Notice that I use the term 'coastal route' senso lato, meaning a migratory route in Asia (from ultimately Africa) via Arabia and South Asia, rather than Central Asia and Siberia. How strictly 'coastal' this migration was is subject to debate but the evidence is strong and growing in favor of the tropical route in any case.
The links were provided to me by a reader (carpetanuiq) in the comments section of this previous post. They are all very valuable but specially the three I want to introduce here. All links are PDF.
Analysis of lesser cost routes into South Asia and through it
Fields et al., The southern dispersal hypothesis and the South Asian archaeological record: Examination of dispersal routes through GIS analysis. Journal of Anthropological Archaeology 2007. [LINK]
Abstract
This research advances a model for coastal-based dispersals into South Asia during oxygen isotope stage (OIS) 4. A series of GIS-based analyses are included that assess the potential for expansions into the interior of South Asia, and these results are compared with known archaeological signatures from that time period. The results suggest that modern Homo sapiens could have traversed both the interior and coastlines using a number of routes, and colonized South Asia relatively rapidly. Use of these routes also implies a scenario in which modern H. sapiens, by either increased population growth or competitive ability, may have replaced indigenous South Asian hominin populations.
Key maps from this paper (there are some others also interesting):
 |
| Cut from fig.2 Location of linear origin and the results of the least-cost route analysis into South Asia. Least-cost route is indicated by the grey [black] line. |
 |
| Fig. 3 Results of the wandering path analyses into and through South Asia. Least-cost routes are indicated by the grey [black] lines. |
The Red Sea in the Pleistocene
Geoff Bailey, The Red Sea, Coastal Landscapes, and Hominin Dispersals. Published within M. Petraglia & J. Rose (eds.). The Evolution of Human Populations in Arabia (2009). [LINK]
(No abstract)
Key maps in this paper are:
 |
| Fig. 1 showing among other things 'a simplified distribution of Lower and Middle Paleolithic archaeological sites in the Arabian Peninsula'. |
 |
| Fig. 5 reconstructed shoreline of Bab-el-Mandeb at different Late UP dates (cal BP) |
The Persian Gulf "oasis"
Jeffrey I. Rose, NEW LIGHT ON HUMAN PREHISTORY AROUND THE PERSIAN GULF OASIS. (2009 or 2010?) [LINK]
Abstract
The emerging picture of prehistoric Arabia suggests that early modern humans were able to survive hyperarid climatic conditions that periodically caused widespread landscape desiccation by contracting into environmental refugia around the coastal margins of the peninsula. This paper reviews new archaeological and palaeoenvironmental evidence from the richest of such zones in eastern Arabia: the “Persian Gulf Oasis.” These data are used to assess the role of this ancient alluvial plain, which, prior to being submerged beneath the waters of the Indian Ocean, was well-watered by the Tigris, Euphrates, Karun and Wadi Batin Rivers as well as subterranean aquifers flowing beneath the Arabian subcontinent. Inverse to the amount of annual precipitation falling across the interior of Arabia, reduced sea levels periodically exposed large portions of the Persian Gulf, equal at times to the size of Great Britain. Therefore, when the hinterland was desiccated, populations could have contracted into the Gulf Oasis and exploited the freshwater springs. This relationship between environmental amelioration/desiccation and marine transgression/regression is thought to have driven demographic exchange into and out of this zone over the course of the Upper Pleistocene and Early Holocene, as well as having played an important role in shaping the cultural evolution of local human populations during that interval.
Key maps are:
 |
| Fig. 2 Map of ancient drainage systems in Arabia showing the Ur-Schatt River Valley as the primary recipient of runoff within the regional catchment zone. Numbers indicate known MP/UP sites in eastern Arabia and southern Iran. |
 |
| Cutoff from fig. 5 (Palaeo-shoreline configuration, drainage systems and archaeological sites around the Persian Gulf basin during the Upper Pleistocene and Early Holocene). Shown 74-24 Ka map only. Dots are archaeological sites. |
Other relevant links mentioned in that discussion are:
- Gahnim Wahida et al., A Middle Paleolithic Assemblage from Jebel Barakah,
Coastal Abu Dhabi Emirate. (Chapter in M.D. Petraglia and J.I. Rose (eds.), The Evolution of Human Populations in Arabia, Vertebrate Paleobiology and Paleoanthropology). [LINK] - Marta Mirazón Lahr et al., Project: 'Searching for traces of the Southern Dispersal'. General description of the archaeological research project0 by the 'Petraglia team'. [LINK]
- Ancient Indian Ocean Corridors. Blog of the same team, inaugurated this spring. [LINK]
- Synopsis of Paleo India (2007). James B. Harrod. [LINK]
The Arabian peninsular details are interesting. But I definitely like figure 3 of Fields et al. - because it succinctly shows what I have been contemplating about and what could be the first separation into three of the modern groupings of humans having their beginning in the Indian sub-continent.
ReplyDeleteThat is, the Indus River/ Hindukush population eventually became the "Caucasian" one, the Ganges/SE Himalayan one was the seed for the Mongolian/Native American one, and the coastal Brahmaputra group seeded the SE Asian/ Austronesian/ Australian one.
I agree that the GIS map of South Asia is very revealing, specially because it largely corresponds to actual archaeological sites (Narmada-Son-Low Ganges and West coast-Krishna/Penner rivers). We lack AFAIK of archaeological data for Pakistan (lack of research) and for most of the East coast (under water?) but it's reasonable to assume these less cost routes were also used. There are also sites at the eastern fringes of the Thar desert (out of the suggested routes), just for the record.
ReplyDeleteBut I do not think your conclusions about 'races' are warranted. For two reasons:
1. 'Racialization' is a long term process of regional homogenization, based of course on founder effects but also largely a process of phenotype convergence that can even be socially biased (cultural preference for certain looks). This may have some genetic backing, as appearance-related genes have been occasionally reported as more different than other "invisible" genes among populations. In any case none of the modern phenotypes are clearly apparent until much later in time (Late Paleolithic in Europe, maybe Neolithic in East Asia).
2. These people of the Indian Middle Paleolithic are surely ancestors of all Eurasians but they must be ancestors also of modern South Asians, naturally. The genetic evidence suggests also a quick expansion to SE Asia (from where to NE Asia and Near Oceania, also quite fast) and I understand that the origins of the Eastern phenotypes should be searched for in those areas, rather than India (even if you may be able to trace some appearance elements back to India these would not be too important).
I do not think in any case that Northern Indians (excepted those populations that appear to be Neolithic immigrants from SE Asia) look 'Mongoloid' at all. If you compare Bengalis with world populations surely West Eurasians are closest.
I am rather of the opinion that the main appearance of early Eurasians was in what I call the Caucasoid-Australoid continuum, with the Mongoloid phenotype being exceptional and specific to East Asia. However SE Asians often retain some deal of this Caucaso-Australid element, apparent also in other isolated populations such as the Ainu of Japan.
On the other hand Northern Caucasoids (Northern Europeans essentially) appear to have some influences from a proto-Mongoloid phenotype, occasionally also apparent in genetics. Obviously blondism is not such a trait but partial epicanthic fold is instead. But the bulk of the Caucasoid phenotype's origins is surely in South Asia, specially in Northern SA probably but not exclusively in Pakistan surely (even if it may have played a coalescent role prior to the colonization of West Eurasia).
Also notice that the same that the easiest route from West Eurasia (and Central Asia) to India is via the coast, the reverse must also be true. Of course, we cannot exclude errors in the GIS analysis, and I would certainly think that some groups went through Khyber Pass but the bulk probably back-migrated by the same coastal route used to reach SA first of all.
I'm struck by how many new archaeological sites there are in the middle east.
ReplyDeleteFrom the article:
ReplyDelete"Use of these routes also implies a scenario in which modern H. sapiens, by either increased population growth or competitive ability, may have replaced indigenous South Asian hominin populations".
Doesn't the evidence indicate continuity across the Toba level? I haven't checked the link yet, but does the author supply dates for the various sites?
"We lack AFAIK of archaeological data ... and for most of the East coast (under water?) but it's reasonable to assume these less cost routes were also used".
The lack of sites on the east is unlikely to be because they're under water. Sea level rose on both coasts.
Your first quote, Terry, is from the abstract of the Fields paper, which does not deal with any factual archaeology but with a computer modeling. So I cannot really answer your first questions because that paper does not talk of archaeology but of a computer model which shows the easier routes by walking.
