Ottoman Turkish in Armenian Script.

It’s been a while since I’ve featured Poemas del río Wang, but the latest post is right up my alley:

The Kumbaracı yokuşu, that is, “Bombardier descent” runs down to the always crowded İstiklal Avenue at its end near the sea, not far from the Passage Oriental, which housed the Café Lebon, the once famous café built in Art Nouveau style by the Istanbul-born French architect Alexandre Vallaury, not long after returning home from his studies in Paris. There were several similar passages on the İstiklal, the former Grand Rue de Péra, the main avenue of the European quarter of Istanbul from the Galata Tower up to Taksim Square, some of them are still open nowadays.

But if you also wander into the small streets and alleys opening from the İstiklal, you can find other, more neglected heralds of old Istanbul, a world gone almost a hundred years ago. On the Kumbaracı, not far from the fountain of Miralem Halil Ağa built as a pious gift in 1729, there is an interesting fin-de-siècle house. Arriving from the İstiklal, the French inscription on the left side of the doorway catches the eye first: “Fabrique et dépôt de meubles”, furniture factory and depot. The inscription on the right side is indecipherable, but the ones on the street front are mostly still there, defying time and weather, advertising the wares of the former owner in three languages and three different scripts.

The one on the left side seems to be the most interesting of all of them. The script is Armenian, but the language is Ottoman Turkish: ՄԷՖՐՕՒԶԱԹ ՖԱՊՐԻՔԱՍԸ mefroizat fabrikası, in modern orthography mefruşat fabrikası, “furniture factory”. It may sound strange today, but Ottoman Turkish was often written with Armenian script until the alphabet reform in 1928, after which the Latin script has been used for Turkish – even the very first Turkish novel, the Akabi’s story was published in Armeno-Turkish script in 1851. For most people it was easier to learn and the language itself could be rendered more precisely than in the otherwise used Ottoman Turkish script, a modified version of the Perso-Arabic alphabet. Precision depended on the language user him/herself too, however. In the inscription of the furniture factory two peculiarities can be observed: first, the ՕՒ oi standing in ՄԷՖՐՕՒԶԱԹ mefroizat instead of the properly used Ու u, which seems to be an influence of Greek. Then, the use of Ք k in ՖԱՊՐԻՔԱՍԸ fabrikası is quite uncommon as it usually stands before front rounded vowels. Before back vowels its almost mirrored counterpart, a Գ should be used (the difference between the two might be more palpable if one looks at their counterparts in the Ottoman script: ك and ق‎, respectively).

It turns out I actually posted about Armeno-Turkish in 2017, but in the interim I forgot all about it, so this is a useful reminder! It’s well worth visiting the río Wang post for the gorgeous images; I will make one pedantic correction: İstiklal Avenue is not “at its end near the sea” but at the upper end; the one near the sea is Kemeraltı Caddesi, as you can see from the map.


  1. ə de vivre says

    I wonder how the sign maker decided which letter to use for the ‘k’ in ‘fabrika.’ A cursory wiki search indicates that Armenian doesn’t use ‘fabrika,’ so it wouldn’t be an influence of standard spelling. If they were improvising from the spoken language, ք and գ were homophonous /kʰ/ in Western Armenian, so maybe it was a random selection (en passant: voiceless aspirated for front vowels, voiced for back – interesting!). Or were they also literate in Arabic-script Ottoman, and were influenced by ‘fabrika’ being an Italian loan? I can’t find any references to how ‘fabrika’ was spelled in Ottoman. Nişanayan gives a normalized “Avrupa emtiˁāsının iˁtibārına ise fabrikalar sāyesinde …,” which doesn’t help much, and the online dictionaries only give ‘kârgâh’ as the Ottoman equivalent of modern Turkish ‘fabrika,’ perhaps because it’s the one that would be opaque to a Modern Turkish speaker. The mind boggles!

  2. Good questions all, and I join you in the boggling!

  3. In general, /k/ in Ottoman Turkish was spelled with a qaf in non-Arabic words. In general, kaf was used for the palatalised k.

  4. John Cowan says

    Similarly, ק qof is used for /k/ in non-Hebrew words, which means that it is the normal way of writing /k/ in Yiddish (Hebrew borrowings are written exactly as in Hebrew).

  5. ə de vivre says

    In general, /k/ in Ottoman Turkish was spelled with a qaf in non-Arabic words. In general, kaf was used for the palatalised k.

