language

Macedonian, which is normally written in the Cyrillic alphabet, has a standard Romanization. It’s a very similar system to that used by Serbian, which is perfectly reasonable since Macedonia has long felt and to a certain extent continues to feel strong Serbian influence over its popular culture, politics, and academics.

Serbian represents its Cyrillic letters that don’t have simple Latin equivalents by using the haček, the word for a diacritic that conveniently enough appears within the word itself.

The haček was invented by Jan Hus, a Czech religious dissident of the 1400s who had terrible timing. He took on the Catholic Church about ethical abuses a hundred years before Europe was really ready for the Protestant Reformation. The Inquisition burned him at the stake.

But before he went he wrote De Orthographia Bohemica, a treatise on spelling the Czech language that suggested the rule “one letter for one sound” and proposed diacritics, including the little wedge haček, as a way of extending the Latin alphabet to accommodate the fact that most European languages have more meaningfully distinct sounds than there are simple letters in the original writing of the Romans. He specifically wanted to extend the number of consonants, which Slavic languages are famously rich in. Adding the haček to consonants such as c and s allowed him represent sounds that in English we need two letters (or more, sometimes) to represent, namely ch as in check and sh as in shoot.

Jan Hus died, but the haček lived on, and it was adopted by Vuk Karadžić, Ljudevit Gaj, and Đuro Dančić in the 1800s as they sought to create a standardized Serbo-Croatian language that could be written in either Latin or Cyrillic.

Gaj, in proposing a Croatian orthography, followed the “one letter for one sound” principle faithfully except in the cases of the Cyrillic letters њ and љ, which Gaj chose to represent with the two-letter combinations of nj and lj. This follows from the fact that “j” represents a palatal glide (like the “y” in yes), and the Cyrillic letters mentioned above are really just palatized versions of n and l. Otherwise, correspondance was one-to-one and modeled on Czech, so the haček became part of Serbo-Croatian’s Latin orthography.

Serbo-Croatian as a standard language was dying due to fights between Serbian, Croatian, and Bosnian academics well before the breakup of Yugoslavia, but the Yugoslav Wars slammed the lid shut on the coffin and left the language pushing daisies. The standard Croatian language, which does not differ very radically from standard Serbian, but whose differences are fiercely protected by modern Croatian academics and some politicians, uses only the Latin alphabet.

Standard Serbian, however, still uses both Latin and Cyrillic. They are both taught to school children and you can see them both in use on the street. Latin has a more modern feeling to it, showing up in advertisements, glossy magazines, and shop signs. Cyrillic has a more academic, perhaps national or religious feeling to it, and thus is used for serious books, public signs, and things related to the Serbian Orthodox Church.

Standard Macedonian from the beginning was written just with Cyrillic, a form of Cyrillic very close to that used by Serbian except for a few letters. The dominance of Cyrillic in Macedonia continues to this day, where in contrast to Serbia it is very rare to see Macedonian written with Latin letters. Advertisements for stylish new boutiques and jetsetting credit cards that would be written in Latin in Serbia are still written in Cyrillic in Macedonia.

There is a standardized Romanization scheme for Macedonian Cyrillic, and it is quite similar to Serbian Latin, hačeks and all, but there is the pesky matter of the Cyrillic characters ќ and ѓ. In equivalent Serbian words, these are written with the Latin characters ć and đ, but ć and đ do not represent quite the same sounds as ќ and ѓ. Instead, the standard Latin transcription scheme for Macedonian calls for ǵ and ḱ, letters that your browser might not even be able to render. What you should see is a “g” with an accent mark on top of it and a “k” with an accent mark on top of it.

So there is a way of writing Macedonian in Latin. People simply don’t use it very often, and almost never if they have to transliterate something that continues ѓ or ќ. The Latin equivalents ǵ and ḱ are extremely rare. You’re not going to find them on typewriters, computer keyboards, or among the basic international character sets of Windows or OS X. No one without specialized Slavistics or linguistics training is going to know how to pronounce those two characters at first sight. It’s easier just to write in Cyrillic.

There are many circumstances these days when you can’t write in Cyrillic, though. So what do you do then?

