Out-of-India - From Theory to Truth: Part 2

[A_Gupta]

This is what I meant when I said that we need to use that model. Stand on whatever post/pedestal/tower you get and jerk off and let your seed spread far and wide. This is the way of science now.

Or else the baton of science and knowledge-production is ripe for being grabbed back from the institutions of the West.

[kit]

• Reproducibility is the ability to test a result using independent methods and alternate choices in data processing. This is akin to a different laboratory testing an experimental result or a different climate model showing the same phenomena etc.
• Replicability is the ability to check and rerun the analysis and get the same answer.

This is fundamental to ALL science. For example - a person has the flu and eats chalk and gets better and says "Chalk works'.

Chalk must be tested in 100 patients with flu and should work consistently and work better than say water alone. Reproducibility and replicability.

But when people who call themselves scientists like Vagheesh publish a non peer reviewed paper and publicize it and discuss on Disqus and Twitter they know damn well that they can talk shit because the vast majority of people are not doing science. This has become a common thing in the west with even books being written for the lay public passing off shit as truth. David Anthony's "Horse Wheel and Language" is one such book an David Reich's book (name?) is another. Quick fame by a name of authority, some money earned on the side, and the ability to make shit into truth by spraying your seed all over the public domain.

This is what I meant when I said that we need to use that model. Stand on whatever post/pedestal/tower you get and jerk off and let your seed spread far and wide. This is the way of science now.

Boss the Chinese got there before you .. did I hear someone talking about plagiarism .. their umpteen scientific achievements include a very good percentage of "self-flagellation"

[gandharva]

The genetic prehistory of the Greater Caucasus

What does it mean?



https://www.biorxiv.org/content/biorxiv ... 7.full.pdf

[Anshuman.Kumar]

It only means the Paper is neither here nor there.
It subtly raises the question on association of R1a/R1b to the so called IE languages and also shows that the Yamnaya through the Maykop have probably southern ancestry as MayKop Y Haplogroup profile is very similar to the southern contemporaneous Cultures/Civilization.But it still allows the conclusion that R1a/R1b were native to Steppe/Europe so not much of an effect on the old Order.

[shiv]

This is what I meant when I said that we need to use that model. Stand on whatever post/pedestal/tower you get and jerk off and let your seed spread far and wide. This is the way of science now.

Or else the baton of science and knowledge-production is ripe for being grabbed back from the institutions of the West.

In fact this is what I mean. It is possible to publish honest stuff that goes against the crap being thrown at us. But trying to use western portals, western validation and western peer review will not work. We have to use our own portals to have our voices heard.

[shiv]

It only means the Paper is neither here nor there.
It subtly raises the question on association of R1a/R1b to the so called IE languages and also shows that the Yamnaya through the Maykop have probably southern ancestry as MayKop Y Haplogroup profile is very similar to the southern contemporaneous Cultures/Civilization.But it still allows the conclusion that R1a/R1b were native to Steppe/Europe so not much of an effect on the old Order.

If you use genetics to disprove language migration, it means that genetics can also be used to prove it. This is actually rubbish

When I say that we Indians are science obsessed - I believe that we are fighting windmills by fighting over genetics when the actual rubbish was foisted on us by linguists. Let me explain the "science obsession" that I accuse Indians of having. To the lay reader genetics papers are as arcane as any linguistics ref. By fighting genetics we are searching where the light is because we can understand genetics arcane as it is. Why do we understand genetic methods? Because there is science there.

Why do we wear blinkers about linguists lies? Because we cannot understand it and we are unwilling to put in the effort. If someone with a science background puts in modest effort into linguistics methods - you will find that those methods are simply unscientific trash, full of assumptions and rules to explain other rules all made up on the spot.

I urge people to look into historical linguists. Those lies have to be torn down. If we happen to lose a genetic argument it still does not mean that it connects to languages - but we are simply giving the impression that by "defeating" a genetic paper we are going to overturn the language migration theory.

