Yeah. The group isn’t any ‘older’ just because we think they’ve been hanging around the same place for the same time that the ancestors of the Pacific Islanders were traveling.
In related news, researchers today released an analysis of the genetic intermingling of 95 population groups. Here is one article about it.
No racism implied, folks. Just trying to describe a (mostly remembered) article in a now-lost issue of the magazine. The article describes how the various ethnic groups spread across the globe over the millennia.
But South Pacific Islanders didn’t spread to their location very recently at all. Europeans arrived on all those islands far later. So why wouldn’t the Europeans be the most recent?
I remember the article stated quite specifically that human presence in the South Pacific occurred only within the past 2500 years.
Yes, and European presence occured in the South Pacific only within the past 250 years. So why aren’t Europeans the youngest group?
The thing I have always found fascinating about the Hadza and the Khoisan people is that in addition to being so genetically far apart*, they both have click languages. Strongly suggesting that the mother tongue of all humans had clicks!
*As the above genetic study shows, both groups have intermingled with Bantus in recent millennia, so they are becoming more closed related via that.
How does it suggest that? Clicks are no more genetic than any other sounds of human speech. Given that all the evidence we have about language suggests that sounds change over time, I don’t see any positive evidence that they have had clicks ab initio. They might equally have arisen at any point in the past.
In cladistics, when two primitive groups share a feature that is not shared with any derived groups, we assume the feature is itself primitive. That’s because any alternative explanation requires that the feature has evolved twice, coincidentally both times in primitive groups. That’s much less likely than the trait having been preserved alongside other primitive traits.
This is the reason why we assume that, for example, the earliest modern birds were ground dwelling omnivores with precocial young. It is possible that all the primitive bird groups (the tinamous, rattites, waterfowl, gamebirds etc) evolved those traits independently. It’s possible that the ancestors of ostriches were aerial insectivores and the ancestors of chickens were diving seabirds. But it’s much more likely that they are both ground dwelling omnivores with precocial young because they both evolved form ground dwelling omnivores with precocial young and they haven’t changed those feature just as they haven’t changed many other features.
And ditto for language. It’s possible that the Hadaza and San independently evolved a click language from completely distinct non-click languages, and it’s entirely co-incidental that this evolution happened to occur in two groups that retained a whole suite of other genetically, culturally and phenoptypically primitive traits. But that is much less likely than that the two groups retained a primitive language alongside a whole lot of other primitive traits.
It’s basically Ockham’s Razor at work.
You have to be careful applying this postulate in the real world, because cause and effect can run either way. It’s possible, for example, that a click language is especially useful for hunter-gatherers who live in the midst of hostile agriculturalists. In which case it may well be more likely that it did evolve independently in these two groups. But without evidence that this is the case, the default assumption that it is itself a primitive trait.
Here you are incorrect. Historical linguistics has to account for borrowing in a way that biology does not, and individual phonemes are readily borrowed from one group to another. Just look at the sounds of French compared to Italian and Spanish on the one hand, and Dutch and German on the other. The cladistic model works best in language with complex features that are unlikely to be borrowed, such as the s-preterite or the dual number.
And here you are incorrect.
Firstly the exact same effects apply in biology. We don’t look for other reasons why Red Wolves share so many Grey Wolf traits, for example. We accept that it is because those traits are introgressions from Grey Wolves. But we do that because we know that grey wolves and red wolves are physically able to share traits.
Likewise with language, we accept that French shares sounds with it neighbours because they *are *neighbours. If French shared sounds with an Aboriginal or Argentinian language, that would be something that requires explanation. When two languages that are not in contact share features, as Hadza and San languages, that requires an explanation. It can not be explained, as you seem to be suggesting, by borrowing between the two groups because they are not in contact.
I saw a fascinating chart about a year or so ago that showed the admixture rates of various Europeans. The Finns, well-known for their linguistic separateness, were also very genetically distinct from their geographic neighbors. The only ones they had even the* slightest* bit of admixture with were the Swedes. All the rest of Europe genetically looks like a gradient, with populations slowly blending and fading into their neighbors, and then… there’s Finland, happy to keep its DNA to itself.
I found it amusing that, according to that study, the Finns hadn’t mixed with their other neighbors, the Russians, to any significant extent. I get that for political and religious reasons the Russians and Finns were unlikely to intermarry, but apparently there was next-to-no Finn-on-Russian or Russian-on-Finn sex going on! I imagine thousands of years of Russians and Finns sizing one another up from across their border and both parties simultaneously going “Ugh, no way.”
(I don’t think that chart included any Sami or Estonian samples. I imagine even the cousin-loving Finns have shared a little bit of DNA with those groups at some point).
Agnatic descendants of Vladimir II, Grand Prince of Kiev (1053-1125) have been tested and show DNA Y-chromosome haplogroup N1c1 – the same haplogroup as Finns annd Saami have. This suggests that Rurik, the famous Varangian Viking “invited” to rule Russia, may have been of Finnish rather than Germanic stock.
However, there are also many living agnatic descendants of Vladimir’s uncle, Swjatoslaw (1027-1076), and these do not have the N1c1 chromosome. Apparently some of the Rurikids are not real Rurikids!
Are you not ignoring known interbreeding between ancient humans and Neanderthals for the Norwegians? That would instantly break the link.
And is there any evidence of interbreeding between ancient humans and Homo Erectus?
