The North Africa connection

New DNA results on Scots ancestry reopen a century-old theory about a North African language connection.

The Scotland’s DNA project has found that no less than 1% of the Scots tested carry a genetic marker which originated in North Africa. The researchers say that the gene, common today amongst the Berber and Tuareg people, is estimated to have originated around 5,600 years ago.

And for over a hundred years there have been linguists who have maintained that there is a link between the Celtic languages and those of North Africa, such as Berber and Egyptian Coptic. The group saying this has been very much in a minority, but it’s contained some brilliant scholars.

The Celtic languages

The Celts of history were a people in central Europe, and little of their language has survived. But some of them migrated to the British Isles, and various languages that developed there have survived. The so-called Insular Celtic languages are six in number. There are the two Gaelic languages – Irish and Scots – along with Welsh, Manx, Cornish and Breton.

The Insular Celtic languages have some peculiar features, which few other languages share. Most linguists believe that these are survivals of an original parent Celtic language, a kind of linguistic analogy of what happens in genetics. The minority group argue that these features came not by inheritance but by environment. They say that the features are the imprints of an older language with which the original Celtic language was in contact.

The idea of a structure in one language transferring onto a neighbouring language may seem strange, but we have examples in English. For instance, in the dialect of the south-west of England familiar from many rural television serials, we may hear them say: “There be I a-singing.” It sounds a complicated way to indicate that I sing, but there is a good linguistic reason. To say it in Old Welsh would be Yr wyf i yn canu, which translates word-for-word as ‘There be I in singing’.

So the structure of the Old Welsh verb has carried over into the English dialect in a zone of contact between the Angles and Saxons, whose language become modern English, and the older Welsh-speakers.

That rather roundabout way of saying things is widespread in the English of people who lives in parts of Scotland, Ireland and Wales where the older Celtic languages were spoken. Instead of ‘I sing’, they might say ‘It is the singing that I was at’ – which indeed is a more interesting way of expressing it. The word to describe this kind of language pattern is periphrasis, meaning a roundabout way of speaking.

Although the periphrastic structure is more obvious in the Celtic regions, it in fact runs through all of English. We can see this by comparing English with Latin. If I love today in Latin, it is amo. If I love tomorrow it is amabo, and if loved last week, the word is amabam. To change the tense from present to future or past, we alter the word-ending – we inflect the verb. In English we do some inflection, as when we go from ‘I love’ to ‘I loved’. But for the rest of the time we take a roundabout approach, to bring in additional words and say ‘I will love’, ‘I did love’.

Where does it come from?

How did we start doing this? The suggestion is that it was through contact with the Celtic languages, so that their habits of periphrasis imprinted themselves on English. And how did the Celtic languages acquire the habit? Again, say the linguists, we must look to a contact with an older language.

These linguists, though much in a minority, were expert in their field. Amongst them was Heinrich Wagner, Professor of Celtic at Queen’s University Belfast. Wagner, born in Switzerland and Professor of Germanic at first Utrecht and then Basle, had an encyclopaedic knowledge of the Celtic languages. He worked for more than twenty years on the four-volume Linguistic Atlas and Survey of Irish Dialects, published 1958-64 and still the standard work today.

As the work took shape, he noticed how dialects could form an imprint on separate neighbours. “I found that each major dialect and each minor subdialect of Gaelic is dependent on its geographical position, all the dialects forming a chain in which two neighbouring dialects always have certain features in common not shared by more distant dialects.”

While working on it he learned that there were still some very old people in the Isle of Man speaking the old Manx language which had been thought to be defunct, so he went there to study it. He noted the periphrastic forms. Instead of ‘I have not seen him’, the Manx phrase was ‘I am not after seeing him’.

He had previously studied Basque, and noted certain similarities in structure. “Although this language is not related to Celtic, its verbal system as such shows a striking similarity to that of Manx.”

Wagner went on to Wales, to learn spoken Welsh from an old woman in Anglesey who could neither speak nor understand English, and again he noted particular periphrastic forms.

Seeking the origins

Where did the periphrastic way of speaking come from? Heinrich Wagner looked at various languages, and found that periphrasis, rare or absent in most languages, is prominent in North Africa. The Egyptian verb had developed in this way, and the medieval language of the Christian Copts showed it prominently. To say ‘he killed the man’ in Coptic is ‘did-he-killing-of-man’. He then noted that the ancient Semitic languages, such as Classical Arabic or ancient Akkadian in Mesopotamia, did not have periphrasis – but that the Cushite languages of Ethiopia, Somalia and Sudan did.

So a remarkable picture built up, of the spreading of periphrastic forms from one language to another through geographical contact. The spread progresses from the Cushite languages to the North African ones. It then goes from the North African languages to the Basques and to the Insular Celtic languages. And from the Insular Celtic languages it moves to the English language itself.

And how could there have been geographical contact between North Africans, Basques and Celts? Clearly by sea, with the Berbers being expert sailors who in Neolithic times reached the Canary Islands. The Basques were also known for their seafaring and shipbuilding, and so contact between the two was very likely. In medieval times, the Basques used to sail north to the Icelandic fishing grounds, and a kind of mixed language developed between them and the Icelanders, enabling the speakers of two very distinct languages to communicate.

