Names have their own stories

It is unsurprising that the two main ideologues of Pakistan, Muhammad Iqbal Lahori and Muhammad Ali Jinnah were both from recently converted Hindu families. Iqbal’s grand father Rattan Lal Sapru was a Kashmiri Pandit who went rogue (i.e. married a Punjabi Muslim and converted), whereas Jinnah’s grand father Meghji Thakkar a Gujarati Lohana (Khatri/trader class of Saurashtra) converted, as far as we know, of his own volition.

I am told (by fellow blogger Omar Ali) that there seems to be some confusion around Jinnah’s name. I was not aware of this. However, hopefully this post would clarify any confusions that may exist and raise other interesting questions.

Muhammad Ali Jinnah was born “Mahomed-ali Jinnahbhai” to one “Jinnahbhai Poonja”. Apparently the name change (dropping the -bhai) was deliberately done by Jinnah himself:

From BR Nanda’s “Road to Pakistan”, biography of Jinnah

Note that it is customary in the Indian regions of Gujarat and Maharashtra to use one’s father’s name as the middle name followed by a caste/occupation denoting surname. So, say, the Indian PM Narendra Modi’s full name is Narendra Damodardas Modi, where the middle name takes after his father’s first name: Damodardas Modi. The surname Modi is the Gujarati caste of shopkeepers/traders (common amongst Parsis too).

Furthermore, Gujarati language uses the suffix -bhai as an honorific. Its use in polite discourse is similar to Sindhi -saeeN or Japanese -san. So, Narendra Modi would be formally referred to as Narendra-bhai Modi in formal speech, say, when addressing the person on a letter. Even within close family, people can be referred to as “bhai” (or “behen”) when the addressee is not actually a brother (or sister) – leading to hilarious results in some situations.

Frequent formal usage of “bhai” (especially as a part of one’s registered name) is rather antiquated in urban areas and I’d be hard-pressed to find many such examples in city-dwelling Gujaratis of my generation. As is the norm for most social conventions in the Subcontinent, however, things take longer to change in the rural hinterland. My own anecdotal understanding is that the practice survives in mofussil towns and villages of Gujarat and nearby areas. The use of “bhai” in the Mumbai underworld (and now in the vernacular and popular culture) to refer to local crime lords also takes after the same custom due to the preponderance of Gujaratis in Mumbai.

So, it is rather obvious that the suffix -bhai in Jinnahbhai is a common Gujarati honorific. The same suffix can also be found in its Anglicized form -bhoy within Gujaratis (cf. Rai > Roy is an equivalent Anglicization in Bengali surnames). But what of the root morpheme “Jinnah”? The Gujarati context clarifies this too, as Jinna (pronounced jiNa, with a retroflex N) simply means “small” or “little” in Gujarati and is often used as a diminutive. So, the name “Jinnabhai” would really imply “little Sir” or “little mister” and is a well-attested name amongst Gujaratis. E.g. see this (excerpt below):

Jinabhai as a Gujarati first name

Finally, as far as I am aware, there is no native IA etymology of Gujarati jiNa. The apparent lack of a phonetic correlate in Sanskrit makes me conjecture that the word is actually a Dravidian lexical borrowing (maybe part of the Dravidian substrate). Sure enough Tamil (and sister Dravidian languages Telugu & Kannada) has the word chinna with an equivalent meaning and similar usage in nomenclature (e.g. chinnappa lit. “little father” or “little lord”, being a common South Indian name / surname). The word-initial /ch/ <> /j/ phonetic shift between affricatives is entirely plausible. However, we would need more examples to see if this is a systematic effect in Dravidian loanwords to IA (or vice versa).

On the “Aryan” debate – the linguistics POV

There has been a recent flurry of activity online (mostly on Twitter and mostly by Indian Twitter trolls, not counting yours truly) around the Aryan invasion/migration issue sparked by one piece in particular – namely written by Mr Tony Joseph in the Hindu. The original piece can be accessed here. Since I have a few substantive points to make from a linguistics standpoint, and lacking any expertise in genetics whatsoever I will focus of the former.