ReplyDelete"The lack of sites on the east is unlikely to be because they're under water. Sea level rose on both coasts".
It is possible. I haven't really checked the orography yet but as you know the extensions covered by water now and firm land in the past vary a lot, depending on the steepness of the continental platform.
Anyhow, IF people migrating by the East coast did so ONLY by beachcombing and not penetrating the interior at all, as could be the case in a "rapid coastal migration" model, senso stricto, then their sites would be covered underwater now. We cannot discard that.
Of course sites may not have been found for a number of other reasons too (localized lack of research, bad luck, etc.) It's not like just for being there an archaeological site is automatically found and suddenly known in all its extension and depth, you know. Active research is needed.
"It's not like just for being there an archaeological site is automatically found and suddenly known in all its extension and depth, you know".
ReplyDeleteNo. But beliefs unsupported by such evidence can be very suspect.
"the Fields paper, which does not deal with any factual archaeology but with a computer modeling".
Exactly. It starts with an assumption and then looks at the possibilities. However the authors make some interesting comments on various aspects of their theory.
"IF people migrating by the East coast did so ONLY by beachcombing and not penetrating the interior at all, as could be the case in a 'rapid coastal migration' model, senso stricto, then their sites would be covered underwater now. We cannot discard that".
The Field paper makes the surprising comment:
"The Makran coast and the coastline of the Thar Desert ... were extremely arid during OIS4, and consisted of a narrow coastal corridor backed by rugged desert hills".
I have known this to be so for some time, yet the authors still suggest that humans could have moved 'rapidly' along that coast. They would certainly have needed to move very rapidly. What did they do for fresh water, for example? A few other quotes from the link:
"but at the moment it is impossible to discern a uniquely modern H. sapiens signature from any deposits prior to 28kya".
The technology of moderns and pre-moderns is obviously very difficult to distinguish, unlike the situation you claim in one of the comments on your New Guinea post. The authors also agree with points I have consistently made:
"The presence of the Indus and Ganges-Bramaputra deltas would have served as full of partial barriers, keeping populations more or less within the confines of the South Asian subcontinent".
And:
"Further population growth would probably have spread modern humans north and south, to the limits of the savannahs".
So they too consider the savanahh region as being prime human habitat.
"The Makran coast and the coastline of the Thar Desert ... were extremely arid during OIS4, and consisted of a narrow coastal corridor backed by rugged desert hills"
ReplyDeleteOk. Which is the chronological frame of the OIS or MIS 4 period? I understand that did not begin before 90 Ka. Please check but I'm pretty sure that until those dates it was the OIS 5, which is the same or includes the Abbassia Pluvial (120-90 Ka. ago), the optimal period for the Out of Africa episode.
"yet the authors still suggest that humans could have moved 'rapidly' along that coast. (...) What did they do for fresh water, for example?"
I understand from the quote that the "narrow coastal corridor" was not so arid and had some freshwater resources. But for food it certainly calls for beachcombers or fisherpeople.
"The technology of moderns and pre-moderns is obviously very difficult to distinguish"...
I am not so sure, these things are largely cultural (you learn from the adults in your community), so same tech may well indicate same species and even same ethnos. H. erectus, last reported in India 200 Ka ago, used Acheulean tech. There's no report of any Neanderthal nor their associated Mousterian industry in the subcontinent.
It is very possible that the arrival of "Middle Paleolithic" (above Acheulean) to India means arrival of H. sapiens but further clarification is needed. Per the synopsis I mentioned in the other thread, MP with stone blades could be as old in North India as 150 Ka ago and not more recent than 100 Ka apparently.
We may well be before a rapid expansion not c. 60 Ka but c. 100 Ka ago (or more). It needs clarification but it's a strong possibility.
"The presence of the Indus and Ganges-Bramaputra deltas would have served as full of partial barriers, keeping populations more or less within the confines of the South Asian subcontinent".
But only by walking... which is all what they measure with the GIS modeling. It would have been no barrier for people able to make rafts or boats. If the much wider and dangerous straits of Indonesia or the Red Sea were no absolute obstacle, I see no reason why the Ganges Delta would be it.
"H. erectus, last reported in India 200 Ka ago, used Acheulean tech. There's no report of any Neanderthal nor their associated Mousterian industry in the subcontinent".
ReplyDeleteThe authors actually claim a specimen in the Narmada Valley dated at 250-300 kya as having been identified as Homo heidelbergensis. Interesting if the identification is correct.
"Which is the chronological frame of the OIS or MIS 4 period?"
I don't know, but I guess the information is somewhere in the paper, if I can be bothered going back. The authors seemed to think the time frame was significant for any postualted time of modern human entry to India. But even today the region could be described as 'arid'. Entry to India from the west has never been straightforward, even historically.
"It would have been no barrier for people able to make rafts or boats".
Again it is the authors who claim the deltas as being obstacles. I happen to agree with them.
"If the much wider and dangerous straits of Indonesia or the Red Sea were no absolute obstacle, I see no reason why the Ganges Delta would be it".
The straits in Indonesia were crossed only much later (especially if you're going to claim a very early H. sapiens presence in India), and we have no evidence at all that ancient humans crossed the Red Sea.
"It is very possible that the arrival of 'Middle Paleolithic' (above Acheulean) to India means arrival of H. sapiens but further clarification is needed".
Quite possibly. But that places the OoA long before any widely accepted date.
"We may well be before a rapid expansion not c. 60 Ka but c. 100 Ka ago (or more). It needs clarification but it's a strong possibility".
I agree completely that any OoA was earlier than currently assumed. However I'm also sure that, like everything in biology, it was nowhere near as simple as most imagine it to have been.
"The authors actually claim a specimen in the Narmada Valley dated at 250-300 kya as having been identified as Homo heidelbergensis".
ReplyDeleteRight, I was not aware of that detail.
Would seem to be a hot issue because it seems to be the only hominin worldwide that has a modern cranial capacity at that time. Per the Synopsis of Paleo-India of Harrod it has a cranial capacity of 1150-1420cc., which is larger than any Heidelbergensis ever. It's a total anomaly and Harrod classifies it as "archaic Homo sapiens", whatever that means.
Homo sp. used to have 700-1050 cc., slightly increasing through time. Excepting this mysterious individual only H. sapiens and H. neanderthalensis are above that range, in what is a qualitative evolutionary jump. These two species are in the 1100-1900 cc. range.
Not sure what to think. Some online stuff to chew on: The Telegraph and an archive.org page.
Maybe she's the true antecessor or both Neanderthals and Sapiens... or maybe is a distinct third species... or maybe it has been wrongly dated. No idea, really.
"Again it is the authors who claim the deltas as being obstacles".
ReplyDeleteFor walking. The paper only deals with walking routes and unless you are some mythological person you do not walk on water, you swim in it or sail on it.
"I happen to agree with them".
Again, how old are you?
"The straits in Indonesia were crossed only much later (especially if you're going to claim a very early H. sapiens presence in India)"
I wouldn't think of "much later" because the genetics don't really seem to allow for long periods between the colonizations of India and Sahul, right? It may also be the case that we are missing fossils in Sahul, as we do in India.
"But that places the OoA long before any widely accepted date".
Not that I really care. Unless there are strong reasons for such dates, what I am not aware of. "God" and Capitalism are widely accepted concepts and I could not care less (except for the annoying sociological pressure for thoughtless conformism).
"Again, how old are you?"
ReplyDeleteSixty- five. But what's that got to do with things?
"Maybe she's the true antecessor or both Neanderthals and Sapiens..."
Quite possibly so. But that puts the common origin of the two species as much more recent than you've argued for. On the other hand in Geoff Bailey's paper he suggests, 'However, this does not rule out ... the possibility ... that earliest hominin populations originated over a wider zone that encompased Africa, Arabia and Asia, or that movements between Africa and Asia may have been in both directions and not just one way out of Africa'.
"For walking. The paper only deals with walking routes"
The authors are apparently prepared to accept that at that stage humans didn't have boats. In fact Geoff Bailey's paper has some interesting comments regarding the Red Sea crossing:
"especially at the southern end, where the sea-channel is shallow enough that it might have been closed or easily crossed at low sea-level stands".
