During the course of the 20th century, the official alphabet of the Azerbaijani language changed multiple times— from an Perso-Arabic script used before 1922, to one version of a Latin script used from ’22-’39, and then several iterations of Cyrillic between ’39 and ’91, and finally a new Latin script after the fall of the Soviet Union.

Every day in Baku, though, I walk past this sign near my apartment: it says, “Open Door,” written in Cyrillic Azerbaijani on a piece of cardboard (гапы ач). The author of this sign feels more comfortable writing in the Cyrillic alphabet than the Latin one, almost 25 years after the official shift. And, at the Akhundov National Library of Azerbaijan where I’ve been working this year, large parts of the Azerbaijani paper catalog are, understandably, still in Cyrillic Azerbaijani. These changes take time, although the Latin Azerbaijani publication industry has exploded in recent years.

In any case, I’ve been organizing a database of Azerbaijani archaeological sites for my dissertation, and find myself in a thicket of toponym transliteration–  developing a standard list of place names that I’ll use in the dissertation, drawing on publications of archaeological sites in Cyrillic Azerbaijani, Latin Azerbaijani and Russian. Between all the alphabets and the related transliterations and variations of names, it gets pretty complicated.

Here’s a simple example. There is a well-known site from the Shamakhi region excavated by Fazil Osmanov in the 1960s. The site is known by the following variant spellings (occasionally more than one spelling appears in a single publication!):

• Нүјди or Нүйди (Cyrillic Azerbaijani)
• Нюди or Нюйди (Russian)
• Nüydü or Nüydi (Latin Azerbaijani)

Or, take for example one of the most famous sites in Azerbaijan, excavated in advance of the construction of a hydroelectric dam along the Kura river:

• Минҝәчевир (Cyrillic Azerbaijani)
• Мингечаур or Мингячевир or Мингечевир (Russian)
• Mingəçevir (Latin Azerbaijani)

And there are more complicated examples. Place names that use the Azerbaijani Cyrillic letters К, Ҝ, Г, and Һ  cause particular problems.

Add to these alphabet changes the frequent use of Russian versions of place names– even occasionally in Azerbaijani texts– and the habit of renaming cities both during the Soviet period and in the post-Soviet transformation, and you end up with a muddy world of place name variations.

For my database, I’ve been assembling a concordance so that I can track archaeological sites through their various name spelling changes, and I’ve also been deciding which transliteration to use in my own English writing. There are a few different transliteration systems for Azerbaijani, but there’s no perfect system for the complex environment. I’m striving to add the least possible confusion– especially for an audience likely unfamiliar with Turkic, Slavic (and, since I also have Georgian sites, Kartvelian) languages. I’m using a modified transliteration of the most common or most widely-accepted modern Latin Azerbaijani place name for each site, and ignoring the history of Cyrillic Azerbaijani and Russian forms. I’m replacing letters not in the standard Latin alphabet, like ə and ı with rough phonetic equilvents (a and i), and I’m replacing false cognates like x and c with transliterations that reflect pronunciation (kh and j). So, the first site above becomes Nüydi, and the second Mingachevir.

I’m still tweaking the system, and I’m not sure that I’ll ever be really happy with it. But, I actually think that this complexity is an important part of the story of the region. It speaks to the layered nature of life in the South Caucasus, where overlapping historical and linguistic narratives continue to shape daily life.