Computers and cellphones are designed mostly with English-speakers or Western Europeans in mind and until recently they did not have good support for Cyrillic. If you wanted to write on these devices in your native language, and your native language uses only Cyrillic, you had to make due with Latin letters. Even today, when Cyrillic support is much more widespread, it can be finicky to turn it on. Even when it’s easy, the habit of writing Cyrillic-based languages in Latin has become deeply rooted in the younger generations and there’s just no way to weed it out now.

For Russians, the solution was to use the Latin letters that look similar to Cyrillic letters (like T, M, C, and K, which when capitalized are the same in Cyrillic and Latin) and combine Latin letters, numbers, or symbols together to make visual approximations of the leftover Cyrillic letters (like 3, 4, *, and W, for З, Ч, Ж, and Щ). The ad hoc writing system is called Volapyuk, a joking reference to the 19th-century constructed language Volapük that Russians considered equally as queer-looking as writing “*EHW,NHA” for “ЖЕНЩИНА”.

Macedonians didn’t have to make Latin letters look like Cyrillic letters, since they already had Serbian as a reference point for a standard way of Romanizing Macedonian Cyrillic. But there was (and still is) the problem of the hačeks (on č, ž, and š), ǵ, and ḱ. How to represent these letters?

The Macedonian youth’s answer to this question: don’t.

All of those letters are simply written c, z, s, g, and k, or very rarely gj and kj for the last two. So a Facebook status update or an SMS that would be strictly written as “Of lele kolku e žeško vo stanov, ḱe umram” is actually written “Of lele kolku e zesko vo stanov ke umram.”

Thus you have Macedonian chat romanization.

For a Macedonian, the missing hačeks and accents are easy to fill in, but for a foreigner this sort of writing takes getting used to. These days, I know Macedonian well enough that I can read and write fluent MCR, and I even have enough of a sense for the natural sounds of the language that I can guess where the hačeks should go in an unfamiliar word. But I still think the language makes a poor show when it’s written in MCR. Without the diacritics, it looks childish, too simple and repetitive.

But since anyone with enough command of the computer and time to write out full diacritics will probably just write in Cyrillic anyway, and there are still many people who don’t bother with either despite knowing how, I think MCR’s going to remain a feature of the language for decades to come.

I’ve finally been able to up-shift my activity level. I recorded my first Pear Story in Romani with my friend T, who was also patient enough to work through the whole three minutes of monologue with me and transcribe it, then translate it. Without his help to separate out the words, it would’ve been hopeless. Once I had the basic lexical items of the story and I’d puzzled out the key verb endings though, I could break up the transcription into its discrete chunks myself. I even managed to correct some errors in the way we had transcribed it.

As an aside, you may wonder why a native speaker, hearing his own speech he’d just made a few minutes ago, would err in telling me what he had said.

First, because we tend to clean up our reported speech when given the chance to listen to it again. Speech involves different grammatical rules, different constraints on continuity, and different expectations of consistency than written word. When we talk, we know these rules intuitively, so our speech sounds normal to each other. When we listen to recordings of ourselves in informal situations, however, and if we write out literal transcriptions of that speech, it can sound sloppy, even stupid. It is neither sloppy, nor stupid, merely different, but that’s the power of a written standard language. It colors your perception so thoroughly that the words that came out of your mouth only five minutes ago don’t sound like something you would say. The temptation to alter the transcription of your speech to fit a standard written language’s grammatical rules and expectations for fluidity is enormous.

Second, because without training, it’s easy to slip into summarization rather than explicit, word-for-word transcription. That’s true even when you have a recorded reference point. I would argue (other linguists would disagree with me) that this is because we do not think in words–the structure of our speech is not synonymous with the content of our speech.

So with corrections applied, today I started hacking away at the grammar of his particular dialect. T is a Rom (colloquially known as a gypsy, but they refer to themselves as Rom, pl. Roma) who lives in Šuto Orizari, or Šutka for short. Šutka is a ward of Skopje, with its own mayor and municipal government, and it holds the distinction of being literally the only district in Europe where Romani is used as an administrative language under official approval.

This situation is due to the Ohrid Framework Agreement, a treaty signed at the end of the 2001 Civil War between Albanians and Macedonians that established rights for ethnic minorities, most prominently language rights. According to the agreement, communities in which a particular linguistic minority constitutes at least 20% of the population may use that minority’s language coofficially with Macedonian.

