Guest post: Why do Indians speak English with a strong accent?

The retroflex consonants or mUrdhanya (as known in Sanskrit phonology) are there in the earliest Vedic speech. They are found in Eastern Iranic languages too, e.g. in Pashto, spoken inside the sub-continent. The best explanation for the ubiquity of retroflexes in the Indian sub-continent is the substrate effect from now-extinct northern Dravidian languages.

In short, the retroflex consonants are our identity in as much the same way as rhotic-lateral conflation (overlap of /l/ & /r/ articulation) is East Asian and they bind us together – across vast distances and differences of caste, creed and religion – like nothing else does, except maybe our brownness.

 

(Disclaimer: This blog is moderated by its owners. The views expressed are my own and may not reflect the views of the blog owners.)

42 thoughts on “Guest post: Why do Indians speak English with a strong accent?”

  1. native speakers also speak with an accent. at some point in the future presumably some # of indian accented english speakers will grow up seeing indian english as their native language and lose touch with their ‘ancestral’ language.

  2. Slapstik, what is the “dravidian” language? Is it ancient Tamil? What language might the Indus Valley Civilization spoken?

    Some Tamilians have told me that ancient Tamil and ancient Sanskrit coexisted alongside each other for thousands of years and are extremely ancient. The great Tamilians spoke both. Including Agastya (called Agastyar by Tamilians).

    1. AnAn, a general comment: You tendency to read Indian civilization into deep history amazes me, but I have to say it is a futile exercise. Often times knowing something is as much (if not more) about throwing information away as it is about imbibing it from everyone/everywhere. Most information has too much error.

      More to the point, you can call “Old Dravidian” language ancient Tamil (or ancient Malayalam for that matter).

  3. Some “Indians” (meaning Pakistanis as well) happened to be educated in schools run by Irish Catholic nuns. The Convent of Jesus and Mary is a famous example. These convent-educated girls spoke English with a better accent than many people in England. My mother tells me the nuns were very big on pronouncing words correctly, especially on differentiating between “v” and “w”. The young women produced by Kinnaird College also spoke very good and unaccented English.

    Upper-class Indians who go abroad tend to have accents which people (at least in the US) think sound vaguely British. I would guess that those who went to lower quality English-medium schools or Hindi-medium schools would have much more peculiar accents.

  4. Something that perhaps reinforces the Indian accent in highly fluent speakers is the idea that Indian English is its own reference point. A continental European may see one of the British or American accents as a final goal to emulate, whereas South Asians would have reached a high level of fluency in their home country and see English as one of their own languages with its own standards and conventions.
    Regarding retroflexion, do you guys sense that Assamese and Bengali have notably less of it compared to others, and is this a clue to the original spread of the substrate languages that had a high inventory of retroflex consonants? Do Munda languages have significantly less retroflexion than Dravidian ones?

    1. Good point.

      Standard Axomiya has no retroflexion at all. So it is well and truly less Indic in some sense than even Pashto. Standard Bangla about the same as, say, Kashmiri or Nepalese Pahari (which is not much).

      Munda (split-off of Mon-Khmer) has some retroflexion, but perhaps as superstrate (much like qafs and ghayns in Urdu). Witzel et al think that Munda is the best candidate for IVC speech, which I frankly find fantastical (no offence to the good professor).

      Lingistically speaking, India’s East and North East have far less echoes of Indic past than the rest of the Subcontinent.

  5. Good to see you back, Slapstik!

    Just a fun fact of interest to you (since linguistics is a perfect part of your element): retroflex consonants exist in nearly all East Iranian languages, with the understandable exception of Ossetian.

    Essentially, Pashto (Afghanistan/northwestern Pakistan), Bargista (Afghanistan/northwestern Pakistan), Wakhi (Tajikistan/Afghanistan), most of the East Iranian languages of the Pamiri highlands (Tajikistan/Afghanistan), etc.

    The persistence of retroflex consonants in Pashto and most of the other Central Asian East Iranian tongues seems to indicate, at least to my mind, that the BMAC civilization (and related cultures in Afghanistan and western Pakistan, like “Jiroft”) may have been home to a group of languages which included a rich inventory of retroflex consonants.