No need for boats. And:
"Moreover, while it is true that the facility to cross the southern end of the Red Sea would have broadened the possibilities of contact and movement between Africa and Asia, such a dispersal route is not essential to populate Arabia or to initiate movement thence further to the east. The whole of the Arabian Peninsular could have been filled with human and other mammalian species derived from Africa, or western Asia, via the northern end of the Red Sea, more or less instantaneously within the chronological resolution of existing dating techniques".
Obviously the author sees no need to postulate a Bab al Mandab route. In fact he stresses several times that there is no evidence for it.
Ultimately, of course, it doesn't matter which of the two routes humans took because both the Bailey and the Rose papers shows humans eventually filled the Arabian Peninsular anyway. If the Arabian Peninsular was relatively lush at times the same would certainly be true for much of the Iranian Plateau. People could have emerged from Arabia onto the Plateau anywhere between Lake Van and the Strait of Hormuz. Both maps show human habitation along the southern slopes of the Zagros. In fact the Hadramawt is extremely unlikely to have been the route to India. And there's certainly no evidence to support it being so. So there's nothing at all 'coastal' about the 'great southern coastal route'. It remains to be seen whether the route of the first arrivals in SE Asia was even 'southern'.
"But what's that got to do with things?"
ReplyDeleteUnsure. There's something in your way of arguing that sounds immature to me: more precisely taking the part for the whole. So in this sense I am surprised by your answer. On the other hand young people are not normally so opinionated, stubborn nor cling so much to old ideas against evidence but actually much more open minded in most cases. And in this sense your answer does not really surprise me.
In any case I suspected a generational gap, really. And some things make better sense now. See: my father is roughly your age and is quite liberal (in the European sense of the word: center-right). I never in my life heard from him a racist slur or any racist opinion until one day when I happen to mention that Brazil was a growing power and a possible soon-to-come first global power (speculating a bit, I know) he said "but they are blacks" (what is incorrect but anyhow).
Whatever respect I had left for him went to the trash bin that day.
It's probably something of the culture of the 20th century before the late 60s, a moment when the sociology changed a lot. It may also be that as we grow older we tend to get stuck in old ideas, even if we know they make little sense and have never liked them anyhow. Unsure.
Well, anyhow, I was really intrigued because some comments didn't make too much sense to me. Now with the age reference I can tolerate more easily your stubbornness and your clinging to old clichés. It's something.
"Quite possibly so".
ReplyDeleteOr not. We have to admit we don't have the slightest idea of how to fit that partial skull in our understanding of the evolution of the Homo genus.
So far it's an out-of-place artifact. And the general academic approach to such things is to distrust because, unless confirmed by further findings, there may be some sort of error in the data.
"The authors are apparently prepared to accept that at that stage humans didn't have boats".
They do not even consider that possibility. That paper does what it does and nothing else.
"In fact Geoff Bailey's paper has some interesting comments regarding the Red Sea crossing:
"especially at the southern end, where the sea-channel is shallow enough that it might have been closed or easily crossed at low sea-level stands".
No need for boats".
See: you are interpreting this sentence to your convenience. Nowhere he says anything about boats and, in my understanding, they are implicit. Whole families with children and baggage do not cross sea or rivers by swimming. Children can easily die in much easier water bodies. My sister at the age of two almost died in a water collecting canal that was perfectly walkable by adults and even slightly older kids (I am just a little more than one year older, did not know how to swim yet, and I was puzzled how she could be at any risk there, go figure!) The rivers I am familiar with can be crossed by a young swimmer easily (not with much baggage though) but would be impassable for families without boats or rafts.
One thing is individual athletic crossing and a very different thing crossing such a water body with all the clan (pregnant and breastfeeding women, children of various ages, and even old grandpa or grandma) as no doubt was done. Elephants are smart but the can also walk and run (and swim) almost after birth, just like your usual gnu or zebra. We just cannot.
"Obviously the author sees no need to postulate a Bab al Mandab route".
I do think it is necessary, partly for the lack of India-Palestine techno-cultural connection.
But specially I think that it was at least as easy to cross the Red Sea as to cross the Sinai Desert. And I think also that it's plainly stupid to believe that people who were just like you and me could not do what you and I can. I have never made a raft in my life but I'm sure I would figure out if I lived in coastal, riverine or lacustrine conditions in Stone Age parameters. You just need boats like you need ropes, baskets or spears.
It's a matter of how each one imagines people some 100 years ago or whatever date happens to be the correct one. I imagine myself there, not some sort of retard.
"There's something in your way of arguing that sounds immature to me"
ReplyDeleteBecause when I discuss things with you you come across like a grumpy old grand-dad who is totally convinced of the literal truth of some religious myth in spite of there being no evidence in support of that myth. Perhaps that is a result of your Catholic upbringing. I also get frustrated with those who, as a consequence of their rigidly held beliefs, cannot see what is so blindingly obvious to me.
"I have never made a raft in my life but I'm sure I would figure out if I lived in coastal, riverine or lacustrine conditions"
Of course you would these days. You're aware that it can be done.
"your stubbornness and your clinging to old clichés".
MY 'stubbornness and clinging to old clichés'? Far from supporting your belief in 'the great southern coastal route via the Bab al Mandab' all three papers you posted are rather sceptical of it. Even Bailey, the author of the paper providing the most promising line of evidence for the possibility, has doubts as to the veracity of the theory. Rose, in the Persian Gulf paper, actually supports what I have claimed about the origin of the theory:
"To explain these genetically predicted dates for human expansion out of Africa, in light of relatively early dates for the colonization of Sahul between 60,000 and 40,000 years ago (O'Connell and Allen 2004), researchers have suggested that the successful AMHS colonists were coastally-adpated groups that moved rapidly along the continental shelf rimming the India Ocean".
And later:
"Thus, the discovery of haplogroup M1 in the Horn of Africa, just across the Red Sea and within sight of Arabia, suggested to scholars that the 'Arabian Corridor' (i.e. Yemen, Oman, and the U.A.E.) served as a conduit for the first populations moving out of Africa and into Asia".
I'm not sure if you've read all of that paper, but if not you will be interested to read the last section: 'The Genetic Conundrum'.
"The puzzle pieces do not quite fit; there is wide disagreement between archeological, genetic, and fossil scenarios of modern human emergence".
"In all three cases, the predicted time spans for each demographic 'event' are so different they hardly overlap. Either there are flaws in the dating, and/or the data sets are measuring fundamentally different things".
"If we consider that the population bottleneck release(s) branching from the common ancestral group emanated from Southwest Asia, not Africa, the threads of archeological, genetic, and fossil evidence agree. The model described in this paper proposes that the AMHS population movement out of Africa during MIS 5 was not a failed wave of expansion; rather, these carriers of the mtDNA L3 marker were bivouacked (so to speak) in Southwest and/or Central Asian refugia, such as the Persian Gulf Oasis, until the ameliorated conditions that set in during MIS 3 permitted subsequent range expansions".
"Marks (2005:81) arrives at a similar conclusion: 'the immediate origins of the explosion of "modern behavior" seen in Europe, but not in Africa, might be found at the contact between Eastern Europe and Western Asia'".
"Because when I discuss things with you you come across like a grumpy old grand-dad who is totally convinced of the literal truth of some religious myth in spite of there being no evidence in support of that myth".
ReplyDeleteI accept the grumpy thing but your insistence on accusations of "belief", "myth", etc. are a total nonsense and I understand you use that only as defensive trick. I understand that you use such tricks, irregularly but often enough, because you lack arguments and prefer to disqualify.
"Perhaps that is a result of your Catholic upbringing".
Well, I'm surely the less Catholic of all Catholics. I quit Sunday church at the age of 11, I was considering myself out of the Catholic Church at the age of 12 (specially because I could not understand their obsession against divorce), I was atheist before reaching 14 and I was sabotaging religion class and compulsory mass since that age... until they kicked me out (before I was 16). Unlike all my peers I never took confirmation. And I rejected joining a prestigious Jesuit university and chose the public one instead.
So, hmmm, forget it. I've argued with many Jesuit priests and left them without arguments once and again, I'm almost as good as Jesus (per the NT's story of the boy Jesus discussing or rather teaching theology to the Rabbis) in pushing priests towards disbelief and doubt.
I was punk when all my friends still went to discos, I was active in social movements at the age of 15.
And at the age of six or seven I already daydreamed of bombing that hateful Jesuit school, which felt like a prison.