The intent was to give municipalities and districts with significant Albanian populations the right to educate, organize, and conduct government business in Albanian. However, the condition was written broadly to include any linguistic minority. The Roma make up 76.7% of Šutka’s population, so they argued that under the Ohrid Agreement they had a right to use Romani as an official language of their municipality.

This provided a huge impetus for the standardization of Romani, which is an ongoing project. There are obstacles to the creation of a written standard particular to Romani that have made this a slow, winding path. Romani is spoken everywhere in Europe, with its core grammar (native verb conjugation, noun declension, gender and adjectival agreement) and lexicon (words for body parts, family members, basic activities) derived from a shared origin in Middle Indo-Aryan languages. That is to say, the Romani language is most closely related to langauges spoken in India.

But speakers of Romani have moved very far over the centuries and the language has a long history of borrowing both words and grammatical structures. The language of a Roma minority in any particular place in Europe today tends to be heavily influenced by the language of the majority that surrounds it. Thus Sinti (German Roma) speak Romani with a heavy German overlay, Arli Roma speak Romani with a heavy Macedonian or Albanian overlay, Vlax Roma speak Romani with a heavy Romanian overlay, and so on.

The fact that different Romani dialects look to vastly different sources for new vocabulary and grammatical constructions means that it’s difficult to make a bridge dialect between all of them that can be used as a standard. Proposals include picking native Indic-origin words from as many dialects as possible and mixing them to create a ‘pure’ Romani, settling on a single means of ‘Romizing’ pan-European words to create a literary vocabulary, or turning to an established literary Indic language like Hindi as a source of neologisms. How to balance these different approaches remains a topic for debate. In any case, the declaration of Šutka as a Romani-language community at an official level took a discussion that had been mostly theoretical and gave it practical significance.

Something I’ve learned while transcribing narratives Macedonians tell in informal speech: my Macedonian textbook is full of lies!

• Macedonian doesn’t have vowel reduction: LIE! Spoken Macedonian reduces ‘o,’ ‘a,’ and ‘e’ with giddy abandon. Detenceto, which should be pronounced something like deh-TEN-tseh-toh, actually comes out as duh-TEN-tsuh-tuh.
• Macedonian stresses native words on the antepenultimate syllable: BIG LIE! Stress shift all over the place.
• Macedonian is written as it is pronounced: WHY DO YOU LIE SOOO MUCH? Those consonants in words like kako “how,” koga “when,” and kade “where,” are fanciful abstractions invented by silly daydreamers flittering fancily away in unreality. The cold hard truth is kaə, koə, and kaj. Didn’t quite catch that vowel plus teensy-tiny microscopic schwa offglide? TOO BAD. SO SAD. GUESS YOU’LL JUST HAVE TO COMPELTELY FAIL TO COMPREHEND THE MAIN POINT OF THE SENTENCE HAHAHA YOU SUCK.

Okay, actually, now that I’m done ranting, there’s nothing too surprising about what I’ve found.

The textbook teaches the standard form of the language, and it’s true that there’s no vowel reduction, stress is antepenultimate, and pronunciation follows the written form in the standard Macedonian language. These differences I’m noting in the recordings of informal speech are totally within the normal range of changes that occur in the colloquial forms of languages.

It’s actually quite pleasant to sit down and hammer out the exact reasons why I still have trouble understanding people my age: they tend to speak colloquially, and the colloquial clearly adjusts the rules of pronunciation a bit. Grammar too, but I haven’t looked as closely at that yet.

I can draw one preliminary conclusion: colloquial Macedonian has an explicit indefinite article that the standard language mostly lacks. Eden shows up all over the place where it shouldn’t.

Now if I can just keep doing this sort of work I won’t feel like such a clump of mud.

Over the weekend I visited Tirana, the capital of neighboring Albania. Seth and I braved the ten hour bus ride together and came to the agreement that Tirana is magnificent. We spent the first day simultaneously stopping to snap photos of the same buildings. Conversation was mostly a series of startled declarations about how much more cosmopolitan, cleanly, and sophisticated Tirana felt in comparison Skopje. Palm trees! Tended parks! Brightly colored, cheerful apartment towers! Walkable, airy boulevards! Big, open central squares with beautiful architecture! Good Italian food! Better coffee!

