuradewa IA? This is quite likely in my view” (ibid.).
Doubting identity of the rice grains after an eminent palaeobotanist has examined them is, in my view, a sort of undue overconfidence in one’s view, if not academic ‘arrogance’. Perhaps this audacity emanates from his framework of agricultural archaeology based on farming/language dispersal hypothesis that he champions.
Let us examine it. I will not talk about the flawed linguistic models he shares, for I have already spoken enough about them.
Farming/language dispersal hypothesis
The hypothesis was proposed by Colin Renfrew in 1987 (in his book Archaeology and Language: The Puzzle of Indo-European Origins), though the idea was mooted earlier by Heine-Geldern (1932) and Romney (1957).
It was readily received because it was an advance over earlier ‘elite dominance’ model explaining language dispersal by conquest.
Renfrew was influenced by Ammerman and Cavalli-Sforza’s (1984) ‘demic diffusion’ hypothesis, which he firmly tagged with farming.
Though proposed to explain Indo-European dispersal, it is considered by some scholars (e.g., Jared Diamond and Peter Bellwood) to have general validity/applicability for all language dispersals.
But, the hypothesis has lost much of its appeal today. And, there are reasons for it.
Its general applicability is challenged. According to Campbell (2003), it fails to explain why certain language families (as Indo-European) do and others (like Uralic) do not correlate with prehistoric agricultural areas.
It takes the process of language develop-ment as primarily revolutionary which, as we have noted, is basically evolutionary.
It brings down the antiquity of all the languages (except the ‘Proto’ ones) to Neolithic times, though latest researches based on lithic-linguistic correlations speak to the contrary.
It is now surmised that “the rudiments of language capacities may have been present in early Homo and that language capacities may have been fully developed in Neanderthals and other archaic Homo sapiens” (Gibson 1996:121).
Researches have established that languages had passed through three stages of evolution from isolating through inflecting to agglutinative during the Lower Palaeolithic times itself.
These stages, it may be noted, are distinguished on the basis of lexemes used (monosyllabic, polysyllabic or long) and the method employed to express grammatical functions.
Peter Bellwood shares perceptions that are on the threshold of becoming outdated.
What is more lamentable is the fact that he does not seem to be sensitive to researches being done in the field of crop genetics.
Genetic Studies on the Origin and Spread of Rice
The earlier perception that rice was domesticated only once has changed.
Recent genetic researches have positively established that cultivated rice has multiple independent domestications.
Oryza rufipogon is identified as the wild ancestor of cultivated rice. However because O. rufipogon is a perennial species and cultivated rice is an annual species, scientists now propose that the annually occurring form of O. rufipogon, Oryza nivara, may represent the most recent ancestor of Oryza sativa.
Problem lies with the genetic (taxonomical) relationship between O. sativa japonica and O. sativa indica.
Scientists engaged in plant breeding research have always experienced difficulty in hybridizing between japonica and indica cultivars. The crosses are found to be wholly or partly sterile.
It was suspected since long, therefore, that O. sativa japonica and O. sativa indica are genetically distinct from each other, and that they may have arisen from different ancestral gene pools.
Two major hypotheses about the origin of cultivated rice have been proposed.
According to one, there was a single domestication of O. sativa indica from O. rufipogon populations growing in the lowland regions of southern China. Later, O. sativa japonica varieties were developed in upland growing regions, selected from the indica rice.
According to the other, there were at least two separate domestication events leading to indica and japonica rice. This is suggested by genetic distance studies.
A ground-breaking new research published very recently
In a paper (published June 20, 2006) in PNAS (Proceedings
of the U.S. National Academy of Sciences), biologists from Washington University in St. Louis and their collaborators from Taiwan (Londo et al.) have examined the DNA sequence family trees of rice varieties and have determined that the crop was domesticated independently at least twice in different Asian locales.
They chose samples across the entire range of rice and looked for DNA sequences that were shared by both wild and domesticated types. They found that not only the cultivated but the wild rice group too clustered out geographically.
This analysis, based on clusters of cultivated and wild haplotypes, indicates that O. sativa japonica (including its tropical cultivars often referred to as javanicas) was domesticated in South China.
O. sativa indica, on the other hand, was domesticated in a separate region south and southwest of the Himalayan mountain range in India, Myanmar and Thailand.
Possibility of a third independent rice domestication in India is suggested for Aus rice cultivated exclusively in India during the short growing season.
Dorian Q. Fuller too has provided several significant pieces of evidence in support of genetic distinctions between indica and japonica in his paper ‘Non-human genetics, agricultural origins and historical linguistics in South Asia’ in Petraglia and Allchin (eds.): The Evolution and History of Human Populations in South Asia. Springer, 2007.