The controversy, dating back to the colonial period and weighed down by a lot of colonial baggage, has essentially been around the origin of the various peoples of India, primarily in the North & the North-West. The idea that the basis of what we now call Hindu (or more generally Indic) culture is actually European in origin (and brought to India via the Aryans) was first mooted during the colonial period. With the expansion of the British Empire, British orientalists starting from Sir William Jones (one of the founders of IE linguistics and founder of the Royal Asiatic Society in Calcutta) and followed by people like James Prinsep (decipherer of the Kharoshthi and Brahmi scripts), Sir Marc Aurel Stein, Sir Olaf Caroe, Col James Tod, Alexander Cunningham and the suchlike, came to India and contributed to this general theme in various ways. Note that most, if not all, of them were first-rate scholars of history and driven by a genuine desire to research their subject with due diligence. However, even the best researcher has a context in which (s)he operates and for these colonial historians the idea of an exogenous origin of Indian culture had a strong pull. Furthermore, all this historical research work done on the general topic of the “Aryan invasion” was, by necessity, devoid of any substantiation by population genetics – simply because the field was not invented until the 1930s. The idea of noble Aryan invaders of the hoary past who brought civilization to barbarians clearly resonated with the 20th century Fascist regimes* too, who imbibed a half-arsed notion of Aryan-ness and usage of symbolism like the Swastika, also from Sanskrit svastikaH, a compound (or samAs) form of the phrase su-asti-karoti iti (lit. good-is-doing that).

It is the linguistic and cultural notion of what Aryahood really is (the old problem) that I am interested in. While population genetics can certainly shed some light on magnitude and timing of population transfers into the Indian subcontinent, it really cannot say very much about cultural and linguistic development because that information is not encoded in our DNA but rather in our literature, in our everyday language and to some extent in our socio-religious traditions.

did Indo-European language speakers, who called themselves Aryans, stream into India sometime around 2,000 BC – 1,500 BC when the Indus Valley civilisation came to an end, bringing with them Sanskrit and a distinctive set of cultural practices? Genetic research based on an avalanche of new DNA evidence is making scientists around the world converge on an unambiguous answer: yes, they did.

Therefore, when Mr Joseph answers the second clause of his question in a ringing affirmative based on genetic evidence, he is on really thin ice. Did these self-avowed Aryans (henceforth Arya, as that is the correct Sanskrit term) actually bring Sanskrit with them? Can they be called outsiders in any meaningful sense? Is the oldest extant literature composed by people who self-designated as Arya non-Indian? The answer is an emphatic no! to all three questions.

  • Let’s start with the they-brought-Sanskrit-with-them spiel first. It is well-hypothesized that Proto-Indo-Iranian (the putative ancestor of the Indic and Iranian language families) split off from Proto-Indo-European around 2500-2000 BCE, quite possibly a result of a drawn-out process of a feudal elite immigrating, influencing or inter-marrying with tribal chieftains across Central Asia. These people clearly had a technological edge in horse domestication and use of horses yoked (cf. Skt. yoga, lit. to join together, past-participle yukt) to chariots (Skt. ratha cognate with Latin rota or Old Saxon rath, i.e. wheel). The process of largely cultural transmission took around a good 500-1000 years, before we can date use of Sanskrit in India from ~1500 BCE.

Does that mean Sanskrit isn’t native to India? Of course not. Languages aren’t things fixed in time and space, but evolving speech patterns. What we call (Vedic or pre-Classical) Sanskrit is a time snapshot of the language of Northern India and (what is now) Pakistan from around 1500 BCE (composition of the earliest Veda) to roughly 500 BCE (roughly contemporaneous with Panini) with a strong local substrate effect visible all through this period. This implies that whenever the native speakers of the old substrate language switched to a newer one, it was long before the existence of speech forms we now label Sanskrit, and Sanskrit itself evolved entirely on the subcontinent. Saying that Sanskrit is exogenous to India is as foolish as claiming that French is exogenous to France – which is obviously silly because even though (vulgate) Latin was picked up by the local Gallic-Celtic population of France under Roman rule, the French language developed entirely within what’s now France. The evidence of Sanskrit ever being used outside modern-day Indo-Pak geographical boundary is absolutely zilch!


