7

When languages borrow words from other languages, they sometimes deliberately distort words to make them phonetically easier to pronounce.

For example, when Japanese speakers are taught the word "circle", it is taught as sa-ku-ru. Similarly, "beer" is taught as bee-ru and "bottle" is taught as bo-te-ru.

Japanese probably has hundreds of such examples. Chinese also does this when it borrows words from English.

Is there a specific linguistic term for this that I can read more about?

How common is this? Is there a specific reason why this happens? Does this have something to do with how accents are developed? Do we lose the ability to pronounce some sounds as we age?

  • 7
    @sumelic Bīru is from Dutch bier. Bia from English beer also exists but usually only in phrases like bia hōru 'beer hall'. – Nardog Sep 24 at 15:41
  • 12
    It's somewhat concerning that you say "some" languages (not "all"), and only cite Japanese and Chinese (foreign to you, I assume) as examples. Surely you don't think English pronunciation of sushi, tsunami, ramen, tai chi, and kung fu are faithful renditions of the original sound? – jick Sep 24 at 16:09
  • 3
    @jick My favorite is karaoke [kɛɹiˈoʊki], which sounds nowhere near the original [kaɾaoke]. I tend to find loanwords originally ending in [e] (sake, boke...) particularly salient, as English does not have a mid front monophthong that can end a word so it replaces it with /eɪ/ or more often /i/ or /ə/. – Nardog Sep 24 at 16:18
  • 11
    @joe It's all subjective. Most Japanese speakers probably wouldn't even recognize [kəˈmoʊnoʊ] as kimono upon hearing it for the first time. Also let's not confuse loanwords and attempts at pronouncing a word in a different language. If you find English speakers' pronunciation of Japanese words closer to original than the other way around, that's most likely because 1) you are more familiar with English than with Japanese; and/or 2) Japanese has a simpler phonotactics so more derivation is required when borrowing from English to Japanese than vice versa. – Nardog Sep 24 at 16:28
  • 13
    @joe You think that "ramen" is a faithful rendition because you are a native English speaker, so you literally cannot recognize the difference between that and ラーメン: they sound the same to you. (Different r, different n, and vowels are also somewhat different.) A native Japanese speaker will likewise think sakuru is a faithful rendition of circle, because they cannot hear the difference. Your question can be paraphrased as "Why do speakers of other languages ignore some sound distinctions found in English?" - put this way, the answer would be obvious. – jick Sep 24 at 20:11
22

The term is loanword adaptation.

It happens every time someone tries to use a word from a different language when speaking another. It's because every language has a different set of sounds that can be recognized as part of that language. A tongue click can be part of ordinary words in some languages in southern and eastern Africa just like any other consonant but not in others. A difference in the length of a vowel can create a difference in meaning in some languages but not in others. The presence or absence of a puff of air following a sound like p can create a difference in meaning in some languages but not in others. Native speakers of languages with such sounds and features perceive and produce them with incredible precision without ever even thinking about them, while others find it difficult.

Because of these differences, two competing goals come into play when a language borrows a word from another. One is to preserve the original form as faithfully as possible. The other is to conform to the rules of the sound system of the target language as much as possible. So when a sound used in the original word is unavailable in the target language, the closest possible sound is chosen. What you often end up is a compromise between these two forces.

Languages also differ in conditions under which each sound is allowed to occur (known as phonotactics). Languages like Japanese allow few consonant clusters. So a word like strengths is often an impossible sequence of sounds in those languages. There are two overall strategies a language can take when borrowing such a word: It can delete existing sounds or insert extra sounds. Japanese usually takes the latter approach (known as epenthesis), so e.g. strong [strɔŋ] becomes sutorongu. At the other end of the spectrum, there are languages that allow longer consonant clusters than English does. In Polish, wszczniesz [fʂt͡ʂɲɛʂ], which may be approximated by fshchnyesh in English spelling, is a totally possible word that native speakers of Polish don't find difficult to pronounce. But if you had to say fshchnyesh in the middle of an English sentence, do you think you would find it easy or the listener would easily understand what you said? I hope that gives you an idea of where the "distortion" may come from.


