The Lost Vedic River Sarasvati

[Mahratta]

http://www.nature.com/news/a-turkish...guages-1.11270

[barrykul]

With computer modelling I can arrive at any conclusion I want. Depends on the assumption, method of modelling etc. Let us say they used Bayesian theory to arrive at the conclusion. What prevents someone to do another model with a reverse Markov chain based on probability transitions and undo the migration spread.

As an example I can prove that Slums migrated from either the South of India or from the North, depending on the assumptions and model I use.



Here is one comment about Turkish language

I think there is some kind of confusion here. It can be very possible that Indo-European languages are originated from Anatolia but that does not mean it has anything to do with Turkish. The first Turkic tribes arrived Anatolia after 1000 A.D. Before that the Anatolia had thousands years of history. So the Turkish language arrived really really late to Anatolia hence it is impossible to have any connection. It belongs to Ural-Altai language branch which is also far away from Indo-European languages.

If you look at the ancient languages of Anatolia like Persian(farsi), Greek, Armenian you can see a lot of connections to European languages or Hindu.

Again, all these researchers miss the key point. Rg Veda in Sankskrit mentions Sarasvati River umpteen times. The river is smack dab in India / Pakistan. This also happens to be the very same home of Indus Valley Civilization. All the evidence is pointing to India/Pak, but these researchers who hung their hat on "Indo - European" classification are wildly flailing in the wind. Tis horses and Anatolia / Greece. If they can usurp Sanskrit to Greece then they can claim glory of their superior languages. Romance and Germanic languages are far removed from Sankskrit.

[Marathaman]

The current Turkish population is indeed very recent. Those before the Turks were probably the I-E speakers. They were not European either.

[Licit Mortal]

I am surprised that no one followed this news.








I am trying to follow this story up, but I could not find even a single news paper report on this story.

[satchitananda]

Originally Posted by Licit Mortal

I am surprised that no one followed this news.


I am trying to follow this story up, but I could not find even a single news paper report on this story.

Not surprising my friend. If they accept Saraswati's existence, it automatically endorses the wisdom of that civilization. It also indirectly endorses the antiquity of our nation. If such a hard admission happens then the current government and its coterie will face renewed resistance from the right. Do you think Cong is that dumb ??

Also all the media are played up puppets to highlight what the master (Power) wants.

Thanks for the newsbyte. Though the automatic upwelling of water is nice news fodder, all the info is stale. Neither is the insight nor the perspective refreshing.

Lots of Paleo channels have been mapped. It just highlights the exciting scientific discoveries that are waiting to be made in this region.

[skganji]

Originally Posted by satchitananda

Do you think Cong is that dumb ??

Lots of Paleo channels have been mapped. It just highlights the exciting scientific discoveries that are waiting to be made in this region.

Well said. Cong will goto any extent to stay in power ?. Nehru, Indira and Sonia are examples of how they abused the system to stay in power. Under them they will definitely deny the truth like Saraswati River, Vedic civilization etc.

Coming to this video and the paleochannels, it is mind blowing to see such huge water channel near Jaisalmer oozing sweet water in the middle of the desert. I was confused on where River Saraswati flowed after it crossed Rajasthan. It seems that it continued flowing in Rajasthan and flowed into Gujarat and into Kutch at one point of time. On the other hand many archaeological sites in Sindh region also point to Saraswati continuing as Nara river in ancient times and draining into the sea.

Medium Resolution Imaging Spectrometer (MERIS) image. Please read the annotations on the image. The red arrows trace the path of the river. Especially notice the delta fan. There is a section, denoted by an Oval, where the image does not show a river. But it just seems to be a limitation of the image. River must have flown through the region encompassed by the Oval.

[Mahratta]

Originally Posted by barrykul

With computer modelling I can arrive at any conclusion I want. Depends on the assumption, method of modelling etc. Let us say they used Bayesian theory to arrive at the conclusion. What prevents someone to do another model with a reverse Markov chain based on probability transitions and undo the migration spread.

Computer modelling is just like any modelling. Yes, you can mess around with parameters and definitions to get a good result, but one can do the same (and more covertly, at that) in any other model. Just using computer modelling means nothing, linguists have been doing it for decades (relevant).