Western universalists have hoodwinked us again and are continuing to make jackasses out of us - fighting over genetics when the lies have been told by people cooking up protolanguages and putting them in convenient places. Genetics is allowed to find anything and reach any conclusion. But ultimately they are depending on linguistics refs. Those are the refs that need to be torn down. Bad news for science obsessed Indians. Note that Talageri, kak, Kazanas, Oak etc all deal with texts, linguistics and history. Khonde and others also deal with the science aspects of old texts. Over here we are going apeshit over genetics which is a diversion from the issue. It is a bogey. It can never be the main target and no one will ever find a gene that links to a specific language because language function in humans is a basic neuromuscular function that can adapt to various languages, not specific ones. We are actually jackasses because no one asks the geneticist which genes code for language. When you ask them they will point to linguists/archaeology papers and say 'Language migration has been demonstrated by linguists who tell us that PIE was in steppe and those people moved to India and Iran, creating Indo-Iranian language first and later Avestan and Vedic Sanskrit"

What argument do genetics obsessed Indian have against this? By fighting in the genetics arena we are simply wasting time and effort without demanding an answer to a simple question "What is the proof that genes link to language?" Fact is they don't. If they don't what the hell are we arguing about?

Please understand that genes are being linked to migration. And migration has been cooked up by linguists to say that language moved this way or that. wtf are we all fighting about?

[JE Menon]

https://news.nationalgeographic.com/201 ... ience-spd/

85,000 year old human footprints found in Saudi Arabia

[A_Gupta]

^^^ If we accept the idea that language spoken in one area can subsequently be spoken in another area, then human migration is one means by which this can happen; and genetics can show us whether human migration occurred.

Of course, what languages were spoken in what area and which era is a matter of archaeology and decipherment if written material is available from that time in that area. When only material culture but not written material is available, only inference is available. Lastly, we have the AFAIK unparalleled case of the Rg Veda, the evidence of the antiquity of which is not written material, but its content, and tradition. Let us note that both are subject to the assumptions made in the interpretation.

Then there is the relatedness of languages, when are two spoken species dialects of the same language versus different languages, how are two languages related to each other, how much is a "descended from" relationship and how much is a "borrowed from/mixed with" relationship - all of which are subjective to some degree. Yes, Indian eyes must recapitulate all of this. This cannot remain "received wisdom", it needs to be examined, and all the assumptions need to be laid out.

[Suresh S]

Or else the baton of science and knowledge-production is ripe for being grabbed back from the institutions of the West.

In fact this is what I mean. It is possible to publish honest stuff that goes against the crap being thrown at us. But trying to use western portals, western validation and western peer review will not work. We have to use our own portals to have our voices heard.

That bolded part is so important I can not overemphasize it shiv. I have been thinking in my dreams if i get awarded the Noble prize I want to tell these ba**rds where to shove it. In the medical field we must have our own standards , stop attending these western conferences , trying to publish in their Journals etc for starters to be "recognized". And to think of it trying to put FRCS and American Board of Surgery on your name plate prominently to appeal to the mental coolidom of our people.

[shiv]

Rg Veda, the evidence of the antiquity of which is not written material, but its content, and tradition.

I am working on an article on ancient India-Iran relations for Infinity foundation and I am reading even more material about Iran and Zend Avesta than I did a few years ago when I started reading.

If you look at texts from the 1800s to 1900s there is one common theme. These buggers had an argument about "tradition" versus text.

That means: In order to find out about a culture, should we depend on the traditions of the people who tell us about their texts or should we depend on the texts themselves? this is not only there is older texts but Witzel also refers to the same topic.

Eventually it turns out that western scholars - right from Max Muller to Witzel to Aurdrey Truschke have decided in favour of texts as opposed to tradition. It started with Anquetil du Perron who spent time with Parsi priests and wrte what they said traditionally. Much of this was "corrected" by looking at texts written 2000 -3000 years later. Lookig at texts allows Muller to sit in Cambridge to translate the Rig veda. It allows Truschke to sit on her ass in Amreeka and talk about the Ramayana. These people do not give a flying fuk for living traditions that maintain a connection. They reach their own interpretations of the text in their university libraries.

