หลักเกณฑ์การทับศัพท์ภาษาโปรตุเกส

“จง​ขอ​เถิด แล้ว​ท่าน​จะ​ได้​รับ จง​แสวงหา​เถิด แล้ว​ท่าน​จะ​พบ จง​เคาะ​ประตู​เถิด แล้ว​เขา​จะ​เปิด​ประตู​รับ​ท่าน เพราะ​คน​ที่​ขอ​ย่อม​ได้​รับ คน​ที่​แสวงหา​ย่อม​พบ คน​ที่​เคาะ​ประตู​ย่อม​มี​ผู้​เปิด​ประตู​ให้”

Mt 7:7–8

To my surprise, I’ve found out that the Royal Institute of Thailand has actually issued rules for the transliteration of foreign words into Thai! From all I’ve read about them, I should actually be surprised I didn’t think of looking it up before.

From what I can gather, and just as I’d imagined, they’ve come up with systems that attempt to reflect as much of the phonology of the original languages and yet represent as much of the original orthographic features and idiosyncrasies by the Thai script.

The languages whose standards have been published seem to be:

• Arabic;
• Chinese;
• English;
• French;
• German;
• Hindi;
• Italian;
• Japanese;
• Korean;
• Malay;
• Russian;
• Spanish;
• Vietnamese.

That means that, contrarily to what the title of this post says, there doesn’t seem to be an official system for Portuguese.

For many features of Portuguese, I could actually come up with a reasonable basic system derived from common rules taken from the published standards that would also be valid for Portuguese – the representation of /ʒ/ by ฌ, for instance; for others, however, I wouldn’t be able to trace any parallels at all and would have to rely on examples found online and, well, my own whim – such as how to deal with unique sounds or raised (neutralised) unstressed vowels (to represent phonology or to reflect orthography?).

ฌ

From a past post of mine:

And why ever did I start spelling my second name as โฌแซ instead of, say, โจแซ? I just can’t remember.

I do remember now – I got it from French!

I obviously know very little about how Thais reason when transcribing foreign names. The few bits I do know I’ve learnt from observing details here and there as I go along.

Thai lacks both /ʒ/ and /ʑ/, but has both /tɕ/ and /tɕʰ/; my thinking of spelling โจแซ /tɕoː sɛː/ was to substitute จ /tɕ/ for Portuguese /ʒ/.

However, it’s only taken me a couple of further examples to realise that Thai (surprisingly?) seems to prefer using ฌ /tɕʰ/ for /ʒ/ instead; Portuguese names are harder to come by, but French offers more than a few:

• ฌ็อง /tɕʰɔŋ/ for ‘Jean’;
• ฌูว์ลี /tɕʰuː liː/ for ‘Julie’;
• ฌัก /tɕʰák/ for ‘Jacques’;
• แฌร์แม็ง /tɕʰɛː mɛŋ/ for ‘Germain’;
• ฌอร์ฌ /tɕʰɔ̂ːt/ for ‘Georges’; etc.

I could only wonder why that’s so though. I mean, for /tɕʰ/ alone, Thai also has ฉ and ช; assuming consonant classes have a role to play, ฉ is a high consonant, but both ช and ฌ are low consonants, and the former is, for all I know, far more common in Thai.

Some more examples seem to shed light on the matter, implying that’s the case because Thai saves ช for foreign /tʃ/ and /ʃ/ (e.g. ริชาร์ด /ríʔ tɕʰâːd/ for ‘Richard’ and ดูว์ช็อง /duː tɕʰɔŋ/ for ‘Duchamp’). Since I’ve indeed always read that Thai transliterations try to preserve as much of the original foreign spelling as possible, that’s an interesting way to do so.

O mal do mau

I’ve been マルセル ⟨Maruseru⟩ for over a decade, so I couldn’t help getting really surprised to learn that fellow countrymen have apparently decided to go fully phonetic in their renderings and become マルセウ ⟨Maruseu⟩ themselves.

Not that it doesn’t make sense, but it’s a type of transcription that I myself have always avoided when spelling names in other scripts, for both visual/subjective and practical reasons. I’m even sure I’ve written something about it once, a long time ago.

Take a surname such as Silva, for instance. In a simple, ‘traditional’ transcription into Japanese, you might have シルバ ⟨Shiruba⟩. The Japanese, however, can (and apparently often will) have an option for ⟨v⟩ (even when they still pronounce it as ⟨b⟩), which might give シルヴァ ⟨Shiruva⟩. But why stop at that if you can (at least visually) indicate a ‘proper’ /si/? You might go on and have スィルヴァ ⟨Siruva⟩ after all. And finally, if you’re to represent the semivocalic nature of the ⟨l⟩, you may end up with スィウヴァ ⟨Siuva⟩. Convoluted much?

I agree it’s difficult to reach a compromise between spelling and pronunciation, but, considering all the variations in the Portuguese-speaking world, I myself usually go for the former in such situations, since orthography is far more likely to represent a broader (if not an absolute) standard.

I wonder whether a Portuguese Luís would go to any lengths to be a ルイシュ ⟨Ruishu⟩ in order to be distinguished from a Brazilian ルイス ⟨Ruisu⟩.

Thai provided me an interesting exception to this, though. I’m มาร์แซล /maːsɛːn/, but I’ve always found it cute to be just แซว /sɛːw/ as a nickname too!

Fita & izhitsa

Those brief moments when apparently useless knowledge comes in quite handy! I was searching for websites with the words for Catholic prayers in Bulgarian, and one of them happened to have a background image of two saints. Sure, being Bulgarian, guessing who the two saints were would’ve been somewhat obvious, but I didn’t really need to think of it, because the titles above their heads gave their names, and I was surprised at how easily I happened to read them: Св·Меѳо̀дїй ⟨Sv. Mef̀òdïj⟩ and Св·Кѵрї́ллъ ⟨Sv. Kỳrḯllʺ⟩! Not sure why they were spelt exactly like that (compared to other spellings I’ve seen or orthographic rules I’ve read about), but I could read them instantly.

Incidentally, I’m also amazed at how easily I could find websites with those prayers. I did think it’d take a lot more work to.

ISO 9

Padronização

I like standards. Even when they’re useless. How come I’ve never actually listed my personal transliteration/romanisation standards? I should work on that.

DIN 31635?

Apparently I used to transliterate Arabic following DIN 31635. It’s hard to know for sure because my early posts with Arabic words had very few letters for me to draw a full picture. I know for sure I used ⟨ḏ⟩ for ذ, and I do remember favouring ⟨ʿ⟩ for ع; ISO 233 uses those, too, but I can’t really believe I’d use ⟨ẗ⟩ for ة, so DIN seems to be far more likely (it uses ⟨h⟩, except when in construct state, when it uses ⟨t⟩.

On the other hand, Hans-Wehr transliteration is based on DIN 31635, except that, instead of ⟨h⟩, it doesn’t transliterate ة when not in construct state, and I kind of like that. Hmmm…

A vida em Homero

I find it really interesting to compare characters’ names in different translations of stories and legends, and sometimes even deal with the different variants within a single language. These days, with everything rotating around Homeric Greek in my head, it’s obviously got to do with the Iliad and the Odyssey (including learning how to pronounce the less common names properly in English).

Most of the characters, both major and minor, pose no particular problem in Portuguese — Aquiles, Heitor, Helena, Páris, Príamo, Menelau, Ulisses, Andrômaca, Pátroclo, Eneias etc. It’s basically just a matter of retracing the (Latin) etymology, and the phonetic rules of Portuguese evolution take care of the final result. There is, however, a problem with one of the central characters, Agamemnon, because his name has at least half a dozen variants depending on translator and/or time of translation. The most familiar variant to my ears is Agamenon, with Agamênon probably coming second; I’ve probably already seen all other possibilities there are, though, which are Agamemnon, Agamêmnon, Agamenão and Agamemnão.

Agamemnon’s matter, by the way, closely parallels that of Poseidon I actually think I once wrote about — Poseidon, Posêidon, Posseidon, Possêidon, Posidão and Posídon are all forms I’ve met.

Back to Homer, Ajax comes to mind, too, with both Ajax and Ájax as possible variants; a parallel is also seen with Astianax, with both Astianax and Astíanax existing. And then again, there is yet another parallel pair regarding Chryseis and Briseis, who seem to vary between Criseis and Briseis, and Criseide/Criseida and Briseide/Briseida.

Visitando o Camboja

I’d forgotten how much I love the Khmer script, even though I’ve long crossed the languages itself off my learning list. Today I’ve finally added the family’s names to my list of ‘foreign counterpart’ names: ម៉ារសែល ⟨Marsêl⟩, លេទីស៊ីយ៉ា ⟨Létisiya⟩, មីហ្គែល ⟨Migêl⟩, and ដាណ្យែល ⟨Danyêl⟩. I’d still like to check with a native speaker, but I’m sure I’ve at least gone the right way with them.

Only recently have I started to pay attention to how Daniel’s name often gets reduced to two syllables instead of three in certain foreign scripts. I wonder whether I should just pay attention to the possibility of more ‘detailed’ transcriptions instead.

«Авеню Бразилия»

Armenian transliterations of Portuguese bring me back to that pet hate of mine – the way Portuguese reduced vowels get transcribed in certain different languages/scripts. As I’ve said before, I’m used to treating any reduced vowels as if they were levelled back to full vowels before being transliterated, as if that was some sort of ‘neutral state’; other languages/scripts, however, seem to take a more phonological approach, so to speak, trying to reflect that actual ‘standard’ pronunciation at the expense of orthography, and that always looks out of place to me.

I remember seeing that a lot in Russian, for instance, particularly because Brazilian soap operas seem to be popular enough there for articles on Brazilian actors being easily found in Russian around the Web — Murílo Benício, for instance, shows up a lot as Мурилу Бенисиу (although the Wikipedia article itself, as a curious example, alternates between that and Мурило Бенисио). Bulgarian seems to go the same way as Russian, although there’s a lot less material for me to check online.

Japanese, on the other hand, seems to follow an orthographic approach almost always (only recently have I found a couple of diverging examples, but I assumed the disparities were big enough to make it obvious that the person transcribing the names didn’t really seem to speak any Portuguese – they clearly treated Portuguese ⟨j⟩ as if it was Spanish instead!), and so does Greek. Korean also seems to present very little variation on that.

Now Armenian started to present a couple of such examples, and, for the look of it, it seems to go either way, at the whim of the writer. As with Bulgarian, though, there won’t really be that much for me to explore, so let’s see what I’m yet to find out.

Some languages, by the way, can’t really be classified in such respects because of the very nature of their scripts. Such fine distinctions would be too much to ask from e.g. Arabic (and Persian) or Chinese.

Oh, and, although I’ve focused on vowels, even consonants can get into the game – using Russian once again, it’s really exotic to find Marcelo Novaes referred to as Марселу Новаеш, Марселу Новаис and a handful of other combinations!

Respelling and renaming

Sou transliterated my name into Thai many years ago, but I’ve decided to tweak its spelling a little now:

มาร์แซล โฌแซ มาโรว์โบ เกรูบีนี
/maːsɛːn tɕʰoːsɛː maːroːboː keːruːbiːniː/
<Masaen Chosae Marobo Kerubini>

The changes are important, and reflect more an ‘academic’ than probably a native speaker’s approach, but I’ll get comfortable with them.