May 29, 2012

Climate change dried up the Sarasvati sending Harappans eastward to the Ganges

Before and after
That's what new research seems to claim: before c. 1900 BCE there was a stable moonson on the Thar Desert feeding what is now known as Ghaggar-Hakra River (seasonal) but mythologically referred to as Sarasvati River (a major hub of the Indus Valley Civilization or Harappa culture).

This, according to the researchers of Woods Hole Oceanographic Institution (WHOI) caused that people already living in the banks of that river had to emigrate eastwards towards the Ganges, where monsoons were still stable but where the fertility was not good enough to support large cities as in the West.

Source and more details at the press release and at WHOI.

See also: Dispatches from Turtle Island (Andrew has a nice review of some of the wider implications).

20 comments:

  1. I'd love to see the evidence arguing for the particular subpoint about emigration to the Ganges rather than elsewhere (or simply collapsing and not migrating). The details in the press release focus more on the climatological cause.

    ReplyDelete
    Replies
    1. If you click on the map I borrowed from WHOI, you can see that red dots (IVC sites) "move" towards the Ganges area. However they seem to mean Punjab/Haryana, not much further Eastwards (unsure because the map is cut and there are dots at the very edge of that cut).

      Delete
  2. Either a large amount of new archaeological evidence has come to light just in recent years or that article assumes a bit more than the physical evidence may support. As I understand it, there has been little substantive investigation of settlements along the Ghaggar-Hakra and what we know of Afghan archeology wouldn't quite fill a small notebook. If they're implying that the Indus Valley and Helmand civilizations are the same, I should very much like to see the supporting evidence.

    ReplyDelete
    Replies
    1. "Afghan archaeology", "Helmand"? They are talking about a seasonal river in India, near the Pakistani border. Not sure what you mean.

      In the map above we can see that, before 3900 BP, there is a thick area of settlements (red dots) along what is labeled as "Sarasvati?", which is no other than the Ghaggar river (known as Hakra further downstream).

      I can't say anything else but they are clearly not talking about Afghanistan at all (unless I'm missing something).

      Delete
  3. From the article at the Woods Hole site: "Once extending more than 1 million square kilometers across the plains of the Indus River from the Arabian Sea to the Ganges, over what is now Pakistan, northwest India and eastern Afghanistan, the Indus civilization was the largest—but least known—of the first great urban cultures that also included Egypt and Mesopotamia."

    Eastern Afghanistan would be Helmand and environs, right? I had been led to believe that the Helmand culture was distinct. I could be wrong and as I say we know precious little of Afghan archeology, really.

    If you dig into the literature, Maju, you soon discover that, while sites have been identified, few on the Ghaggar-Hakra have been extensively investigated or, if they have, the data still awaits publication. At least this is what I found.

    ReplyDelete
  4. "Eastern Afghanistan would be Helmand and environs, right?"

    Not really. Helmand is rather West, South West Afghanistan. I see some vague, badly drawn, maps that include parts of Afghanistan in IVC (including maybe parts of Helmand) but the most serious ones do not.

    I'm not sure where IVC enters the modern borders of Afghanistan (if it did) but does not get much into it in any case.

    So what you are saying is that there is a related culture in Southern Afghanistan (Helmand culture) that may be related to IVC but is not exactly the same, right? I had never heard of it before but I'll take your word on it.

    "If you dig into the literature, Maju, you soon discover that, while sites have been identified, few on the Ghaggar-Hakra have been extensively investigated or, if they have, the data still awaits publication. At least this is what I found".

    Vale, good to know. However it's clear that this entry's map is not the only one that shows IVC settlements along a line East of the Indus, on what must be the Haggar seasonal river.

    The authors do not deal with the arcahaeology of the sites but with paleo-climatology. What they have found is a climatic divide that may explain both the abandonment of these sites and the settlement of other in Haryana instead.

    ReplyDelete
  5. Some points that I feel need to be addressed;

    — The article seems to imply that the idea of a once flowing Saraswati river in North-Western India (Punjab-Haryana) is not geologically valid. The original and obvious inference regarding the Saraswati river was that if there are numerous cognate river-names across South-Central Asia (Haraxvaiti and a Haroiiu river found in Afghanistan; paralleled by the now dry Saraswati river in Haryana [apparently non-existent by this very article]; and a Gomati and a Sarayu river found in Uttar Pradesh) alluded to the fact that, the migrating pastoral/semi-nomadic Indo-Iranian speakers travelling in South-Central Asia named rivers in areas they were getting acquainted with in a manner reminiscent to rivers in regions they had already explored further west. This study has gone onto assert based on their findings that the purported Saraswati river that the Harappans relied on was a monsoon-fed river.