Among linguists (outside of the closed and hierarchical cult of historical linguistics) q.c.l. is well known, in any case, so it doesn't seem that the qualms you have with the method -- whatever exactly they are, as you didn't specify -- don't cause too much of a fuss in linguistics departments.

In any case, it's a far more sensible wager than going by a single source of Iron age poetry with little attached temporal-historical information and then relying on a cultish institution to decide which path among such texts and readings is canonical on murky grounds. Of course these little empirical tidbits are of use, but there's not much to be gained from trying to "see", guided by a theological reading of some old poems.

The obvious point that models require motivation is moot: the real argument is between this new application of q.c.l. and the antiquated linguistic taxonomy practised by historical linguists.

Again, all these researchers miss the key point. Rg Veda in Sankskrit mentions Sarasvati River umpteen times.

Uh, so what? As for the comment on Turkish, sure...nobody said that Turkish has anything to do with the roots of IE -- the modern Turks only got there millennia afterwards. People move around over time, you know. Anyway, the section you quote makes this point itself.

on "Indo - European" classification are wildly flailing in the wind. Tis horses and Anatolia / Greece. If they can usurp Sanskrit to Greece then they can claim glory of their superior languages. Romance and Germanic languages are far removed from Sankskrit.

Yes, the particular point in time is arbitrary (that's the reason cladistic methods are interesting). They are imposing the existing historical linguistic structure on the results just as the Linnaean structure was imposed on cladistic results -- so other people in the field can understand them. Given time, these methods may bring us to question the imposed structure. However, for the time being, the textual data suggests a Turkish origin for IE, whatever IE is or is not.

I'm sure that another accumulation point could have been selected, but IE is well known in the present historical linguistics community -- it's the same reason cladistic biologists use old taxonomic terms like "tree" instead of cladistic terms.

[barrykul]

Originally Posted by Mahratta

Computer modelling is just like any modelling. Yes, you can mess around with parameters and definitions to get a good result, but one can do the same (and more covertly, at that) in any other model. Just using computer modelling means nothing, linguists have been doing it for decades (Irrelevant).

Linguists are the worst at understanding computer modelling techniques like Bayesian and Markov, I doubt whether they understand the math/computer based theories. The entire Indo - European language tree ( is it a tree ?) is drivel with false assumptions like PIE as the root node. There is no evidence to support this poppycock first cooked up by British/German authors.

If you read your reference above:

The probability of relatedness of languages can be quantified and sometimes the proto-languages can be approximately dated. The topic came the attention of the popular press in 2003 after the publication of a short study on Indo-European in Nature (Gray and Atkinson 2003).

and surprise what do they use for obtaining this stunning result....

Bayesian inference is used to search for the optimal tree. A Markov Chain Monte Carlo algorithm [32] generates a sample of trees as an approximation to the posterior probability distribution.

The I contend with "reverse Markov chain based on probability transitions" this is negated. Kapische.

You make an assumption then you use modelling to come up with cute results. But what about the assumptions. Who believes in these linguistic assumption except Brits/Germans since their language would become sub-standard compared to Sanskrit. No one questions the assumption as if it is some gospel truth.

A better precise science like math shows the way.
Bertrand Russell wrote an entire book Principia Mathematica to prove 1 is not 2. He questions all assumptions in the number theory, Peano's axioms, in the end he convinces everyone that 1 is inherently 1 and 2 is inherently 2 and they can never be the same. Where is rigor by linguists on their assumptions. They dance around like blind men nodding to the temple of PIE.

The rest of your post is some inane rambling which requires no response, thank you.

BTW this is not the Turkish thread it is the "The Lost Vedic River Sarasvati"

[Mahratta]

Originally Posted by barrykul

Linguists are the worst at understanding computer modelling techniques like Bayesian and Markov, I doubt whether they understand the math/computer based theories.

Having met a few linguists and talked to them about mathematical (proof theoretic, model theoretic) methods, the impression I've got is -- surprise surprise -- that we can't paint them all with one brush. In any case, if at least some of them can understand fairly sophisticated model theory, baby stuff like Monte Carlo shouldn't pose much of a problem.

Again, it's better than basing theoretical formulation on "divine readings" of texts of dubious character and origin. You've got nothing besides a few passages of poetry from RgVeda and a lot of hate for Witzel. There is, in fact, a world that exists apart from India, whether you accept it or not.