Most significant is Figure 1 of Fuller’s paper which is a phylogenetic representation of modern rice cultivars and wild populations based on studies of chloroplast and DNA variants called SINEs.
It is worth reproducing.
These genetic researches put to rest all controversies regarding the origins of domesticated rice.
The domesticated rice that has diffused from South china (japonica) has a different ancestry from that of the domesticated rice that has been grown in India (indica) since the very beginning of farming in this country.
So, the question of Austroasiatic, Munda or any people carrying domesticated rice grains from China to India does not arise. And, the possibility of its reaching Ganga Plain from Myanmar or Thailand is ruled out by the date of domesticated rice that Lahuradewa has given.
The last point.
What language did the Lahuradewans of the 7th millennium BCE speak?
After a thorough study of agricultural terms currently prevalent in Ganga Plain, Colin P. Masica had found (long ago in 1979) that most of them do not belong to any known language. He coined the name ‘Language X’ for the language to which they pertained.
It is this Masica’s ‘Language X’, not Munda, that was the language of the Lahuradewans of 7th millennium BCE and their contemporaries in Ganga Plain.
http://www.hindu-tva.com-a.googlepages.com/mlecchitavikalpa.doc Mleccha and mlecchita vikalpa: notes by Srinivas Tilak (Nov. 5, 2007) At one stage, mleccha referred to an alien or an outsider. According
to the Bhavishya Purana, it was King Shalivahana who demarcated
Sindhurashtra as the land and nation of the Aryas that lay east of
the Sindhu River effectively separating it from the land of the
mlecchas on the west of the Sindhu River(sthapita tena maryada
mleccharyanam prithak prithak. Sindhu sthanam iti jneyam rashtram
aryasya ca uttamam. Mleccha sthanam param sindhoh kritam tena
mahatmana (Pratisarga adhyaya 2).Mimamsa, usually dismissed as the most orthodox school of Indian
philosophy, nevertheless paid more attention to the mlecchas and
unhesitantly lauded their accomplishments in secular matters than
any other darshanas. For instance, commenting on Jaiminisutra
(1:3.10), Shabara raised and discussed the problem whether the
meaning of certain Vedic words like pica or nema (which were not
common among the Aryas but well known among the mlechhas) should be
derived from Sanskrit roots or from their actual usage among the
mlechhas. He advocated the linguistic usages of the mlecchas in
secular matters and encouraged their incorporation at the Prakrit
(lokavani) level.Kumarila (ca. 700), another great Mimasa philosopher, granted them a
potentially superior competence in worldly and secular (laukika)
matters. In his Tantravartttika he discusses the mlecchas at length
and advises to engage with them in empirical transactions
(drisharthavyavahara) and learn from them such secular professions
and skills as agriculture, astrology, and drama. Acknowledging that
the mlecchas were more qualified in fields like building houses,
producing silk products, and making harnesses he credited them for
providing appropriate terminology and words in these areas (I am
wondering if Dr Kalyanraman’s reference to and discussion
of ‘Mlecchita vikalpa’ would be relevant here)? Kumarila also
invited Indians to explore countries inhabited by the mlecchas (see
Tantra Varttika # 150, 153 on Jaiminisutra 1:3.10).Prabhakara, another leading exponent of the Mimamsa school, also
rejected parochial attempts to (1) derive all mleccha words from
Sanskrit roots and (2) construe their meanings `etymologically’
regardless of their actual usage by the mlecchas (see Shabara and
Kumarila on Jaiminisutra 1:3.10)(also Wilhelm Halbfass 1990: 179).
As a result, there has been a long tradition of Sanskrit scholars
who were diglossic (i.e., bilingual = dvaibhashika)(see Wilhelm
Halbfass, India and Europe: An Essay in Philosophical Understanding,
Delhi: Motilal Banarasidass, 1990:185).Such an early positive perception of the mlecchas however changed over the centuries. Some of the reasons may be found in Bodhayana’s Dharmasutras where he defined the mleccha as one who eats beef, records his disagreement repeatedly [assertively?], and is devoid of righteous behaviour (Gomamsa khadako yastu, viruddham bahu bhashate, sarvacara vihinasya mleccha iti abhidhiyate).