    I’m just spitballing, and I surely don’t possess even a quarter of the linguistic knowledge which you have at your intellectual fingertips, so I won’t be surprised if I’m wrong about this.

    But, going by my admittedly amateur level of knowledge concerning the subject matter in question, this seems like a sensible possibility.

    It wouldn’t be surprising, considering the recent aDNA data; BMAC and Jiroft, while highly distinct from “INP” samples, did share some very deep ancestral streams with “INP”. This was perhaps mirrored by the linguistic situation, if that makes sense.

  6. Thanks for the kind words, Commentator jân.

    I don’t know much about the genetic landscape (and someone more knowledgeable can comment) but it makes sense. Some of E Iranian speakers like Pashto (esp in N Pakistan) certainly were IA speakers before they switched to Pashto. So the substrate was re-substrated so to speak. But existence of retroflexes in descendants of Soghdian, like Wakhi, or other Pamiri-Badakhshani dialects – totally Iranic zones that were never inhabited by IA speakers AFAIK – does lend further credence to what you are implying.

    I would just be careful about retroflexes in BMAC though (if you were making that point). The signature of linguistic origin is the region of highest variation. That region is S/SW India or at least that is what is left of it.

    Perhaps the OIT guys are in their own way right about some South > North cultural diffusion in the hoary past, except they got the timing wrong. That is, diffusion of a culture speaking retroflexes happened in the Northerly direction *before* the steppe dudes messed up the Old World. Or maybe there was a lot more variation of retroflexes in the Northern marches of India that has since attenuated with natural drift and acculturation, and what surives of it is in the peninsular South. I do not know.

    1. Slapstik wrora, it goes without saying that you were missed! You’re an essential voice here, and I’m sure that many others are glad to see you back at it.

      “Perhaps the OIT guys are in their own way right about some South > North cultural diffusion in the hoary past, except they got the timing wrong. That is, diffusion of a culture using retroflexes happened in the Northerly direction…”

      In my own clumsy manner, this is exactly the notion I was trying to convey.

      For example, when looking at the genetic evidence with regard to BMAC, we do see robust evidence of increased gene-flow with populations from the northern quadrant of the sub-continent.

      Sarazm_Eneolithic has essentially 0% AASI, and belongs to a distinctive stream of West Eurasian ancestry witin the broader Iran_Neolithic-related family of ancient populations (of course, as inferred from genetic data; I try to avoid intellectual reification and abstraction as much as possible).

      Namazga_Eneolithic has essentially 0% AASI, and is a mixture between western Iran_Chalcolithic-related and Sarazm_Eneolithic-related ancestries.

      But once we start looking at later BMAC sites (Sappali Tepe, Bustan, etc), we see 5%-8% AASI, and a small trickling of a West Eurasian element highly unique to South Asia (exemplified by Shahr_I_Soktha_BA2, a diasporic IVC sample which is approximately 90% West Eurasian, but of a highly distinctive kind, with around 10% AASI).

      So, later BMAC samples do show tell-tale signs of IVC-related gene-flow. Probably indicative of substantial cultural and economic linkages, as already documented by archaeologists. It would be no leap on our part to imagine some sort of ancient sphere of socio-cultural interaction.

      That being said, one should note that even after the transformations that came with steppe-related incursions, southern Central Asia/eastern West Asia/northern South Asia have almost always constituted a highly interconnected zone of mutual socio-cultural influence.

      In truth, the spread of Islam, followed by the Turko-Mongolic transformation of Iranic Central Asia, disrupted this very ancient dynamic.

      In a world without Islamification, and without Turkification, the shared roots of Iran, Turan, and Hind would be far more obvious to all of us.

  7. English is a stress oriented language – i.e. different parts of a word are given more or less stress, which Indian tongues don’t have . That makes a large part of ‘indian’ accent ; not to speak of grammatical habits or syntax. Retroflex does play a part .OTOH Scottish English also is heavy on r . Aberdeen accent on first hearing may be incomprehensible to most Indians. (even to most English 🙂 ).

    Each word in English has to be learnt individually – there are no easy ‘rules’ . While English can be easy at many levels like word formation or grammar , it is relatively difficult to learn as the natives speak.