Anyway, the editorial board for Nature is the most rigorous (along with Science) among scientific journals, so I'd expect their judgment to be better than both of ours.

The entire Indo - European language tree ( is it a tree ?) is drivel with false assumptions like PIE as the root node.

(1) Stating that something is the case, however forcefully, does not make it the case.

(2) That was the point of the majority of my last post. The "root node" is arbitrary, it was merely selected to cohere with existing historical linguistics theory.

The I contend with "reverse Markov chain based on probability transitions" this is negated. Kapische.

What is negated? If you want to use terms from Stochastics 101, you ought to try to use them coherently.

You make an assumption then you use modelling to come up with cute results. But what about the assumptions.

That's the point I made in the last post which you didn't seem to pick up.

A better precise science like math shows the way.

Ahahahahahahahahahahahahaha, as someone actively researching in mathematical logic I find this very funny indeed.

Are you seriously comparing mathematics to linguistics? That betrays a profound ignorance of mathematics. Mathematics is no more a science than poetry is.

Anyway, while the status of linguistics as a science is dubious, it is far better than the crazy religiously inspired text reading you seem to favour. Do you believe that the existence of Agni is scientific fact as well, as he features in RgVeda?

Bertrand Russell wrote an entire book Principia Mathematica to prove 1 is not 2. He questions all assumptions in the number theory, Peano's axioms, in the end he convinces everyone that 1 is inherently 1 and 2 is inherently 2 and they can never be the same.

Indeed, and then Godel showed PM to be either inconsistent or incomplete. There's no basis to its supposed rigour.

Perhaps the example you would have liked to use would be the various independence/forcing programs in set theory, which compare the relative strength of assumptions (with assumed rules of inference, as in all other mathematics) -- indeed, model theory in general provides a nice template for this sort of work. But I guess you didn't find that in your intro logic textbook.

[barrykul]

Originally Posted by Mahratta

Having met a few linguists and talked to them about mathematical (proof theoretic, model theoretic) methods, the impression I've got is -- surprise surprise -- that we can't paint them all with one brush. In any case, if at least some of them can understand fairly sophisticated model theory, baby stuff like Monte Carlo shouldn't pose much of a problem.

You meeting people and having an opinion does not make up for the shoddy research that linguists produce.

Again, it's better than basing theoretical formulation on "divine readings" of texts of dubious character and origin. You've got nothing besides a few passages of poetry from RgVeda and a lot of hate for Witzel. There is, in fact, a world that exists apart from India, whether you accept it or not.

Ya, like the word of Nazi Witzel is more "divine". They did create a world unto themselves based on Moron Max Mueller's interpretation of "Arya" into a race called "Aryan".

(1) Stating that something is the case, however forcefully, does not make it the case.

(2) That was the point of the majority of my last post. The "root node" is arbitrary, it was merely selected to cohere with existing historical linguistics theory.

I am glad you agree in (2). When the root is suspect so is the rest of linguistic theory. BTW (2) contradicts your own stmt (1). So much for your mathematical logic.

What is negated? If you want to use terms from Stochastics 101, you ought to try to use them coherently.

Dude, you have no idea of computer based modelling.

Are you seriously comparing mathematics to linguistics? That betrays a profound ignorance of mathematics. Mathematics is no more a science than poetry is.

Giggle, poetry is science? And the same as mathematics. Vow. I finally understand where you are coming from.

Anyway, while the status of linguistics as a science is dubious, it is far better than the crazy religiously inspired text reading you seem to favour. Do you believe that the existence of Agni is scientific fact as well, as he features in RgVeda?

If you believe "crazy religiously inspired text" what do you think Witzel et al are doing? A clue: the same "crazy religiously inspired text".

Indeed, and then Godel showed PM to be either inconsistent or incomplete. There's no basis to its supposed rigour.

Again complete ignorance on your part. Godel just proved that there are stmts which cannot be proved true / false in a system with axioms. This is Comp Science 101, wherein a Turing machine cannot be built to solve all problems. He did not disprove PM that 1 is not 2.

BTW Stop polluting this thread with your BS linguist arguments and introducing extraneous Turkish stuff. This thread is about
"The Lost Vedic River Sarasvati".