    Right, which usually corresponds to the front/backness of the following vowel (Ottoman Turkish also used different letters to represent consonants that didn’t have palatalized allophones, with the emphatic one getting the back vowel, which is why I mentioned vowel quality rather than palatalization). You’d expect a qaf in fabrika, but since the Armenian alphabet spelling doesn’t follow the way they usually express the same distinction, and without being able to find any examples of Arabic script fabrikas, I wondered if there was something specific about this word that made it not follow general rules.

  6. The author of the post at Poemas del Río Wang writes ՄԷՖՐՕՒԶԱԹ. This would be transliterated as mēfrōwzatʿ in the Hübschmann-Meillet system. (I have transliterated in lowercase letters for simplicity.) However, the transliteration mefroizat (i.e. mēfrōizatʿ in the Hübschmann-Meillet system) that the author gives in his post would instead correspond to an Armenian ՄԷՖՐՕԻԶԱԹ with Ի, not Ւ. There is some confusion of uppercase Ւ (transliterated w) with uppercase Ի (transliterated i) in the post.

    I don’t understand the point about Greek influence. I would not expect Greek spelling habits to introduce an Armenian ՕԻ oi in this word, since Modern Greek οι spells [i], not [u]. If there is any Greek influence, is it in the use of Օ rather than Ո in the sequence ՕՒ ōw in the author’s rendering ՄԷՖՐՕՒԶԱԹ? The vowel [u] is spelled ՈՒ ow in Armenian, in the same way that [u] is spelled ου in Greek with an omicron. Is the author suggesting that the appearance of Armenian Օ (rather than Ո) here is due to its resemblance to the Greek omicron in the digraph ου?

    The lower portions of some letters in the Armenian text are damaged. Are the extant traces consistent with ՄԷՖՐՈՒՇԱԹ mēfrušatʿ, with the digraph ՈՒ ow transliterated as Roman u and representing the vowel [u] in Armenian?

    In any case, there seems to be a clear Շ (š) instead of Զ (z). This reading of the Armenian as ՄԷՖՐՈՒՇԱԹ would correspond exactly to Ottoman مفروشات and Republican Turkish mefruşat. For the particular shape of uppercase Շ (š) in the typeface of this inscription (somewhat resembling the Chinese character 己 ), compare for example the 20th-century shape on the lower left in this image taken from the Armenian Soviet Encyclopedia.

    I can’t find any references to how ‘fabrika’ was spelled in Ottoman.

    As for the spelling of Ottoman فابریقه fabrika that ə de vivre wonders about, we can consult the entries in several Ottoman dictionaries, such as the Lehçe-i Osmânî (1876), and in the Kamûs-ı Türkî (1901), and in the Kamûs-ı Osmânî (1910).

  7. Fantastic comment, thanks!

  8. Somewhat OT but close enough for a comment, I think-
    The truly wonderful documentary Honeyland, which has been nominated for Academy Awards in two categories – Best International Feature Film and Best Documentary – has a protagonist who ives in rural North Macedonia and, from what I’ve read, speaks Turkish to her aged mother and Macedonian to everyone else. Anyone with an interest in the Turkish spoken in the Balkans might find it worthwhile from that angle, aside from its extraordinary merits as a film.

  9. I’ve heard good things about it, and that makes me want to see it even more.

  10. Very interesting, thanks for that!

  11. Zabel Yesayan is the greatest Istanbul-born writer most Turks never heard of.

    Judgments of comparative literary merit are scurvy things, but surely that accolade goes to a Constantinopolitan Greek?

  12. Who did you have in mind?

  13. None in particular, but since there were Greeks writing there for almost two millennia, and without doubt Turks have heard of few or none of them, it seems less than probable that an armenophone however great is actually The Greatest.

  14. That’s like saying “I don’t believe Shakespeare is the greatest English writer, because there have been many more English writers who weren’t Shakespeare than who were, so surely it’s more probable that the greatest is one of them?”

  15. Yes, much as I admire, say, Michael Psellus I wouldn’t assume he’s “greater” than a writer I don’t know. He wasn’t that great.

  16. Perhaps Cavafy would appeal more to you as a candidate.

  17. Except 1) he was from Alexandria, and 2) Turks have heard of him.

  18. Apart from those trivial details, comrade, you are of course perfectly correct!

Speak Your Mind