Do we lose the ability to pronounce some sounds as we age?

Yes, we start losing the ability to recognize fine shades of sounds when we're a few months old, and we complete this process before puberty. This is natural and essential as part of first language acquisition. If you think about it, the fact we can recognize the same word as the same word when uttered by different people, male or female, hoarse or high-pitched, in a loud room or not, having cold or not, etc., is an incredible ability that computers are still trying to catch up with. But as we start recognizing a group of similar sounds as the same sound, we also lose the potential to develop the ability to produce them distinctly in exchange.

  • +1. "when a sound used in the original word is unavailable in the target language, the closest possible sound is chosen." - and it will be "the closest" according to the borrower's ear, which (the choice of the substitution) might actually puzzle the speaker of the "lending" language. – tum_ Sep 24 at 20:59
  • 2
    It would be more accurate to say "Yes, we start losing the ability to recognize fine shades of sounds not in our native language when we're a few months old, and we complete this process before puberty." – Azor Ahai Sep 24 at 23:18
  • 3
    Azor you have it backwards. The point is we naturally create equivalence classes of shades of sound that are all equivalent in our language in order to be able to not be distracted by this variation. – Apollys supports Monica Sep 24 at 23:49
  • 3
    @usee45266 Major debate. The short version is that accent is one of the hardest things to master if a language is learned later on – even highly fluent, articulate learners sometimes retain only a crude approximation of some of the target language's sounds. In my experience as a teacher and tutor, the actual recognition can be learned. If I ask you to tell which of two sounds I'm pronouncing enough times, you will correctly distinguish them. What is rarely acquired is distinctions or pronunciations rapidly and accurately on the fly. But it's certainly not a a universal "Nope, too late."! – Luke Sawczak Sep 25 at 4:12
  • 2
    @Orion Not more definitely pinpointed than Nardog said, unfortunately. The infants are easy to test through repeated signal response but the overall decline in language acquisition / sensitivity around puberty has mixed results in studies. As for continuing to recognize different shades, yup, though you'll notice people's ability to mimic accents accurately varies widely. I even have a friend with a particularly faint Russian accent that's unambiguous to me and my linguist friends but invisible to about half the people she meets! – Luke Sawczak Sep 25 at 11:03
11

It's not "deliberate" – it's the automatic, nigh-inevitable result of fitting a set of sounds from one language's inventory into a different inventory.

It's like changing a photo from RGB to CMYK or changing the encoding of text that includes special characters. Most values will transfer but some will just be approximated. Sometimes the change is obvious, but even values that appear very similar may be represented or realized slightly differently.

As for frequency, it's a universal phenomenon. Because no language has an inventory of sounds and combination rules that is a superset of all other languages' inventories, every language will need to approximate some sounds and sequences when borrowing words. The greater the overlap between a given pair, the less approximation needed.

  • 4
    While I agree it's usually not deliberate, there is a degree of arbitrariness tied to the cultural traditions in an area. For instance, I was checking various dubbed languages in a DVD, and I noticed that an English name was pronounced with an English R in German, but an Italian R in Italian. I can't speak for German, but I know that in Italian, if you pronounce a name with its original R sound, you just sound pretentious and ridiculous. It doesn't matter if you know the sound or not, you're just not expected to. I suspect in German, it's the other way around, or at least it is in dubs. – LjL Sep 24 at 15:01
  • @ljl there is no one English R there are various realizations, some of which sound ridiculous in German, others sound like some realisations of a German R. Anything too far back is ridiculous, so "Ronald" with a roled R kind of works in moderation, but "Raegan" is impossible because the vowel Anlaut simply isn't backed (and can't be with the tongue muscles I've grown) – vectory Sep 25 at 3:55
  • @LjL in Russian it's the same as in Italian - all English names are pronounced with a trilled (the same as Italian) R. Pronouncing English R usually means imitating the English accent. – trolley813 Sep 26 at 12:13

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.