22 June 2008

What the Romans spoke

Everybody would agree that what we call Romance languages (Spanish, French, Italian, etc.) are languages that derive from Latin. The Roman legions fought hard to make the Empire bigger and bigger, and as a result Roman civilization and language reached a vast territory. In the traditional view, it was by way of conquest and acculturation that Latin became the language of these new provinces, substituting the pre-existing ones, which would only remain as the substratum.

The explanation is quite simple and logical, but...

Let's take a closer look at the map of the Roman Empire:

(Source: Wikimedia Commons. Click here for a larger image and further details).
And now the map of the Romance languages:

(Source: Wikipedia. Click here for a larger image and further details).
We can see that Romance languages are basically located in western Europe: Italian Peninsula, Gaul, Iberian Peninsula. Rumanian seems to be the only exception. What about the rest of the Roman Empire? Greece, the Danubian area, Anatolia, the African provinces... In some of these areas, like present-day Tunisia, the level of Romanization was as deep as in any western European province of the Empire. Impressive Roman remains can still be seen in many North Africa locations, and in the old days many important figures of Roman history were born there... Therefore...


I don't know if there are many linguists who have asked themselves this question, but I find it interesting.

Traditional linguistics can offer no satisfactory explanation for this fact, because of its chronological constraints and its dependence on historical conquests and migrations. But fortunately, there are some authors (e.g. Untermann, Villar, Ballester, Alinei) whose contributions are helping to build a completely new framework for Pre-Roman languages and their connection with present-day languages. It is obvious that the Roman conquest is still the main factor in the formation of Romance languages, but it cannot be the only one. Otherwise, we would expect Romance languages in many other places, for example in northern Africa.

Very probably, the answer to the question must be found in pre-history. And also in the application of some basic principles, like this one: the more you go back in time, the less likely it is that a language spreads by way of conquest or colonization. The words of Mario Alinei (in Origine dell lingue d'Europa, Vol. 2, p. 813) are quite relevant:

"nella preistoria come nella storia, l'ibridazione linguistica è la regola, la sostituzione linguistica l'eccezione".

It is obvoius thet in the pre-history of Italy, Iberia and France there were elements that facilitated the birth of what we call Romance languanges. They can be socio-economic factors, but also linguistic, such as the presence of IE languages (mainly Celtic or Italic) in these areas already in pre-historic times. But if we look at some of the peoples that the Romans subdued (Etruscans, Iberians, etc.) we see that they are non-IE, or at least their inscriptions were written in non-IE languages. However, it is becoming quite clear that when we talk about Etruscans, for example, we are mainly talking about an intrusive elite that ruled over a mainly Italic-speaking population. Mario Alinei has given an abundant amount of evidence (from archaeology and dialectology) to prove the continuity of the Italic-speaking population in Etruria and other places, even outside Italy. The Romans of northern Africa and the Etruscans of Etruria had something in common: they were ruling elites without a capacity to alter significantly what the original populations spoke. Something similar could be said about the Iberians. They were the ruling elite in some areas. But what do we know about the peoples they subdued?

Historical linguistics must go beyond easily-explainable historical facts and enter the realm of real multidisciplinary analysis.

In future posts I'll focus on the languages of Pre-Roman Iberia, which are particularly interesting.


JoseAngel said...

Nonetheless, elites are important (I just wrote on this in a commentary in Babel's Dawn). I suppose a main factor to take into account is the presence of the Church as a cultural elite in Europe but its disappearance in Africa from the 6th century. Notice that the pattern of Roman colonization did not favour the universal adoption of Latin as a communal language - merely as a lingua franca and as an elitist language. Under the Romans, cities and villages tended to remain very much in a kind of ethnic (and I suppose linguistic) apartheid - Iberian cities for the Iberians, and separate if neighbouring Roman cities for the Roman colonists. There must have been much exchange of course, with the concomitant use of Latin as a lingua franca and prestige language, and pidgin phenomena likewise. But the presence of the Church in Europe, and its disappearance in northern Africa, remains a major factor. Elites are a major element in language formation.

Jesús Sanchis said...

José Ángel, what you say about the influence of the church is interesting. In fact, the answer to the original question ("Why is it that there are no Romance languages in northern Africa?") must be a combination of factors. The new thing in this old debate is the idea that it is essential to take into account the pre-historical record in connection with present-day dialects. This is Mario Alinei's original approach. He applied it in his study of European languages, with extraordinary results. I suppose that it would be interesting to carry out similar research on North African languages or elsewhere.

Bayndor said...

As for the disappearence of Romance languages from North Africa, there is one crucial factor that cannot be ignored: namely, the conversion of populace to Islam, and the arabization of urban population. The same process can also be observed within the framework of Mozarabic language, that was spoken in the medieval Iberian Peninsula. It was unlike Spanish in many ways: Mozarabic was much more conservative in terms of its grammar, but heavily arabized on the other hand. Its extinction shows the general direction of language development in this area: On the territories conquered by Castilian, Aragonian or Portugese kingdoms, it was replaced by their respective languages, while in Granada, it was supplanted by Arabic.

I do not find any hardship, therefore, to explain its disappearence. Like the Coptic Egyptian language - that is exclusively maintained by the Christian minority in Egypt (being more conservative), the Romance-language speakers were likely limited to the Christian minority in North Africa (Morocco, Tunisia, Algeria) as well. But unlike Egypt, that minority - gradually dwindling - disappeared around the 11th-12th century. Their Romance language likely died out with them. As these communities left little if any written records, we are unable to reconstruct the way their language developed...

Jesús Sanchis said...

One of the things I wanted to emphasize in this post (and in this one: http://languagecontinuity.blogspot.com/2009/04/populations-and-languages-strait-of.html) was the role of affinity in situations of language contact. It is one of many possible factors, but I think it's an interesting one, which is sometimes overlooked. Romance-related language sin northern Africa were spoken by a minority of people maybe because the rest of the population saw it as somehting too different, or clearly different. Arabic, which belongs to the same language group as the languages of northern Africa (the Afro-Asiatic group) was perceived as something more familiar, which didn't happen in other territories in the Muslim Empire like Persia, with a non-Afro-Asiatic background. As I said, I'm just suggesting that the degree of affinity might be relevant in explanations of language 'substitution' or hybridization.

Owain Barry said...

The geographical separation of Romanian from the other Romance languages is an interesting challenge for the PCP. While I myself maintain a strong scepticism towards most ancient language shift theories, it seems the most logical situation here is that the entire Balkan region outside of Greece was Romance speaking before the coming of Slavic speakers during the dark ages; a pretty standard mainstream non-continuity explanation.

Any thoughts on the matter?