Arabic and Persian loans in Turkish?

Question

Both Persian and Turkish - as have other "islamic" languages - have a great deal of Arabic vocabulary. Due to an (initially) favourable vowel inventory (and maybe due to the same script being used) these loans have come to Persian in much of their original shape, while their Turkish counterparts do differ quite a lot. There seems to be some regularity, but I did not, up to now, figure it out completely.So the question is: how do (especially vowel) sounds change, when a word travels from Arabic to Turkish? How about the same thing for Persian words in Turkish? To understand these processes better: what was the general way of Arabic words coming to Turkish, I mean by way of which idiom or even language did Arabic words come to modern Turkey-turkish?Now what I figured out up to now is this:The short vowels a, i and u in Arabic can be realized in Turkish as a/e, i/ı and u/ü respectively. This accords with the vowel harmony and it seems, that there is some connection with a kind of "consonant harmony" which is familiar to me from Old Uyghur, to the effect that the dark vowel is chosen in circumstances of dark consonants like q, 'ayn and the retroflex l. And of course vice versa, the light vowel being used in appropriate consonant surroundings.Does anybody know more about that, since it does not seem to be the be-all and end-all?

Answer 1

The question posed is a difficult one to answer here using a few words, and perhaps requires a thesis or even book to answer completely. I am not an expert on the topic, but I have done some research on the borrowings from Farsi into Turkish. So, let me express what I discovered through my search. (If you are looking for a short definitive answer scroll down to the section in bold letters towards the end.)

One difficulty in addressing patterns in Arabic to Turkish borrowings is that Farsi was the intermediate language between the two long before Turkish came into contact with Arabic. That is, Turkish could have borrowed Arabic words from Farsi that was already altered in some fashion to fit the Farsi phonological inventory then reshaped again to fit the Turkish one. Indeed this did happen as Perry provides a great example of this process: ta-ziyat (Ar) > Ta’ziat (F) > Ta’zie (T) > ta’zie (F).

Another difficulty is that the state of the Turkish language spoken during the Ottoman era. I will quote a paragraph from one of my papers on this issue (Demir, 2017; unpublished):

"The Ottomans continued the tradition of using Farsi as the language of the literature. So much so that Farsi was a “sine qua non of Ottoman education” (Tietze & Lazard, 1967). By the end of the Ottoman era Farsi was (and had been) an integral part of the language. By 19th century written language of the Ottomans had become a mixture of the three very different languages from three different families: Turkish from Altaic, Farsi from Indo-European and Arabic from Semitic. Although the syntax of the language was very much Turkish at its core, the lexicon of the educated Turk was made largely of Arabic and Farsi words (Lewis, 2002). So much so that Hagopian devoted half of his Ottoman-Turkish Grammar book to “The Elements of Arabic and Persian” (Hagopian, 1907). Lewis states that by the end of the Ottoman, Turkish was “the only language that ever came close to English in the vastness of its vocabulary” (Lewis, The Turkish Language Reform: A Catastrophic Success, 2002). This hybrid of a language on the other hand was vastly unintelligible for the everyday subjects of the Empire, not only to Greeks, Arabs and many others, but also to Turks (Lewis, 2002)."

This suggests that the elite was literate enough in Farsi (and possibly Arabic) so that when borrowings occurred through this population, there is a good chance the phonology was preserved as much as possible. The common folk on the other hand probably borrowed them in the form that they could “hear” from Farsi speakers (categorical perception). There is indeed evidence for this from Tietze. I will provide another quote from my paper:

“One of the interesting findings in the data provided by Tietze is the regional variations observed both in the phonology and the semantics of some of the borrowings. For example, under the entry for Pers. /araqtʃ͡in/ ‘a handkerchief; a little cap’, he lists more than a dozen different regions and that many different meanings. Similar patterns of variation can also be observed in borrowings from Greek as well (Tietze, 1955).”