  • What about Vedic literature’s cultural/geographical moorings? The actual content of Sanskrit compositions shows no cultural dislocation unlike, say, Turkish or Persian compositions by speakers of those languages who immigrated to India or by the bards of Old-Saxon in what’s now England. Old English epics like Beowulf are culturally and geographically located in Northern Germany and regions of Scandinavia further North. On the other hand, even the oldest compositions in Sanskrit can’t get enough of the Indus and its tributaries or of the Himalayas or the flora and fauna of Northern India. Sanskrit shows a very strong Dravidian substrate, which includes a entire series of consonants called retroflexes (or murdhanya in Sanskrit) which clearly are correlated with the Dravidian language family. Sanskrit speakers not only got the retroflex substrate but innovated on it – leading to aspirated retoflexes /Th/ and /Dh/ (where aspiration is a purely IE feature). This again is further evidence that the Sanskrit language could not have existed outside India. Further, Sanskrit also includes tonality characteristic of Austro-asiatic (of which Munda or Burmese are modern day forms). Latter-day North-Eastern IA Prakrits have Tibetan and Tai-Kadai substrate too (cf. Nepalese or Axomiya). Nonetheless, existence of substrates is not a weird or exotic feature of Sanskrit but a general natural condition of all languages. E.g. around 20-30% of all Germanic vocabulary is attributable to a substrate non-IE langauge that no longer exists.


  • Finally, I contend that the old use of the term Arya in the Indian context has primarily been a marker of culture and language use rather than racial classification. It is akin to the Classical use of the word Roman, which signified citizenship of the Roman state (senatus populus que romanus) and knowledge of (and fluency in) Latin literature and language. I do not know of a single unambiguous citation from the earliest of the Vedic scripture (which predates the oldest Avestan Gathas by half a millennium, give or take a century) that uses the term Arya for family or tribe – e.g. like the Rg Veda talks about the Bharatas, Pakhtas, Bhalanas etc. The term Arya is squarely used to define a linguistic culture and knowledge of or adherence to a specific canonical tradition, not as a tribal ethnonym. So one speaks and behaves like an Arya, if one’s educated in Sanskrit speech (vAk) and adheres to the orthodox Vedic ritual (vrata). The people who could not speak proper Sanskrit and had little/no knowledge of the Vedas were variously termed anarya, barbara (lit. stammerer, cf. Hindi verb baRbaRana to utter meaningless noise, Gk. barbaros uncivilized) or mlecchha. Going by that definition, even the Persians and Greeks were non-Aryans for the Indians – and the Mahabharata epic (probably composed originally, as Jaya, sometime around 900 BCE, with later additions up to 3rd century BCE) says so very explicitly. It terms the pArasikAH (Persians), yavanAH (Ionians/Greeks), chInAH (Chinese) etc as barbarians irrespective of their skin-tone or “racial” classification. E.g. Mahabharata Book 6 (bhISmaH parvaH), Chapter 10:

Among the tribes of the north are the Mlecchas, .. O best of the Bharatas: the Yavanas, the Chinas, the Kambojas, the Darunas, and many Mleccha tribes; the Sukritvahas, the Kulatthas, the Hunas, and the Parasikas; the Ramanas, and the Dasamalikas.

We should be very careful in reading our modern-day biases into ancient history generally, and both the far-Left and Right in India have been quite guilty of it. Of course, they all have their own pet periods of Indian history to read their views into but the ramifications are similar. Pakistanis, on the other hand, have no skin in this hot Aryan-invasion controversy because they’re Arabs and Turks after all 🙂

I don’t really think the (more recent) question of genetics of the Indian subcontinent is very germane to the socio-politics of the subcontinent. Evidence that the composers of the Vedas had patrilineal descent – separated by roughly 20 to 40 generations – from people of (what is now) Eastern Europe can be an interesting factoid and quite possibly correlated with the spread of IE languages in this part of the world, but it really adds little to the study of the Indian language or culture from the Vedic period onwards (which both the Left and Right have strong opinions on). E.g. it cannot be used in any meaningful sense to dent claims of cultural nativism made by the Hindu Right. There are other effective ways to counter such pernicious chauvinism, but that’s a topic for another day.



[*] in which I include the Iranian Pahlaviyan Shahdom along with the Third Reich, who changed the name of the country to Iran in the 1930s (from Old Persian Airiyanam-khshathra lit. dominion of the Aryans).