{"id":274,"date":"2008-06-01T22:37:06","date_gmt":"2008-06-01T22:37:06","guid":{"rendered":"http:\/\/localhost\/wp\/wordpress\/remarks-on-the-romanized-kurdish-alphabet\/"},"modified":"2008-06-01T22:37:06","modified_gmt":"2008-06-01T22:37:06","slug":"remarks-on-the-romanized-kurdish-alphabet","status":"publish","type":"page","link":"https:\/\/kurdishacademy.org\/?page_id=274","title":{"rendered":"Remarks on the Romanized Kurdish Alphabet"},"content":{"rendered":"
By Vladimir F. Minorsky<\/a>,
\nThe Journal of the Royal Asiatic Society,<\/i> London, January 1933, pp .<\/i><\/div>\n

Mr. C. J. EDMONDS’S "Suggestions for the use of Latin characters in the writing of Kurdish" merit the attention of all those interested practically and theoretically in Kurdish, for no one probably has had better opportunities for studying the practical side of the question than Mr. Edmonds in his surrounding of Kurdish intelligentsia.<\/p>\n

<\/p>\n

By Vladimir F. Minorsky<\/a>,
\nThe Journal of the Royal Asiatic Society,<\/i> London, January 1933, pp .<\/i><\/div>\n

Mr. C. J. EDMONDS’S "Suggestions for the use of Latin characters in the writing of Kurdish" merit the attention of all those interested practically and theoretically in Kurdish, for no one probably has had better opportunities for studying the practical side of the question than Mr. Edmonds in his surrounding of Kurdish intelligentsia.<\/p>\n

The inconvenient side of all Semitic alphabets is their disregard of vowels (not only short ones, but some of the long ones and the diphthongs). Those alphabets are sufficiently adapted to the languages for which they were invented and in which the consonantic frame (cf. Arabic, mostly tri-literal, roots) forms the real backbone of the word of which the basic sense is more or less recognizable from the consonantic symbols.<\/p>\n

This system is entirely unsuitable for languages with a developed vocalic system where vowels are not accessories of the consonantic frame but integral parts of the stem. In Kurdish d\u0101r “tree” and d\u016br “far" have nothing to do with each other in spite of their similar consonantic frame (d.r). Here the vowels make all the difference of the basic meaning, whereas the vocalic system itself is considerably complicated by the existence of \u0113, \u014d (> ü\u0113) which the Arabs in their own terminology call majh\u016bl, i.e. “unknown" to themselves.<\/p>\n

The Arabic script has been occasionally used for writing many different languages (Albanian, Turkish, Malay, numerous Caucasian, African, and Indian idioms and occasionally even Spanish and Serbian), but whenever the considerations of direct convenience of the writing were no more obscured by any reflexions of political and religious order, phonetic alphabets have triumphed all along the line (We leave for the moment out of the question such languages with developed literatures closely associated with Muslim (Arabic) culture, as Persian, for instance.).<\/p>\n

Nothing can be said against the special phonetic alphabets of long standing, such as Greek, Russian, Armenian, Georgian, well adapted to their object, but as the Latin script in the most widespread in the world and has reached the highest technical perfection in its printed form (artistic consistency ­of the outer form of the whole scale of signs, lack of confusion in characters, existence of different varieties of type), only Latin script comes into question when a new form of phonetic script is under consideration for a language just acquiring a literary importance.<\/p>\n

For the success of the reform in Kurdish it is essential, that the Latin alphabet should be utilized in its most simple form with as few additions of conventional signs as possible. In this respect Mr. Edmonds’s effort to remain within the possibilities of the ordinary type seems quite comprehensible and well founded. The Kurdish alphabet as a practical instrument need not aim at an absolutely rigorous application of the principles: "Each sound to have a single and non-compound sign, each sound to be pronounced only in one way For example, there is no practical inconvenient writing sh (\ufeb6) instead of the Czecho\u2011Slovakian š (whatever, its well\u2011known scientific convenience in connection the other special signs), or the Turkish \u015f (borrowed obviously from Rumanian).<\/p>\n