I rejected to be the godfather of my younger brother. I have not gone to the baptism of any of my nephews. I don't step in churches, not for marriages not for funerals: I hate the Catholic Church like you could not understand.
What can I say: there's probably very few "Catholics" as anti-Catholic (and generically anti-Yahvistic) as myself.
So wrong: absolutely wrong.
"Of course you would these days. You're aware that it can be done".
ReplyDeleteSuch awareness is as simple as to see a piece of wood floating on water. After you experiment with single log models for a little while you must soon realize that something flatter will be more stable. And the rest is just refinement, improvements of that raft made with a few logs.
It's terribly simple. I find Acheulean or Levallois tech much more complicated, sincerely. A raft is like chopper industry, almost chimpanzee level tech. Notice that I say almost, I know that chimps do not make rafts, but their brains are several times smaller than ours. They could learn to make rafts, specially bonobos, but inventing them is probably too much for them.
"To explain these genetically predicted dates for human expansion out of Africa, in light of relatively early dates for the colonization of Sahul between 60,000 and 40,000 years ago (O'Connell and Allen 2004), researchers have suggested that the successful AMHS colonists were coastally-adpated groups that moved rapidly along the continental shelf rimming the India Ocean". (Rose)
Well, I understand that the main reason for the rapid migration is genetic: you have almost 30 coding region mutations from "Eve" to L3 and almost 30 CR mutations from M or N to present day haplogroups, but between L3 and M there are only three mutations and four between L3 and N.
As an African origin for L3 seems beyond doubt, the time for the coalescence of M and N, which appear as such in South Asia and SE Asia respectively, is very small.
It has only a diffuse relation with any absolute dates: that is a misunderstanding. The true reason is relative time lapses.
If [1] Age(M)=60 Ka, then Age(L3)=66 Ka. But if [2] Age (M)=100 Ka, then Age(L3)=110 Ka. All very rough figures but the point is that only so much time can have occurred between L3 and M (and N too). So if the archaeological data suggests [2], then the migration may have lasted some 10,000 years, not 50,000 in any case.
Coastal or by land, earlier or later, the migration between the African coast of the Red Sea and the South Asian coast of the Arabian Sea happened in surely less than 15,000 years in the most ancient scenario we can consider.
"I'm not sure if you've read all of that paper, but if not you will be interested to read the last section: 'The Genetic Conundrum'".
In truth I have only read the second and third papers quickly (shallowly). I spend a lot of time doing other things, including replying to you. I will read them this week for sure.
"The puzzle pieces do not quite fit; there is wide disagreement between archeological, genetic, and fossil scenarios of modern human emergence".
They can be conciliated, I'm pretty sure. Beware anyhow of the "genetic scenarios" because those depend on who you read. Specially beware of those putting too much emphasis on the molecular clock.
The "fossil" (skeletal remains, not really true fossils) record is of course very imprecise. Sadly enough. We can't do much about that but wait for further discoveries. As in the case of Europe, where archaeology is pretty good but there are huge gaps in the skeletal record, specially notable in the key period of MP-UP transition, we can't but do some guesswork based on other archaeological aspects and correlations with other areas where skeletal remains and tools are found together.
Of course this leaves some room for different opinions, specially in the case of "who made this particular industry?" In Europe this allows for some 20,000 years of uncertainty about when the actual colonization by H. sapiens happened: was it in the proto-Aurignacian c. 48-44 Ka? In the true Aurignacian c. 40 Ka? Or was it only with Gravettian c. 28 Ka?
In less well documented parts of the world the uncertainty is even greater.
...
"The model described in this paper proposes that the AMHS population movement out of Africa during MIS 5 was not a failed wave of expansion; rather, these carriers of the mtDNA L3 marker were bivouacked (so to speak) in Southwest and/or Central Asian refugia, such as the Persian Gulf Oasis, until the ameliorated conditions that set in during MIS 3 permitted subsequent range expansions".
ReplyDeleteThat would require at least 10,000 years of "bivouac" (what a word, had to look it up at the dictionary!) or even as much as 40,000 years if the L3 migration to West Asia happened c. 120 Ka.
Much more logical would be to think, it seems to me, that the migration to South Asia (and beyond) happened also at those dates more or less, as suggested by the blade industry of Patpara that Harrod considers the first evidence of H. sapiens (dated 100-150 Ka). However there is a lapse between that site and the next ones, which occur only soon before Toba.
We are still lacking some key evidence, I understand. But there is no real contradiction in the data, just in some of its interpretations. In fact the data cannot contradict itself: a dinosaur is a dinosaur, regardless of whether it had feathers or not.
Reminds me of the story of the blind men who tried to guess how was an elephant touching different parts of its body. You know that one, right? The trunk, the tail and the leg may seem contradictory but they are not in fact.
"'the immediate origins of the explosion of "modern behavior" seen in Europe, but not in Africa, might be found at the contact between Eastern Europe and Western Asia'".
That's a nonsense: the oldest evidence of the so-called "modern behavior" (a stupid arrogant concept, IMO) happens in Africa and Palestine, not in Europe. And also in India (cf. Harrod again): rock paintings with Acheulean dated to c. 100-106 Ka ago at Bhimbekta.
Then you have South Africa, c. 70 Ka. And only then Europe (oldest would be Neanderthal use of ochre c. 50 Ka in Murcia). But I call all this evidence of symbolic behavior, with emphasis on the word evidence, because we cannot expect to find evidence for music or dance or art made with/on perishable materials, such as leather, feathers, wood, the soil, etc.
One thing we have to accept is that, even in the best case, we will only have part of the evidence: bones and stones are important but can't tell it all.
I cannot access anymore the Bailey paper and I did not save it to my computer. Did anybody save a copy? If so please send me one to lialdamiz[AT]gmail[DOT]com, thanks in advance.
ReplyDeleteBy the way, Terry, one detail of note in the Rose (Persian Gulf Oasis) paper is that he mentions that the MP tech of the Persian Gulf is clearly different from that of Palestine (and intrudes at least in one site towards the Iranian Paleteau). They also mention a site as having African-related technology of the same age as the Skhul/Qahfez.
ReplyDeleteIn all this I see at least circumstantial evidence favoring a southern migration unrelated (or mostly unrelated) to the Palestinian offshoot. Petraglia also thinks that the migratory route went through Arabia even if he doesn't like the term "coastal migration".
However the Rose paper makes a strong case for whatever populations at the Persian Gulf forcibly being driven by increased aridity towards the now submerged coastal/lacustrine oasis in the MI4 period and later again in the MI2 hyper-arid period:
The transition to aquatic subsistence and “beachcombing” is often invoked to explain the rapid modern human expansion across the Indian Ocean rim (e.g. Stringer 2000; Field and Lahr 2006; Mellars 2006; Field et al. 2007). In this case, the Gulf Oasis model provides an environmentally driven mechanism that removed big-game hunting as a viable food source during MIS 4, forcing the adoption of aquatic subsistence as biomass and freshwater resources became concentrated on the exposed continental shelf.
Of course such coastal orientation may have already happened earlier in Africa, maybe also forced by arid periods. There is some evidence of coastal habitation and foraging strategies c. 150 Ka. ago in Eritrea, as you already know.
"What can I say: there's probably very few 'Catholics' as anti-Catholic (and generically anti-Yahvistic) as myself".
ReplyDeleteOnce a Catholic always a Catholoic, they say. Your view of species and evolution seems very much centred on Old Testament ideas about the creation of 'kinds'. You have several times claimed that evolution merely acts through drift, not selection. This belief seems to be a consequence of your fundamental belief in the fixity of species. You seem to believe that each individual species is absolutely separate from all others, and this has been so since the moment each particular species first appeared, or was created.
"If [1] Age(M)=60 Ka, then Age(L3)=66 Ka. But if [2] Age (M)=100 Ka, then Age(L3)=110 Ka. All very rough figures but the point is that only so much time can have occurred between L3 and M (and N too)".
I thought you were sceptical of the molecular clock? I tend to accept its validity over very long periods but I certainly become sceptical when it's applied over short periods. Although mutations in the chromosome may be reasonably regular there are many reasons why a mutated version of an mtDNA chromosome might or might not survive to replace its parent haplogroup. Pre-M or pre-N could have either stayed together in a single place, or could have each moved a huge distance, by the time any mutated versions survived to replace their parent hapolgroup. Likewise for the daughter chromosomes and their own expansions.
