-1

I'm a not a native speaker and I just begun studying English.

I'm not used to phonetics at all. I know that is has something to do with pronunciation, but I really don't have any idea.

I'll be grateful for any help.

RegDwight
  • 113
  • 9
obviousfirstsemestre
  • 29
  • 1
  • 4
    There is a lingists' joke about regarding [h] and [ŋ] as allophones of the same phoneme in English, because they are in complementary distribution, but I don't think anybody has ever advanced this seriously. – Colin Fine Nov 20 '13 at 14:10

2 Answers2

6

In any language (or most languages, anyway) there are sounds which speakers of some other languages will regard as different, but in the specific language they are regarded as variants of the same sound: usually, most speakers of that language will not even be aware that there is a difference.

A group of sounds which are not distinguished in a language are referred to as a phoneme, and the different realisations of that phoneme are its allophones.

Your example, of [l] and [ɫ], is a case in point: for most English speakers the choice between them is entirely occasioned by the phonetic environment (what sounds come before or after them), though this choice varies for different dialects of English.

The standard test for whether two objectively different sounds are the same phoneme in a language, is to look for oppositions: words in which both occur, and are regarded by the speakers of that language as different sounds.

However, that test is not the whole story. [h] and [ŋ] are not found in contrary opposition, yet are clearly not the same phoneme; the explanation, as Aspinea says, is that in English they are each limited to environments which don't happen to overlap.

An interesting case in English is the dental fricatives, [θ] and [ð]. Many English speakers do not even notice until it is pointed out to them, that these are different sounds (the fact that they are both spelt 'th' often misleads people). They are different phonemes, but there are only a few minimal pairs: "wreath"/"wreathe"; "sooth"/"soothe", and (marginally in the modern language) "thigh"/"thy".

Colin Fine
  • 7,454
  • 22
  • 28
  • 3
    Don't forget "ether" and "either." – James Grossmann Nov 23 '13 at 06:40
  • 2
    Another example are the /ŋg/ in "finger" vs the /ŋ/ in "singer". I had a British friend once who pronounced all such words the former way and he claimed not to be aware other people pronounced them differently. This also happens with morphemes where most English speakers think plurals and third person singular present verbs end in an "s" sound, unaware there there are three variants: /s/, /z/, and /əz/. – hippietrail Nov 23 '13 at 11:53
  • And then there are languages like Mandarin Chinese where multiple sounds are in complementary distribution, thus allowing multiple analyses. The concepts of phoneme and allophone are not really natural concepts after all, and they fit some languages better than others – hippietrail Nov 23 '13 at 11:55
  • 1
    @JamesGrossmann: well, they aren't a pair for me ;-) – Colin Fine Nov 23 '13 at 13:40
  • 1
    @Hippietrail: yes pronouncing /ŋ/ as [ŋg] (even finally) is a characteristic of the North Midlands and the North-West of England. For those dialects you can reasonably say that /ŋ/ isn't a phoneme, but that then requires you to say that "sing" is phonemically /sɪng/, with a different structure from standard English /sɪŋ/ – Colin Fine Nov 23 '13 at 13:44
  • I always have some doubts about the minimal pairs test. There appear to be very few minimal pairs that distinguish [θ] from [ð]: what if those all disappeared, like "thy" is disappearing? Woulld [θ] and [ð] suddenly just be allophones? I think they wouldn't, as they'd still be found in similar environments, even if they wouldn't be found in exactly the same words. In other words, unless I'm missing something, I think minimal pairs are useful to illustrate how two sounds are separate phonemes, but if there aren't minimal pairs, it doesn't, per se, show they aren't. – LjL Mar 02 '19 at 16:12
0

If I understand your idea of opposition correctly:

[l] and [ɫ] are allophones, which menas they are varieties of the same phoneme. Only two different phonemes can be in opposition with one another.

Here I'm guessing:

[h] is a phoneme you usually find in an onset position; [ŋ] is one you'll usually find in a coda position. Two linguistic entities can't be in opposition unless they turn up in the same place.

Aspinea
  • 699
  • 5
  • 6
  • 1
    Only two different phonemes can be in opposition with one another. - is this true? I'm pretty sure Mandarin Chinese breaks this assumption. – hippietrail Dec 23 '13 at 03:01