« Hayek on neural networks and free will | Gene Expression Front Page | Pan sapiens? Homo troglodytes? »
May 20, 2003

Journey of Men

Just read Spencer Wells' book Journey of Man, a brief but thorough survey of human population genetics in the vein of Cavalli-Sforza's The Great Human Diasporas and Bryan Sykes' Seven Daughters of Eve. While Sykes focused on Europe and mitochondrial DNA lineages (the mother line) Wells puts the spotlight on Y chromosomal lineage (the father line). Wells gives a few reasons why the Y chromosomal lineage can yield more information-there are more points for mutations to build up and human patrilocality tends to skew toward male genetic localization and diversity.

The book has less fluffy filler and more "red meat" than Sykes' work. Like Sykes, Wells is trying to make money, though he is an independent consultant rather than a businessman, so he has incentive to make this field "exciting." Basically Wells is doing archeogenetics-tracking movements of populations using genetics. Human beings have an inordinate amount of interest in their origins-so it makes sense that we would care about our recent antecedents at least as much as which branch of the more ancient hominid lines we issue from. Much of the data that Wells draws upon comes straight from the Human Population Genetics Laboratory at Stanford University (follow the links and see detailed publications which offer more than the sketches given in the book). Of course Wells draws from the work of other laboratories to fill in the gaps but the major selling point of the book seems to be his insider point of view, and in contrast to Sykes he doesn't get tedious about the interpersonal details and let it overwhelm his narrative. The fact that Wells made a film on this subject for PBS probably helped him frame the story in a more compelling fashion.

There are a few obligatory chapters that serve as a genetics primer (though he continues to insert these elucidations throughout the work after the introduction which can be distracting) and descriptions of the scientific methods that have made analysis of Y chromosomal lineages far easier than in the past. To lay persons the methodology will be uninteresting and to those in the know it will be redundant and simplified, so though these sort of chapters are necessary to make a book of this sort seem complete, I doubt many will read them closely.

You hit the payoff near the end of the book, especially chapter eight, titled The Importance of Culture, which is a detailed exposition of the current knowledge in archaeogenetics, slanted a bit toward Wells' own researches and travels (chapter eight also has a nice map of the movement of markers that transmits the gist). But before I hit this, I have to address a few minor political points. Like his mentor Cavalli-Sforza Wells innoculates himself against any charge that he is studying race. On the back cover there is a blurb from one Dick Lewontin and a paean within to his study back in the 1970s that showed that most human genetic diversity was not between races, in fact less than 10% [1]. Wells also attacks C. S. Coon, the physical anthropologist who published The Origin of Races and elaborated a theory that pre-dated the modern multiregional camp, arguing for five human subspecies that attained sapiency independently [2]. Milford Wolpoff rebuts the naive charges of racism against Coon in Race & Human Evolution and highlights the differences between modern ideas of multiregionalism and their percursors, such as Coon's theory [3]. Wells several times attempts to paint the multiregionalists as deluded cranks. I personally do think the evidence for Out-of-Africa is compellling, but Wells does not make it clear enough in my opinion to the lay audience that it is still circumstantial, genetics is after all not physics, and multiregionalists are not Steady Staters [4]. An argument can be made that Wells is trying not to muddy the waters in a book that is only 200 pages, but that is no excuse for giving the public a misimpression of the situation on the ground.

There are also a few attempts by Wells to explain human adaptations and delve into evolutionary psychology-he lauds Toobey & Cosmides in one section-but they are sleepers compared to the rest of the book. But there was one section that he seemed to dwell on, and which ties in to his focus on the north-south difference in the Out-of-Africa migrations. On page 117 he talks about the "central Asian" clan forged in the fires of the brutal tundra (so to speak):

The Eurasian interior was a fairly brutal school for our ancestors. Advanced problem-solving skills would have been critical to their survival, which helps us to understand why it was only after the Great Leap Forward in intellectual capacity that humans were ready to colonize most of the world. During their sojourn on the steppes, modern humans developed highly specialized toolkits...The problem-solving intelligence that would have allowed Upper Paleolithich people to live in the harsh northern Eurasian steppes and hunt enormous game illustrates something that could called the 'will to kill.'

