Article in The Guardian on the surge of translation business

I just found an excellent article in The Guardian about 2011 becoming the ‘year of translator’.  It’s an interesting read, touching on topics such as the Tower of Babel, the King James Bible, Haruki Murakami, Noam Chomsky, Google Translate and even hot dogs.

Posted in Seen elsewhere

Translator as a grammar nazi

It takes some level of obsession to be a good translator. Sure, any translator must be enthuastic about their subject areas and languages in general. This time I’m specifically talking about obsession with grammar, especially target language grammar.

I don’t know how normal people cope with the ever increasing rate of linguistic inaccuracies and general floppyness in their respective mother tongues. For I beleive this is a worldwide phenomenon.  How many times have I seen poorly constructed source text sentences, “its” and “it’s” mixed up, “should of done” used instead of “should have done”?   Young Spanish and Italian speakers tend to use “ke” instead of “que” – perhaps as a result of text messages.  German and other languages are increasingly infested with English words.

As a translator and proofreader, whenever I see a new text, any text, I cannot immediately start reading and mentally processing the message, as at the very glance I’m inadvertedly looking for mistakes. Spelling mistakes, grammatical inaccuracies, you name it. More specifically, in any Hungarian text I’m looking for signs of any incorrect translation (since many texts are actually translated texts).

One recent and widespread phenomenon in Hungarian that really annoys me is the incorrect usage of suffixes. As a general rule, all suffixes are attached to the word stem without any hyphen.  Brüsszel is “Brussels”, Brüsszelben is “in Brussels”. Mazda is a Mazda, Mazdával is “with (a) Mazda”.

There are some exceptions, of course. If the word stem is an acronym, the suffix is preceded with a hyphen: NATO-val means “with NATO”. Also, if the word stem is a foreign word where the last letter is not pronounced, the suffix must conform with the pronunciation rather than spelling, and the suffix is also attached with a hyphen. Typically these words are French proper nouns. Mitterrand-nal is “with Mitterrand”.  (The -val/-vel suffix changes its first vowel to conform with the last consonant of the word it is attached to. The suffix retains its -val/-vel value if the last sound of the preceding word is a vowel: Peugeot-val: “with a Peugeot”.)

These days I’m increasingly confronted with the incorrect, hyphenated usage of such suffixes: in blogs, movie subtitles, forum comments, status updates… and sometimes even in printed press.

And this poses a more general dilemma. Is this a sign of a “deterioration” of the language or simply an “alteration”?  Should translators and other purist fight the new winds or turn around and go with the flow?

I know, no rules are meant to be forever. Early in my translation career the Windows operating system was still new, with no established Hungarian localization available. I was abhorred to see that some other translators translated “file” as fájl. To me, and to most other translators at that time, a simple phonetic transcription seemed a horrible idea. We simply used file, file-t, file-lal, etc. Back then there were some attempts at introducing a Hungarian word for this: állomány. This word already existed with a different meaning, and somehow it didn’t succeed in replacing file. After all, using a monosyllabic word is always much more economical than using a trisyllabic word.

A few years later Windows 95 came out, with a proper Hungarian localization… and “file” was now “officially” translated as fájl.  Yes, it certainly looked awful at first, but I had to realize that it’s actually a much more user-friendly word than file.  The single biggest benefit of the phonetic transcription was that now any suffix could be attached without a hyphen: fájlt, fájllal, etc.

Finally, back to my headache, the recent tendency of hyphenating suffixes unneccessarily and incorrectly. This usage goes against the underlying logic of any linguistic tendency – simplification, that is.

I stand by my right to be a grammar nazi as a translator.

Posted in fun, Hungarian | Tagged , ,

How to get a 10 quid per word rate?

Just as I was sinking in an onslaught of translation jobs that will definitely affect my  plans for the weekend, I received a small assignment from a favourite British client specializing in creative adaptation.

Needless to say it’s super urgent…. but it’s only three words, with a total of seven (7) letters: “up to 50% off”.

Within a few minutes I duly provided the Hungarian translation, with additional explanatory notes and tips on typesetting.

The shortest ever job I’ve had in my freelancer career was also for this client, a few years ago. Then they needed the words “yes” and “no” as they would appear on a website as buttons.

Breaking: As I was writing this blog post, the client called me on the phone, saying that the actual copy to be translated is “up to 50% off the original price”.  That still leaves 5 pounds per word – not bad for ten minutes of work.

Posted in business, fun | Tagged

How to translate “to”?

Every so often I receive an Excel file with hundreds of short words to translate. Most of the time these short words are items on a graphical user interface – or GUI for short. As a rule, such lists are much harder to translate than an ordinary texts that consists of actual sentences.

GUI items often include variables that may stand for certain values (e.g. currency code, country name, client name, etc.) or actual numeric values (e.g. amount, date, length, etc.).

In some cases GUI items contain such variables. For example “Dear <client name>, ” or “Your password expires on <date>.”   Such items are relatively easy to translate.