"the MP tech of the Persian Gulf is clearly different from that of Palestine"
And different from anything in Africa. Any time that the Hadramawt is going to be lush enough to support a migating human population, surely the Mediterranean coastal route would be equally lush, and capable of supporting a huamn migration.
"I cannot access anymore the Bailey paper and I did not save it to my computer. Did anybody save a copy?"
I got to it via your link. I think that probably still works.
Yes, it works:
ReplyDeletehttp://eprints.whiterose.ac.uk/10261/1/RedSea_Bailey_Pre-Pub.pdf
"Once a Catholic always a Catholoic, they say".
ReplyDeleteThat's absolutely ridiculous. People make choices in their lives. I presume you were also raised as Christian, yet you are not.
According to that, Luther and Calvin were also "always Catholics" (certainly much more than me, as they remained Christian and only reneged partly in their late lives). Things are not that way.
"Your view of species and evolution seems very much centred on Old Testament ideas"...
Catholics do not believe in the Old Testament, that's Protestants and Jews. Catholics believe what the Pope says, which right now is Darwin and the Big Bang. The NT also plays a role but the OT is considered more or less like the Illiad. Jesuits in particular, who are traditionally the "intellectual elite" of the Roman Church, advocate for teleology: the idea that evolution is subtly directed by "God" (de Chardain, etc.) Jesuits had absolutely no problem with the Theory of Evolution, Catholics in general have no problem with that nowadays (I know a couple of ultramontane creationists but they are plainly dumb, extremist fascists and very old). The Genesis is like the tooth mouse: a story without any real significance. Now if you question Jesus' miracles they may be more belligerant, but even in that the sensible ones emphasize the message rather than the magic.
You do not only insult me by accusing me of "Catholicism" but you also insult Catholics by accusing them of being Protestant in a Biblical-literalist sense. That's not that way: Catholics have their hierarchy and the hierarchs dictate what to believe, which is often totally unrelated to what the Bible says, specially the OT.
Besides that the most important element of Catholicism is Marianism, i.e. the veneration of Mary as a quasi-goddess, and also the saints, which take the place of ancient gods, much like in Santería. In fact, I'd say that together with Jesus and the fancy Holy Spirit, they leave almost no room for Yaveh ("God the Father") in all that mythology.
While the roots are Jewish, it's very much degenerate and only diffusely Yavihstic. Creation in the Genesis sense is irrelevant in their beliefs, just a mythological common place.
"You seem to believe that each individual species is absolutely separate from all others"...
ReplyDeleteInstead to try to interpret my "beliefs", well beyond what I actually say, why do not you illustrate us with your own beliefs. It's cheap to go to the throat of your debate opponent with made-up accusations like those. But helps nothing to the discussion.
I do not think that Homo species are totally impermeable to other Homo species but they are clearly distinct entities anyhow. You do not get an Homo sapiens beginning with a Neanderthal, much less in a few thousand years. Evolution needs time and is not convergent.
What I do believe (understand, think) is that multirregionalism is as dead as Dillinger. This regardless of minor admixture events.
"I thought you were sceptical of the molecular clock?"
I am. Not so much of the general principles but of the results. I distrust the models in use, largely because I understand that there is wide room for older chronologies and that the usual MC models have been more than once put in evidence when contrasted with the facts (for instance new world monkeys or chimp-bonobo divergence). It is not in any case a radiocarbon-like dating method with the precision and well demonstrated efficiency that is required. Not at all.
But the generic principle of proportionality should stand, with whatever corrections needed to factor in drift (favoring the stability of larger haplogroups) and other factors such as variable population size, different possible effective mutation rates, etc.
The method is IMO still very young and highly unreliable. Also different authors use different methods and they tend to pressure each other towards what can well be a consensus in the error instead of a well proven methodology.
I do not understand anything else you say in the paragraph on M and N. The case remains for the period of coalescence of these two clades being almost 10% of what happened before or what happened later. This is a clear and strong time constraint not really allowing for a too long "bivouac" in West Asia.
My best guess is a much older arrival to South and East Asia than usually considered. However archaeology will have to demonstrate this if true. There are some indications in favor of this but they are too weak by the moment.
"And different from anything in Africa".
They mention affinities with Africa in at least one site and they mention Nubian-style artifacts too.
"Any time that the Hadramawt is going to be lush enough to support a migating human population, surely the Mediterranean coastal route would be equally lush, and capable of supporting a huamn migration".
Probably yes but the African Mediterranean coast goes nowhere that at that time was not populated by Neanderthals, a clear competitor in equal terms. IMO it were Neanderthals who prevented a direct migration into most of West Eurasia, directing the flow towards the east along the Indian Ocean coasts.
"I got to it via your link. I think that probably still works".
It does not work for me anymore. Possibly some Linux-related odd glitch. The page appears but the Adobe plug-in cannot read it. Care to send me a copy?
"I presume you were also raised as Christian, yet you are not".
ReplyDeleteNo, I was not raised in any religion.
"why do not you illustrate us with your own beliefs".
I've summed up my own beliefs regarding the mechanism of evolution through comments at Dienekes blog:
http://dienekes.blogspot.com/2010/10/deep-ancestors-of-human-dna-compatible.html
Quote:
"The study actually supports what many geneticists have been trying to tell everyone for years. Each individual gene in a species has its own individual origin. Genes advantageous for the survival of the particular species move through that species like ripples through a pond. Of course individual genes are often carried through the species by the expansion of a subspecies, carrying the relevant gene as well as another particular combination of genes".
"You do not get an Homo sapiens beginning with a Neanderthal, much less in a few thousand years. Evolution needs time and is not convergent".
Evolution needs time and is not convergent? I'm sure you've claimed that the reason ancient humans show evolution in the direction of modern humans over a wide region of the world is not a result of gene flow, but of convergent evolution.
"My best guess is a much older arrival to South and East Asia than usually considered".
Why is that necessarily more likely than a long 'bivouac' in West Asia before entry to South Asia?
"There are some indications in favor of this but they are too weak by the moment".
I agree that the case for Y-hap F and mtDNA is reasonable. But even in those examples the haplogroups could have entered South Asia already diversified to some extent.
"the African Mediterranean coast goes nowhere that at that time was not populated by Neanderthals"
I was under the impression that Neanderthals didn't reach the Levant, or even SW Asia, until the climate cooling of around 70,000 years ago.
"Care to send me a copy?"
Should I email it to Tim? Did you try this link:
http://eprints.whiterose.ac.uk/10261/1/RedSea_Bailey_Pre-Pub.pdf
IDK, the link is exactly the same one as in the article, yet one worked and the other did not (says "file does not start with %PDF").
ReplyDeleteConfusing, really. (And anyhow, I'm sure I listed my email two posts ago, which is also available in my profile - Tim is in London now and there's no reason anyhow to bother him with that - but not needed anymore, thanks).
"I'm sure you've claimed that the reason ancient humans show evolution in the direction of modern humans over a wide region of the world is not a result of gene flow, but of convergent evolution".
ReplyDeleteParallel evolution, aka convergent evolution towards similar traits by different evolutionary pathways, is not the same as converging towards a single species. Pinguins and seals, for instance, show parallel evolution but they will never ever converge. In fact they tend to diverge into many different subgroups, as do all other species. Occasional horizontal transfer of genes between "imperfectly" separated species does not mean convergence.
But well, I see that you are an old school multirregionalist. It took me a while to understand. Sorry but that hypothesis is nonsense - regardless of minor admixture episodes. You don't get Neanderthals evolving into Sapiens overnight, and if that would happen, then there would not be single haploid ancestors so close to present in Africa but we'd have something like what was apparently spotted (will we see more research now?) in X-DNA: one lineage very much apart from the rest at low frequencies but also concentrated in some populations particularly, those that directly would descend from those miraculously transformed Neanders or H. erectus.
We do not see that. Even if X-DNA haplotype B006 can be confirmed as non-African, it does not show magical transformation of one species into another but, again, residual absorption of some minor genetics of one species by the other.
IDK it's like saying that, just because some genes occasionally are transferred between, say, leopards and lions, that means that you can get a lion from a leopard. Nope, it's not that way. Nor you get an mammoth from an Indian elephant, nor a dinosaur from a chicken.
Let's see:
ReplyDeleteFrom the mtDNA Root to L3 there are, I count now, 23 C.R. mutations.