Wells' own ancestors did come from the interior of Eurasia, as did some of his wife's (she is mentioned as Hong Kong Chinese). But these were not the ancestors of most Africans or Australians, and few southern or western Asians. I will let readers connect the dots-but Wells talks about Calvin's idea of the Ice Age being a sort of pump that pushed humans back and forth and served as an impulse for higher intellectual development.

So after Spencer makes a few half-hearted attempts at evolutionary psychology and natural history, he gets to the real meat, the preoccupation with ancestry, and the "clans" of humans (let's give him a good wink here). I'll sum up the main points in the book, with my own personal interjections. I suggest GNXPers read chapter 8 in the bookstore, the rest of it most who were well-read in evolutionary sciences already know (and to be sure much of chapter eight is cobbled together from the work of others).

Africa: That's where we come from. All of us. Those of us from outside Africa come from the northeast corner-in two major streams, one out along the Somali coast and another up through the Levant.

Europe: Wells quotes the 80/20 Paleolithic/Neolithic split as definitive. Basically Wells asserts that Europeans are mostly the descendents of Cro-Magnon people who have their origins in the central Asian cauldron. He has done research relating the remnants of the Sogdian people in Tajikstan (Yangbonis) to the English rather than the Uzbeks of the lowlands. Wells is obviously onto something here, but I did not find the updated figure that others are pushing, that Europeans are 50% descended from Middle Eastern farmers. In either case, there is a NW to SE gradient, while Greeks even in the 80/20 scenario possibly being closer to Levantine peoples than to Swedes. The European "bioculture" as some would say is the byproduct of a cocktail of Levantines and Cro-Magnons, the only question is the portion of each. I don't know where that leaves the Indo-Europeans, but I suspect that they were neither Cro-Magnons nor the first farmers-I think elite dominance is the way to go in Europe (and Wells seems to settle on this position as well).

East Asia & America: The northern wave that passed through central Asia and the southern wave via India and south east Asia seems to have met up in China. Some of the southern wave mingled early enough to push their genes over the Bering Strait to the Americas when the Siberian hunters populated it. Cavalli-Sforza's results that clustered northern Chinese with Europeans and southern Chinese with southeast Asians probably result from the higher portion of central Asian ancestry in the north. Rushton has wondered as to this result as the south Chinese have rather high IQs, but in this case, recent cultural practices might have had a stronger selective pressure than ancient adaptations during the Ice Age, as that aspect of phenotype converged in both north and south under the influence of the same culture. Unlike Europe, the agriculturalists in eastern Asia swamped the peoples of southeast Asia that were related to the Australian Aborigines and are to be found among the remnants like the Negritos and Andaman Islanders (even under the 50% Middle Eastern scenario northwest Europeans are mostly Cro-Magnon). Wells does not dwell on this difference, but obviously the descendents of Cro-Magnons were more flexible and more difficult to overpower than the scions of ancient beach-combers.

Australia: This ancient land is the most pure repository of the "southern wave." Henry Harpending, whose work Wells knows because he mentions it, suggests that the Great Leap Forward did not impact the Aboriginal peoples of Australia much at all. This whole area is filled with political land-mines, so of course Wells would tend to be careful here.

India: There is a strong signature of the "southern wave" in the female line though little in the male line. An early branch of the "northern wave" pushed into India around 30,000 years ago and is like an equivalent to the Cro-Magnon's of Europe. There were later "agricultural" waves from the Middle East around 10,000 years ago. Finally there is one marker Wells mentions that is prominent in much of India and eastern Iran called M17 that is also prominent in eastern Europe and seems to be from southern Russia. It, like many of the other markers, shows a strong NW-SE grandient, in this case, being found in 35% of men in Delhi but fewer than 10% of men in the Dravidian south. The dots are easy to connect.

Middle East: Not much to say here. Western Iran and eastern Iran are more different than one would think and there is little evidence for the "southern wave" in the Middle East, indicating that they didn't push farther inland. Seems the Middle East is important more as the starting point for migrations, primarily because its early role as agricultural innovators, sending culture & people north, west, south and east.

I could say more-but read the book. The details will change over the years, but I think Wells has presented a good hint of the outline that will be discerned about the history of human movements.

Update: Here's a version of the map in question (this is simplifying the picture of course!)....