The Western sea-lane from the Mediterranean northward to the Irish Sea and the Minch thus can be seen as a zone of contact between people and languages. The news that a genetic element from North Africa has also survived suggests that the amount of sailing and trading in the Neolithic was considerable, and that there was also some settlement by North African traders.

Meanwhile Wagner’s work continues to fade on the margins of linguistics. He died in 1988 and his whole approach has been regarded as flawed by others in the field. He built on the work of several predecessors, including Sir John Morris Jones, who wrote some of the standard works on the Welsh language. Could the latest DNA result lead to a long overdue revival of interest in their work?

Back to OISF

This entry was posted in General science. Bookmark the permalink.

15 Responses to The North Africa connection

  1. Kamal says:

    Im berber from Algeria we have few words in comon like,canu in old welsh for singing,which we use chanu and Foot which we use in berber Fudh which is quite similar in pronuncition.
    Very interesting research i should conclude. Thanks

    • howiefirth says:

      I’m very glad to hear from you and very interested to hear about these words. I do think that the evidence suggests that the berbers were the people who sailed up the coasts of France and Iberia and up the Irish Sea in the neolithic period. I have found some other possible connections and will write them up in the next day or so and will be interested to hear your views.

  2. Archer says:

    Very interesting post. Britain and Ireland seem very much tied to the neolithic expansions (mainly-also some earlier and some bronze age movements happened too) of people along the western Atlantic coast, which of course includes North Africa. The Beaker people were also in North Africa, and they have been found to carry y-dna R1b, the most common dna type in western Europe but also the overwhelming type in Britain and Ireland, the Basque country and elsewhere in the west. Alhough most Berber men carry clades of y-dna E, a minority do carry the most western R1 as well.

  3. Borut says:

    Just a call from Slovenia!:)


    Here’s a link, possibly of some general interest to the discussed topic!?:)

  4. Recently, I had a DNA test done because I had little knowledge of where my ancestors came from, other than the British Isles in general. The DNA results were surprising, showing that my background is 97% mixture of North African Berber and Basque. The other 3% was a various mix of Portuguese, Spanish, Finnic, and a smidgin of the middle east. The linguistic pathway, just seems to follow the genetics, I’d say.

    • I’m a Berber ( Amazigh), from kabylie in Algeria. I assure you that you will be so excited to discover the life of your ancestors. I’d love communicate with you if possible. Lately started studying welsh, Irish, Manx and Norwegian because I find them easy to be used to their rhythm. My Facebook is: amazigh agraklan amdyaz

  5. mara says:

    had dna done — 2 granparents Irish, one scot-irish another sicilian———– my test results were: mostly berber and spanish and some small percentage of scot and scandinavian—– and native american (the latter was a surprise, had to be the scot-irish been in US over 300 years and lived in wilderness areas!) The Irish great grandparents lived in the wild west or ireland (galway, mayo)—so berber and spanish didnt surprise me –wow what about the Spanish Gate in Galway City—those Spanish sailors got around! Spent some winter months in Galway (few tourists) and saw many native men that looked like my sicilian relatives/friends —pretty good looking ha

  6. same says:

    What’s a mere 1% got to do with anything? (it is commonly held that ALL peoples originate from Africa). The fact that ONLY 5% of No. African DNA is genetically Arab, tells the story of acculturation rather than a true blood-line. Also Ireland’s Trinity College DNA study tells a clear story of No. Africans who settled in the Basque region before finally settling in IRELAND.

    Scot’s are genetically Scandinavian and Celtic… Celts are the people of GAUL! The Welsh are also of Celtic origin. Outside of the geographic proximity to Scotland, the and Ulster, isn’t time the ‘Black Irish’ begin to discover it’s unique lineage.

    • Ultasila says:


      I’m Berber from the Middle Atlas Mountains of Morocco. Althought the men in my father’s side are brown/tanned skinned with mongol features but with Caucasian features (for some reason) the women in my father’s side are white skinned, and brown/ashen haired and hazel eyed. I’m personally white pale skinned, asian eyes and dark hair. My mother is berber too but the entire family from my mother’s side are red haired or blonde and white skinned. My grandmother has blue eyes, white skin and red haired.

      So I guess only time will speak our origin. And who knows? Time may never have a chance to speak it because there always be people trying to change the truth.

      Here is some really nice Celtic/Berber music for those interested.

      And with vocals:

      Blessed Be,

      • The features you mentioned are usually common among Imazighen of the North and the Middle Atlas.. and I think that says something about the relationship between Berbers and Europeans throughout history…

  7. Jeffrey Banner says:

    I am 100% Scottish Irish and Welsh but the DNA test I just took show concrete connections to North Africa, Sardinia and the Basque. After watching Otzi the Iceman reborn – on Netflix – I realize that all of these people were the same at one time. They were isolated pockets of people that the new waves of immigration couldn’t reach due to mountains or being on islands so they remained the same such as the Berber in the mountains, the Irish the Sardinians. They are all related.

  8. Jeffrey Banner says:

    I think the author is missing the point by saying Celtic DNA is what is linked to the berbers they should be looking at the pre- Celtic DNA. I also took this test being primarily Scottish and Irish and I had about a 5% Berber match

  9. Pingback: ‘Black Star’ lyrics… David Bowie’s last message – Wake The Dragon

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s