    1. Verb conjugation is much easier in English than in the Romance languages. Also, you don’t have to remember the gender of random objects. The gender of people is obvious enough. In French, I’m constantly forgetting whether it is “le table” or “la table”.

      Urdu and Hindi also have gender and I remember when my brother was trying to learn Urdu he had a hard time with why “kursi” (chair) is feminine and “kapra” (cloth) is masculine. On some level, it is just arbitrary.

    2. You may be changing goal-posts here a little bit. The original point wasn’t that Indians have an accent (and the French or Italians don’t), but that Indian accent is characterised as strong even after much fluency has been achieved by the speaker.

      Note that the English stress-accent throws a speaker of any of the Romance languages just as easily. Let alone Romance, Welsh speakers have a problem with it to date – sparking off many caricatures of how the Welsh from the Valleys speak. As an aside, old English TV shows usually had a Welsh actor (with a darkened face and a turban) playing Indians.

      The reason for why that accent is considered “strong” is due to the retroflex-mapping according to me. Accents themselves are simply manners of enunciation and any language, especially in/near its place of origin, shows a lot of diversity in manners of enunciation. Scotch has both Celtic and Nordic substrate.

      1. ‘Strong’ is a subjective assessment depending on the listener. The Time used to call Henry Kissinger’s strong German accent even though he went to the US as a boy. Accent is a mixture of any peculiarity of speech , stress, retroflexion, intonation, singsong and much else. Actually too much of retroflexion can result in dialectical drift rather than accent.

        Indians have a strong accent because they did not learn English from native speakers and neither their teachers.

  8. Another possible explanation is that the great retroflexion began with the farmers who disseminated agriculture east from Iran. If Elamo-Dravidian is a thing, ancient farmers from Iran could neatly account for retroflexion in both Indic and East Iranian.

    Thia begs the question as to why the languages of Iran proper generally lack retroflexes, but there are a few good reasons for why they might have been lost in the western part of the Iranian plateau and not the east.

    Also, the absence of retroflexion in Ossetic might not be due to Caucasian interference, but because Ossetians’ Sarmatian ancestors retained the old II nomadic lifestyle, and never settled down in the areas with a retroflexing substrate.

    1. Another possible explanation is that the great retroflexion began with the farmers who disseminated agriculture east from Iran. If Elamo-Dravidian is a thing,

      I think this is not a parsimonious explanation precisely because it creates a bigger problem than it tries to resolve. Besides I cannot aver enough how speculative the Elamo-Dravidian thingy is. We have absolutely zero proof of that, rather convenient, unification.

      Secondly, retroflex usage has a very strong geographical locus in India – too strong a signal to look for exogenous origins in my humble opinion. I’d love to plot out isoclines of retroflex usage frequency in South Asia (or love to get hands on any research that does this – I don’t know of any). I expect the clines to clearly indicate regions of high-frequency in South / SW of India, with a local maximum near the old Indus delta in Gujarat / Kutch.

      1. Thinking about it more, there’s an even better argument against pervasive retroflexion originating in India. I think we can assume that Indo-Aryan developed its retroflexes due to contact with the people of the IVC. But the people of the IVC probably didn’t speak an indigenous Indian language. It’s extremely unlikely that an agricultural population would adopt the language of neighboring hunter gatherers, and even though the IVC individuals tested so far have a large AASI component, that’s probably due to millennia of interbreeding, with a pace slow enough to limit the linguistic impact of the AASI hunter gatherers on the farmers. The language of the IVC people probably comes from the part of their ancestry tracing back to Iran.

        So even if Dravidian, which definitely developed retroflexion on its own, comes from India instead of Iran (and it might, even if Elamo-Dravidian is bogus), retroflexion in Indo-Aryan probably doesn’t come from India, at least originally. I suppose you could argue that even if the IVC language is Iran-derived, it could have adopted retroflexion from the indigenous hunter gatherers who the farmers from Iran mixed with. But that would mean that retroflexion in Indo-Aryan is from a substrate of a substrate (AASI language > IVC language > Indo-Aryan). I suppose that’s possible, but I don’t know of any other examples of such a distinctive feature being passed on through a series of substrates.