I should formulate the principles underlying Mr. Edmonds’s scheme as follows: \u2011<\/p>\n

Avoidance of any unusual signs which would embarrass the Kurdish presses.<\/p>\n

Use of double signs for "long" vowels [only in Mr. Edmonds’s first article!].<\/p>\n

Use of h after some consonants to connote some aberrant use of these characters.<\/p>\n

To these points I should add the desideratum of the slightest possible variance from the established use of the original Latin script. All alphabets are conventional and even if instead of a, b, c we write respectively k, l, m (as in some unsophisticated schoolboys’ cipher) it can be learnt after some practice, yet any queer functions of the familiar signs are apt to mislead the Kurds in the scientific study of their language in comparison with the other Iranian languages. In this respect the new Turkish alphabet, which gives a practical solution for local use, is certainly inconvenient for comparative purposes, such words as gelecek necessitating their re-transcription into gelejek, etc. It is likewise un­desirable to introduce new peculiar spellings for the words belonging to international scientific vocabulary.<\/p>\n

The following are my more detailed observations on, and suggestions in regard to, the systems proposed by Mr. Edmonds in his two articles which hereafter will be respectively referred to as E1 and E2.<\/p>\n

As regards the "long" vowels their exact duration as compared to that of the "short" ones may need some further investigation, but there is no doubt that the respective sounds of the two classes\u2011\u0101, \u012b, \u016b and a, i, u\u2011are felt as distinct phonemes, and, in the case of \u0101 and a, differ in timbre; \u0113 (closed sound palatalizing the preceding consonant) has no corresponding short sound; and o in dost and xosh (xwosh ?) (though entirely of distinct origin) seems to be confused in Kurdish while the typical treatment of the original long \u014d in Kurdish is the diphthong ü\u0113 (with palatalization of the preceding consonant), e.g. k’ü\u0113r (< k\u014dr) "blind", g’ü\u0113z < g\u014dz "nut". There is consequently no practical need for intro­ducing a distinction of \u014d and o but the sign ö (E 2) will be quite welcome as a comparatively simple conventional expression for ü\u0113, and find its justification in the etymological origin of this sound (from \u014d).<\/p>\n

Following the principle of reduplication of the characters in order to express the length of a vowel, I should write aa for Kurdish long \u0101 and leave simple a for its corresponding short sound. Such a system is one of the practical characteristics of the Dutch script. As a matter of fact, short Kurdish a sounds like ä (cf. English "man"), or even as a real short \u0103, while with the use of e (E1 and E2) we are distinctly drifting to a different class of sounds. The proposed use of aa and a will allow us to restrict the use of e to the real e (see above). This unique e will be written without any diacritical sign (as against E1 and E2: ê), just as in Sanskrit transcriptions e stands exclusively for a long \u0113.<\/p>\n

The signs ii and i are quite natural, but there exists in Kurdish a characteristic sound of an extra\u2011short i perfectly distinguishable on account of its dull timbre. It somewhat reminds one of Russian \u044b (Polish y) and Turkish \u0131 (i) in ald\u0131 (\u0627\u0644\u062f\u0649), but is a furtive intermediate sound which for an untrained English ear would perhaps resemble the vowel in "but". In E1 and E2 it is conveniently expressed by y (cf. Polish y), but it would be very desirable to reserve to y the obvious function of \u0649 (English and French y). One could think then of the new Turkish \u0131 (without dot), even the Turks admit now that this sign is conducive to confusion and seem disposed to replace it by ï. As we have obtained the elimination of one character with diacritic sign (ê) by ï [the special signs in our alphabet would consequently remain restricted to two: ï and ö] a simple one, we could afford to introduce in the present case ï [In E2 y has a threefold use for expressing consonantic y, short \u012d, and the length of \u012b (iy)], but perhaps it would be more advantageous to adopt for our case \u1ecb (with a dot underneath) which would be better distinguishable from both ii and i and in case of emergency could be easily improvised by the printers; it would suffice for them to place an ordinary i upside down.<\/p>\n