This has to be called out. Balu was right in sayiong what he said about the Bible. Long ago Christians figured out this trick. they wrote some stuff about Christ and now say it is all true. Anything else is questionable. Tradition is questionable. While this is fundamentally wrong what is the argument you have to offer against it?

Of course as I see it - by sitting in university libraries they miss out some of the things that are nost obvious to us. But I can only expand on that after I have done my Infinity foundation article.

[A_Gupta]

Tradition is questionable. While this is fundamentally wrong what is the argument you have to offer against it?

The problem is that tradition is malleable. But one should be able to show consistency, e.g., if the archaeologically established material culture and the content of the texts and the traditions are consistent, etc., etc.

[shiv]

Tradition is questionable. While this is fundamentally wrong what is the argument you have to offer against it?

The problem is that tradition is malleable. But one should be able to show consistency, e.g., if the archaeologically established material culture and the content of the texts and the traditions are consistent, etc., etc.

That is not a problem any more than evolution is a problem. That is the nature of human behaviour and rejecting one in favour of the other is a deliberate decision to obfuscate by rejecting insights that might come from tradition. It is the desire for consistency that allows linguists to cook up languages.

For example Anquetil du Perron, after living with Parsi priests wrote his book the Zend Avesta and said that "Zend" means more ancient texts. This was rejected and there has been total confusion. They are now saying (in the west) that it should be Avesta+Zand - where Avesta is Gathas and Zand is commentary. This goes against Jatinder Chaterji and others who point out that Zand=Chhand=chants of veda, while upastha, while meaning penis in Sanskrit is actually derived from a Zoroastrian language word for book which became "ubestag" in middle Persian - cognate with "(u)pustak". Turns out that Perron was right all along.

[A_Gupta]

Perron wins on consistency - tradition, etymology match.

[Primus]

^^

I am reminded of a quote from Sir William Osler that was part of the classic surgical textbook edited by Bailey and Love IIRC (Shiv can pipe in with more on this I am sure):

"He who studies medicine without books sails an uncharted sea, but he who studies medicine without patients does not go to sea at all."

In this context all these 'experts' never sailed the sea at all, i.e. never went to India or met anyone who spoke Sanskrit or was living the life of the Vedas every day. And yet they have had the audacity to lecture the very people who were sailing the sea all the time about what it meant to be a sailor!

It boggles the mind how the world can take the word of these jokers to be the gospel truth.

[shiv]

^^

I am reminded of a quote from Sir William Osler that was part of the classic surgical textbook edited by Bailey and Love IIRC (Shiv can pipe in with more on this I am sure):

"He who studies medicine without books sails an uncharted sea, but he who studies medicine without patients does not go to sea at all."

In this context all these 'experts' never sailed the sea at all, i.e. never went to India or met anyone who spoke Sanskrit or was living the life of the Vedas every day. And yet they have had the audacity to lecture the very people who were sailing the sea all the time about what it meant to be a sailor!

It boggles the mind how the world can take the word of these jokers to be the gospel truth.

Primus - I have been reading about what was happening in Zoroastrian and pre-Zoroastrian Iran an the signature of Vedanta is clear and present. But you have to have clue as to wtf Vedanta means. No library in Cambridge can explain that to you

Just check this out - but the wording is from an Abrahamic viewpoint - I will call it interpretatio Christae and explain that later

[Nilesh Oak]

Too late here in Amarika. Came back from ~10 day long trip. Met with Vagheesh and spent 3 hours with him. Will provide brief summary in next few days.

I had made a list of predictions (about things) before meeting with Vagheesh. All of the predictions came out correct. Some of them with R-square much higher than even I had expected.