From L3 to M there are only 3 mutations, to N it is 5 (not 4, as I wrongly said before).
From M and N to the last large nodes (K1a1, T2b) there are 19/18 mutations, more to present day lineages (private or shared).
While you can, I presume, introduce some corrections to account for population dynamics, there should still be a proportionality.
So, if the L3(pre-M and pre-N) bivouac in West Asia lasted 50 Ka, mtDNA Eve becomes something like 383,000 years older than L3, placing the origin of shared human mtDNA lineages c. 500,000 years ago, well beyond what anybody I have read, including myself (I tend to push ages towards the long end, more than most authors), accepts.
After comparing with Neanderthal and Denisova mtDNA, we'd be talking of millions of years of Homo sp. divergence. Neanderthals would have diverged more than 2 million years ago and Denisova (H. erectus?) maybe 5 million years ago. This does not fit the archaeological nor palaeontological record at all. Five million years ago there was simply no Homo genus at all.
If we do the same apportioning towards the present, all currently existing low level haplogroups just could not exist, would only appear many thousands of years into the future.
You'd need such extreme "corrections" in all parts of the tree that it makes no sense whatsoever.
In other words: the "bivouac" must have been much shorter. Or the peoples of the Arabian "bivouac" were not the same ones migrating to South Asia (and beyond), what again would place us in the 'rapid coastal migration' scenario in its purest form.
"I agree that the case for Y-hap F and mtDNA is reasonable. But even in those examples the haplogroups could have entered South Asia already diversified to some extent".
I've been working mostly with mtDNA, which I consider more reliable and easier to manage for an amateur like me (no need to use extremely large amounts of microsatellite data) and the large star-like structures of M (very specially) and of some other early haplogroups (N, M4"64, M30, R) can only be explained as signature of very rapid demographic expansion. This expansion does not correspond with the limited fertility we infer from the parts of West Asia occupied apparently by our species (Arabia Peninsula mostly). The geographic spread and diversity levels also talks of a South or SE Asian origin, not Arabian or West Asian.
In West Asia, specially Arabia Peninsula, we see old L(xM,N) lineages, specially in the L0 branch but also a few others (L3i, L4a, L6), which are much smaller - and then, of course, back-migrated M or N sublineages. But no signature of the M or N roots, which are clearly located farther East.
All I do with Y-DNA nowadays is comparing its patterns with those of mtDNA and finding some reasonable parallels, what give me some nice clues on when (relative concept of "when") each Y-DNA lineage may have coalesced.
"I was under the impression that Neanderthals didn't reach the Levant, or even SW Asia, until the climate cooling of around 70,000 years ago".
You are right per the skeletal fossils (though in some cases at least the uncertainty of dates goes beyond 100 Ka ago). Excellent point, thanks. I stand corrected in this aspect.
However there are no H. sapiens sites either north or east of Palestine. So maybe it was just the cold climate what kept them out?
OK, I've just finishing re-reading Bailey's paper and all in it is suggestive of the importance of coastal areas around the Red Sea. However for lack of archaeological-specific data he may result inconclusive. Whatever the case it provides a good frame for understanding the possible Red Sea crossing.
ReplyDeleteA detail of some potential importance is the suggestion that at least sea levels under the -50 m. mark would be "required" for an easy crossing of the water body with the limited raft/boat capabilities presumed for those early members of our species. Assuming this is correct, it would imply that crossings could not happen BEFORE 90 Ka., what makes the Arabian/coastal route less likely to be contemporary with the settlement of Palestine c. 120 Ka. ago.
However I do not see this reasoning as strong enough: while these lowest sea levels would have facilitated the crossing, there is no absolute need for them because we know nothing for sure of the boating abilities of peoples who were clearly exploiting coastal areas since at least 150 Ka. ago and that later managed to cross the much wider water bodies of Wallacea.
The case for the exact route remains open, what is clear is that there are no remains of H. sapiens nor related industries North of the Galilee-South Iran line. So coastal or not (or something in between), it was a very southernly, tropical flow, it seems.
Good find by the person who made these 2 papers known to you.
ReplyDeleteI've linked them at our forum:
http://s6.zetaboards.com/man/topic/8684908/1/
I love you Maju!
I see how much you "love" me, ahem:
ReplyDelete"Maju was alerted to it by another person, so, as usual, he never finds anything interesting but just recyles others' ideas (like how has recyled all my ideas and manage to impress his followers)".
Enfin... keep that insulting tone up, it makes me laugh.
"But Maju is no Dienekes".
Nor I want to be, so thanks... I guess. I pass of being homophobic, sexist, racialist bordering racist, ethnocentric and conservative bordering fascism. And I do not have money to subscribe to pay-per-view journals.
"just recyles others' ideas"
ReplyDeleteWhat's wrong with that? That is the advantage of the Internet: sharing information.
"So, if the L3(pre-M and pre-N) bivouac in West Asia lasted 50 Ka"
Who has said anything about 50,000 years? Humans can travel huge distances in 5000 years. To return to something you wrote the other day:
"I do not understand anything else you say in the paragraph on M and N. The case remains for the period of coalescence of these two clades being almost 10% of what happened before or what happened later. This is a clear and strong time constraint not really allowing for a too long 'bivouac' in West Asia".
You have several times mentioned the three or four mutations between mtDNA L3* and the respective roots of haplogroups M and N. It's extremely unlikely that these three or four mutations all occurred simultaneously. So that gives us an L3preM* haplotype (1 mutation) and an L3preN*. I have in the past referred to these haplotypes as M and N as shorthand, but I see why that confuses you.
Today no haplogroups survive having only that first mutation. Haplotypes L3preM* and L3preN* have each left just one surviving descendant haplogroup. And those descendants have replaced any other offspring lines they may have had.
We next have haplotypes L3preM** and L3preN**. Again only one descendant line each. Any other descendant lines have become extinct. Then L3M*** and L3preN***. Same again. Who else was present in their communities?
This period of bottleneck, or period of restricted haplotype diversity, indicates a population confined to some region of limited extent, rather than any extinction of most individuals in the population. And almost certainly L3preM** lived in one region and L3preN** in another, otherwise one would have replaced the other just as their daughters' lines have eliminated them. But by no means need either have lived in India at that stage.
It is not until the haplotype L3preM**** had appeared that we find a huge expansion through India. To me this indicates the arrival of a new haplogroup, exploiting the environment in some new way. It is certainly not evidence for a speedy arrival from Africa.
What about L3preN****? Haplogroup N actually has two expansions: N* and R*. You appear to believe that these two expansions are actually the same expansion. Correct? Surely the various N* haplotypes had already finished up in separate regions by the time R* expanded. Where, again, they became isolated into populations where daughter haplotypes replaced parent haplotypes. Haplotype A remained in just such a small region for a very long time. I think seven mutations from N*, with no surviving daughter haplotypes, but one. So we have NpreA******* before the haplogroup managed to emerge from where-ever it had been hiding. India? Expanding with N* and R*?
"Who has said anything about 50,000 years?"
ReplyDeleteIf I understand correctly, the hypothesis of the "bivouac" means that H. sapiens arrived to West Asia c. 120 Ka. and to South Asia only c. 80-75 Ka. It's 40-45 thousand years.
My estimates (mentioned before in this thread), based on the genetic distance between L3 and M, suggest rather 10-15,000 years (for the 120 Ka date of arrival to West Asia, less if later). The "bivouac" means 3-4 times what I consider reasonable. This could be solved if effectively people arrived to South Asia c. 110-100 Ka, as Harrod suggests based on the existance of blade tech (mode 4) at Patpara (Son valley). He also suggests the Narmada-Son, instead of the coastal route, but it's a mere note with a question mark.
Btw, I have added the link to his Synopsis of Paleo-India PDF to the article because it's coming up all the time.
"You have several times mentioned the three or four mutations between mtDNA L3* and the respective roots of haplogroups M and N".
I corrected myself above: it's 3 for M and 5 for N. Or 4 and 5 if we consider the only HVS mutation involved at the root of M. For the record.
"So that gives us an L3preM* haplotype (1 mutation) and an L3preN*. I have in the past referred to these haplotypes as M and N as shorthand, but I see why that confuses you".
Yes, please; they were not yet M nor N. Pre-M or L3(M) and pre-N or L3(N), or something like that makes things much more clear.
"Who else was present in their communities?"