[Mahratta]

Originally Posted by barrykul

You meeting people and having an opinion does not make up for the shoddy research that linguists produce.

You missed the point:

"Anyway, the editorial board for Nature is the most rigorous (along with Science) among scientific journals, so I'd expect their judgment to be better than both of ours."

I am glad you agree in (2). When the root is suspect so is the rest of linguistic theory. BTW (2) contradicts your own stmt (1). So much for your mathematical logic.

Uh...(1) was addressed to your "argumentative technique" of passing something off as drivel without any proof. Has nothing to do with (2), which is why I marked them off as separate in the first place.

As for your cheap shot regarding my work, if you would like I could send you links to some publications of mine in logic via PM.

Giggle, poetry is science? And the same as mathematics. Vow. I finally understand where you are coming from.

Again, you missed the point, which is that they are both not sciences. There is nothing scientific about mathematics. Given your largely superficial name-dropping of popular mathematical methods/theorems, it's not surprising that you don't get it.

If you believe "crazy religiously inspired text" what do you think Witzel et al are doing? A clue: the same "crazy religiously inspired text".

I never said that Witzel was right. Again, if you read: "You've got nothing besides a few passages of poetry from RgVeda and a lot of hate for Witzel." Saying that you have nothing but a lot of hate for Witzel is not the same thing as saying that I support Witzel. Not even close.

Again complete ignorance on your part. Godel just proved that there are stmts which cannot be proved true / false in a system with axioms.

That's funny, since what you said is entirely the same as what I said, but you have simply stated it in a convoluted and vague manner.

The relevant direction is that if the Godel sentence G is true in our model of arithmetic M (i.e. a model of PM), then PM is inconsistent as ~G is a consequence of G by elementary logical operations. If G is false in M, then while M is consistent, it is incomplete, as G asserts its own falsity and is as such true (outside of M).

This is Comp Science 101, wherein a Turing machine cannot be built to solve all problems. He did not disprove PM that 1 is not 2.

Ahahaha, I never said he disproved that 1 is not 2. (subject to the system of PM, 1 is not 2 -- the problem is that PM itself is either incomplete or inconsistent, if you know what those words mean)

Rather, I said that there is no "rigour" in PM, which is an easy corollary of Godel's theorem. While the halting problem (what you vaguely try to mention with your comp sci bit) is a corollary to Godel's incompleteness theorem, it is not the same thing. You seem to be completely out of your depth here, so I understand your defensiveness. I'll leave you to preach to those who find your "Logicomix"-based understanding of mathematics convincing.

Originally Posted by barrykul

BTW Stop polluting this thread with your BS linguist arguments and introducing extraneous Turkish stuff. This thread is about
"The Lost Vedic River Sarasvati".

Those "BS" arguments you mention have been published in the most well known scientific journal. Your ideas, meanwhile, are limited to a crackpot section of an Indian-nationalist subforum of SSC.

However, I do agree that this is not relevant to the topic. I'll leave you guys to conspiracy theorize.

[barrykul]

Originally Posted by Mahratta

You missed the point:

Clearly not. Don't pretend that you have some great logic in defending what is essentially wrong premises/axioms from a cult of linguists. It is not about these individuals or me, it is about the rank inaccuracy of PIE, Aryan, Language Tree, modelling to suit ones own opinion, etc. All the coffee turned theorems of yours cannot defend such boringly inaccurate conclusions. BTW I would be least bothered with knowing about your great prowess/or lack thereof in the field of mathematical logic.

For a logician, you display a remarkable lack of understanding of your own writing. Here you are trying to show mathematics as incomplete, without rigour using Godel but you fail to apply the same standards to the editorial board of Nature in publishing a specious article that locates the mythical Aryans in Anatolia/Greece. Nature is without rigour and incomplete says Godel, right? Or do you have a cofee inspired theorem to rebut this conclusion as well?

[Marathaman]

...an Indian-nationalist subforum of SSC

Is that bad? Oh dear...