More soon

Nilesh

[Prem Kumar]

There are 2 recent papers that have pretty much killed the Steppe-homeland theory.

1) The absence of Steppe DNA in Anatolia during the time IE was spoken there

2) The Maykop culture being ancestral to the Steppe, implying South to North DNA movement

This has caused a lot of heartburn among the Steppe lovers. Now linguistics are scratching their testimonials, trying to interpret the results. Fun times!

[shiv]

In fact this is what I mean. It is possible to publish honest stuff that goes against the crap being thrown at us. But trying to use western portals, western validation and western peer review will not work. We have to use our own portals to have our voices heard.

That bolded part is so important I can not overemphasize it shiv. I have been thinking in my dreams if i get awarded the Noble prize I want to tell these ba**rds where to shove it. In the medical field we must have our own standards , stop attending these western conferences , trying to publish in their Journals etc for starters to be "recognized". And to think of it trying to put FRCS and American Board of Surgery on your name plate prominently to appeal to the mental coolidom of our people.

https://www.theguardian.com/science/201 ... or-science

Is the staggeringly profitable business of scientific publishing bad for science?
The way to make money from a scientific article looks very similar, except that scientific publishers manage to duck most of the actual costs. Scientists create work under their own direction – funded largely by governments – and give it to publishers for free; the publisher pays scientific editors who judge whether the work is worth publishing and check its grammar, but the bulk of the editorial burden – checking the scientific validity and evaluating the experiments, a process known as peer review – is done by working scientists on a volunteer basis. The publishers then sell the product back to government-funded institutional and university libraries, to be read by scientists – who, in a collective sense, created the product in the first place.

[A_Gupta]

On Twitter, manastaramgini:
Indo-Aryans: If we go by mainstream indological models the RV should have been composed in the Panjab no earlier than ~2000-1900 BCE - stretching things to the limit butm more like ~1700 BCE \pm 100 yrs. There are some serious problems with this which have been overlooked. Hence, we conclude that the core RV, meaning a certain archaic kernel of it was definitely composed outside India and probably much earlier even if the final redaction and compilation happened later in India. We see no other way out.

—— this is how “interpretation” works.

[Anshuman.Kumar]

The only way out is understanding their methods and then come out with our own models which beat their models hollow on proven scientific methods.

Most polemic or even scientific work which at best raises questions on their conclusions will never pass muster. You can't let them be comfortable in their zone.

Despite attacks on linguistics it still holds strong in its turf.so where are we lacking.?
As suggested here previously by some senior members OF BRF..not everyone can have that rigour but those who have the capacity will have to understand what needs to be done.we cannot work in silos.

[shiv]

Anshuman Kumar ji, other than me and a few other mice no one has attacked linguistics yet so there has been no sustained attack on linguistics. And as far as my knowledge goes no one else has launched as much of a tirade against linguistics as I have done so please don't imagine things. That we must not be in silos is a banal truism but unless someone joins me in my love affair with linguistics my battle is going to be a lonely one. But I have cast the first or at least one of the earliest stones at that pseudoscience. It has to be done. Linguistics is one of the legs that western worldvoew stands on and it needs to be attacked to see how they will respond. And respond they will, given time.

[Nilesh Oak]

On Twitter, manastaramgini:
Indo-Aryans: If we go by mainstream indological models the RV should have been composed in the Panjab no earlier than ~2000-1900 BCE - stretching things to the limit butm more like ~1700 BCE \pm 100 yrs. There are some serious problems with this which have been overlooked. Hence, we conclude that the core RV, meaning a certain archaic kernel of it was definitely composed outside India and probably much earlier even if the final redaction and compilation happened later in India. We see no other way out.

—— this is how “interpretation” works.

This is Vagheesh Narasimhan's interpretation, too.

Vagheesh is also forced to interpret writing of Ramayana and Mahabharata sometime after 1500-2000 BCE in India, based on folk knowledge/memory of Indian continent, but written in Sanskrit (which OF COURSE existed only after 2000 BCE).