Mlecchas in Early India
Munishiram Manorharlal, 1991
Hardback, 334 pp
Exploring Identity and the Other in Ancient India
Whether it is called register, mode, style, or trope, every historian now takes seriously the idea that the way in which facts are presented is as important as the actual facts themselves. And with the complex issue of mleccha’s there are so many ways it could have been presented badly. The temptation to prove a point could descend into polemic. Extended criticism would become dry and technical (philological detail is rarely anything else). Instead Parasher has taken the word Mleccha as a thread, around which she has woven a series of explorations of different aspects. This is precisely the right approach as it conveys to the reader the multiple resonance, and the ambiguity of words that are used to define identity.
Mleccha (and its equivalent milakkha) are usually tranlated as foreigner or barbarian. A translation which is inadequate in so many ways but not least because it implies that it was a word used by Indians to describe non-Indians. In fact it is a term used by some writers who lived in certain parts of India to describe people native to what we think of as India but who lacked some important criteria the writer felt defined his cultural identity (language, religion, geographical location, ancestry etc.). Most often it was used by Brahmanical writers to describe those outside of the aryavarta (the cultural community defined by the caste system, the sanskrit language, and Brahmanical ritual).
Parsher begins with a discussion of the etymology of Mleccha. As the earliest reference occurs in the Satapatha Brahmana, which is part of an oral tradition dating to before 500 BC, scholars have usually looked for various origins in the bronze age societies of the first and second millennium BC. The chapter on this is well-referenced and fairly comprehensive, covering a bewildering array of possible theories. The temptation to ridicule what are inevitably speculative attempts at a reconstruction is avoided and there is sufficient here for the reader to get a good grasp on the problem (in fact, probably a bit too much, non-philologists will suffer no harm by skipping the chapter entirely).
The next chapter begins an exploration of the relationship between mleccha and language (vac). It is hardly surprising that since Brahmins placed so much weight on proper speech for their own position in society, that language should feature heavily in the definition of Mleccha. In fact in early texts it is clear that mleccha status was defined largely in terms of language (either the inability to use Sanskrit, or the inability to use it correctly). Language was central to identity in ancient India, as evidence by the process of Sanskritization in the early centuries AD, the importance of the Grammarians from Panini onwards. Readers interested in this aspect should also consult the very good collection of essays by Madhav M Deshpande, Sanskrit & Prakrit: Sociolinguistic Issues (Mohilal Banarsidass, 1993).
Chapters 4 and 5 tackle the issues of inclusion and exclusion. The sources provide very different views on the need to exclude, or the opportunities for incorporating mleccha. The chapters are intrigueing, though more could have been made by placing some of the sources in a chronological context. For example, the Arthasastra suggests that mleccha would make valuable mercenaries, in fact it prescribes their use for a number of activities (assassination, espionage, poisoning) which might be considered beneath arya. This is a not entirely positive view, but it is a pragmatic one. The epics, which Parsher takes as generally later in tone, also portray the mleccha as valuable mercenaries. On the other hand, the Dharmasastra literature generally takes a theoretical (but not consistent) view of non-contact with the mleccha, and the Mudraraksana a similar position, portraying Malayaketu as depending on mleccha mercenaries in contrast to Chandragupta. If the sources are taken in this order, they suggest a shift towards a rhetoric (if not reality) of mleccha exclusion. It is an interesting thread, and would perhaps be interesting to explore if the developing rhetoric of exclusion was accompanied by a development in the systems of inclusion she looks at in chapter 5. Unfortunately, while the two ideas are well developed these possibilities are obscured by the lack of a chronological context.
Chapter 6 is unfortunately the weakest. It covers the use of mleccha to describe tribal groups (especially those of central India). While there is clearly some overlap between peoples described as atavika (forest dwelling) and mleccha, Parasher completely fails to demonstrate that it was their material culture or habitation (rather than say, lack of certain rituals, caste institutions, or improper speech) which was responsible for this classification. The assertion that ‘aboriginals were apparently ostracized because of their backwardness and repulsive habits’ (213) is unsubstantiated and surely extremely suspect. Given that mleccha developed a prejorative sense in the early historic period (with the rise of militant Brahmanism) it is equally likely that aboriginal groups were presumed to be backward and have repulsive habits because they were defined as mleccha. Parasher vacilitates ‘… they were all listed together as mlecchas. This is not difficult to understand and can be explained by the fact that to the brahmin writers these people were all outside the varnasramadharma’ (214).
Chapter 7 covers foreigners, which as mentioned above is the most common translation of the term. Unfortunately it is extremely difficult to define what we mean by foreign. Some cases (such as the early Bactrian Greeks) seem obvious enough, but what of the Kushans, the Indo-Parthians, or the Western Ksatraps? The Ksatrapa rulers were supporters of Brahmanical institutions, aspired to Indian ideals such as the Cakravartin, and stood at the fore-front of trends such as Sanskritization. While their political were Scythian, very few of their ancestors would have been. Yet the Western Ksatraps are usually thought of as a foreign dynasty, in contrast to the ‘Indian’ Satavahanas. In fact the both have equal claim to be mleccha. Parasher is aware of this difficulty though is never quite able to get a grip on the problem of how to analyze an ancient notion of identity using a modern one.