And yet, you could still see the past and present impoverishment of the country. Little things, like how some buildings’ courtyards had thick walls with pieces of broken glass from old beer bottles cemented into the top, an ad hoc way of discouring people from climbing over. Seth suspected these fortifications were remenants from the period in 1997 when Albania descended into economic anarchy, to the extent that warlords and criminal gangs were running little fiefdoms in the vacuum left by the collapse of the government’s authority. Even less well-protected buildings are more rundown when you look beyond mayor Edi Rama’s technicolor painting project. Traffic is about what you’d expect in a country where maybe only a few percent of the country’s population was allowed to drive before the 90s. There seems to be more international industrial and service investment in Tirana, but fewer big-name corporations like, say, McDonalds. Cows grazed on the lawns in front of major German engineering firms on the city edges. The route from the Macedonian border to Tirana is half highway, half little more than dirt road. The only thing you could anticipate was startling contradiction.

In any case, my verdict is that Tirana is gorgeous. I want to live there some time. I liked the people, I had a lot of my (often Macedonian-informed) preconceptions utterly exploded, and I saw a lot in the city I’d like to learn more about.

Pictures forthcoming, but I wanted to relay an odd incident coming back from Albania. Seth and I took a bus that was supposed to stop in Skopje before continuing to Prishtina, in Kosovo. However, the bus actually went around the city of Skopje and dropped us at a gas station many miles outside of the city. The attendants at the station were ethnic Albanians. I’d spent the whole weekend speaking Albanian in Tirana, but it was 3am, I was exhausted, and I couldn’t understand the men’s accents in Albanian, so it was with an awkward mix of Macedonian and Albanian that I asked one of them to call a taxi for us.

He pulled out a cellphone to call and talked to the driver in Macedonian, but only to tell him the location. Then he handed the phone to me. I was confused and surprised–I didn’t really see why I needed to be part of the call since all the necessary information had already been provided. What came out of my mouth was an exhausted broken Macedonian with lots of Albanian conversation holders that I had to backtrack and correct. When I’d finally squared with the taxi driver what we wanted and where we were, I gave the phone back to the gas station attendant and asked him why he’d wanted to me to talk on the phone. I pointed out that I’d probably confused things more.

He said that, since he was an Albanian, and since the taxi driver could hear it in the man’s Macedonian-language accent, the driver wouldn’t have been willing to come outside the city. It was better for the driver to hear my broken, but clearly foreigner-accented Macedonian than to hear an Albanian speaking perfectly fluent Macedonian he has to use every day.

I understand the social constructions that support this sociolinguistic contortion, but I still find it irrational and preposterous. Here’s a frequent complement I get from ethnic Macedonians: you speak Macedonian better than some natives of this country! You speak it better than many Albanians!

And here’s my own experience: I have never, ever, not once, met an Albanian who does not speak Macedonian whole orders of magnitude more fluently than I do.

I’m not saying there isn’t an accent or there aren’t grammatical mistakes or that there isn’t something wrong in the word choice of some Albanian speakers of Macedonian. I’m not the best judge of this, obviously, but I’ve sat in on classes for Albanians to learn standard, college-level Macedonian and I’ve seen the professor carefully highlight some common confusions of Albanian speakers due to the influence of their native language. But when it comes to communicative ability, fluency, and exactness, I have yet to meet an Albanian I could even dream of surpassing in my Macedonian ability.

There is a sociopolitical dimension to this compliment I keep getting that I speak “better than Albanians,” I’m almost certain. Language perception is powerfully influenced by the way we perceive the social category of the speaker, so the disparity between my own perception of ethnic Albanians’ Macedonian and ethnic Macedonians’ perceptions is not something unique to this country. But it does puzzle and sometimes anger me. I could not come up with a good reason why the gas station attendant’s perfectly comprehensible Macedonian should be inadequate for calling a taxi that didn’t involve some reflexive ethnic bigotry against him.

Ethnic bigotry which, by the way, goes both ways. When he’d finished calling a taxi for me, the attendant started asking the standard biographical questions: where are you from? what are you doing here? why do you speak Macedonian? From there though, he pivoted into asking: why are you bothering with Macedonian? don’t you know it’s not a real language? Then he talked about how Macedonia is a country without a history, that Macedonians are just Greeks or Bulgarians or Serbs who had an artificial identity created for them by Yugoslavia, that Macedonia is going to break up soon, that this area would be absorbed into Greater Albania, and that another war was coming.