When I brought up the nature of Sanskrit names of all character and places of these two epics - he did not realize what I was saying for a while - but then recovered and said the same thing would apply to the names! When I asked what he meant by that.....he was not sure how to solve the problem....so I went into his shoes and asked if he meant, "the writers of the epics began with non-Sanskrit names of actual historical incidents from folk knowledge/memory and converted them to 'Sanskrit' based names ---individuals (Dasharatha, Sahadeva, Krishna, Balarama, Kumhbakarna, Dashagriv, Sugriv and Agastya, etc.) /places (Hastinapur, INdraprastha, Kurukshetra,Ayodhya etc.) /rivers (Sarasvati, Yamuna, Iravati, Narmada, etc.) /mountains (Himavat, Mahendra, etc.)/ etc---- before transcribing......He said, "Yes".

Fun times ahead!

[Nilesh Oak]

Anshuman Kumar ji, other than me and a few other mice no one has attacked linguistics yet so there has been no sustained attack on linguistics. And as far as my knowledge goes no one else has launched as much of a tirade against linguistics as I have done so please don't imagine things. That we must not be in silos is a banal truism but unless someone joins me in my love affair with linguistics my battle is going to be a lonely one. But I have cast the first or at least one of the earliest stones at that pseudoscience. It has to be done. Linguistics is one of the legs that western worldvoew stands on and it needs to be attacked to see how they will respond. And respond they will, given time.

+108

When Vagheesh asked me who are other good Linguistics experts besides Talageri? He did not explicitly state it but wanted to know who were the names from the other camp?.. The first camp being likes of Witzel.

I responded, "Dr. Shiv Sastry" and also added that 99.99% of linguistics experts do not know what the word 'SCIENCE' means. And then said, Dr. Shiv is an ACE.
--
Vagheesh Narasimhan meets Li. Witzel (not unlike Dr. XYZ, Li. Witzel is my invention) often. He was going to meet Li. Witzel in next few days and I told him to convey my regards and also to check with him if he (Witzel) had a chance to read my Mahabharata book I sent to him in 2011.

I gave Vagheesh, author-autographed books of mine (both of them) and he told me he is eager to read them.

Nilesh

[shiv]

^^

[A_Gupta]

Maybe off-topic, maybe not. How the mind plays tricks with perception. Maybe the Greeks did actually hear "Sandrocottus", this is not just a Greek-ization of an Indian name.

https://www.theatlantic.com/technology/ ... ns/560693/

PS: also, maybe off-topic, maybe not: the crisis in science:
http://backreaction.blogspot.com/2018/0 ... ysics.html

[RoyG]

On Twitter, manastaramgini:
Indo-Aryans: If we go by mainstream indological models the RV should have been composed in the Panjab no earlier than ~2000-1900 BCE - stretching things to the limit butm more like ~1700 BCE \pm 100 yrs. There are some serious problems with this which have been overlooked. Hence, we conclude that the core RV, meaning a certain archaic kernel of it was definitely composed outside India and probably much earlier even if the final redaction and compilation happened later in India. We see no other way out.

—— this is how “interpretation” works.

This is Vagheesh Narasimhan's interpretation, too.

Vagheesh is also forced to interpret writing of Ramayana and Mahabharata sometime after 1500-2000 BCE in India, based on folk knowledge/memory of Indian continent, but written in Sanskrit (which OF COURSE existed only after 2000 BCE).

When I brought up the nature of Sanskrit names of all character and places of these two epics - he did not realize what I was saying for a while - but then recovered and said the same thing would apply to the names! When I asked what he meant by that.....he was not sure how to solve the problem....so I went into his shoes and asked if he meant, "the writers of the epics began with non-Sanskrit names of actual historical incidents from folk knowledge/memory and converted them to 'Sanskrit' based names ---individuals (Dasharatha, Sahadeva, Krishna, Balarama, Kumhbakarna, Dashagriv, Sugriv and Agastya, etc.) /places (Hastinapur, INdraprastha, Kurukshetra,Ayodhya etc.) /rivers (Sarasvati, Yamuna, Iravati, Narmada, etc.) /mountains (Himavat, Mahendra, etc.)/ etc---- before transcribing......He said, "Yes".