No it wouldn’t. If my Aunt had married a Neanderthal (and I think she did :D), my siblings and I would *still *be more closely related to my cousins than I am you you. My siblings and I would still share ~25% of our genome with our cousins and we would still share less than 25% with you. And we would still share a common ancestor two generations removed, and a common ancestor more than two generations removed with you.
When comparing two groups with common ancestry to an outgroup, the degree of relatedness can’t affected by introgression by even *more *distantly related outgroups.
No, but probably only because we haven’t been able to suitable DNA samples from Erectus fossils.
Many paleontolgists accept that we are *directly * descended from erectus, in which case we must have interbred. But even if we accept some sort of intermediate species, that intermediate would have been so closely related to erectus that it would be almost unbelievable if it didn’t occur.
(Bolding by CP, for emphasis)
The “last common ancestor” is one way to look at the notion of “relatedness,” but I’m not sure I’d word the Norwegian example exactly that way. I realize you understand the difference, but I just want to clarify that the idea is that any given Norwegian might have some fraction of their geneset in common with an ancestor, and not that any given Norwegian might have all of their geneset in common…
As you point out (I think), there’s a difference in the concept of the last ancestor who contained all the original genes belonging to the pool of the current individual, and a common ancestor in the sense of tracing a lineage back to an ancestor who may have contributed part of the pool of the current individual.
(For the moment let’s discard new variations of genes in descendant pools.)
Because migration gates (as you mention) open and close with climate patterns and geographical limitations (not to mention some groups keep to themselves for other reasons), it’s tricky to deduce a most recent common ancestor. And of course, a common ancestor of some part of your geneset is not necessarily a good measure about how related you are. But neither is separation in time necessarily a direct correlate of how unrelated you are. If you’re in a population that split off 50,000 years ago but had a very small number of founders, you might be surprisingly closely related to the population off from which you split ( sorry Winnie) if your peeps have not indulged themselves with foreigners for whatever reason. Other populations may have partied quite a bit more, even to the point of dallying with lineages (Neandertals or Denisovians, e.g.) yet more ancient than the (mtDNA) L0 groups who founded anatomically modern humans. Assuming Pääbo is right, that would push the common ancestor for the full Eurasian gene pool (but not their MRCA) back a few hundred thousand years til you reach the common ancestor for L0 and Neandertals.
Anywhere, I just popped in to suggest that if one does want to use the Most Recent Common Ancestor as one way to look at “relatedness,” Doug Rohde’s mathematical modeling is worth reading. (Reminder to the non-Blake group less familiar with the topic: the MRCA is not a suggestion of how much genetic material you got from Great Great Granpa Sanji; it’s a way of saying any two individuals might be able to trace a lineage back to GGG Sanji. The actual amount of genes they got from him could be close to zero.)
When Rohde makes educated (but still WAG) modeling that tries to include actual patterns of migration and intermingling, among the (mathematically simulated) groups least likely to contain ancestors to each other are a central african sim and a south american sim–i.e. a central african’s probability of having a south american ancestor is essentially zero (P 27, top, in that link). These are just mathematical models, but I think they do shed an interesting perspective on the topic. If we focus on ancestral and not genetic inheritance, we can bring the MRCA forward to as close as 15,000 years or so (probably way too close to comfort for some SDMB members when they are distancing themselves from me 
).
Click languages are extremely rare. The only other group (outside of those mentioned and their neighbors) I’ve heard of is an Australian tribe where it is used by the males for ritual purposes. Not for hunting or other public use.
If there were an intrinsic usefulness for click languages that would cause them to arise independently, you would expect them to be substantially more common. OTOH, clicks are an easy to produce sound in the early stages of language evolution when more subtle sounds are not yet feasible.
Just FYI, latest researchshows that the Khoisan haven’t been as genetically-isolated as all that…
No, you wouldn’t. I don’t know why linguistics seems to attract more amateur woo than other fields, but we can at least try to keep it out of GQ. Consonant genesis and loss are not caused by a sound’s “intrinsic usefulness”.
There are a limited number of consonants produced by the human mouth. Some consonants are just plain rare. Others are common. When the rare ones spontaneously arise in unrelated languages, that suggests…nothing. It would be weird for these sounds to not pop up somewhere else. There are multiple sounds that are rarer than click consonants that appear in unrelated languages across the globe, yet no one is suggesting anything about primitive language.
And why are clicks supposedly easier for early languages? What “subtle sounds” do you speak of? Perhaps the ones that babies make? And the ones that show up in nearly every language?
Blake wants to talk about Ockham’s Razor. We could have a set of extremely lossy sounds (clicks are frequently dropped after click words are borrowed) that have been, without precedent, conserved without language shift for tens of thousands of years.
Or, they could have borrowed consonants from a now-defunct neighboring language.
Or, their ancestors’ languages could have spontaneously generated rare consonants.
Yes, let us strop that razor and cut away.
And who knows? Maybe I’ll find out that Sindhi and Mayan have retained primitive implosive consonants, and that the first human languages had linguolabial consonants, as evidenced by the shared features of some languages of Guinea-Bissau and Vanuatu. And then I can get a job in the Department of Anthropological Sciences at Stanford.
But do they share sounds because they are inherited from a common ancestral language (inheritance, genetic model) or because one borrowed from the other (borrowing, cultural model)? Both are equally likely at first glance, and Occam’s Razor doesn’t help decide between them.
The Welsh ll sound is very rare among Indo-European languages and comparatively rare in world languges, and yet we know it arose in the medieval period due to an abundance of data.
You can’t really use language to establish anything much about a specific group of people diachronically unless you have a great deal of evidence, which we don’t for Southern Africa.