It was a tense conversation. I do not believe any of those things and I think they represent an unfair and inaccurate attack on Macedonians, but I didn’t know how safe it was to argue with him. Seth and I were on an isolated road too far outside of Skopje to walk, with little money between us, very late at night. Instead of arguing, I tried to make it clear that I’m interested in and respect the languages and cultures of both Macedonians and Albanians and that I hoped there would never again be any armed conflict in Macedonia. I argued that violence between the communities would ultimately hurt everyone. I think the attendant sensed my discomfort and was willing to back off a bit, having voiced his frustrations. He agreed that further armed conflict would damage the prospects of all and that it was only politicians who were really stirring conflict: thieves and opportunists, all of them!

No matter whether I’m talking to a Macedonian or an Albanian, everyone can agree on bashing politicians.

The taxi driver eventually arrived, exchanged a few greetings with the attendants in Albanian (A je mirë? Mirë, mirë, natën e mirë!), then eventually told us on the drive back to Skopje that he was a Montenegrin who grew up speaking Serbian and moved to Macedonia. Then he ripped us off for the cab fare.

Oh Balkans, even when you’re being awful you light my mind on fire.

Something that I found attractive about the Macedonian language right when I started learning it was the consistent stress pattern. Unlike some other Slavic languages like Russian or Serbo-Croatian, Macedonian has a predictable stress system. Antepenultimate syllable receives stress, which means the third syllable from the end. If there are less than three syllables, the first syllable receives stress. So, to rattle off a series of nouns based on one root and show how the stress works, kup, kupi, kupuva, kupuvanje, kupuvanjeto, otkupuvanjeto, etc. While stress may move as you perform derivations on a root, attaching prefixes and suffixes, you can always predict where it will land.

Except when you can’t.

There is a set of exceptions I was quite aware of, and that’s foreign words. Many words, especially those taken or derived from Latin roots, will take penultimate stress. Balkanizam or telefonira, for example. Sometimes they’ll even take final stress, as in intervju or bagaž. But these exceptions are easy to remember because they’re usually words that already exist in English, due to French influence.

Another set of exceptions, though, I don’t understand. These are Slavic-origin words that by all rights should conform to the standard Macedonian stress pattern, but fail to. I’ve tended to just force them into the antepenultimate stress pattern in any case, but ironically it’s one of the signature features of my speech that marks me as having a foreign accent, because they are very common words. For example, izvini, the informal way of saying “sorry,” should be pronounced izvini according to the stress system, but it is actually pronounced izvini, at least as far as I can hear. Uživaj, a multipurpose word that means both “enjoy” and “go ahead” or “feel free” among other things, is pronounced uživaj not uživaj. And, trickily enough, makedonski in everyday speech is stressed on the don, not the ke. So in the very act of naming the language I give myself away as a foreigner.

I don’t know why stress shifts on these words. I don’t know whether this is limited to Skopje, or extends across Macedonia. I don’t even know whether it is part of the standard Macedonian language and I simply didn’t learn it, or whether the antepenultimate rule is honored more in the breach than the observance. If I were to take a guess, I would suppose the stress shift occurred under the influence of Serbian, since penultimate stress sounds more Serbian to me, but that could be completely irrelevant.

Unfortunately, once I get my hands on a rule, I’m very reluctant to let it go. I’m a incorrigible systems-builder. So it’s been difficult for me to break the habit of stressing all native Slavic-rooted words on the antepenultimate syllable.

People compliment me on my accent in Macedonian, saying that I have a much better accent than other Americans they’ve met. But, keeping in mind my countrymen’s famous deficiency in foreign language abilities, that may be damning with faint praise. If I ever want to get past more than one sentence without the taxi driver asking me where I’m from, I’ll need to learn to let go of my dearly beloved stress rule sometimes.

Today was a good Macedonian day! Except for that one sentence where I confused the word for ‘curves’ with the word for ‘whores’! Which was okay, because I had everyone around me almost falling on the floor in hysterical laughter as I talked about taking the differential of a function to find lines lying tangent to whores! But otherwise my Macedonian today was James Bond suave!

I also managed to say more than a dozen sentences in Albanian! Some of them were even coherent!

I am drinking an enormous amount of tea to treat my cold! Which is what Macedonians recommend! Tea has caffeine! Caffeine is a stimulant!

I am stimulated!

!

!!!