        1. Fraxinicus, why do you assume IVC didn’t speak Sanskrit? The artifacts from IVC strongly indicate a Hindu civilization. It is possible that IVC spoke multiple languages.

          Do you think Tamil was spoken in IVC? It might have been. The 18 Siddha Tradition might have once been popular in Pakistan and IVC. Which suggests Tamil might have been spoken in IVC. It is also possible that now extinct languages that we don’t know were spoken in IVC.

          1. AnAn
            Tamil does not have a history (of literature anyway) pre 300 BC. Tamil is a historical language like any other undergoing the same change. IVC language is anybody’s guess and no one has come up with anything remotely coherent about it since not a single letter or a single word has been deciphered. WE really don’t know any specific details of IVC society , persons, kings , priests, religion, langauge/s , etc . So, I would not bet anything on it.

          2. VC, one of my favorite books is Tirumantiram. When do you think the book was first composed? I am not asking when it was last modified or edited but rather when it first began to be composed?

            Many Tamilians say that Tirumantiram and Tirumular are older than the Mahabharata and Krishna. Do you think this is true?

          3. @AnAn
            “Many Tamilians say that Tirumantiram and Tirumular are older than the Mahabharata and Krishna. Do you think this is true?”

            Many Tamilians say lot of different things starting from Lemurian Tamil to Tamil as the Mother of all languages to what have you , we can’t be taking everything seriously. People are free to say anything Taking a considered view of available material , we cannot say anything about IVC .

          4. Because all science supports the AIT. You really need to stop taking your idiosyncratic understanding of history for granted in discussions with others. It’s not worth having a conversation with you if you operate on an entirely different epistemology from the rest of the world.

            But I’ll respond this one last time. The religion of second millennium BC Aryans forms only a small part of modern Hinduism, and it’s no surprise if some aspects of Hinduism go back to the pre-Aryan peoples of India.

            Tamil is first attested in the last centuries BC. Tamil’s ancestor might have been spoken in the IVC, but that isn’t certain. Another candidate for proto-Dravidian’s culture is the South Indian Neolithic.

            It’s possible and even probable that at least some IVC people spoke a language which has no descendants today.

        2. @Fraxinicus

          So even if Dravidian, which definitely developed retroflexion on its own, comes from India instead of Iran … retroflexion in Indo-Aryan probably doesn’t come from India, at least originally

          Your preference for explanations to locate IA retroflexes outside of S Asia, even when you are willing to consider (or concede) autochthonous retroflexes in Dravidian, is a little befuddling to me. Can you give me a good reason why you think IA picked retroflexes outside India?

          I don’t personally mind it if IA retroflexes are exogenous, I just find it very improbable given how central retroflexes are to Dravidian phonology and IA speakers have been in contact with them for a seriously long time.

          I don’t know of any other examples of such a distinctive feature being passed on through a series of substrates.

          But that is the whole point of the piece above 🙂

          The retroflex substrate in IA gets mapped to Germanic when Indians speak the latter.

          1. Hello Slapstik,

            I tried to control myself a lot not to comment on this thread for personal reasons but had to succumb ultimately 🙂

            What do you think about the hypothesis of Hans Hock school of thought that Dravidian may not have been the source of Indo-Aryan retroflexion? (One of his accessible papers on the matter is at https://www.ideals.illinois.edu/bitstream/handle/2142/26503/SLS1993v23-2_hock.pdf?sequence=2 ) He argues that since the triggers for Indo-Aryan retroflexion are apparently its sibilants (with sibilant-stop clusters like *-zdh- changing to *-Zdh- and *-ZDh- which on loss of -Z- then becoming -Dh- leading to the emergence of retroflex stops as phonemes in Indo-Aryan) which are not even present in at least the known Dravidian, the source may have been another, non-Dravidian, language in the northwest.