I should rather not follow E2 in transcribing \u016b by uw and \u012b by iy for the "Dutch" principle of doubling letters of the long sounds seems to me to possess all the advantages of clearness (), but I should admit the use of uw\u2011 and iy\u2011 in, the cases when the long \u016b\u2011 and \u012b\u2011, being followed by a vowel, phonetically become a group composed respectively of u + w or i + y. This orthographical rule would be conditioned in this special case by the phonetic modification.<\/p>\n

Coming to the consonants I should reserve simple j and c respectively for \u062c and \ufb7a in conformity with the very clearly established use (see the hallowed Sanskrit transcription) and the historical tradition of c which in all the systems derived from Latin stands for voiceless k, \u010d, or ts. The only exception is the new Turkish alphabet, but we have mentioned its philological inadequacy for scientific purposes.<\/p>\n

Zhand sh seem to be quite suitable expressions of \ufb8band \ufeb6 logically consistent with z and s for \ufeaf and \ufeb1<\/p>\n

The use of h as an auxiliary sign in lh and rh as differentiated from l and r is a happy idea already realized in Albanian script. Kurdish lh is a hard cerebral \u013c pronounced with the tip of the tongue upturned (a characteristic very distinct from Turkish and Russian hard \u013c (\u043b); rh is the rolled r pronounced with the tip of the tongue (a similar distinction between r and \u0157 exists in Armenian and Albanian).<\/p>\n

As regards the harsh guttural sounds, the use of x for \ufea5(as in Spanish, Greek, and Russian) would be consistent with the general scientific practice. As we connote the corresponding voiced \ufecd by gh, it was first suggested (E1) to express this sound with xh, but as \ufea5 is frequent in Kurdish the new simplification (E2) will be very welcome. On the other hand, Mr. Edmonds feels inclined to disregard the \ufea1 sound, occurring in Kurdish, and not only in Arabic loan\u2011words, but also in some purely Iranian words as \ufea4\ufeee\ufe96hawt "seven". This sound, though rare, is very characteristic of Kurdish and I should allot to it precisely the conventional xh, where \u2011h, following our practice, will indicate an aberrant use of the original symbol x.<\/p>\n

Contrary to the Turks and Persians, the Kurds very naturally pronounce \ufec9 (and prefix it even to such an Iranian word as asp "horse" which in Kurdish sounds (\u0639\u0633\u0628). It would be helpful to express \u0639 with an apostrophe whenever the Kurds pronounce it: ‘ajbat \u0639\u062c\u0628\u062a but there is of course no question of simply reproducing Arabic forms: if \u0639\u0628\u0627\u0633 and \u0639\u062b\u0645\u0627\u0646 are pronounced Habb\u0101s, and Watm\u0101n they will be spelt accordingly [In handwriting \u0639 could be expressed still better by spiritus asper ]. On the contrary, there is no need to transcribe the Arabic hamza in the beginning and at the end of words (\u0631\u062c\u0627\u0621 \u0623\u0646\u0633) though in the middle of words it would be helpful to express it by a hyphen \u0647\u064a\u0626\u062a hay\u2011at.<\/p>\n

Likewise no special mark of elision seems to be necessary, in such words as l\u0113r\u0101 <l’\u0113r\u0101, any more than in separating the locative ending \u2011da, but, if so desired, the same hyphen could be used for such purposes as well.<\/p>\n

We need not be more precise about Kurdish sounds, as time will show what particular nuances and sandhi phenomena will be discovered by specialists in phonetics. Under this ruling come the Sul\u0113m\u0101n\u012b spirants δ (\u0630) and θ (\u062b), which can hardly be considered as real phonemes and do not represent a general phenomenon even in southern Kurdish.<\/p>\n