But sometimes Excel files include GUI items without any visible variables. In such cases the dedicated translator prepares a detailed list of questions to clarify what variables may come before or after a specific GUI item.

My all-time favourite is “to”. Yes, on its own, without any context or reference.   Now, I really need to know what possible words or values may come before or after this lonely “to”. Is it followed by a place name? A person? A time period? Does it come between two numeric value to denote a value range?

A good client will furnish the answers within a day or two. (Well, usually two days, as they have to ask their client that is often in a completely different time zone…).  But not all clients are created equal, and some simply reply “Don’t bother about those variables, simply translate the word.”  Yeah right.

I always have to explain that the structure and logic of the Hungarian language is fundamentally different from English (or German, French, Spanish, you name it). What may be neatly translated into major European languages as zuà or hasta, without worrying about what comes afterwards, Hungarian incorporates this grammatical information in a suffix that is attached to the end of the word that follows.

Depending on context, “to” should be translated into Hungarian as “-ba/-be”: Franciaországba (to France); “-ni”: törölni (to delete); “-nak/-nek”: Andrásnak (to Andrew); “-ig”: szombatig (to Saturday); with several more possibilities.

Now, since the text of the variables cannot be changed, a workaround solution must be found. “to” must be rendered with a word that means something completely different. What’s more, a different solution must be found for every specific case, depending on what type of data the variable represents. In most cases the solution is a noun followed by a colon.

Using the examples above, “to” should be translated in graphical user interfaces with the following words.

Localized text Literal meaning Example
Úti cél: Destination Úti cél: Franciaország
Címzett: Addressee: Címzett: András
Záró dátum: Closing date: Záró dátum: szombat (2011. nov. 26.)
(nil) törölni (in this case infinitives must be used as possible values)
-18°C … +25°C (best solution for ranges of numeric values)

Needless to say, my other favourite word is a stand-alone of . . .

Posted in Uncategorized

Days of the week in Hungarian

After a busy week here’s a new beginning on Monday morning.   And Hungarians actually mean it – the word for Monday literally means “week’s head”, i.e. “week’s beginning”.

To start with, the Hungarian word for “week” is hét. Incidentally the same word is used for “seven”.  A similar analogy was used in Latin (septimana vs septem). Although it’s not recognizeable, hét is actually an Indo-European loan word in Hungarian. Ancestors of early Magyars adopted the word from an Old Iranian language (septa > seta > heta > hét). Linguistic evidence suggests that the dual meaning of the word (period / number) developed early, before the arrival of Magyars in Central Europe.

The Hungarian word for Monday is hétfő. As I mentioned before, this means “week’s head” or “week’s beginning”. Early Hungarians probably adopted this logic from a Slavic language. In early Croatian, “prvi dan” (“first day”) was used for Monday, which, in turn, must have originated from late Latin “primus dies”.

Tuesday is kedd, which is archaic form for expressing “second”. The development must have been kettő (“two”) > ketted (archaic “second”, later “half”) > kedd (shortened form). Try to imagine an English word of “twoth”, and you get it.

Hungarian equivalents of Wednesday, Thursday and Friday are Slavic loan words. Wednesday is szerda, which comes from a Slavic form of sreda that means “middle”.  All modern Slavic languages still use very similar words for Wednesday. A similar logic is followed in German: Mittwoch is literally “middle of the week”.  Early Hungarians adopted hundreds of words from various early Slavic languages, but often found it hard to pronounce several consonants at the beginning of words . That’s the reason for swapping “r” and “e” in this particular case.

Thursday is csütörtök, which may seem alarmingly difficult to pronounce at first. “Cs” is pronounced as “ch” in English, while “ü” and “ö” have the same phonetic values as in German and Turkish.  The word derives from the Slavic word for “fourth”. The Slovak word for Thursday is štvrtok, the Croatian word is četvrtak (now, think again if Hungarian is difficult to pronounce).

Friday is péntek, which again comes Slavic (Croatian petak, Slovak piatok), with the original meaning of “fifth”. The n in péntek suggests an early adoption from Slavic, when many Slavic dialects still had nasal vowels. In modern Slavic languages only Polish retained nasal vowels: for example their word for Friday (piątek) is pronounced as “piontek”.

The word for Saturday is szombat, another example of early Slavic adoption with a nasal vowel. This word entered Hungarian from a Southern Slavic language (perhaps Old Church Slavonic sobot). The word can be traced back further to Byzantian Greek, and ultimately to Hebrew shabbat.

And this brings us to the end of the week. Sunday is vasárnap, which literally means “market day”.  The modern Turkish word pazar günü means exactly the same. (What’s more, Turkish “pazar” and Hungarian “vásár” both derive from the same Persian word that also gave the world the word “bazaar”).

As a pronunciation guide for the days of the week in Hungarian, listen to this song from the early 1980’s. The actress sings about how she’s having the time of her life on Tuesdays, while all the other days are boring or gloomy.