Fun times ahead!

I thought of this too when gaming from their side [Kernel]. Very poor argument and didn't expect them to use it. The fact he is meeting with Witzel means that we've been battling him all along. Nothing has changed. He has polluted everything - linguistics, comparative anthropology, genetics, etc. They are purely in the business of invention. The fact they are even accepting arguments like the ones you posed to them makes a mockery of science.

You would expect at least 1 non IE name somewhere in the NW-Gangetic plains that the so called 'Aryan invaders' recycled to mark a topographical feature or settlement. Does he realize how foolish he sounds?

[gandharva]

On Twitter, manastaramgini:
Indo-Aryans: If we go by mainstream indological models the RV should have been composed in the Panjab no earlier than ~2000-1900 BCE - stretching things to the limit butm more like ~1700 BCE \pm 100 yrs. There are some serious problems with this which have been overlooked. Hence, we conclude that the core RV, meaning a certain archaic kernel of it was definitely composed outside India and probably much earlier even if the final redaction and compilation happened later in India. We see no other way out.

—— this is how “interpretation” works.

Looking for redaction history (or layers in a Text), originally developed in the context of biblical criticism by Protestants, is projecting Protestant habits onto the Hindu Scriptures by Western Indologists. Even though the original field of application of this concept was Old Testament criticism, they were found useful in epic studies to separate out an “Urepos” (a primordial epic or an original epic) from the text as extant. This process led to the postulation of two phases (stages or ideologies) in Indian history: a Aryan-Indo-European phase and a Brahmanic-Hindu phase. Knowledge created thus (with all logic & reasoning) is tangential at best and irrelevant at worse (firmly rooted in Protestant theology) to the real content of Hindu Scriptures. The real problem with these imaginations/interpretations is that Indology is an ill-conceived Protestant theology and its "facts" are the typical case of the result of research determined a priori by the requirements of the method.

Abv is just my recollection of reading "Nay Science" by Vishwa Adluri.

[shiv]

On Twitter, manastaramgini:
Indo-Aryans: If we go by mainstream indological models .blah blah blah

The "we" always tickles me no end - especially when used by an individual. This is not a royal we, it is an ass covering we. It always reminds me of that song "Inhi logon ne" from Pakeezah - "humri na maano sipahiya se poocho" - it's not just about me - ask anyone else aljo

[Suresh S]

[A_Gupta]

[sudarshan]

This is Vagheesh Narasimhan's interpretation, too.

Vagheesh is also forced to interpret writing of Ramayana and Mahabharata sometime after 1500-2000 BCE in India, based on folk knowledge/memory of Indian continent, but written in Sanskrit (which OF COURSE existed only after 2000 BCE).

When I brought up the nature of Sanskrit names of all character and places of these two epics - he did not realize what I was saying for a while - but then recovered and said the same thing would apply to the names! When I asked what he meant by that.....he was not sure how to solve the problem....so I went into his shoes and asked if he meant, "the writers of the epics began with non-Sanskrit names of actual historical incidents from folk knowledge/memory and converted them to 'Sanskrit' based names ---individuals (Dasharatha, Sahadeva, Krishna, Balarama, Kumhbakarna, Dashagriv, Sugriv and Agastya, etc.) /places (Hastinapur, INdraprastha, Kurukshetra,Ayodhya etc.) /rivers (Sarasvati, Yamuna, Iravati, Narmada, etc.) /mountains (Himavat, Mahendra, etc.)/ etc---- before transcribing......He said, "Yes".

Fun times ahead!

