alt=""/>
ARABIC-SCRIPT LANGUAGES AND THE WEST.
From the eighth century, throughout most of the Near East, Central Asia, and north India, Arabic script was widely used for literary languages (including Arabic, Persian, western and eastern Turkish, and Urdu). The Arabic “alphabet” is deficient, in that it does not have characters for most vowels. It is particularly unsuited to representing the eight-vowel system of Turkic languages.
Literacy was not—and still is not—widespread in rural and nomadic areas. Thus the early Western rug-collectors had little more than their ears to work with; and since these were attuned to English, French, German, or Russian sounds, the resulting transcriptions of native terms left much to be desired. All the more since the orthography of their own languages (particularly English and French) was—and still is — a chaos of historically conditioned letter-combinations. The final stage of confusion is reached when we try to interpret another’s transliterations in ignorance of the conventions of the original or an intermediate language.
For example, the sound represented in English by j, as in judge (or by dge, of course!) also occurs commonly in Arabic, Persian, and Turkish. It does not occur natively in French, German, or Russian, where it has to be represented by unfamiliar letter-combinations. Thus an Iranian of Central Asia, properly known in English as a “Tajik,” is in German Tadschik, in French tadjik, and in Russian appears with the same dj combination written in Cyrillic characters. Since it was through the Russians that the Western world was introduced to the peoples of Central Asia and the Caucasus, the Russian form was transliterated into English; and since there was no “French j” sound in English, the Russian letter following d was represented as zh—hence the “English” form Tadzhik, with three letters for the value of one. The same cycle of analysis and resynthesis of one sound produces the forms Azerbaidschan, Azerbaidjan, and Azerbaidzhan for “Azerbaijan.”
Similarly, our perception of a Berber term from North Africa may have been filtered first through a local spoken form of Arabic and/or literary Arabic, then through written French (e.g. “Ouaouzquite,” to be pronounced approximately wa-ooz-keet). English itself is notorious for its spelling traps, as the tourist visiting Cirencester (sisister) or Godmanchester (gumster) can attest. Thus a short “a” vowel common in north Indian languages sounds like the English vowel of such words as sung, luck, but, hurry. English speakers therefore represented this Indian vowel as “u” in writing Lucknow, suttee, curry, etc. Lucknow, luckily, preserves the context of the substitution; the unfamiliar “dhurrie” (pronounced darry) is likely to be mispronounced as doory.
To pronounce a term with confidence, it helps to know the word’s linguistic lineage. This is not always possible, but the table of systematic correspondences of sounds and spellings on the previous page will guide the reader in making an educated guess.
It can be seen from several of the examples how, once the language context has been established from one orthographic feature, other conventions become clear. Since in “Chichaoua” (a town in Morocco) ch must be French for sh, the ou lurking between two other vowels must represent w, to yield a pronunciation shishawa. German-processed “Zejwa” and “chalat” will respond to similar reasoning. Occasional words imported into Italian or Spanish will, of course, follow the conventions established for, say, e and g in those languages.
TURKISH.
Since 1928 Turkish (in the narrower sense of the term, i.e. the language of the Republic of Turkey) has used the Latin alphabet with a completely regular orthography. Those few letters that do not approximate English usage are as follows.
c: j as in jog, e.g. cicim.
ç : ch as in chip , e.g. Çanakkale.
ı: e as in the or i in sir , e.g. Topkapı.
ş : sh as in ship , e.g. Uşak.
ğ : silent; lengthens the preceding vowel, e.g. Niğde (nee-deh).
ö: approx. as in nurse , with lips rounded; German böse , French peu , e.g. Gördes (=Ghiordes).
ü: approx. as in few , with lips rounded; German für , French lune , e.g. Yörük.
In Turkish words, each syllable is pronounced separately: Tekke is tek-keh. The vowels a, e, i, o, u are pronounced approximately as in Spanish.
The designation “Turk.” for the language of origin generally means modern Turkish or its ancestor, Ottoman Turkish. However, it may also stand for “Turkic,” an adjective embracing the closely related language family of the Azerbaijani Turks, Turkmens, Uzbeks, Kazakhs, Kirghiz and Qashqa’i. A term designated as “Turk.” may be common to two or more of these varieties of Turkish, and particularly in its transcribed form it may not be possible to assign it a definitive origin. Most of these languages are spoken in the Russian Caucasus and Central Asia, and were originally transcribed with the Cyrillic alphabet: see the “Russian” column in the preceding table.
Some of the Turkic languages have consonants which may be written as q, kh, and gh (see under Persian). These are usually equivalent to the modern Turkish k, h, and ğ respectively (e.g. qanât, kanat). The vowel y in Turkic words transliterated through Russian is equivalent to the undotted i (e.g. “asmalyk”).
PERSIAN AND ARABIC.
Persian was for many centuries both the principal literary language and the spoken lingua franca of the eastern Islamic world, supplementing Turkic, Kurdish, Pashto, Baluch, and other vernaculars. It is still widely used in Central Asia, Afghanistan, and Pakistan as well as Iran. As such it has borrowed freely from other local languages and in turn provided them with vocabulary of its own and many terms taken from Arabic.
Most terms of Arabic origin that appear in this lexicon have been “processed” through Persian. Persian still uses a modified form of Arabic script, and systems of transliteration are almost identical for the two languages. It is thus convenient to treat them together from the viewpoint of unfamiliar sounds and spellings.
dh: (Arabic only) like th as in the, wither, e.g. dhar’ (see “zar”).
gh: like French uvular r, but produced even farther back in the throat (Arabic, Turkic langs, e.g. Ghormaj). In Persian of Iran it is pronounced the same as q, e.g. ghermez.
kh: like ch as in German Bach , Scottish loch . E.g. Khila, Bakhtiari.
q: like k, but farther back in the throat, e.g. Qajar.
â: In Persian, long “a” as in raw or war; to be distinguished from a , as in bat , e.g. Râvar, kenâreh. In Arabic and Turkish, generally the long equivalent of a , e.g. salât, lâle.
The apostrophe is used to represent one of two letters sounded differently in Arabic, but pronounced identically in Persian, as a slight hiatus between vowels (Sa’idâbâd) or between vowel and consonant (Za’farânlu). The vowels a, e, i, o, u are pronounced approximately as in Spanish. The other letters and combinations used are roughly equivalent to English usage or otherwise self-evident.
Adjacent letters found in transliterations of Turkish, Persian, etc., may sometimes stand for separate letters (hence distinct sounds) and not the combinations listed above: thus Akhisar is Ak-hisar, Mashhad is Mash-had.
FORMATIVES.