It must be finally well understood that the suggested Kurdish alphabet has in view principally the convenience and development of printing. As regards the writing in Kurdish considerable simplifications will be introduced due course: for instance, double vowels aa, ii, uu will be easily replaced by some signs like \u0101, \u012b, \u016b or á, í, ú. Many people in Europe instead of double consonants still write only one with a dash over it (as a substitute for an Arabic tashd\u012bd). Kurdish orthography and calligraphy will follow their own ways, while we are trying to find some practical and simple solution of the fundamental problem of the basic alphabet.<\/p>\n

The following is the comparative table of Kurdish sounds as figured in Mr. Edmonds’s two articles and my additional remarks: –<\/p>\n

\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
\n

 <\/o><\/p>\n<\/td>\n

\n

A. VOWELS<\/b><\/p>\n<\/td>\n

\n

 <\/o><\/b><\/p>\n<\/td>\n<\/tr>\n

\n

 <\/o><\/p>\n<\/td>\n

\n

C.J. E. 1931<\/b><\/p>\n<\/td>\n

\n

C.J. E. 1933<\/b><\/p>\n<\/td>\n

\n

V. M. 1933<\/b><\/p>\n<\/td>\n<\/tr>\n

\n

\u0101<\/p>\n<\/td>\n

\n

a<\/p>\n<\/td>\n

\n

a<\/p>\n<\/td>\n

\n

aa<\/p>\n<\/td>\n<\/tr>\n

\n

\u0103 (ä)<\/p>\n<\/td>\n

\n

e<\/p>\n<\/td>\n

\n

e<\/p>\n<\/td>\n

\n

a<\/p>\n<\/td>\n<\/tr>\n

\n

\u0113<\/p>\n<\/td>\n

\n

e<\/p>\n<\/td>\n

\n

ê<\/p>\n<\/td>\n

\n

e<\/p>\n<\/td>\n<\/tr>\n

\n

\u012b<\/p>\n<\/td>\n

\n

ii<\/p>\n<\/td>\n

\n

iy<\/p>\n<\/td>\n

\n

ii<\/p>\n<\/td>\n<\/tr>\n

\n

\u012d<\/p>\n<\/td>\n

\n

i<\/p>\n<\/td>\n

\n

y<\/p>\n<\/td>\n

\n

i<\/p>\n<\/td>\n<\/tr>\n

\n

Ï (dull)<\/p>\n<\/td>\n

\n

y<\/p>\n<\/td>\n

\n

i<\/p>\n<\/td>\n

\n

ï (or) ?<\/p>\n<\/td>\n<\/tr>\n

\n

o<\/p>\n<\/td>\n

\n

o<\/p>\n<\/td>\n

\n

o<\/p>\n<\/td>\n

\n

o<\/p>\n<\/td>\n<\/tr>\n

\n

ü\u0113<\/p>\n<\/td>\n

\n

uy<\/p>\n<\/td>\n

\n

ö<\/p>\n<\/td>\n

\n

ö<\/p>\n<\/td>\n<\/tr>\n

\n

\u016b<\/p>\n<\/td>\n

\n

uu<\/p>\n<\/td>\n

\n

uw<\/p>\n<\/td>\n

\n

uu<\/p>\n<\/td>\n<\/tr>\n

\n

\u016d<\/p>\n<\/td>\n

\n

u<\/p>\n<\/td>\n

\n

u<\/p>\n<\/td>\n

\n

u<\/p>\n<\/td>\n<\/tr>\n

\n

 <\/o><\/p>\n

B) CONSONANTS (disposed by groups, means ”no changes”, and ? “not expressed”.)<\/p>\n