            He also thinks that since Proto-Dravidian did not permit retroflexes in the word-initial position (only my native subgroup, the South Dravidian-II, began to have retroflexes in the word-initial position because of metathesis and very likely considerably later- sometime in the first millennium BC; all the other three subgroups apparently still don’t allow retroflexes and alveolars in the word-initial position), Dravidian retroflexes may have been secondary- he also summons for himself the well-established Proto-Dravidian morphophonemic rules that obtain retroflexes (and alveolars) of the stop kind like -T- (articulated lenis [D], perhaps as a voiced retroflex stop or a voiced retroflex flap) and -TT- (articulated fortis geminate [TT]) and alveolars -t̲- (likely articulated lenis [d̲], voiced alveolar stop in PDr. and in Proto-Central-Dravidian but as voiced alveolar trill [r̲] in Proto-South-Dravidian) and -t̲t̲- (articulated fortis geminate [t̲t̲]) at morph boundaries when root-final retroflex lateral approximant -L (and alveolar lateral approximant -l) and retroflex and alveolar nasals -N and -n interact with the dental stop t- or its geminate tt- in the beginning of many suffixes, and speculates that all retroflex stops in Dravidian may have come about this way and that this may be why Dravidian does not have retroflexes in the word-initial position. In traditional Dravidian linguistics though, it appears that they posit both true retroflex stop phonemes (which cannot occur word-initially) and some which arose by the above mentioned morphophonemic rules. This is apparently because the available data in Dravidian does not permit including the former too in the latter category as alternations between many pairs with morph-final (typically root-final) retroflex and alveolar nasals (-N and -n) and laterals (-L and -l) and morph-final retroflex stops (-T, -TT, -t̲, -t̲t̲) cannot be established. Now, having speculated as above, Hans Hock is left with only retroflex nasals and laterals in Dravidian which he then tries to explain as having arisen in alternation (due to what causes he speculated I can’t recollect at the moment, but it could be contact with some language having true retroflex phonemes) with the more normal alveolar nasals and laterals with some limited amount of data but here too, there is not a large amount of data in Dravidian to support Hock’s proposal unfortunately. He says that the late attestation of Dravidian may have been behind the loss of recoverable alternations of this kind.

            He actually also posits a parallelism between the Indo-Aryan sound changes like *-Zdh- [–>] *-ZDh- [–>] -Dh-, etc. and the above mentioned Proto-Dravidian morphophonemic rules -L + t- giving -T-, -N + t- giving -NT-, -l + t- giving -t̲- and -n + t- giving -nt̲- and that Dravidian retroflex stops may have arisen because of contact with Indo-Aryan which arose in Indo-Aryan because of the retroflex sibilant triggers and in Dravidian because of the retroflex lateral and nasal triggers. This may be plausible as Proto-Dravidian unity is broken by Bh. Krishnamurti also at around 1500 BC (but Krishnamurti himself did not support this contact-hypothesis according to what he wrote here: http://list.indology.info/pipermail/indology_list.indology.info/1998-June/107451.html, stating that the Proto-Dravidian morphophonemic rules date as far back as 3000 BC.) (The earlier Dravidian retroflex nasals and laterals, he perhaps attributes to contact with Indo-Aryan itself (at a further earlier period) or some non-Indo-Aryan Indus Civilisation era northwest Indian language or perhaps a substrate somewhere in India , as I said above, but I have to go through the paper again thoroughly to recollect.)

            This is a very interesting hypothesis (and personally very tempting to me) in my view and I wanted to ask your opinion about it if you have any.

          2. Edit to above:

            It appears Hock posited more of a convergence-and-not-substratum hypothesis regarding prehistoric Dravidian and Indo-Aryan, with retroflex stops arising in both the languages at around the same time due to parallel phonological changes, with sibilants being the trigger in Indo-Aryan and laterals and nasals being the trigger in Dravidian. The origins of the retroflex triggers- the retroflex sibilants in IA and the retroflex laterals and nasals in Dravidian remain unclear though.

          3. @historumsi

            Very enlightening comment, and I must admit I had not heard of this paper before by Hock. I think we should discuss it further.

            The comment format is perhaps not the best way to put your thoughts across. Would you mind writing a longer piece around your comments and I’d be happy to put it up as a post, with due citation of course.

            If you can’t be bothered to write a longer piece, I’m happy to post your comment as is to start a new thread. Do you approve of it?

            I’ll add my thoughts in the comment section on that post.

          4. Sorry if I wasn’t clear. I think IA picked up retroflexes in India from the IVC people. But the IVC people’s language(s), which is what gave retroflexes to IA, was probably brought to India by pioneering farmers whose ultimate origin was in Iran.