[1] See this article by Steve Sailer taking on Lewontin's assertion. Additionally Wells does not even hint at the fact that Lewontin is a politicized Marxist who played a critical role in personally attacking E. O. Wilson during the sociobiology controversies of the 1970s. In fact, though S. J. Gould continued the good fight on behalf of the "anti-hereditarian" Left until his death, Lewontin faded away from public view, but books like Defenders of the Truth: The Sociobiology Debate, pin-point him as the real svengali behind the scientists who made the scientific controversy into a circus.

[2] Caucasoid, Mongoloid, Australoid, Congoid and Capoid were the races (at least from what I remember). Congoid and Capoid refer to the "black" and "Khoisan" peoples of Africa, while the others are rather self-explanatory.

[3] The short of it-Coon spoke out against racism early in his career, but he never developed the hypersensativity to racial questions that other anthopologists did after World War II, so it was easy to paint him as a Neandertal. His scientific theories can be salvaged only through intense modification-but the attacks that Wells and others engage in against him are very personal and accuse him of being an unreconconstructed bigot, I think to cover their own attempts to do what Coon did-reconstruct human racial history. Additionally they use Coon's quasi-multiregionalist ideas to taint modern multiregionalism as retrograde. Never one to not use tools at their disposal multiregionalists like Wolpoff paint the Out-of-Africa camp as vicious propogandists for the racial supremacy of modern Homo sapiens. Oh, and I have to add that Wells makes a big deal about the fact that Coon thought that Africans were an evolutionary "dead end." I didn't really get this from Coon's work, but perhaps it's my memory, but in any case, Wells himself contrasts the relatively easy life of beach-combers (southern wave) with the harsh life of northern Eurasians (northern wave). I can't but help wonder if he's trying to throw people off.

[4] To take the anology further, while it is hard to imagine a synthesis between static and expansionary universes, it does not seem impossible that both multiregionalism and Out-of-Africa hold a portion of the truth.

Posted by razib at 01:51 PM

"I did not find the updated figure that others are pushing, that Europeans are 50% descended from Middle Eastern farmers."

The 50% figure probably comes from this article:

Chikhi, L. et al. (2002). "Y Genetic Data Support the Neolithic Demic Fusion Model." PNAS 99(17), 11008-11013.

I haven't read the article and am not an expert, so I won't comment further.

Posted by: anon at May 20, 2003 09:30 PM

Neanderthals were not primitive, just bad PR: they had larger brains, made birch pitch tar that can be only made at 300-400 degrees C, they made flutes, buried their dead, probably more intelligent than any race today.

The “typical” Neanderthal was a pathological 50 year arthritic. The typical Cro-Magnon also was not typical. What we find is a continuum, a cline where early Neanderthals become less robust more Cro-Magnon – with more African features like longer limbs smaller noses – this continues down though Mesolithic then modern Europeans. Mongoloid genes coming from the East also probably had an effect. Nordics, Scots, (the genes for red hair have been traced back 200K, where are they in Africa?), Basques, Irish, Jews and Armenians, Kurds may be more Neanderthal. Europeans have features that are not found in Africa, but in Neanderthals. There would have been more Neanderthal impact on the flow on the Y-chromosone; Neanderthals were more intelligent and much, much stronger, but could be outbred; Africans far more than Neanderthal females would have been impregnated. It was a long process over at least 80,000 years, the admixture would have been slow. There would have been some, and has been in historic times, gene flow back to Africa.

Last year Templeton, in a genetic study in Nature “Out of Africa again and again”, using “ten different haplotype trees (MtDNA, Y-chromosomal DNA, two X-linked regions, and six autosomal regions)”, showed migrations out of Africa at 1.7, .84-.42 and .15-.08 myr, and considerable genetic interchange between populations throughout this period.

http://www.rdos.net/eng/asperger.htm and elsewhere

The OOA thesis is even more shaky in East Asia, you can trace flat faces, inward teeth and other Mongoloid features back 500K to Peking man; there has been a lot of white gene flow in prehistory and history, seen in fossils, that have moderated these feature and brought Asians and whites closer together.

Yet in Australia it becomes completely unstuck. Aboriginal skeletons are either robust or gracile, (current aborigines are a varying intermixture) the robust, many late as the Holocene, are closer in cranial morphology to Homo Erectus than modern Europeans or Orientals. Homo Erectus skeletons have been found in Solo, Ngandong, in Java which has been dated at 25 000 to 50 000 years old.

Amost all media presupposed OOA, or mentions MRE as a theory amongst a few scientist but then for the rest of the article/documentary presupposes OOA. All talk about Anatomically Modern Humans as if there was no differentiation. It is all PC, justifying the mythology of human equality.

Posted by: Dan at May 20, 2003 10:15 PM

those Kurds and Basques are sure exhibiting their superior Neanderthal brains ...

Posted by: Jason Soon at May 20, 2003 10:59 PM

yes-i know the study, i just meant that it isn't mentioned in the book though he probably knew about it unless it took a lot longer to get to print that usual. the circumstantial evidence seems to point against neandertal contributions to modern europeans from what i have seen, though it doesn't eliminate the possibility-you can see in the morphology what you want, but perhaps by educational bias is coming through....

Posted by: razib at May 21, 2003 12:07 AM

I wonder how Kurds and Basques would go on IQ tests? They must be pretty tough to survive so long relatively unhomogenised. They may have suffered all sorts of dysgenic affects, warfare, inbreeding, etc.

Posted by: Dan at May 21, 2003 05:43 AM

The Basques do pretty well. Most are monolingual in Spanish by now. The Basque area is one of Spain's most developed, and the Basques are probably the hardest working Spaniards. A fair proportion of famous Spaniards are Basques under Spanish names.

Spain was fully-modern up until 1700, and seems to be recovering from its decadence pretty well. One area of science which has a strong Spanish-language influence is neurology, probably because of the influence of Santiago Ramon Cajal.

Posted by: zizka at May 21, 2003 07:43 AM

Fascinating Zizka to learn about such Basque achievement, I must read up on it more. They are a marvelous people, Paleolithic survivors, and sole inheritors of an ancient Pre-Indo-European language. I would say that Spain suffered from the dysgenic effects of Muslim rule and black admixture, then from dysgenic immigration of intrepid men (mostly) to the New World, many to die from disease and warfare, the surviving now mestizo/mulatto in varying degrees along a racial/economic cline. The Spanish fertility rate, CIA Factbook states 1.16 in 2002, is among the lowest in the world; this will be an increasing strain - I believe there is also political pressure to accept more Arabic immigration.

Posted by: Dan at May 21, 2003 08:32 AM

Dan says:

"I would say that Spain suffered from the dysgenic effects of Muslim rule and black admixture"

Are you a Kempian? you can provide a source, litle nazi?

you can mention some genetic study that mentions the black admixture?

Because I don't know none, in fact in most of the genetic studies on Spain don't mention anything of that.

And "arab" it is not an ethnicity or a race, Spain didn't have Arabs, but Muslims Berbers (and not all the Muslims were foreign), Period!


An Spaniard from Malaga

Posted by: Slayer at May 21, 2003 10:39 AM

dan-you knew that a spanish reader would get pissed off-but seriously, there is surely more *jewish* than sub-saharan african admixture by an order of magnitude because of the moorish period (especially the elite). in fact, as i've mentioned, basque nobles did not, unlike other "spaniards," have to provide a certificate of pure christian blood because of the lack of jewry. this is the problem with making claims having to deal with the g of ancient groups, you assert the brilliance of cro-magnonoid groups (for lack of a better term), but it was among middle easterners that "civilization" began and resided for several millennia (and spaniards are definately of the cro-magnon clan since the berbers of north africa are in large part from that group anyhow).

oh, and as far as sources, don't quote MARCH OF TITANS.

Posted by: razib at May 21, 2003 10:45 AM

Razbis says:

"dan-you knew that a spanish reader would get pissed off-but seriously, there is surely more *jewish* than sub-saharan african admixture"

The problem is that there is not black admixture.

In this study (Bosch et al. "High-Resolution Analysis of Human Y-Chromosome Variation Shows a Sharp Discontinuity and Limited Gene Flow between Northwestern Africa and the Iberian Peninsula."Am. J. Hum. Genet., 68:1019-1029, 2001) says the following:

"No haplotype assumed to have originated in subSaharan Africa was found in our Iberian sample."

Posted by: Slayer at May 21, 2003 11:08 AM

Slayer: That should be no surprise. Moors were not subSaharan African (that is, "black"); they came mostly from the coast, and probably some had come all the way from Arabia. The present population of north Africa probably includes quite a few descendants of subsaharan slaves, but as far as I know the practice of slave-raiding across the desert was started after the Muslim conquest. So in Mohammed's time you'd probably have found more German genes (from Visigoths and Vandals) in the Moors & Berbers than subSaharan. The Spanish of the time were also a mix of Visigoth and darker earlier natives, but it's likely that there were simply more Visigoths in Spain, so Spanish christians tended to be lighter, especially in the upper classes.

And so the light brown Spaniards found it made good propaganda to call the darker brown invaders "black." It emphasized the difference to help maintain their own unity. Possibly distinguishing themselves as "white" versus "black" invaders gave them more of a common bond with the Franks (still German) than the religious issue alone.

And finally, Isabella and Ferdinand kicked out the Moors. Then they gave extraordinary powers to the Spanish Inquisition to investigate whether the Moors and Jews who converted rather than leaving were truly practicing Christians. This eventually turned into a persecution of anyone with Moorish or Jewish ancestry and enemies willing to make up stories (and be rewarded with a share of the victims estates!). I suspect that over a few centuries, this not only eliminated the Moorish remnant, but also anyone who looked too Moorish - say, a descendant of a black Roman slave.

Posted by: markm at May 21, 2003 07:04 PM

the 'spaniards' did not view the invaders as racial opponents at all-it was a religious thing. most of the 'moors' were of mixed heritage, converts from christianity.

Posted by: razib at May 21, 2003 08:24 PM

PS-there were several waves of muslims between 700 and 1200. the later groups, the almoravids & almohads had multiracial origins. the 'moors' were a combination of berbers, with only a few arabs, and native women. additionally-it seems plausible that many of the slavic slaves that were common in al-andalus during the height of the ummyyad caliphate became muslims....

see the book ORNAMENT OF THE WORLD for a physical description of how 'spanish' the muslims of spain were (and many of the moriscos did not leave for morocco but converted to christianity to be reabsorbed)

Posted by: razib at May 21, 2003 08:40 PM

The Arabic, Turkish, and Romance-speaking gene pools have to be extremely diverse, since these languages were spread militarily and politically by relatively small armies into densely-populated areas. Furthermore, none of the conquering armies were ethnically uniform either; they all recruited mercenaries and followers with little regard to origins.

A study looking for Celtic, Germanic, Roman, and other (Basque, Alan) ancestry in France is probably possible now. I personally have no idea whatsoever what proportion of the "French" of 800 AD were descendants of the original Celts. (The Bretons don't count; they came from Britain ~500 AD.)

On the Ottoman founders and other similiar groups:

Lindner, Rudi Paul, "What was a Nomadic Tribe?", Comparative Studies in Society and History, 1982.

Posted by: zizka at May 22, 2003 07:56 AM

zizka-i would bet most north french are descended from celts. but the germans and celts are not that far off-and many of the 'latins' that settled in southern gaul might have been latinized descendents of cisalpine celts from the po valley.

Posted by: razib at May 22, 2003 11:36 AM

having read the tract on Gaul in the Bartleby Encyclopedia a few days ago I have to ask the question whether they were Cispadane or Transpadane Gauls;)

Posted by: Zachary Latif at May 22, 2003 03:15 PM

"Journey of Man" seems like a tremendously interesting book and i really like the map that you posted. But there's one aspect of human migrations that's never studied, it seems : The intra-african migrations . The picture that one gets is that the "rest of humanity" was basically born out of east-africans who went towards the north and the east. But what about the "other half" of humanity, those who went west and became the west-africans ? Is there any work at all being done on their migration patterns ? Being of west-african descent myself I'd like to be able to "make sense" of the various black people of africa ( because the diversity is really immense) and some scientific vulgarisation work on those matters would be very welcome. Any pointers ?