 <\/p>\n<\/td>\n<\/tr>\n

\n

b<\/p>\n<\/td>\n

\n

b<\/p>\n<\/td>\n

\n

b<\/p>\n<\/td>\n

\n

b<\/p>\n<\/td>\n<\/tr>\n

\n

p<\/p>\n<\/td>\n

\n

p<\/p>\n<\/td>\n

\n

p<\/p>\n<\/td>\n

\n

p<\/p>\n<\/td>\n<\/tr>\n

\n

v<\/p>\n<\/td>\n

\n

u<\/p>\n<\/td>\n

\n

u<\/p>\n<\/td>\n

\n

u<\/p>\n<\/td>\n<\/tr>\n

\n

f<\/p>\n<\/td>\n

\n

f<\/p>\n<\/td>\n

\n

f<\/p>\n<\/td>\n

\n

f<\/p>\n<\/td>\n<\/tr>\n

\n

w<\/p>\n<\/td>\n

\n

w<\/p>\n<\/td>\n

\n

w<\/p>\n<\/td>\n

\n

w<\/p>\n<\/td>\n<\/tr>\n

\n

d<\/p>\n<\/td>\n

\n

d<\/p>\n<\/td>\n

\n

d<\/p>\n<\/td>\n

\n

d<\/p>\n<\/td>\n<\/tr>\n

\n

t<\/p>\n<\/td>\n

\n

t<\/p>\n<\/td>\n

\n

t<\/p>\n<\/td>\n

\n

t<\/p>\n<\/td>\n<\/tr>\n

\n

δ (\u0630)         <\/p>\n<\/td>\n

\n

dh<\/p>\n<\/td>\n

\n

?<\/p>\n<\/td>\n

\n

?<\/p>\n<\/td>\n<\/tr>\n

\n

θ (\u062b)        <\/p>\n<\/td>\n

\n

th<\/p>\n<\/td>\n

\n

?<\/p>\n<\/td>\n

\n

?<\/p>\n<\/td>\n<\/tr>\n

\n

\u0135 (\ufe9d)<\/p>\n<\/td>\n

\n

c<\/p>\n<\/td>\n

\n

c<\/p>\n<\/td>\n

\n

j<\/p>\n<\/td>\n<\/tr>\n

\n

\u010d (\ufb7a)<\/p>\n<\/td>\n

\n

ch<\/p>\n<\/td>\n

\n

ch<\/p>\n<\/td>\n

\n

c<\/p>\n<\/td>\n<\/tr>\n

\n

k<\/p>\n<\/td>\n

\n

k<\/p>\n<\/td>\n

\n

k<\/p>\n<\/td>\n

\n

k<\/p>\n<\/td>\n<\/tr>\n

\n

g<\/p>\n<\/td>\n

\n

g<\/p>\n<\/td>\n

\n

g<\/p>\n<\/td>\n

\n

g<\/p>\n<\/td>\n<\/tr>\n

\n

\ufed6<\/p>\n<\/td>\n

\n

q<\/p>\n<\/td>\n

\n

q<\/p>\n<\/td>\n

\n

q<\/p>\n<\/td>\n<\/tr>\n

\n

h<\/p>\n<\/td>\n

\n

h<\/p>\n<\/td>\n

\n

h<\/p>\n<\/td>\n

\n

h<\/p>\n<\/td>\n<\/tr>\n

\n

\ufecd<\/p>\n<\/td>\n

\n

gh<\/p>\n<\/td>\n

\n

gh<\/p>\n<\/td>\n

\n

gh<\/p>\n<\/td>\n<\/tr>\n

\n

\ufea5<\/p>\n<\/td>\n

\n

xh<\/p>\n<\/td>\n

\n

x<\/p>\n<\/td>\n

\n

x<\/p>\n<\/td>\n<\/tr>\n

\n

\ufec9<\/p>\n<\/td>\n

\n

\u060c<\/p>\n<\/td>\n

\n

?<\/p>\n<\/td>\n

\n

\u060c or ,<\/p>\n<\/td>\n<\/tr>\n

\n

\ufea1<\/p>\n<\/td>\n

\n

x<\/p>\n<\/td>\n

\n

?<\/p>\n<\/td>\n

\n

xh<\/p>\n<\/td>\n<\/tr>\n

\n

l<\/p>\n<\/td>\n

\n

l<\/p>\n<\/td>\n

\n

l<\/p>\n<\/td>\n

\n

l<\/p>\n<\/td>\n<\/tr>\n

\n

\u013c<\/p>\n<\/td>\n

\n

lh<\/p>\n<\/td>\n

\n

lh<\/p>\n<\/td>\n

\n

lh<\/p>\n<\/td>\n<\/tr>\n

\n

r<\/p>\n<\/td>\n

\n

r<\/p>\n<\/td>\n

\n

r<\/p>\n<\/td>\n

\n

r<\/p>\n<\/td>\n<\/tr>\n

\n

\u0157<\/p>\n<\/td>\n

\n

rh<\/p>\n<\/td>\n

\n

rh<\/p>\n<\/td>\n

\n

rh<\/p>\n<\/td>\n<\/tr>\n

\n

m<\/p>\n<\/td>\n

\n

m<\/p>\n<\/td>\n

\n

m<\/p>\n<\/td>\n

\n

m<\/p>\n<\/td>\n<\/tr>\n

\n

n<\/p>\n<\/td>\n

\n

n<\/p>\n<\/td>\n