            I accept that retroflexes are indigenous in Dravidian, but not necessarily that Dravidian is indigenous to India. If Dravidian was spoken in the IVC, and is the source of retroflexes in IA, then it probably isn’t indigenous to India, because the IVC language(s) were probably not indigenous to India. If Dravidian originates in the South Indian Neolithic and is not related in any way to the IVC, then it means that IA didn’t get its retroflexes from Dravidian, and that pervasive retroflexion originated in both India (Dravidian) and prehistoric Iran (the IVC language(s) and IA substrate).

            Like you, I consider that unlikely, which is why I’m now leaning more towards Dravidian being exogenous to India. I think that the most parsimonious explanation for retroflexes in Dravidian and in the substrates of IA and East Iranian is that they originated once, somewhere in Iran, and that they were spread to South Asia and what is now Afghanistan and Tajikistan by pioneering farmers from Iran.

            I’d be curious to know how central retroflexes are to Burushaski phonology. If they’re as important in Burushaski as they are in Dravidian, that’d be good evidence in favor of the retroflex-ex-Iran theory, as the Burusho are the best candidates for direct linguistic heirs to the neolithic farmers coming out of Iran. If retroflexes in Burushaski look more like retroflexes in IA than retroflexes in Dravidian, then I’d consider it more likely that retroflexes in Indo-Iranian languages are derived from India, probably via AASI languages having a substrate effect on the IVC language(s).

            “But that is the whole point of the piece above 🙂

            The retroflex substrate in IA gets mapped to Germanic when Indians speak the latter.”

            Can’t believe I didn’t catch that 🙂 I still doubt that the indigenous Indian languages of AASI hunter gatherers would have had much of a substrate influence on the IVC language(s). I’ll change my mind if ancient DNA shows that at some point there was a large and fast injection of AASI ancestry into Indus Valley farmers, instead of just a gradual drip feed over millenia of contact.

            It’s also interesting that we see a downgrading of the importance of retroflexion when we add a layer of substrate. Correct me if I’m wrong, but as far as I know, in Indian English retroflexion is just a phonetic detail, and has no phonological significance. Whereas Indo-Aryan acquired a whole new series of retroflex phonemes from its substrate.

          5. @Fraxinicus.

            Sorry for the delayed response (busy with work) but I see your point clearly now. And I am in general agreement with it, barring some specifics 🙂

            I agree that Dravidian is a candidate for the IVC language, but no last word has been said on this yet (and likely never will be). I also think that the origin of Dravidian culture is probably foreign. It is a clearly agrarian culture with a lot of deep similarities to other agrarian cultures of West Asia and beyond (eg JallikaTTu bull run which Mahadevan traces back to IVC). Goes without saying that Dravidian language too is ultimately a likely import with early agriculturalists from what is now called Iran.

            However, that said, retroflexion does not need to be imported from outside even if the language is. Eg the Great Vowel Shift in Middle English added all the diphthongs we hear in English today – a vowel inventory quite unlike its Germanic counterparts in the Netherlands or NeiderSachsen.

            So exogenous Dravidian (which maybe is/isn’t isomorphic with IVC speech) does not rule out indigenous retroflexes. They could be an even older substrate, or a local IVC innovation.

            Burushaski is an interesting case. There are some Burushaski speakers in Srinagar (where I was born and brought up) and the retroflex usage in their language is much higher than in Kashmiri (which is IA).

            As historumsi mentions, they have affricate retroflexes too – both voiced and unvoiced – which no Kashmiri native speaker can pronounce properly. Heck! we can’t even do voiced dental affricatives. Their (unvoiced) aspirates seem to be the same as Dardic/NW IA group and I wager occur only in borrowings from IA.

            So clearly a lot of innovative retroflexion in Burushaski from what I have heard and know.

          6. Hello Fraxinicus,

            Please forgive me if I’m being intrusive- it’s just that I find this discussion very interesting.