    — Here is the piece of the puzzle that's seems to be missing from the picture, in my humble opinion. From a linguistic perspective, retroflex phonemes are present in perhaps all the language families of the Indian subcontinent. No other Indo-European language exhibits this influence except Indo-Aryan languages sans the Middle-Iranian Khotanese language. As I mentioned in the Dravidian languages thread, it seems only logical to assign that specific time-frame in history wherein Sanskrit was influenced by this seemingly characteristic South-Asian retroflex phoneme prior to the conception of the Vedas as Vedic Sanskrit exhibits a retroflex reminiscent (but not exact) of the same. That is apart from some few but very obvious Dravidian loan words. All this logically pushes the geographic area of the conception of the Vedas much more southerly than Afghanistan, parts of which were extensions of the Bactria–Margiana Archaeological Complex and probably did not speak a language related to Dravidian but possible some sort of Middle-Eastern language, going by the few seemingly Middle-Eastern roots of the few words inferred by linguists of the BMAC's language. Perhaps, all the way down to the area that houses the so called Sapta Sindhu, which play a prominent part in the hymns of the Rig-Veda. This is really the only area where the original Indo-Aryan speakers could have come in proper contact with Dravidian speakers at such an early time-period and thereon undergone some phonetic change and influence and also acquired few Dravidian loan words.

    — Hence, it seems as if the perpetual mention of the legendary river Saraswati, which flowed from the Himalaya and emptied finally into the Gulf of Kachchh, does not seem to be located in Haryana but instead Afghanistan, as the former seems to have been replenished by monsoons and the latter via a Himalayan glacier source. So the writers of the Vedas were simply reminiscing about the river? What's odd here is that the Ghaggar-Hakra river is an exclusively monsoon flowing river, but the Indus tributaries that are in geographic proximity with it, i.e Jhelum, the Chenab, the Ravi, the Sutlej, and the Beas all ultimately have their source in the Tibetan Plateau which lies between the Himalayan range and Kunlun range. It look as if there seems to be a huge discrepancy between the geological nature between the two rivers as far as source is concerned.

    ReplyDelete
    Replies
    1. Not sure if I follow you in every detail but I think I can agree grosso modo for the two first two paragraphs: the Vedic scriptures surely speak of Punjab and nearby areas and not Turkmenistan and it was there where the Indoeuropean immigrant minority met the Dravidian-speaking IVC people creating as result the Indo-Aryan ethno-culture and language subfamily.

      That's what I infer from your first two paragraphs. But then, in the third one, you go back and miraculously you claim that the Sarasvati would not be one of the Sapta Sindhu but a remote river of Central Asia which however could not flow into the Gulf of Kachchh at all. What's the deal?

      I could understand if you said that the Sarasvati would be another river such as the Indus itself or some other one. But in the BMAC area?

      I see that Avestan (Zoroastrian) tradition also mentions that river but I can only imagine that this is because, prior to their invasion of Iran, the Iranian branch was influenced by the Indo-Aryan "superior" culture, including religion. However, retaining their "classical" IE pastoralist culture, they found highly undesirable to accept in the mid-run South Asian customs like vegetarianism or yoga-style ascetism, producing as reaction Zoroastrianism (which is in so many aspects a negative of Hinduism, its rejection made a new religion).

      For that to happen BMAC would-be-Avestan peoples must have been influenced by Hinduism, what we should not be surprised about at all: pastoralist barbarians near a center of civilization, how could it be otherwise?!

      Delete
    2. Maju, let me clarify my post.

      — The study in question has proved that the Ghaggar-Hakra river and the purported Saraswati river which was once also flowing through the Thar desert were primarily monsoon-filled rivers. However, the Vedas describe the Saraswati river as arising in the Himalayas.

      — It was a typo on my part when I mentioned that the Vedas describe the Saraswati river as draining into the Gulf of Kachchh. The wide river bed (paleo-channel) of the Ghaggar river is said to suggest that the river once flowed full of water during the great meltdown of the Himalayan Ice Age glaciers, some 10,000 years ago, and that it then continued through the entire region, in the presently dry channel of the Hakra River, possibly emptying into the Rann of Kutch.

      — Thus, my main doubt was whether this study effectively proves that the Vedic-Aryans were in fact referring to the Helmand river of Afghanistan, which does derive primarily from Himalayan glaciers, consistent with what the Vedas describe. Let me quote Wikipedia on this to further elaborate on my point - "Helmand historically besides Avestan Haetumant bore the name Haraxvaiti, which is the Avestan form cognate to Sanskrit Sarasvati. The Avesta extols the Helmand in similar terms to those used in the Rigveda with respect to the Sarasvati: the bountiful, glorious Haetumant swelling its white waves rolling down its copious flood." The Saraswati river is mentioned a whopping 42 times in the Vedas.

      — Are you asserting that Zoroastrianism is an off-shoot of early Vedic Hinduism? I was always under the impression that the Iranic and Indo-Aryan split took place somewhere around the BMAC and these parallels and opposite parallels (such as Devas being positive deities in Vedic Hinduism and Asuras being bad; while it is the opposite for Zoroastrianism) simply point to the same ancient cultural root.

      — Just to point out, vegetarianism is a recent phenomenon among Brahmins. It was fueled by the advent of Buddhism which had a disdain towards unnecessary animal cruelty and in particular, the much prevalent animal sacrifices of the time. Some did indeed give it up voluntarily prior to the Buddhist times as they felt that meat eating wasn't conducive to spiritual actualization but it wasn't the norm, and many sages who were deemed spiritual masters were described as eating meat in scenarios such as, where a given king was entertaining them with a feast.

