Blood & Continuity

Genetics cannot prove anything about a language. Languages are learned, not inherited, and a population can change its language without changing its ancestry or change its ancestry without changing its language. Anyone who tells you a genetic marker makes someone Amazigh is selling something.

What population genetics can address is a narrower factual question that comes up constantly in the argument: how long have people been living in North Africa, and is there a record of them being replaced? The honest answer is that the Maghreb has been continuously inhabited for many millennia, and that the people living there have never been genetically isolated. Both halves of that sentence matter, and the second half is the one this page is most careful about, because overstating it is the easiest way to lose an argument that was already won.

Fifteen thousand years at Taforalt

The oldest sequenced human genomes in Africa come from the Iberomaurusian burials at Taforalt (Grotte des Pigeons) in eastern Morocco, roughly 15,000 years old (van de Loosdrecht et al. 2018). People were living, burying their dead and maintaining a funerary tradition in the Maghreb deep in the Pleistocene, long before any historical migration into the region. The archaeology of that cemetery is treated separately.

That is where the simple version of the story ends, and it is worth being precise about the rest, because the paper is frequently cited for a claim it does not make. The Taforalt individuals carry roughly 63.5% Natufian-related ancestry and 36.5% sub-Saharan African-related ancestry (van de Loosdrecht et al. 2018). The study’s own title describes these genomes as linking Near Eastern and sub-Saharan African populations. It documents deep antiquity and deep connection at the same time.

So Taforalt is strong evidence for a very old human presence in North Africa. It is not evidence that the region was ever a sealed room, and nobody serious claims it was. Neolithic-period genomes from Morocco show the local Iberomaurusian-related ancestry persisting while ancestry from both the Levant and Europe arrives alongside it (Fregel et al. 2018). A population with 15,000 years of local depth and a history of contact is simply what an inhabited crossroads looks like.

The maternal line

On the maternal side the characteristic North African lineage is mitochondrial haplogroup U6. Its phylogeny has been rebuilt from 230 complete mitogenomes and tracked through 761 partial sequences, and the reading is that U6 marks a Palaeolithic return to North Africa from southwestern Asia, with its later expansions and contractions inside Africa following the continent’s climate (Secher et al. 2014).

That hypothesis has since been tested against an actual ancient genome rather than inferred from living people. The mitogenome of Peştera Muierii 1, an individual who lived in what is now Romania about 35,000 years ago, carried a basal U6* lineage not previously found in any ancient or present-day human, while the derived branches of U6 are concentrated today in northwestern Africa (Hervella et al. 2016). A basal form of the lineage sitting in southeastern Europe 35,000 years ago is what a back-migration into North Africa should leave behind, and it is what was found.

That cuts both ways, and this page is not going to pretend otherwise. U6 is strongly characteristic of North Africa and has been there for tens of thousands of years. It also did not originate there. Those two facts sit together without difficulty, and anyone who needs the second one to be false has misunderstood what indigeneity means.

A paternal lineage, and a dispute about its age

The Y-chromosome lineage now labelled E-M81 is strongly characteristic of Amazigh-speaking populations across the Maghreb. How old it is in North Africa is an open question, and the two readings have very different implications.

Bosch and colleagues attributed about 75% of the North African paternal gene pool to an Upper Palaeolithic colonisation, one that in their reconstruction had originated in eastern Africa (Bosch et al. 2001). On that reading the depth of the paternal record is tens of thousands of years.

Arredi and colleagues read the same landscape differently. They found haplotype variation decreasing from east to west as frequency rises, and interpreted it as a Neolithic demic diffusion from the east, on the order of a few thousand years rather than tens of thousands (Arredi et al. 2004).

This site does not pick a winner, because the field has not. What both readings agree on is that this lineage is native to and concentrated in North Africa, not imported by any historical conquest. The disagreement is about whether it arrived in the Palaeolithic or the Neolithic, and either answer places it in the Maghreb thousands of years before Rome, let alone before the Arab conquest.

What “indigenous” can and cannot mean

Put the strands together and a pattern appears that is worth naming, because it is the thing most often got wrong, in both directions.

The paternal lineage probably came from the east, in the Palaeolithic or the Neolithic depending on whose reading you take (Bosch et al. 2001) (Arredi et al. 2004). The maternal lineage came back from Eurasia in the Palaeolithic (Secher et al. 2014) (Hervella et al. 2016). The oldest genomes in the region already carry ancestry from two directions (van de Loosdrecht et al. 2018), and later farmers brought more (Fregel et al. 2018).

If indigenous is taken to mean a population that sprang from the soil and never mixed, no such population exists anywhere on Earth, and holding the Amazigh to that standard is a trap rather than a question. If it means a people long established in a place, with a continuous presence measured in millennia and no record of displacement, then the evidence here is about as clear as population genetics gets.

The second meaning is the one everybody uses for everybody else. There is no reason to apply a different one here.

What this pillar is actually for

Read on its own, none of this says anything about Tamazight. Its only job is to close off one specific bad argument: the claim that the population of North Africa was swept away and replaced, and that today’s Amazigh are latecomers to their own region. There is no genetic record of such a replacement. There is a record of people living there continuously for fifteen thousand years, trading ancestry with their neighbours the entire time, exactly as populations everywhere do.

The evidence that the language is ancient is linguistic and epigraphic, and it lives in the language pillar and the archaeological record. That is where the argument is actually settled.

---

Every citation on this page was checked against the original source. Where the literature disagrees, both readings are cited above rather than resolved. What was checked, and what was not, is recorded in each entry in the bibliography.