How far does Turkic-speaking area stretch?
As I’ve already pointed out in several previous postings, good books on languages and linguistics, that is, books that are simultaneously accessible and accurate, are rare. But the problem extends beyond books specifically on language matters. Often, authors — who happen to be experts in other disciplines such as history, philosophy, anthropology etc. — make comments on languages. And very often these comments are linguistically uninformed, biased or outright wrong.
Here’s just one example. In an otherwise very interesting book The Middle East. A Brief History of the Last 2,000 Years, Bernard Lewis, a historian and an Emeritus Professor of Near East Studies at Princeton, writes:
The Turkish or Turkic languages, a closely related group of which Ottoman Turkish is the westernmost representative, are spoken in a vast region extending from the northern and southern shores of the Black Sea, across Asia, to the Pacific.
Now, Lewis may have “no living rival in his field” and combine “profound depth of scholarship with encyclopedic knowledge of the Middle East” (all quotes from the back cover of the book), but something’s gone awry with his knowledge of basic linguistics and geography.
First of all, “Turkish” and “Turkic” is not the same thing: the former is a language and the latter is a large family of languages. A similar confusion exists between “German” and “Germanic” languages, and “Romanian” and “Romance” languages. Only isolate languages that have no living relatives can be equated with a family (or a branch of a family), as is the case with Greek (= Hellenic), Albanian, Armenian — all three are isolates within the Indo-European family, having no close relatives. Each of the three is the sole representative of its own branch of Indo-European. Similarly, Basque and Burushaski are isolate languages. They do not belong to any language families. Thus, for example, one can talk about “the Basque language” or “the Basque (or Vasconic) language family” as one and the same.
Second and more importantly, Turkic languages (and even less so Turkish) do not extend to the Pacific. The map below shows the geographic distribution of Turkic languages.
The easternmost Turkic language is Yakut (self-name: Sakha), spoken by some 400,000 people in the Sakha Republic (also known as Yakutia) of the Russian Federation. This is an area known for its climate extremes, with the Verkhoyansk Range being the Northern Hemisphere’s Pole of Cold (i.e., the coldest area in the northern hemisphere). There, the temperatures reached as low as −67.8 °C (−90 °F) in 1892, and the average January temperature is −47 °C (−52.6 °F).
But it would be a long stretch to say that the Sakha Republic stretches all the way to the Pacific. In fact, a stretch of 500 miles long! That’s the distance from Yakutsk (its capital city) to the Sea of Okhotsk. And as the map above indicates, it’s an even greater distance to the Pacific Ocean proper.
Thus, it would be more correct (and, I think, more impressive) to say that Turkic languages are spoken in the area that stretches from the warm waters of the eastern Mediterranean to the frozen shores of the Arctic Ocean.
Note also the gap between the Altai-Tuva-speaking area and that of the Sakha/Yakut. Here, languages of the Mongolic and Tungusic families are spoken (as well as Russian, of course). There is a debate among scholars as to whether Turkic languages can be combined with Mongolic and Tungusic languages into an even larger family, the so-called Altaic family. This proposed family could, according to some scholars, include Korean and Japanese, and therefore take us to the shores of the Pacific Ocean.
However, the Altaic family hypothesis remains quite controversial and not all scholars subscribe to the idea. The alternative hypothesis is that the languages in this region comprise not a family bound by common descent by a linguistic area where languages share similarities through language contact. In a forthcoming posting we’ll consider some arguments for and against the Altaic language family hypothesis.