\n

n<\/p>\n<\/td>\n

\n

n<\/p>\n<\/td>\n<\/tr>\n

\n

z<\/p>\n<\/td>\n

\n

z<\/p>\n<\/td>\n

\n

z<\/p>\n<\/td>\n

\n

z<\/p>\n<\/td>\n<\/tr>\n

\n

s<\/p>\n<\/td>\n

\n

s<\/p>\n<\/td>\n

\n

s<\/p>\n<\/td>\n

\n

s<\/p>\n<\/td>\n<\/tr>\n

\n

\u017e (\ufb8b)<\/p>\n<\/td>\n

\n

zh<\/p>\n<\/td>\n

\n

j<\/p>\n<\/td>\n

\n

zh<\/p>\n<\/td>\n<\/tr>\n

\n

š (\ufeb6)<\/p>\n<\/td>\n

\n

sh<\/p>\n<\/td>\n

\n

sh<\/p>\n<\/td>\n

\n

sh<\/p>\n<\/td>\n<\/tr>\n

\n

y (\ufeef)<\/p>\n<\/td>\n

\n

y<\/p>\n<\/td>\n

\n

y<\/p>\n<\/td>\n

\n

y<\/p>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n

P.S.\u2011The above suggestions are based on the assumption that, for the facility of Kurdish printing, signs with diacritical points must be avoided as far as possible. On the other hand, as shown by the latest experiments in Erivan and Damascus, this practical consideration need not be over estimated. Under such conditions, a more liberal use of diacritical points would very likely represent a convenience and simplification in Kurdish writing.<\/p>\n","protected":false},"excerpt":{"rendered":"

By Vladimir F. Minorsky, The Journal of the Royal Asiatic Society, London, January 1933, pp . Mr. C. J. EDMONDS’S "Suggestions for the use of Latin [\u2026]<\/span><\/p>\n","protected":false},"author":1,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"footnotes":""},"categories":[],"tags":[],"class_list":["post-274","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/kurdishacademy.org\/index.php?rest_route=\/wp\/v2\/pages\/274","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/kurdishacademy.org\/index.php?rest_route=\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/kurdishacademy.org\/index.php?rest_route=\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/kurdishacademy.org\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/kurdishacademy.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=274"}],"version-history":[{"count":0,"href":"https:\/\/kurdishacademy.org\/index.php?rest_route=\/wp\/v2\/pages\/274\/revisions"}],"wp:attachment":[{"href":"https:\/\/kurdishacademy.org\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=274"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/kurdishacademy.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=274"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/kurdishacademy.org\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=274"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}