Probably they tended relatively rapid to drift towards a single dominant or a few dominant haplogroups. However as the Arabia Peninsula population was surely made up of several more or less distinct groups, it's possible that other L(xM,N) survivals we see in that area are also descendants of the original settlers. If so, that would be three L0 clades and three L3'4'6 clades: L3i, L4a and L6 as a whole possibly. There could also be others that have vanished in genetic drift; the presence of one L2 case in Malaysia Negritos offers an interesting extra possible OoA clade, which might be related to the controversial detection of L(xM,N) in one Australian aDNA from very early.
It's not impossible at all that there were other haplogroups, surely at low frequencies, that have eventually vanished (drift works that way). In fact it's probable that they existed but somewhat anecdotal rather than informative.
"This period of bottleneck, or period of restricted haplotype diversity, indicates a population confined to some region of limited extent, rather than any extinction of most individuals in the population".
Yes but we also talk of a very small population in comparison with what there was in Africa, right? So how much is "bottleneck" (drift) and how much is simple founder effect (small number of founders, always in proportion to the original African macro-population)? I'd say that both and that we cannot easily discern what is which.
But from that already restricted genetic pool, an even smaller group was the one that finally made it to South Asia. It was still large enough to carry several lineages (2, at least, in the mtDNA line, 3 in the Y-DNA one), but less diverse than the Arabian ancestral group and much less diverse than the African one, of course.
"And almost certainly L3preM** lived in one region and L3preN** in another, otherwise one would have replaced the other just as their daughters' lines have eliminated them".
Maybe. It's not an exact science. Maybe the two groups lived just 100-200 Km away, maybe more, maybe less. Maybe it was a plural group, formed by remnants of several neighboring communities. I just can shrug unless we get better evidence, either archaeological or aDNA or both.
...
...
ReplyDelete"But by no means need either have lived in India at that stage".
No. Probably M arrived at South Asia fresh (i.e. already with the three mutations downstream of L3 that define the clade) and pre-N arrived still at intermediate stage: pre-N*** or pre-N****, to use your terminology. Pre-N may have arrived to South Asia when the expansion of M was already strong because it seems it found no room to grow there.
And that would be a reason to think that the small pre-N clan could have got a fast-track ability, such as coastal specialization with boating, so it managed to go across M lands (South Asia) without leaving trace and arriving to SE Asia early enough as to leave a stronger mark there than any M clan (though not than M as a whole). It was a "tight race" where M had the advantage of an slightly earlier start (the pool position) and hence much larger numbers soon after, while pre-N had some "speed hack" that allowed it to take some advantage East of India.
This advantage may or not be related to whatever advantage the derived R clan had at a later moment. We are speculating on thin air after all and hence we can only reach so far.
"It is not until the haplotype L3preM**** had appeared that we find a huge expansion through India. To me this indicates the arrival of a new haplogroup, exploiting the environment in some new way. It is certainly not evidence for a speedy arrival from Africa".
IMO the M clan (not anymore pre-M) arrived first to South Asia, so it had suddenly all the room it wanted for a very rapid expansion, the details of which may be inferred possibly by interpreting the many M subclades, some large and some small.
...
...
ReplyDelete"Haplogroup N actually has two expansions: N* and R*. You appear to believe that these two expansions are actually the same expansion. Correct?"
No, that's not what I think. They may be related but the geography and time of each are different: N expanded in SE Asia most likely, with some subclades heading west (N1'5, N2, X and R - add "pre-" as needed). Maybe it was just one N subgroup that branched out in South Asia without leaving any trace because there was no time for a mutation to accumulate (i.e. a very rapid back-migration of a small group, which would be consistent with a spread from the Gulf of Bengal rather than farther East).
Of course there's always the possibility that the inverse is what happened: the N clan, upon arrival to South Asia from Arabia (after M had already begun its expansion) found some minor room in South Asia and the nearby West Asian frontier but also migrated Eastwards fast enough (and in small enough numbers) as for a mutation not having time to accumulate. In this case we'd have two unattested N nodes: one in South Asia (N-west) and another in SE Asia (N-east).
This is not impossible but the genetics just say nothing about it. This could happen for instance if the migration happened in a single millennium or less (or not much more), assuming something like 3 Ka. for mutation (as I did in the past and fitted reasonably well). It's perfectly possible but genetics is mute about that. The only thing we can attest is that South and SE Asia were clearly dynamically linked in that early period, a pattern that later disappeared, with both regions becoming distinct, surely as result of almost total occupation.
"Surely the various N* haplotypes had already finished up in separate regions by the time R* expanded".
That we cannot attest clearly either. After all R is just one CR mutation (two if we count HVS) downstream of N. In order to make a systematic reconstruction I conveniently suppose that all mutations happened simultaneously as if they were ticks of a "clock". But obviously this is a simplification, convenient but totally uncertain. If there are several mutations we can think that the estimate is correct because the randomness of each one should balance the others but one mutation alone can happen at any moment, so we cannot really know when the stems are so short. We enter the indeterminate zone. R could have happened any time from the daughter or granddaughter of the first N carrier to maybe 5000 years later (or even more).
"Haplotype A remained in just such a small region for a very long time".
I was imagining you'd end up in A. MtDNA A had a long coalescence, IMO as a private (minor) lineage in some larger clan dominated by some other haplogroup/s. At some point it found a niche for expansion and that happened in NE Asia. Its expansion seems simultaneous to those of M8a, Z and Y (also F but this one seems geographically unrelated), so I'd guess that pre-A migrated northwards with M8 and N9 (and after D did).
"IMO the M clan (not anymore pre-M) arrived first to South Asia, so it had suddenly all the room it wanted for a very rapid expansion"
ReplyDeleteThat is exactly my reasoning too. Star-like expansions tend to indicate recent arrival in previously unexploited regions, or the development of new technology or culture that leads to the same effect.
"MtDNA A had a long coalescence, IMO as a private (minor) lineage in some larger clan dominated by some other haplogroup/s. At some point it found a niche for expansion and that happened in NE Asia".
It's very difficult to imagine such a scenario. Minor haplotypes tend to have just two options: expand if have some advantage or be rapidly drifted out.
"Pre-N may have arrived to South Asia when the expansion of M was already strong because it seems it found no room to grow there".
It seems very stange, then, that it managed to move through India to emerge in SE Asia, even taking into considerstion your idea of 'a fast-track ability, such as coastal specialization with boating, so it managed to go across M lands (South Asia) without leaving trace and arriving to SE Asia early enough as to leave a stronger mark there than any M clan'.
"Another issue is the known length of downstream branches, even in some well studied lineages, which appear almost 'frozen' since their expansion. This seems to happen in particular to large star-like lineages like M and H (not sure why) but, in any case, the high variability in the length of the lineages towards the present is an anomaly that I would rather not have to face".
The 'frozen' lineages survive as a result of the population expansion, and consequent lack of 'drift', or 'bottleneck', or whatever you wish to call it. A few days ago you wrote:
"The case remains for the period of coalescence of these two clades being almost 10% of what happened before or what happened later".
I'm sure you already know the following, but I'll comment anyway. The trouble with the molecular clock does not just concern the dating of the separate mutations. It's with the idea that the clock is regular in the first place.
We need to consider mitochondrial DNA itself. A group of women does not suddenly all possess a newly mutated version of their mtDNA. The mutated version has first to happen in an egg cell, presumably when it is first formed. Females probably usually have many egg cells with mutations in the mitochondria, but for the mutation to survive at least two things need to happen. First of all the egg must be fertilised, then it must be carried by a mature female who in turn produces offspring who survive to produce more offspring etc. That's before the mutation can even begin to take over and replace its parent haplotype.
Consider a village of, say, 10 women, all with the same mitochondrial DNA. It doesn't take many generations for a village of such a size to contain just one haplotype. Suppose one of these 10 women produces a daughter with a mutated mtDNA. That will result in just one woman in the next generation who has the new mtDNA. It could take many generations for this mutation to replace the parent haplotype in the village. So there's actually no reason at all why a non-mutated mitochondrial DNA could not survive unchanged almost indefinitely (as you've mentioned for M and H), let alone be regularly replaced by a mutated daughter version. That is the problem with the molecular clock idea, and why we can't assume a regular replacement of mtDNA haplotypes. Over the long period the changes average out, but over just a few mutations the molecular clock is almost meaningless.
"Minor haplotypes tend to have just two options: expand if have some advantage or be rapidly drifted out".