            Regarding the question of the place of origin of retroflexion- whether it was remote Iran or India, what do you think of the interesting fact that some languages of the Andaman islands have retroflex and dental contrast? I read that it is the case, in the 2016 book “The Languages and Linguistics of South Asia”. One of the groups of the Andaman islands, the Onge particularly, are routinely used as a proxy for the AASI in all these aDNA studies. I have seen minimum estimates by some bloggers that separate them from the mainland at 10000 years ago. Is there a possibility that they may have got the distinction from the mainlanders beginning from the British period?

            That said, you view, if I’m reading correctly, that retroflexion arose at once in east Iran and northwest India in not so distant past (? as in paleolithic Iranians but later neolithic and chalcolithic era northwest Indians) is probably supported by Hans Hock also, though perhaps not to the last detail, who seems to say that retroflexion arose in Indo-Aryan, first in the sibilant, in the northwest, and that retroflex stops emerged in both Indo-Aryan and Dravidian which he likely considers to have been present somewhere in the peninsula when it first interacted with Indo-Aryan, due to phonological convergence, at around the same time, after Indo-Aryan moved into inner India. I don’t know if he addresses the origins of the retroflex sibilant itself in the northwest, as his recent paper on the topic, titled “The Northwest of South Asia and beyond: The issue of Indo-Aryan retroflexion yet again” (https://www.degruyter.com/view/j/jsall.2015.2.issue-1/jsall-2015-0005/jsall-2015-0005.xml?format=INT) is not accessible to me, but I’m wont to speculate that he might have made use of the fact that retroflex sibilants are present as phonemes in Burushaski. And unfortunately, from the abstract, I can’t see what explanation he may have suggested for the first emergence of Dravidian retroflexion in laterals, approximants and nasals, according to his scheme that is (which is that all the Dravidian retroflex stops are secondary).

            And then, you write here,
            “I’d be curious to know how central retroflexes are to Burushaski phonology. If they’re as important in Burushaski as they are in Dravidian, that’d be good evidence in favor of the retroflex-ex-Iran theory, as the Burusho are the best candidates for direct linguistic heirs to the neolithic farmers coming out of Iran. If retroflexes in Burushaski look more like retroflexes in IA than retroflexes in Dravidian, then I’d consider it more likely that retroflexes in Indo-Iranian languages are derived from India, probably via AASI languages having a substrate effect on the IVC language(s).”

            I don’t know the relevance of Dravidian in particular here but it appears that at least the current Burushaski phonology indicates that its retroflexion is much stronger than in both Indo-Aryan and Dravidian. For example, from here (https://en.wikipedia.org/wiki/Burushaski), it can be seen that Burushaski even has retroflex affricates which Proto-Dravidian most likely did not have and as far as I know, Proto-Indo-Aryan did not too; it has the voiceless retroflex sibilant present in Proto-Indo-Aryan (or closely on the path to emergence) and at the same time it also has a retroflex approximant which Proto-Dravidian had (though it is also mentioned that this last phoneme has several other realisations also, like “a voiced retroflex sibilant with simultaneous dorso-palatal narrowing”- I don’t really know what that later part means lol, not very knowledgeable in even basic phonetics). Not sure what phonology is reconstructed for Proto-Burushaski though.

            Also,
            “It’s also interesting that we see a downgrading of the importance of retroflexion when we add a layer of substrate. Correct me if I’m wrong, but as far as I know, in Indian English retroflexion is just a phonetic detail, and has no phonological significance. Whereas Indo-Aryan acquired a whole new series of retroflex phonemes from its substrate.”

            Thanks for this! I did not realise that this is the case at all until you pointed it out. What role do you think the fact that Indian English is still not spoken as a native language by large sections of Indians may have in this? Things appear to be changing recently and English is emerging as a native language in some pockets in cities and not just as an adstrate/superstrate (I honestly am quite confused how to view English as- probably it is a superstrate as far as its impact on spoken registers of all Indian languages is concerned) in which one has to be fluent, along with one’s mother tongue, i.e. the historical phenomenon of just bilingualism with limited language shift to English.

          7. Hello Slapstik,

            Thank you very much for the response and the opportunity! I am inclined to write a post which I hope will be much tidier and less erroneous in all overall details than the comments I wrote here before but I think I require some time for it- perhaps about a week. Is it okay that I submit that piece at about this time next week? Thanks again.