      Delete
    3. Alright. I see your first three points. All I can say is that this study seems to support that there was once a dense IVC population at the banks of the Ghaggar river and that, coincident with a climatic shift, this population decreased and migrated eastwards to the Ganges basin, where it did not build cities.

      Whether the Ghaggar is the Sarasvati or not... I can't say for sure but I read in Wikipedia that there are some powerful reasons to suspect it being the Ghaggar-Hakka:

      · It's mentioned in geographic order between the Yamuna and the Sutlej
      · It's said to die in "samudra" (translated as "ocean")

      However it's true it attributes it mountain origins. Can't say.



      ...

      As for Zoroastrianism being related in negative with Hinduism, this seems quite clear for what I have read: "deva" means god in Hindu tradition (what is consistent with general Indoeuropean, i.e. deus, theos) but demon in Zoroastrian one. The main god of Zoroastrianism is known as Ahura Mazda, which etymologically derives from "ahura" < "asura" = demon (for the Hindus).

      [I've also read some speculations about Germanic gods, the aesir, being more related to the concept of asuras than that of devas but it's less straightforward. If true it would suggest that the duality has deep IE roots, rather than being specifically Indo-Iranian].

      As for vegetarianism, I got it wrong (my memory), what Zoroaster despised and forbade was fasting, a virtue for many Hindu schools (notably yoga-related) but a weakness in Zoroastrianism.

      "I was always under the impression that the Iranic and Indo-Aryan split took place somewhere around the BMAC and these parallels and opposite parallels (such as Devas being positive deities in Vedic Hinduism and Asuras being bad; while it is the opposite for Zoroastrianism) simply point to the same ancient cultural root".

      It's possible but to my eyes it seems to be more: there was ongoing interaction of some sort and Zoroastrianism is partly (and quite clearly) a rejection of Hindu influences. I can only imagine that Hindu influences (Vedic and even pre-Vedic) reached Southern Central Asia but were rejected by means of Zoroastrianism becoming the dominant religion.

      Delete
  6. This comment has been removed by the author.

    ReplyDelete
  7. All this logically pushes the geographic area of the conception of the Vedas much more southerly than Afghanistan

    But retroflex consonants are also found in the Indo-Iranian languages (including the Iranic branch) of Afghanistan and even some further northern areas in Central Asia (except Persian/Tajik/Dari, which is a language that was introduced to the region from Iran during the Islamic period):

    "The Nuristani languages of eastern Afghanistan also have retroflex consonants. Among Eastern Iranian languages, they are common in Pashto, Wakhi, Sanglechi-Ishkashimi, and Munji-Yidgha."

    http://en.wikipedia.org/wiki/Retroflex_consonant

    ReplyDelete
    Replies
    1. And Polish and Spanish. However the latter must be because of Basque/Iberian substrate influence (hence unrelated).

      Delete
    2. Swedish and Russian are also mentioned in the examples.

      Delete
    3. That is interesting. I though that retroflex in IE languages was restricted to Indo-Aryan derived languages and Khotanese. Are there any studies on what degree this retroflex is present in the Polish/Spanish/Russian/Swedish relative to Indo-Aryan languages? Is there any true measure of such linguistic and pronunciation aspects?

      Delete
  8. Onur, but other Iranic languages lack it for the most part. Nuristhani's retroflex could be explained by it's intimate relatedness to Kalash.

    ReplyDelete
  9. other Iranic languages lack it for the most part

    But not the ones in Afghanistan and some further northern regions except the late comer Persian/Tajik/Dari. So a retroflex and possibly Dravidian substrate in Afghanistan is equally likely as the one in Pakistan and North India. Brahui may be a remnant Dravidian language in its local area.

    ReplyDelete
  10. This comment has been removed by the author.

    ReplyDelete
  11. I though that retroflex in IE languages was restricted to Indo-Aryan derived languages and Khotanese.

    All the East Iranic languages of Afghanistan, Pakistan and some further northern areas in Central Asia have retroflex consonants. Also, as Maju explained, retroflex consonants are not restricted to the Indo-Iranian branch of the Indo-European language family among the Indo-European languages. In the case of the Indo-Iranian languages, retroflex consonants may be from a pre-Indo-European substratum in South Asia, Afghanistan and some further northern areas, and the strongest candidate so far is the Dravidian languages. The Dravidian languages might have been spoken in a much larger territory before the Indo-Iranian migrations and Brahui may be a remnant language.

    ReplyDelete
  12. a pre-Indo-European substratum in South Asia, Afghanistan and some further northern areas, and the strongest candidate so far is the Dravidian languages

    and also in parts of Iran

    ReplyDelete

Please, be reasonably respectful when making comments. I do not tolerate in particular sexism, racism nor homophobia. Personal attacks, manipulation and trolling are also very much unwelcome here.The author reserves the right to delete any abusive comment.

Preliminary comment moderation is... ON (your comment may take some time, maybe days or weeks to appear).