jump to navigation

Phonology of Yaghnobi July 25, 2007

Posted by آستان in Phonology, Yaghnobi.

Concerning phonology of Yaghnobi language, there are many aspects not clearly solved yet. The only available work about this problematique is an old article by V. S. Sokolova in Očerki po fonetike iranskix jazykov, unfortunately I haven’t seen this work yet. From two main articles by A. L. Xromov I tried to compile the basis of Yaghnobi phonological structure. I tried to add some actual observations I made during the Summer School of Yaghnobi in Ravenna – some features I consulted with Sayfiddin Mirzozoda to improve my knowledge of present phonology of the language.

The main difference in the system as was presented by Xromov is a different pronunciation of pharyngeals, they do not keep its original Arabic sounds but merge with sounds native to the language. Other interesting feature is that difference between short and long vowels seams to dessapear slowly, instead of quantity the quality is becoming more important – Vinogradova notes, that in present language vowel quantity is of less importance and a short vowel can become long under the influence of stress. Also an /ү/ sound seams to be only a dialectal feature instead to be a part of the vocalic system of “common Yaghnobi”

The result of my try can be seen here. An attempt to unite the systems of transliterations used for Yaghnobi is available here.


A. L. Xromov: Jagnobskij jazyk. Moskva 1972.

A. L. Xromov: Jagnobskij jazyk. In: V. S. Rastorgujeva (ed.): Osnovy iranskogo jazykoznanija. Novoiranskije jazyki II – Vostočnaja gruppa. Moskva 1987; pp. 644-701.

V. S. Sokolova: Jagnobskij jazyk. In: V. S. Sokolova: Očerki po fonetike iranskix jazykov. Moskva – Leningrad 1953, pp. 59-57.

S. P. Vinogradova: Jagnobskij jazyk. In: Jazyki mira. Iranskije jazyki III – Vostočnoiranskije jazyki. Moskva 2000, pp. 290-310.

Update: I updated the tables according to present state of IPA, the only changes I did are the phonetic symbols for /š/ and /ž/ and a reinterpretation of phonology of /w/.


1. Bahrom - July 26, 2007

Hi Luboss, thanks for this post. It’s great to see someone else working on the phonetcs and phonology of Yaghnobi, since this is not my area of expertise.

I have a couple questions about the paper on phonology. How much is a summary of Xromov, and which parts are based on your own observations? How did Xromov/you determine the phonetic segments- did you do any spectrographic analysis? I assume Xromov’s work was all done by ear. How did Xromov/you determine the phonemes? Were any minimal pairs or complementary distributions found? If so, it would be very interesting to see them.

In the paper on orthography, what does the abbreviation “UNI” mean?

Thanks again for these interesting papers!


2. آستان - July 26, 2007


concerning the system of Xromov I do not know anything about the origins of his observations, what I asked Sayfiddin he couldn’t answer me neither. My analysis was, unfortunately, done only by ear with some help of my colleaguaes in Ravenna.

The basis of the phonological transcription in from Xromov, I myself added only few comments:

1) Aspiration of stops – in this case I also consulted with Sayfiddin and we agreed that /p, t, k/ are aspirated word initially, aspiration word finally is questionable, but I argeed on this fact with my colleagues.

2) The notes about plausible palatalisation of /t, d/ word initially in front of (stressed) /ī/ is only my observation, actually in this case palatalisation is only facultative, but in this case I’m sure that palatalised /t, d/ can appear in Yaghnobi while I’m familiar with those sounds from my native languages (Czech, Slovak).

3) Of the pronunciation of pharyngeals I asked Sayfiddin and he told me, that in (present) Yaghnobi actually does not appear “ha hutti”, it is pronunced as laryngeal. Concerning “ain” Sayfiddin told me there is no such sound in Yaghnobi (exept one word), but I found it in some words. Actually “ain” is pronunced not like in arabic but in a similar way as in Tajik (as “hamza” or as a small break in speech).

4) Problematique of /w/ seams to be the most problematic question. What I and my colleagues heard there is no pronunciation of /w/ in a similar way as it is realised in English. Xromov notes, that /w/ is a bilabial approximant, Vinogradova notes that it is a bilabial fricative. This I can’t tell, but in this case I’d bellieve both authors – Xromov notes the pronunciation as he heard it in the 1950′-1970′, Vinogradova probably notes the pronunciation as she heard in by the end of 1990′ or this can really differ in dialects. What I noticed (but I’m not so sure), Sayfiddin used both variants of /w/ in his speech – in soem cases it was a fricative, in some other an approximant, sometimes one word was pronunced differently.