Posted by: ogunsiron at May 22, 2003 09:12 PM

read cavalli-sforza's work, or jared diamond for that matter. this is what i know about africa:

about 2,000 years ago a group of "bantu" in the environs of ibo-land in eastern nigeria got a hold of a better agricultural technique and exploded across the continent in about 1,000 years. they pushed east on a straight shot along the edge of the rainforest and around the lands of the 'pygmies' toward rwanda/burundi and then south through east africa until they hit the brick wall of south africa. there the agricultural techniques didn't work well past what is now xhosa-land because the climate was dry & somewhat mediterranean, not tropical. another branch of "bantu" pushed south along the coast into angola, but again, the dry desert in namibia slowed them down and the two groups ended up comingling in the high-veld of katanga in southern congo, zambia and zimbabwe.

in west africa there area variety of languages called 'mande' i believe and a lot of diversity, which indicates in this region all the peoples attained agriculture pretty quick so that one couldn't swamp another. there are isolates like hausa in northern nigeria that are afro-asiatic, like arabic. this is the result of back-migration from northern africa-pretty obvious when you look at the hausa->fulani->tuareg phenotype, going from mostly negroid to caucasoid (the taureg speak a 'hamitic' language related to the berber family). there are some markers in west africa that indicate genetic back-flow from eurasia rather deep in time.

the people we know as 'khoisan,' probably ranged along the whole east africa coast 2,000 years ago. the 'queen of punt' that is illustrated in the depictions of the trade voyage sent out by the 'pharoh' hatshepsut depicts extreme steatopygia, the 'fat ass' syndrome that attains its most impressive levels in khoisan peoples. punt was probably the current somaliland and its environs. the genetics of ethiopians often shows a clustering with the khoisan people, showing that that substrate still exists. additionally the hazda people of tanzania speak a 'click' language-distantly related to bush.

along the east africa coast there are non-bantu elements we term 'nilotic,' basically the steppe adapted tall thin peoples we know like the dinka, nuer, etc. in recent centuries a fair number of these people have pushed into the 'bantu' agricultural zone, bringing their rangy physiques and cattle-based lifestyle. the tutsis are one such group, as well as the kalenjin people of tanzania (the great runners). additionally there were arabs and indians on the coast. on a minor note the malagasy people of madagascar speak languages that have close resememblences to some spoken in borneo, and their appearence suggests asian admixture-BUT their language also indicates they spent some time on the east african coast.

finally-the people of ethiopia in some studies show a strong y chromosomal signature that is 'caucasoid,' they speak a language that is clearly semitic and have long has associations with the peoples of yemen (they think they are the heirs of sheba-the sabaean kingdom). of course their appearence could also be a good adaptation to the highlands in which they live.

that enough for you? i think that's pretty much correct though i might have gotten a few details wrong....

Posted by: razib at May 23, 2003 12:25 AM

From someone who takes the easy option of throwing the Nazi taunt loosely, Slayer’s hostility to the possibility of having black heritage was amusing. Razib, if I am correct interpreting you, I think you're the closest when it comes to the Spanish and black genes, not none but very little, from the Berbers and the much smaller numbers of Arabs.

Black gene flow into Arabs, mostly through the Slave trade – see below

Source on small sub-saharan heritage in Berbers, though these are modern Berbers, the Berbers who populated Spain may have different characteristics:

“In summary, one-third of Mozabite Berber mtDNAs have a Near Eastern ancestry, probably having arrived in North Africa 50,000 years ago, and one-eighth have an origin in sub-Saharan Africa.”


It was just a passing point about Spain hurriedly written, and I gave more emphasis to the dysgenic immigration to the New World. There was a possible ambiguity in my post; I meant that Spanish speaking Central and South Americans, not the Spanish are mestizo/mulatto in varying degrees along a racial/economic cline”. I don’t know a great deal about Spain and was forgot the extent of the Berbers influence. Also Lynn states Spanish IQ as 99, quite high, close to North European. Spain’s relatively low GDP could also be explained by culture and religion, though this may be in part the result of sociobiological reasons – interesting how Protestantism is predominant in North Europe. A couple of years ago I read somewhere in a National Interest, I think one of the 2000 editions, about how Protestantism tends to lead to higher GDP for various reasons, I won’t try and remember all the arguments. Note: this is not an attack on Catholics; in many ways I prefer their traditional collective approach.

Posted by: Dan at May 25, 2003 03:38 AM

Does anyone know anything about the genetics of the Arab-speaking peoples? Based on what I know of the Muslim conquests, "the Arabs" should be pretty diverse.