  9. VC, many very intelligent Tamilians that I greatly respect believe in Kumari Kandam. What would you tell these Tamilians?:
    https://en.wikipedia.org/wiki/Kumari_Kandam

    Some Tamilian friends also insist that Tamil and Sanskrit have co-existed since the beginning of civilization and the great Tamilians saints of the past (Agastyar, Tirumular, Boganathar etc.) were experts in both Tamil and Sanskrit. How would you respond to this?

      1. Vijay

        Tell them Kumari Kandam is absolute , total rubbish. It is a waste of time to entertain and even oppose it.

        How can you rubbish above claims. Do you have credentials in History. I suspect you are some kind of science guy with post modernist leanings.

        (To be honest I have no clue what post modernist means, but has an nice accusative ring to it).

          1. sbarrkum and VC:
            :LOL:

            One thing I have learned. When a Tamilian says that Tamilians know Sanskrit (now and for many thousands of years in the past) and the scriptures better than non Tamils . . . always and completely agree. Also always agree with Tamilians who say that Tamilians are the fountainhead of all global civilization, culture, technology, good, Sanskrit, Tamil, shastras, art, music, medicine, Sanathana Dharma. Tamilians are number 1 in everything.

            Another important point to express 1000% agreement with. Tamil and Sanskrit are the two bestest languages of all time. And Hindi [Kabir I mean Urdu or Hindustani] isn’t as hot.

          2. I don’t really care about your internal Indian issues, but in my opinion, Urdu is the most sophisticated language and the most sophisticated people are ahl-e-Zabans from Agra, Lucknow and Delhi. My paternal grandmother was from Agra and she had much more elegance and class than the Punjabi side of my family.

            Urdu’s tradition goes back to Mir and Ghalib and that was considered the High Culture of Delhi and of the UP.

  10. Franxinicus,
    You need to break your prior assumptions and look at all the data with an open mind.
    “Because all science supports the AIT.” AIT is consistent with some data. But that doesn’t mean AIT is true. The entire historical record was heavily manipulated for political purposes (Catholic Church) by the 1600s and 1700s. Englightenment scholars in good faith were building on a shaky foundation.

    The written historical records of a billion people and the beliefs of a billion people cannot be discounted out of hand. It has to be carefully considered in addition to other data inputs.

    “The religion of second millennium BC Aryans forms only a small part of modern Hinduism”
    Not true. Many of my friends have memorized much of the Rig Veda, Sama Veda, Yajur Veda. Many of my friends are Sanskrit scholars.

    “and it’s no surprise if some aspects of Hinduism go back to the pre-Aryan peoples of India.”

    Arya is a cultural thing unrelated to Jati. There are many different Jatis recorded in the ancient records. All of them are Aryan even though they are genetically quite different from each other. Why do you assume that South East Asians are not Arya?

    Hinduism has infinite streams. All of them are Arya. All are valid. In fact there is no concept of “religion” in Hinduism. How can there be epistemology without an understanding of Hinduism/Buddhism/Jainism, which again requires meditative or mystical experience.

    “Tamil is first attested in the last centuries BC.” Just because scientific evidence hasn’t yet been found doesn’t mean Tamil isn’t far more ancient than that. You are discounting the entire Tamilian written record. Tamilian scholars would be deeply offended by this lack of rigorous analysis.

    “Tamil’s ancestor might have been spoken in the IVC, but that isn’t certain.” On this we are agreed.

    “Another candidate for proto-Dravidian’s culture is the South Indian Neolithic.” Where in the entire historical record is there any evidence of anything called Dravidian that wasn’t considered Arya? The idea of Dravidian was first proposed in the 1800s. What you can say is that there are now extinct Arya Jatis that played a large role in the ancient past. I actually think this might be true.

    “It’s possible and even probable that at least some IVC people spoke a language which has no descendants today.” This is likely true. There are mentions of many ancient languages in the old Arya historical records.

    When you use the phrase Dravidian are you speaking of Surya Vamsha (versus Chandra Vamsha which I believe might have come from the west or north)?

    Many Indologists from the 1800s believed that Iranians use to worship Daityas and Danavas, which suggests that they were considered Arya by people in the east in the ancient past.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.