Most studies of European genetic diversity have focused on large-scale variation and interpretations based on events in prehistory, but migrations and invasions in historical times could also have had profound effects on the genetic landscape. The Iberian Peninsula provides a suitable region for examination of the demographic impact of such recent events, because its complex recent history has involved the long-term residence of two very different populations with distinct geographical origins and their own particular cultural and religious characteristicsNorth African Muslims and Sephardic Jews. To address this issue, we analyzed Y chromosome haplotypes, which provide the necessary phylogeographic resolution, in 1140 males from the Iberian Peninsula and Balearic Islands. Admixture analysis based on binary and Y-STR haplotypes indicates a high mean proportion of ancestry from North African (10.6%) and Sephardic Jewish (19.8%) sources. Despite alternative possible sources for lineages ascribed a Sephardic Jewish origin, these proportions attest to a high level of religious conversion (whether voluntary or enforced), driven by historical episodes of social and religious intolerance, that ultimately led to the integration of descendants. In agreement with the historical record, analysis of haplotype sharing and diversity within specific haplogroups suggests that the Sephardic Jewish component is the more ancient. The geographical distribution of North African ancestry in the peninsula does not reflect the initial colonization and subsequent withdrawal and is likely to result from later enforced population movementmore marked in some regions than in othersplus the effects of genetic drift.
Dienekes has offered his criticisms, and I generally agree. There's something funny going on here. These are paternal lineages, so this is where you would likely find the highest proportion of genes which suggest North African ancestry. Remember that water makes a big difference when it come to gene flow; it seems that even what seems a trivial water barrier blocks the deme-to-deme migration of individuals which results in small but persistent flows which equilibrate between group differences. So it seems plausible that incontrovertible North African lineages would be Berber provenance and due to the impact of the Muslim conquest. But, the historical data makes clear that the native Muslim populations also derived in large part from local converts, including portions of the Visigothic nobility.
If you read a book like The Victors and the Vanquished: Christians and Muslims of Catalonia and Aragon, 1050-1300 you can see that it is relatively adduce from tax and legal records that substantial Muslim minorities long remained in Christian polities even during advanced phases of the reconquest, on the order of 20-30%. After 1492 the Jews were expelled, while many of the elite Muslims emigrated to North Africa, with those who remained tending to convert to Christianity. Though many Muslims did convert to the Catholic religion, in areas of southern Spain where ex-Muslims were still the majority and Arabic was still the local language conversion was nominal and the Islamic religion persisted for generations. This is why ~1600 the crypto-Muslims, Moriscos, were expelled from the kingdoms of Castile and Aragon in an instance of ethnic cleansing. I use the term ethnic cleansing because Moriscos who were sincere Christians were also expelled, these generally settling in the southern Italian possessions of the crown of Aragon.
So why are there are still the possible signatures North African ancestry? Because large numbers of Muslims, likely the majority, converted to Catholicism in the century between the fall of Granada and the expulsion of Moriscos. In fact, the majority of the Muslims of the Iberian peninsula had lived under Christian rulers since ~1250, and of these many had converted to Christianity even by 1492. The expulsion of 1600 occurred in the context of foreign adventures of the Spanish state in North Africa, combined with the fear of fifth columnists in southern Spain, who had already rebelled several times and received some aid from the co-religions and the Andalusian diaspora in Morocco. Some of the Spanish notables who argued for the expulsion were themselves of documented Morisco ancestry, and it may be that their argument for the removal of communities whose religious loyalties were under question also removed any possible taint against themselves.
It seems that the necessary complement to these sorts of studies would be to examine evidence for Iberian lineages in communities in Morocco and Algeria where it is known that Andalusian exiles settled in large numbers. If it is true that the Muslims of Spain were in large part converts from the indigenous populations, despite their partial North African provenance, then that should be manifest in their genetic heritage.
I seem to recall watching a documentary a bunch of years ago which mentioned large amounts of interbreeding between the Christian, Jewish and Muslim communities prior to the reconquista.
I don't know how they can determine that some hgs are Jewish and not Arab, since, on the paternal side, there is no difference between them. In addition to the Berbers, some of the Arabs who conquered North Africa also settled in Spain. Also, under the Ummayyads, some Arabs came from Iraq, Syria and even Yemen/Oman etc. They were always a minority of the Muslim population, most of which was made up of native Iberians, followed by North African Berbers. They were an elite, so to speak, but they also left their genes and hgs which are not different from those of Jewish men. So I think there are a lot of Middle Eastern origin Y haplogroups in Spain whose origin could be Jewish or Arab, or even Pheonician, so you can not categorize them all as "Sephardic Jewish", while pretending that the only "Arab"/Muslim hgs" were North African. I think this study should be expanded to include some Arabs from other countries. There is no doubt that long before 1492 Jews and Muslims converted or were forced to convert to Catholicism in Spain, as new territories came under Christian control. So it is not surprising to find a genetic connection, but the percentage of Jews was smaller than that of Arabs and Berbers. So logically, there should be more genetic material from the Arabs and North Africans than from Jews, although, as I said, when it comes Arabs and Jews, genetically speaking, it is very hard to tell them apart, just based on their paternal haplogroups.