(I used Gábor Zaicz’s Etimológiai szótár  (Dictionary of Etymology, 2006) to write this article)

Posted in fun | 1 Comment

Project management in the cloud

One of my clients, a Swiss translation agency specializing in financial texts, especially descriptions of investment funds, has been using an online project management system for several years now.  Recently another client introduced a similar system – incidentally or not, this Luxembourg-based company also specializes in the translation of investment funds.

Can we spot a new trend here?

Apart from their specialization in investment funds and other financial documents, what distinguishes these two language service providers from other clients of mine is that they usually send me very small texts at a time. A typical project volume is just a few paragraphs.

Could this be the reason they chose to develop their web based proprietary project management systems? On their websites they describe the advantages of their own platforms that allow for versioning, workflow management, automatic email notifications, you name it.

While such platforms probably benefit translation companies and their end clients, freelance translators may prefer the old fashioned way of doing things.

For starters, project related communication becomes completely impersonal. Freelancers hardly ever meet their clients in person, but at least we got used to emails with greetings and good wishes – these may be clichés, but still retain a comforting level of personal touch. With web based project management platforms this personal element is gone.  You get to “communicate” with an impersonal platform. First you click on a link in an automatically generated email, then you click on Accept, then, after completing your task, you click on Upload and Done.  No more “Hi, I hope this email finds you well” and “Have a nice weekend” in emails.

Second, it takes some additional steps to copy all the necessary files to your computer.  When you receive jobs in an email, you can copy all email attachments with a single click. In web based platforms you have to copy them one by one.

One of these clients I mentioned above actually went one step further. You, as a freelancer, not only have to accept and deliver the tasks online, but you also have to enter the translated texts into their online interface. Now, if you’re a conscientious translator, you want to use your favourite CAT tool to leverage any existing translations as well as maintain terminology consistency. In this client’s web based platform, you first have to copy and paste text from each task into a Word or Excel file, perform the actual translation with your CAT tool, then copy the translated text back into the online platform.

Cloud services may be all the rage these days, but it will take some more time before freelance translators will also enjoy the benefits of cloud nine.

Posted in business | Tagged ,

How to say “eleven” in 11+11+11 languages?

On this auspicious date 11/11/11 it’s time to look at how number eleven is constructed in a number of languages. (To be precise, in 11+11+11 = 33 languages.)

Let’s start from Germanic languages that also include English. “Eleven” comes from an old Old English form of “endleofan” that meant “one left” (over ten). Modern Danish and Norwegian resemble most, both with “elleve”. Swedish “elva”, German and Dutch “elf” all derive from the same root.

Latin “undecim” is a compound of un + decem (one + ten).  Of all Romance languages, Italian deviated the least from the Latin form with “undici”.  In most other Romance languages the first syllable of the Latin “decim” was dropped and the word became “once” (Spanish and Catalan) or “onze” (French and Portuguese). Romanian is different. First of all, a d>z consonant shift resulted in „zece” (ten). The word for eleven is „unsprezece”, literally „one over ten”.

Russian „одиннадцать” (odinnadtsat’) is literally „one above ten”, although the first letter of ten (десять) is merged with the equivalent of „above” (над), the wovel is lost from the first syllable and the „s” sound became a „ts” sound.  Ukrainian одинадцять follows a similar pattern.  Among Western Slavic languages, the Polish word for eleven is “jedenaście”, Czech is “jedenáct”. In all these forms, the full first syllable of „ten” is lost (just as in the case of Spanish and French, remember?).  In the Southern Slavic group, Slovenian uses “enajst”, Croatian “jedenaest”, and Bulgarian „единадесет”. All of these forms can be traced back to a now extinct „-nadeset” form, which literally meant „over ten”. Bulgarian seems to have kept this old form. Other Slavic languages follow very similar patterns.

In the Baltic group, Lithuanian is famous for its archaic forms. Eleven is “vienuolika”, which literally means „one left”, just as in the Germanic languages. The Latvian word is „vienpadsmit”, i.e. „one over ten”.

Hungarian “tizenegy” is literally “one over ten”, with the word for ten coming first. Finnish “yksitoista” is more interesting: it literally means „one of the other/second” (the word for ten is totally different: kymmenen).

Turkish „on bir” is very simple: „ten one”. The same pattern used in some other Asian languages too. Chinese is 十一 (pronounced as Shí yī), Japanese is also written as 十一 (pronounced as jū ichi). Korean uses two distinct forms (십일, shipil, and 열 하나, yeolhana), both are simple compounds following the „ten one” pattern.

Arabic أحدَ عشر is pronounced as aḥada ʿašar , and it’s made up of „one + ten”, with both words slightly different than the two standalone words.  The Hebrew word for eleven is written as אחת עשרה and pronounced as achad asar, making it mutually understandable with Arabic.  Swahili „kumi na moja” is simply „ten and one”.

Georgian uses a 20-base counting system, just like French (and Basque). Eleven is თერთმეტი or t’ert’meti.

And finally let’s turn back to another Indo-European language, Hindi. ग्यारह is pronounced as „gyāraha”, and it’s not related to ten (dasa) nor one (ek).

Posted in fun | Tagged