The guy doesn't realize what a self-goal he scored? He doesn't seem very bright, probably just a tool in somebody else's hand (Witzel comes to mind, like you said). Nice job nudging him along.

Next, you could ask him where his own name "Vagheesh" came from. It finds a mention in the Vishnu Sahasranama (MB), so I guess it was either composed out of India and brought in by the "Sanskrit bearers," or a folk memory of something else in India which the writer(s) of the MB transliterated into Sanskrit, or whatever. A guy like this invites so many chances to tie him up into knots.

[A_Gupta]

For future reference:
https://liorpachter.wordpress.com/2014/ ... rto-rican/
This blog takes computational biology to task.

PCA - detailed explanation
https://liorpachter.wordpress.com/2014/ ... -analysis/

There are many generalizations and modifications to PCA that go far beyond what has been presented here. The first step in generalizing probabilistic PCA is factor analysis, which includes estimation of variance parameters in each coordinate. Since it is rare that “noise” in data will be the same in each coordinate, factor analysis is almost always a better idea than PCA (although the numerical algorithms are more complicated). In other words, I just explained PCA in detail, now I’m saying don’t use it!

Perhaps more importantly, there are issues in using PCA that I have not discussed. A big one is how to choose the PCA dimension to project to in analysis of high-dimensional data. But I am stopping here as I am certain no one is reading at this far into the post anyway

[A_Gupta]

https://www.nature.com/articles/ng.139
2008 paper, so one hopes it is incorporated into modern thinking already.

Interpreting principal component analyses of spatial population genetic variation

John Novembre & Matthew Stephens

Nature Genetics volume 40, pages 646–649 (2008)
doi:10.1038/ng.139
Abstract:

Nearly 30 years ago, Cavalli-Sforza et al. pioneered the use of principal component analysis (PCA) in population genetics and used PCA to produce maps summarizing human genetic variation across continental regions1. They interpreted gradient and wave patterns in these maps as signatures of specific migration events1,2,3. These interpretations have been controversial4,5,6,7, but influential8, and the use of PCA has become widespread in analysis of population genetics data9,10,11,12,13. However, the behavior of PCA for genetic data showing continuous spatial variation, such as might exist within human continental groups, has been less well characterized. Here, we find that gradients and waves observed in Cavalli-Sforza et al.'s maps resemble sinusoidal mathematical artifacts that arise generally when PCA is applied to spatial data, implying that the patterns do not necessarily reflect specific migration events. Our findings aid interpretation of PCA results and suggest how PCA can help correct for continuous population structure in association studies.

[A_Gupta]

^^^ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3989108/

n summary: i) when analyzing data with a spatial covariance structure, PCA produces highly structured results relating to sinusoidal functions of increasing frequency; ii) in as far as PCA results depend on the details of a particular data set, they are affected by factors in addition to the actual underlying spatial population structure, such as the distribution of sampling locations and the amount of data available. These conclusions are supported not only by the results we present here, but by extensive empirical results from other fields where PCA has been applied to spatial data, including pattern analysis of natural images [28], ecology [23] and climatology [25, 29, 30, 31].

Both features i) and ii) above limit the utility of PCA to draw inferences about underlying processes, a fact previously noted in climatology ([25, 29, 30, 31]). In particular, interpreting gradient and wave-like patterns in PC-maps as signatures of historical migration events is problematic because such patterns arise quite generally under a simple condition: that genetic similarity decays with distance. This finding has important implications for interpretation of PCA in population genetics generally, and for Cavalli-Sforza et al’s PCA analyses in particular. Not only is this simple condition likely to be satisfied under many possible models, but, since Cavalli-Sforza et al used spatial interpolation to estimate allele frequencies, their data could satisfy this condition even if absent in the underlying allele frequencies [6, 8]. (Indeed use of interpolation may partly explain the striking similarity between Cavalli-Sforza et al’s PC-maps and those predicted by theory, particularly in Asia where their analysis was based on fewer samples. That said, recent analyses of European data without interpolation also show perpendicular gradients in PC1 and PC2 [32].)