The principle issue in this chapter is the reaction to the repeated invasions of the North-west from the second century BC to the fifth century AD, as typified in the Purana tradition. Including the Kushan rulers:
‘Although it cannot be conclusively established by what name the Kusanas were known in Indian writings, the role they played in the socio-economic affairs of northern India for at least two centuries could not have been totally ignored by the Brahmans. By conquering vast parts of the Gangetic Valley down to Varanasi or even further east they had disturbed the orderly existence of everyday life. Further, the fact that the Kusana kings worked essentially in a Buddhist framework, they may have posed a threat to the Brahmanical supremacy. The Indians were too weak to resist this foreign invasion, even less than the earlier incursions, and thus ultimately the period of foreign domination as a whole came to be described as one of the evils of the Kali Age in their Itihasa-Purana tradition.’ (233-4)
Parasher’s argument in this chapter is that foreign invasions broke down the social hierarchy the Brahmins needed, and thus the period was perceived in this manner by them. There are several problems with Parasher’s description of the Kushanas. Firstly, though the Kushans and Western Ksatraps extended considerable influence over the Gangetic valley, they never extended their direct rule much past Mathura. And the statement that the worked in a Buddhist framework is extremely inaccurate. The vast number of Buddhist (and Jain inscriptions) are actually private donations. Kushan kings tended to patronize their own central Asian cults in the north of the Empire or Brahmanical institutions in India itself (inscriptions 233, 235, 482, 484, 567). There is nothing to indicate that the Kushan emperors did not attempt to emulate already existing cultural norms of kingship and though there is some evidence that Indo-Scythian dynasts favored Buddhism, they did so in Gandhara, an area already peripheral by Brahmanical standards.
So given that the Kushans (and other ‘foreign’ kings) were not foreign in any real sense how can the reaction in the Purana tradition be accounted for. First it is worth establishing exactly what period the Purana tradition belongs. This is hard, as they are stratified texts, but there is an important internal clue. The texts are written as prophecies talking about future dynasties, and the future they predict runs up to the fourth century and the early Guptas – it is a list whose details focus on the Gangetic valley, and brahmanical woes. So the final form is locatable, it belongs to the Gangetic valley of the fourth and fifth centuries, the period of Gupta dominance. The general tone also fits well with other Gupta period material. The Brahmins were reactionary cultural conservatives and their Purana tradition idealises a past which did not actually exist. Part of forging a new sense of identity was creating an ‘other’, to hold up as an example of what aryavarta was not. The concept of mleccha had served that role before and so was ideally suited (when applied to the dynasties outside of the Gangetic valley) to serve the role again. In this respect Parsher makes a fundemental mistake in the final chapter. She treat the Brahmanical description of the Kali age as a reaction to some activity on the part of mlecchas. ‘The other’ does not exhibit agency, its actions, behaviour, and norms are creations of the people who write about ‘the other’, in this case the Brahmins of the Gupta period about the mleccha (Greeks, Huns, Scythians, and Kushans), and they are entirely about the ideals, aspirations, and sense of identity of the Brahmins. It is the motivations, and objectives of the Guptas and Brahmins that need to be understood.
The book ends with an appendix on the sources themselves. This is a well written, though in places there is insufficient detail. Two examples; the Mahabharata is an extremely complex text because it results from the compilation of a very lengthy oral tradition and has no single redactor, but I don’t think sufficient time is spent to justify Parasher’s particular use. Another example is the Mudrarakshasa, which Parasher says in the appendix she has already dealt with in Chapter 4. What she actually means is briefly touched upon it in a footnote on page 148. I also dislike the placement of this section. Though it is traditional to place overviews of sources into appendices it is inappropriate here, and unnecessary – readers who can handle the chapters on etymology or anthropology are going to be adequately prepared for this. This should have been positioned after the first chapter to give the reader an overview of the sources and their dates. There are many subtleties that are missed without a chronological framework, one of the subtleties a reader will miss is that though the chapters are themed those themes are arranged broadly in chronological order.
However, none of this should take away from what is an excellent book It is well-written, and though individual points can be critiicized that is inevitable given the complexity and breadth of the problem. More importantly, it does something all good history should do – it avoids telling them answers, and instead invites them to think about problems.