[dayalbaba]

mahratta, appeals to authority is no grounds for justification of the mumbo-jumbo that passes for paleo linguistics. all kinds of ad-hoc assumptions are created to match the starting dogma of current linguistics research. of course, if you start with an idee fixe, you will always be able to to think up laws to suit them, especially in a field as open as this one. we used to call this back calculation in college days.

having been part of academia for sometime, I am quite aware of the machinations and mutual back scratching that go on during publications.
while I agree that nature is usually better than the rest but even if the mathematics in that article is rigorous, it says NOTHING about the validity of that study or its conclusion. please, who says that the mathematical model of disease spread can be equally applied to spread of languages ?

in any case, even highly rigorous mathematics give rise to incorrect theories all the time, case in point, string theory (none of its predictions have been confirmed, many have been confirmed to be not true).

----------------------

Originally Posted by Marathaman

Is that bad? Oh dear...

of course it is. remember this ?

Originally Posted by Mahratta

Born in India, live abroad. I don't consider myself Indian in the slightest - in fact, I usually take offense when the term's applied to me. In fact, the steady increase in my distaste for the term is pretty well-documented by the progression of the nature of my posts on SSCI over the past few years...

so you can see where he is coming from and his blind support for anything that might be used to diminish India.

[barrykul]

The lost vedic river sarasvati is home to the glorious Indus Valley Civilization. The amount of knowledge accumulated in this era is staggering. Nothing comes close. It is a known fact that India is the fountain head of the current number system - zero, infinity, decimals, 10 based counting. Mathematicians were well versed in geometry, trignometry, PI (not PIE!), quadratic equations, infinite series. Panini's sanskrit grammar rules are the foundation of modern day computer syntax rules.

What is little known in Modern day history is that the usurping of such concepts by the Greeks for Geometry and Trignometry. C. V. Raju drives a Mac truck through Euclid (he is a ficitional character wrought by the translation from Arabic of Indian Knowledge). Europe first learnt about the "Elements" through translations from the Arabic books in the Toledo library during the Crusades against Islam. Possibly, the name “Euclid” was inspired by a translation error made at Toledo regarding the term uclides which has been rendered by some Arabic authors as ucli (key) + des (direction, space). So, uclides, meaning “the key to geometry”, was possibly misinterpreted as a Greek name Euclides.

http://ckraju.net/papers/MathEducation1Euclid.pdf

Meanwhile from the same Arabic translation : Arab sources tell us that Archimedes was a short black man! Not Greek. What a let down for the Nazi racists like M. Witzel claim that Aryans from PIE riding on horses from the Pontic steppes achieved all of modern day thought.

[Mahratta]

Originally Posted by dayalbaba

mahratta, appeals to authority is no grounds for justification of the mumbo-jumbo that passes for paleo linguistics.

I agree; notice that I did not use an appeal to authority as a criterion for judging the validity of what the linguists said.

Look at the context: barry attempted to dismiss the conclusions by a general ad hominem against the authors. I merely used the appeal to authority to show the absurdity of this argument, not to prove anything regarding the truth/falsity of the conclusions themselves.

Keep in mind that the point I was arguing against was entirely one of personal experience ("linguists don't understand math modelling"). As such, it's perplexing why you'd call out the response to an argument that was not well-founded, rather than calling out the initial argument itself, as I don't think that argument by personal experience is any better than argument by authority, and my argument by authority was a response to an argument by personal experience.

having been part of academia for sometime, I am quite aware of the machinations and mutual back scratching that go on during publications.

Again, I agree.

However, what are we comparing this sort of modelling to? Either the more corrupt area of old historical linguistics, represented by folks like Witzel, or the sort of theorizing found on this thread. Can anything better be said for either of these sources? The first is in academia, so your own experience applies to it just as it does to those behind the Nature study.
The second is an online forum of likeminded individuals, and as the hostile personal response to my posts makes clear, opinion which differs strongly is not very well tolerated. What's that if not "mutual back scratching"?

please, who says that the mathematical model of disease spread can be equally applied to spread of languages ?

Nobody says that. Indeed, if you look at my posts from earlier on, I merely indicated that it's an interesting new avenue to take in a field that is completely bogged down in terms of methods.

in any case, even highly rigorous mathematics give rise to incorrect theories all the time, case in point, string theory (none of its predictions have been confirmed, many have been confirmed to be not true).

Again, you're arguing against a straw man here. I never claimed their conclusions were true, but rather that they merited discussion. After the dismissal of the conclusions by a few forumers (without reading the actual paper, as I didn't post that) the argument changed, but I merely was arguing against the reasons for their dismissals, not for the truth of the conclusions themselves.