ReplyDeleteNo. There are minor haplogroups surviving from very deep roots to present. While that you say is the tendency of drift, the final result is not guaranteed to happen in any particular term or conditions. Minor lineages do survive, though they are probably just a tiny fraction of all those that once existed.
Whatever the case A did exactly that, unless you think it was hibernating in some igloo like a trans-millenial polar bear of sorts, while mtDNA D4 expanded instead in that same area.
Besides, A is the only N lineage that seems to exist have a norther expansion center, not even X does that.
"It seems very stange, then, that [mtDNA pre-N] managed to move through India to emerge in SE Asia, even taking into considerstion your idea of 'a fast-track ability, such as coastal specialization with boating".
All other explanations are even stranger.
"so it managed to go across M lands (South Asia) without leaving trace and arriving to SE Asia early enough as to leave a stronger mark there than any M clan'".
Well, what I tried to explore in the previous post is that, if N moved swiftly enough between South Asia and SE Asia (and/or vice versa) in the time of some 25-250 generations (maybe up to double that if in the MI5 time-frame), it could perfectly have branched itself in two or more real people's lineages without leaving any mutation track, while still having several actual "matriarchs" radiating at different locations such as Thailand and Punjab at the same time (or if you prefer, Sindh, Guangxi and Australia all at the same time). But mtDNA A would still be a belated "flower" of the Guangxi or Thailand node in any case, as it's obviously late in comparison to most others.
If so, it would have left a clear, yet secondary trace in South/West Asia. There are at least those two possibilities:
(a) leaves no trace in SA or anywhere else before arrival to SEA, what almost demands the "rapid coastal migration" or some other "odd" explanation like a Toba-caused 'bottleneck'. This demands in turn a rapid back-migration too, in order to explain R, N1'5 and X, which hang from N from very short stems.
(b) it does leave track but it's impossible to describe because mtDNA mutations take long enough to accumulate, so most of it is R and the other three WEA N sub-haplogroups (two of which are also SA).
...
"The 'frozen' lineages survive as a result of the population expansion, and consequent lack of 'drift', or 'bottleneck', or whatever you wish to call it".
ReplyDeleteMakes sense. But they are still not obvious participant of the expansion as such until they "melt".
All you say regarding the relativity of the molecular clock is totally correct. In fact it's part of my reasoning for option (b) in the previous post: if migration (but not so much expansion) happened fast enough, we won't see its traces. And fast enough can be up to several millennia, depending on actual time-frame and random events we can't measure nor evaluate in any way.
"Over the long period the changes average out, but over just a few mutations the molecular clock is almost meaningless".
Yes. Almost my words a few posts above. But if N expanded very dynamically without much movement, we'd see its track anyhow (as we do with M). In fact we see some of that (not a super-star but still a star of 12 branches) but we see the branches scattered all around, what means that it the time of the N matriarchs there was a lot of movement across tropical Eurasia, from WEA/SA to Australia (in whichever direction or even all them).
And I don't think you can expand the time-frame of an evolutionary step such as N-root forever. Specially not when it shows such clear sign of expansion (many "villages", not just one anymore). This constrains the maximum time-frame of N-root to exist without further downstream mutations to a few millennia. My average for a 70-60 Ka Eurasian expansion was of some 2.5-3 Ka per mutation, for an earlier process could be double but it does not allow for N-root to have existed as such for more than 10Ka in the most extreme scenario. IMO it was surely much faster and could well be a matter of just a few centuries or 1-2 millennia.
IMO again, it means that the SA-SEA-Australia flow at the N stage probably happened in some 5000 years or less (up to a few centuries only). This is "rapid migration" - be it with boats, UFOs or marathon runners. N seems to represent a particularly mobile group. And R does too, which is a good reason to consider they managed to belong to the same highly nomadic kind of culture even after divergence, regardless that they do indeed show some sort of non-synchrony and different central geography too. Compared with them no M (or even other N subclade, maybe with the partial exception of X) looks so dynamic and specially mobile (there are some M-derived medium-sized stars but none so scattered across the whole Eurasian continent).
So we can consider the M clan as relatively "sedentary" in comparison with N and R. Nothing absolute, just a matter of degree, of course.
Terry: I decided to make a quick re-review of the quite apparent patterns of spread of mtDNA M, N and R, focusing on those sub-lineages that show very early signs of expansion (i.e. are only one CR mutation under their respective macro-haplogroup's core node). The resulting patterns are most informative, really. LINK.
ReplyDeleteAnd Ren, btw, if you want to see some original Maju(TM) materials, the link I just posted is a good place to start. Feel free to criticize constructively.
ReplyDelete"N seems to represent a particularly mobile group".
ReplyDeleteA preference for an open savanah habitat?
"And R does too"
Possession of the boating technology that necessarily was developed before the crossing to Sahul?
Just speculation, of course.
"A preference for an open savanah habitat?"
ReplyDeleteSavannah people would have problems crossing the many major rivers, swamps and deltas in this journey, not to mention Wallacea. Boating coastal dwellers ('sea gypsies') are a reasonable alternative idea, to say the least.
Still, M also looks very mobile, but with a preference for the East, probably sing of a somewhat earlier expansion. I would think that these mobile lifestyles were not exclusive of any lineage, though is obvious that some groups were much more restless than others.
"Possession of the boating technology that necessarily was developed before the crossing to Sahul?"
I think so. The Wallacea crossing is no easy journey, it implies really experienced boaters, whatever the details of the technology at hand. Considering the speed and repetition of the expansion to Sahul, many different peoples (in the M, N and R macro-clades) had access very early to that boating tech and the related skills. You can of course imagine that the ultimate perfecting of their techniques happened in the Wallacean islands but the overall concept seems older and must date from at least the Arabian (Red Sea, Persian Gulf) period (probably older - remember the Cretan case).
"You can of course imagine that the ultimate perfecting of their techniques happened in the Wallacean islands"
ReplyDeleteExactly what I was getting at. My comment was spefically aimed at your comment about mtDNA R. Their rapid expansion may have been a result of that perfecting. They were able to move rapidly back through the SE Asian islands, up the Ganges and out into SW Asia, and so to Europe. This seems to be very likely, in the light of mtDNA haplogroup P.
Well, what I see is that mtDNA P, like S or M29'Q, reflect that the naval feat was possible for all three early Eurasian "clans". You may I guess interpret them as all happening all at the same time but my reconstruction rather suggests an earlier arrival of the M subclade M29'Q, followed in due time (maybe 6000 years later - two "molecular ticks") by the N subclade S and the M subclade M27 and then (one "tick" later) by R-derived P and N-derived O.
ReplyDeleteSo it looks more like all them had boats all the time, regardless that their use became more intensive in Wallacea.
I don't think that there is any reason to think of a back-migration from Wallacea, though some back-migration from mainland SE Asia (then including Sundaland) is probable - specially on light of the Y-DNA MNOPS subhaplogroup P, which must have coalesced in Pakistan or surroundings.
More intriguingly, when you consider only the earliest offshoots, none of the three macro-haplogroups looks particularly slanted towards the East. N still looks more Eastern-looking than M and R. They have the following ratios in fact:
M: W=0.61 E=0.39
N: W=0.50 E=0.50
R: W=0.63 E=0.37
[By W (West) I mean South Asia and West Eurasia, E (East) means all Eastern Asia and Near Oceania].
So yes, N looks more Eastern oriented since its beginnings but only somewhat. However the coastal habits seem to have existed in all the major "clans" without any clear favoritism. M in fact shows a very early dynamism, not just in Near Oceania but also along the Eastern coast of Asia (which is admittedly less radically conclusive of the use of boats or coastal habits but very suggestive because of the early presence in Japan, apparently). The M clan looks in fact very pioneering in Eastern Eurasia overall in an expansion that looks coastal to me, and in the Western frontier is also coastal (M1).
I don't think there is any strong evidence suggesting a special advantage of N and R. They may have got some but nothing clear in the genetic data setting them apart from M. In the Y-DNA alternative view, mtDNA M and probably N or part of it, must have expanded with Y-DNA C (and D) in the East, but possibly with F in the West. The likely mtDNA R plus Y-DNA MNOPS in East Asia seems to indicate the only clearly defined secondary expansion. But it's important to realize that it's secondary and also in Melanesia surely too: the second (or third?) and not the first wave.