Regarding the specific question of a Neolithic expansion in Europe, we emphasize this paper is not about whether or not such an expansion occurred; a full consideration of this would require a synthesis of multiple types of evidence from many diverse sources (see review in [33] and e.g. [34, 35, 9, 36, 37, 38, 39]). It is true that the NW-SE slope of the PC1 gradient in Europe suggests that this may be the direction of greatest genetic variation in Europe (although a careful analysis would have to take account of the fact that other factors, including the shape of the continent, could also influence the slope direction). However, if a Neolithic expansion could explain this, it is but one of many possible explanations.

For another example of how our results aid interpretation of PCA, consider the data from Linz et al [13] who found that PC-maps from Heliobactor pylori show similar patterns to those in Cavalli-Sforza et al’s human data and who use this as part of an argument that genetic patterns of H. pylori reflect a shared migrational history with humans. There are good reasons to suspect genetic variation in H. pylori will have been influenced by human migrations. However our results show that similar patterns in PC-maps of two groups does not imply a shared migrationary history; indeed, if each group shows an underlying spatial covariance structure, then similar patterns will often occur in the top few PC-maps even if their histories are independent (e.g. fig. S1).

[Nilesh Oak]

^^^ A Gupta ji

Excellent find. Good stuff. I have to read and re-read few times to internalize the content and summary of these studies. Thank you.

[A_Gupta]

^^
I cant claim to have understood it all either.

[sudarshan]

Here's the contrary view. I can't find a corresponding publication/ paper, this is just a .ppt. But the guy seems well-published otherwise.

www.tau.ac.il/~saharon/StatGen2010/PCA-pop.ppt

This document upholds Cavalli-Sforza's original 1978 work, and stresses that while Novembre-Stephens do have a point about PCA artifacts due to local effects, that N-S nevertheless have failed to account for some factors. Also, that the effects noted by C-S in 1978 are much larger than the artifact effects that N-S point out. Therefore (according to this document) Cavalli-Sforza's original conclusions are right.

I know the basics of PCA, but can't fully decode either of the above. However, it seems (from cursory searches) that even when dealing with EEGs, the artifacts generated by PCA are an issue.

[Vayutuvan]

Probably something to do with the interpolating polynomials chosen when the samples are too few in number. One needs orthogonal bases for a proper fit.

[A_Gupta]

^^^ Thanks! Good find!

I do have some reservations about this

We can also make simple, non-asymptotic statements, such as: If the environment is a square and migration only local, then by symmetry the North-South and East-West eigenvalues should be similar

Which is true, but the point is e.g., that geographically north-south and east-west migrations are not the same. East-west keeps one in a similar climatic zone, allows domesticated plants and animals, lifestyle, etc., easier migration. One of the points of Jared Diamond has in Guns, Germs and Steel, is that Eurasia has a east-west axis, while the Americas have a north-south axis. There is no symmetry and it should not be expected.

I accept the idea that possibly Cavalli-Sforza found a real signal above the artifacts. But it seems to me that a PCA of genotypes will expose geographic structure first; and one needs signal above and beyond that natural geographical to infer migration.

[Vayutuvan]

We are actually jackasses because no one asks the geneticist which genes code for language. When you ask them they will point to linguists/archaeology papers and say 'Language migration has been demonstrated by linguists who tell us that PIE was in steppe and those people moved to India and Iran, creating Indo-Iranian language first and later Avestan and Vedic Sanskrit"

Don't these geneticists recognize what is called 'a circular argument' when they advance this kind of an answer?

The entirety of the language science is described in, mathematics, is built on foundations and deductions of the kind 'if A then B' & 'A' then deduce B.

If there are three implications as in

1. if A then B
2. if B then C
3. if C then A

just showing those to be true doesn't say anything about the truth value of A, B or C. On the other hand, if one is able to prove one of A, B, or C to be True, then all three predicates A, B, and C are True.