6) I do not accept the /ү/ (ü) sound to be an integral part of the language – Sayfiddin told me that this is only a dialectal feature, similar fact note Andreyev and Peshchereva.

What is written about the vowels is reproduced from Xromov’s works. In note #2 I mention “current observations” – this is what I thing is happening with Yaghnobi long vowels – they may be pronunced as half-long. But probably the information given by Vinogradova is more precise – the distinction between long and short vowels in Yaghnobi dissapears, all vowels appear as short, but undre the influence of stress they are prolonged. This fact is according to her caused by the influence of Tajik, but this appears aslo in other modern Iranian languages as I was told.

Concerning a spectrographic analysis I didn’t do any, I hope I will do it in autumn after my return from Tajikistan, at present all facts presented by Xromov and/or Vinogradova were “corrected” by my ears – unfortunately I’m not a phonetician, but many facts I tried to consult with some other people. Later I hope I’ll be more clever.

Finally, in the “orthographical table” the field UNI means a “unified” orthography I use whithin my Yaghnobi studies, it is very similar to the system used by Xromov.


3. Bahrom - October 22, 2007

Hi Luboss,

Thank you for the detailed answers!

I have a few more questions:
1) Khromov (1972) describes the sound represented by ɣ in his system of transcription (ғ in Tajik Cyrillic, or /gh/ in English transilterations of Yaghnobi) as a voiced uvular fricative. I had been transcribing this as a voiced velar fricative which is represented by ɣ in IPA (hence some confusion). Do you know if other linguists agree with Khromov on this, or do others describe this sound as velar? Should I be transcribing this sound as ʁ (a voiced uvular fricative) in IPA?
2) This is a similar question. I have been transcribing the sound represented by Cyrillic х as a voiceless velar fricative (x in IPA), but Khromov describes it as a voiceless uvular fricative, which is x in his transcription system, but χ in IPA. Is there agreement on this, or do any other linguists (yourself included) consider this to be a velar fricative (x in IPA)?
3) In the entries in your chart for basic realization, like the one for/i/: “Short vowel in basic realisation, wide spectre of articulation variants [i – ɪ – e]”, does this mean that these three sounds are in free variation?
4) You give the symbol ʕ for the realization of ‘ayn. This is the IPA symbol for a pharyngial fricative, but I would have expected this to be realized in Yaghnobi as a glottal stop, ʔ in IPA, or as you said, just a pause [.]. Did you intend to use the symbol for a pharingial fricative, or is this a typo?
5) Does Khromov’s 1987 article add to, or revise any of the material in his 1972 book? (I haven’t seen the article.)

4. آستان - October 23, 2007

Hi Brian,

ad questions 1 & 2) both /γ/ and /x/ are uvular fricatives as it is noted by Xromov, IPA [ʁ] and [χ] – the transcription used by Xromov is so called common iranist system of transcription of Iranian languages used in USSR. In this case I agree with Xromov as I listened to speakers of Yaghnobi also in the Yaghnob valley…

ad 3) variants of /i/ (and also of /u/ and in a way of /a/) can be in free variation as I understood, but I think the main influence on allophonic pronunciation is influenced by the neighbouring consonants. In this case I’m not so certain, but according to the pronunciation I heard in Yaghnob the variants depend mainly to these conditions…..

ad 4) realisation of ‘ayn: IPA symbol [ʕ] is used to represent the sound in realisation as it should be pronounced, but many speakers do not use this sound (f. ex. I heard a relative of Sajfiddin from Zumand pronouncing zaعī́fa [zæˈʕiːfa]), many speakers of Yaghnobi use glottal stop [ʔ] but it is usually realised just as a pause [.] – problem of ‘ayn in Yaghnobi is made by the cause that is is a foreign sound and its realisation may differ also according to Tajik realisation in a Tajik dialect in vicinity of the Yaghnobi speakers.

ad 5) Xromov 1872 and 1987 present the same description of Yaghnobi phonology, in many aspects it seams that Xromov used the earlier version to describe the phonology in the article of 1987, many sentences are the same, all examples used are the same. Better article is that from Vinogradova, but it is not so detailed as Xromov…

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: