The pronunciation of Vietnamese is hard for several reasons. Every syllable carries one of six tones, indicated by five special accents (à, ả, ã, á, ạ) that turn Vietnamese writing into equal parts text and score. The numerous different vowels and diphthongs are no piece of cake either. To give just one example: the language has a shorter and a longer ‘uh’ sound (spelt as â and ơ respectively).
But while these things look intimidating right from the start, something unexpectedly treacherous lurks behind the seemingly innocuous letter combination ng. As in English and other Germanic languages, this pair stands for the sound heard in sing and singer; the phonetic symbol is ŋ. In Vietnamese, however, this sound can also appear at the beginning of a word, as in ngai for ‘throne’. We don’t do that in English, or any other European language that I’m aware of for that matter. As a result, it’s surprisingly difficult for us, or for me at any rate, to distinguish words beginning with ng from those beginning with n. And since ngai and nai (‘deer’) have different meanings, ignoring the difference is not an option.
Why should this be so? Is there anything about the beginning of a word that makes it harder to properly hear the /ŋ/? I suspect not. More likely, it’s a matter of expectations: my listening brain is just not prepared for a /ŋ/ to pop up in that particular position.
I got evidence for this hunch when earlier today I heard someone talk about a ‘union-style effort’ – you know, the sort of effort trade unions usually make. Except that he said this in the context of psychology, and I soon realised that the speaker had actually said ‘Jungian-style effort’: an effort along the lines laid down by the psychologist Carl Jung. What those lines are is a matter I have only a hazy notion of, but they’re certainly unrelated to the labour movement.
Union (/junjən/) of course is a much more common word than Jungian /juŋjən/, and words ending in /njən/ (companion, opinion, draconian) are generally much more common than those ending in /ŋjən/ – indeed, I have trouble finding any at all. So what happened is that expecting to hear an n, I heard just that, even though the speaker undoubtedly vocalised ng, that is /ŋ/. At the same time, I’m confident I would have no such trouble distinguishing sinner and singer or winger and winner. I know that these pairs exist, and therefore my expectations wouldn’t lead me astray there.
So while the bad news is that the initial /ŋ/ is one more headache for a European student of Vietnamese, there is good news too: once my brain gets wise to the possibility of /ŋ/ appearing as the initial sound of words, it will probably start listening out for them and at some point catch them. And a good thing too, since the Vietnamese word for linguistics is ngôn ngữ học.
You can find the other blogposts about my adventures in Vietnamese here.
People are puzzled about how to pronounce the ubiquitous Vietnamese name Nguyen (or Nguyễn in proper Vietnamese spelling) but when I point out that the ng sound is the same as the ending of sing, they can do it.
There is also the Cantonese name Ng and I myself don’t know how to pronounce it.
Apparently, the Cantonese pronunciation is simply… ng. In other words, ŋ̍.
Another nice reflection on unfamiliar languages. Just to follow up your point about the oddity of words starting with “ng”, I might point out that Welsh does have this feature – albeit in the slightly indirect form of “mutated” words. I won’t go into that here (though I did once send you, Gaston, a description of the process, since you drew attention to it in a chapter in Lingo). Suffice to say that when triggered by certain preceding words, the initial letters of Welsh will change or mutate according to a regular pattern. After the word “fy” (=my” for instance, nasal mutation means that the word “gardd” (garden) appears as “ngardd”; while my cat (“cath” in Welsh) would be “fy nghath” .
Having explained that, I would be the first to say that for a non-native speaker the “ngardd” form does take a bit of handling, and I can see why you are a bit daunted by the Vietnamese version. Whether or not you count this Welsh phenomenon (I believe Breton has the same sort of thing) as a “real” ng opening, it definitely requires a corresponding sound to be made.
I didn’t think of that! Yes, nasal mutations like ngardd definitely fit the bill. Breton and Irish seem to have this initial sound as well.