As I have stated I have not done any work on Arabic-Turkish borrowings, so I will leave that question to someone else. On the subject of Turkish-Farsi borrowings (and vice versa), there are existing studies. Perry has done extensive work on Farsi and some of his work can be used for this purpose (see my references at the bottom).

Tietze is another name that comes to mind that published a book directly addressing this issue. In reference to phonological change, I will use another quote from my paper:

“It is sometimes quite difficult to date the borrowings between the two languages because of the intensive and extended contact between each other and the intermediary languages such as Arabic. However, in some cases, the possibility of recent borrowing can safely be ruled out because of the retention of older forms of Persian phonemes in Turkish loans (Tietze & Lazard, 1967). Generally speaking, the phonemic patterns in borrowings from Farsi are largely predictable. Long vowels which are contrastive in Farsi but not in Turkish, became short in borrowed forms. That is, Persian /aː/ and /uː/ became /e/ and /y/; /eː/ and /oː/ became /i/ and /u/ in Turkish (Tietze & Lazard, 1967). In the following examples, one can observe these patterns along with effects for Turkish harmony both in rounding and fronting (Tietze & Lazard, 1967):

ú  /daːna/   ‘grain’  > /dene/     ‘grain’; a counting word akin to ‘many’

ú  /yaːraːn/ ‘well spoken person’, ‘friend’ > /yeren/

ú  /dʒ͡uːdʒ͡a/ ‘chicken’ > /dʒ͡ydʒ͡e/ ‘chicken’, ‘young chicken’

ú  /aːxur/   ‘stall for horses’> /axɯr/ ‘stall for herding animals’

Change was not limited to vowels. Following phonological processes and patterns can be observed on consonants as well (Tietze & Lazard, 1967):

Word initial and final devoicing:

ú  /xarsang/ ‘road not suitable for donky travel’; from Pers. xar ‘donkey’ and sang ‘stone’> xersek ‘stony area’

ú  /dahra/ ‘a sharp dagger’ > /tahra/ ‘a butcher’s knife’

We also observe de-voicing in environments we don’t expect:

ú  /kaːbin/    ‘matrimony’ > /kepin/    ‘marriage money’

ú  /zambuːrak/ ‘a cross bow’> /sempirek/ ‘bow or sling used by children’“

However, I must caution the reader regarding the samples provided in Tietze. As a native speaker of Turkish (and I suspect this is true for the most speakers of modern Turkish), I am not familiar with the great majority of these words. These words are either extinct in modern Turkish or exist only in regional dialects (which are also disappearing fast thanks to mass migration to Istanbul and wide spread use of media).

Nevertheless, I think you will find Tietze especially useful on the matter if all you are looking for is general patterns. Here are my references:

Hagopian, V. (1907). Ottoman-Turkish conversation-grammar: a practical method of learning the Ottoman-Turkish language. London & Heidelberg: Julius Groos.

Lewis, G. (2000). Turkish Grammar. New York: Oxford.

Lewis, G. (2002, February 11). The Turkish Language Reform: A Catastrophic Success. Istanbul: Jarring Lecture: Swedish Research Institute.

Tietze, A. (1955). Griechische Lehnwörter im anatolischen Türkisch. Oriens, 204-257.

Tietze, A., & Lazard, G. (1967). Persian Loanwords in Anatolian Turkish. Oriens, 125-168.

Perry, J. (2001). Ethno-Linguistic Markers of the Turco-Mongol Military and Persian Bureaucratic Castes in Pre-Modern Iran and India. In Mitteilungen des SFB “Differenz und Integration” 5: Militar und Staatlichkeit (pp. 111-125). Orientwissenschaftliche Hefte.

Perry, J. (2001). The Historical Role of Turkish in Relation To Persian of Iran. Iran and the Caucasus, 5(1), 193-200.

Perry, J. (2007). Persian Morphology. In S. K. Alan, Morphologies of Asia and Africa 2 (pp. 975-1019).