Posted by: zizka at May 25, 2003 10:49 AM


you don't have to look at the genetics-the process of "arabicization" as i'm sure you note is well documented historically. the north african case is easiest to point-arabcization of the "tamizat" speakers ("berber") is still occurring. a few arab tribes DID migrate en mass to the deserts of north africa-but they were the exceptions that proved the rule. even the city-dwelling "arabs" are probably the descendents of one arab man who married a berber, and this process continuing unto the generations.... the first ummayyad emir of cordova was the son of a berber woman, though the arab through the father-line. his descendents were mostly "spanish" in origin-though arab by identity. the egyptians are mostly descendents of ancient egyptians-the the presence of assyrian speaking villages in syria attests to the last mop up operations of "arabicization."

the arabicization of the peoples of the fertile crescent is not that difficult to understand-the semitic languages that descended from aramaic were close to arabic-even jews ended up speaking arabic , after all they had given up hebrew for aramaic.

btw-most upper class arab families have made up geneologies claiming the status of 'sayyids'-descendents of the prophet.

PS-some arabs in areas like the haudrawmat in yemen show strong southeast asian & indian influences because of the indian ocean trade.

Posted by: razib at May 25, 2003 03:46 PM

In my experience Arabs, especially Sunnis, are very ignorant as to what is "Sayyid ancestry" (strangely enough it seems that these numerous Sayyids, who number in the millions going on claim alone, seem to have packed their bags from Arabia and migrated to the Shi'ite regions of Persia & Pakistan).

Most of the Arabs are quite parochial when it comes to their kin in other nations. Whilst genealogies are meant to have great importance they are no longer relevant for the settled populations of the Fertile Crescent (the Palestinians wouldn't have been agitating for their state otherwise and would have been able to assimilate elsewhere had they traditions been of a more nomadic nature).

Caliph Umar admonished the tribal Bedouin armies to remember their lines so as to not be like the soft natives who when asked of their origins "merely answered that they hailed from hereabout". However a millenia of history overrides the dictates of a Caliph in this instance.

As for North Africa it was reinforced by 600,000-700,000 Arabic speaking immigrants from expulsions in the Iberian peninsula (granted many of whom were Mozarabs and Berbers in the first place) so the population there does have a degree of Arab Bedouin ancestry (the Maghreb frontier, or rather the Egyptian-Libyan border, is noted for the sharp shift in the Bedouin dialect.

Arabisation and Islamisation is the culmination of a very slow, but at times rapid, process that involved significant acculturation with the host society. It seems similar to puncuated evolution "where instead of a slow, continuous movement, evolution tends to be characterized by long periods of virtual standstill ("equilibrium"), "punctuated" by episodes of very fast development of new forms." That is analogous to the growth of Islam in the heartland...


Posted by: Zachary Latif at May 26, 2003 10:47 AM

The accuracy of memorized geneologies is suspect. Mary Douglas has concluded that in msot cases geneologies are accurate as far back as it makes any concrete difference -- i.e., an ancestor will be remembered if his descendants are part of your life. She suggested three living and three dead generations as a rule of thumb. Further back than that you have misty figures who can be used to hook in fictitious kin, and further back yet you have legendary figures shared by large groups of people. The Chinese worship ancestors collectively past a certain generation (3 or 4 dead generations).

So anyway those Shaikh lineages are all suspect.

My family has a geneology going back to about 1200 A.D., but the first reliable link is about 1799 or 1770. Geneologists are not paid to be scrupulous.

Posted by: zizka at May 26, 2003 09:06 PM

Hi I would like to make note of the features that Ethiopians/Malagasy/Swahili peoples all sometimes seem to share. My mother is of pure ethiopian descent, yet she is often mistaken for an indonesian. (She has high cheek boned and a small enpathic **is that the right word??** fold on her eyes. I am myself half english (white) and I look distictly oriental. On my most recent visit to ethiopia I took this into note and found that alot of the people in ethiopia have small hints of "mongol" appearance such as the singular fold in the eyes and high cheek bones, along with the small stature of their body height. I thought it was impossible so please help - ive been wondering about this for a long time!!!

Posted by: MJ at September 26, 2003 11:47 AM