As for string theory, sure, but the standards are different in physics than in historical linguistics. An analogously strong (weak) theory would be well accepted in linguistics, I imagine. Even their most well-accepted notions are basically untestable as it stands.

so you can see where he is coming from and his blind support for anything that might be used to diminish India.

The post you quoted was obvious trolling.

Originally Posted by Marathaman

Is that bad? Oh dear...

Indeed not. Surely there is a difference between the interpretations of Indian history one gets from a "mainstream" Western text and an Indian-nationalist forum, just as there is a difference between the interpretations of Tibet of, say, a Chinese-nationalist forum and some "mainstream" Western text.

Sheesh, even you're jumping on the touchy nationalist bandwagon these days? Don't do it, it's a saturated market.

[Marathaman]

Is SSC-I an "Indian nationalist forum"? I thought it was a forum about skyscrapers (among other things) in India.

Also, aren't you contradicting yourself by using "mainstream" and "western" in the same sentence? Shouldn't you be using terms like "westernist" or "Europeanist" or "Anglo-Saxon-ist" etc?

Mind you, I don't agree with anything that barrykul here says, but don't put convenient labels on his views, especially when the "Indian nationalist" perspective is on history is generally considered to be that of Romila Thapar and her ilk.

Also, history is not an exact science. It is always seen from a certain perspective. You can't write a "neutral" history because it doesn't exist.

[Mahratta]

Originally Posted by Marathaman

Also, aren't you contradicting yourself by using "mainstream" and "western" in the same sentence? Shouldn't you be using terms like "westernist" or "Europeanist" or "Anglo-Saxon-ist" etc?

No no, by "mainstream western" I was just specifying a reference point, i.e. SSC-I is classifiable as Indian-nationalist with respect to the mainstream western perspective.

Yes, "Anglo-Saxon" is probably more precise than "Western", but I think that the definition of "Indian nationalist" is nearly the same as the mainstream Anglo-Saxon definition throughout the rest of the Western world.

There is no absolute meaning floating over every classification, hence the reference point.

Mind you, I don't agree with anything that barrykul here says, but don't put convenient labels on his views, especially when the "Indian nationalist" perspective is on history is generally considered to be that of Romila Thapar and her ilk.

"Indian nationalist" has a better ring to it than "educated urban Indian middle class / NRI". Maybe something like "Indian e-nationalism" is a good compromise.

Also, history is not an exact science. It is always seen from a certain perspective. You can't write a "neutral" history because it doesn't exist.

Certainly. But you can look at a view of history from the fixed perspective of some other view; i.e. the common notion of a "nationalist" rendition of a history is from a mainstream Western perspective.

[Marathaman]

"educated urban Indian middle class / NRI"

Aren't you part of that same category? Now we have a problem

Perhaps your view should be classified under NRI-Canadian-Mathametician and his view should be classified under NRI-Computer-Programmer.

See the problem yet?

[dayalbaba]

However, what are we comparing this sort of modelling to? Either the more corrupt area of old historical linguistics, represented by folks like Witzel, or the sort of theorizing found on this thread. Can anything better be said for either of these sources? The first is in academia, so your own experience applies to it just as it does to those behind the Nature study.
The second is an online forum of likeminded individuals, and as the hostile personal response to my posts makes clear, opinion which differs strongly is not very well tolerated. What's that if not "mutual back scratching"?

for the 2nd type perhaps groupthink is a better term.

having said that, at least everyone, barring the utterly uncouth has a voice in an online forum. hostile response in a rather personal way is not unique to internet fora, academia, especially in the highly politically motivated world of humanities has its fair share.
people like witzel himself behave like drunken louts towards people who disagree with them. towards talageri for example. admittedly, talageri's work might be sloppy in parts but it is no worse than witzel's work, who can be downright idiotic in his reasonings ! not to forget that witzel did offer to take in talageri as a student at one time.

my view is this, modern mathematical models applied to linguistics would serve no purpose unless the field of linguistics itself reforms its foundations.
sans that, any fancy model would only result in GIGO.
and until that reform happens journals like nature should refrain from carrying articles on such topics, it amounts to encouraging bad science.