IPA query

Jim   Wednesday, October 06, 2004, 04:17 GMT

A missing word and a missing comma: it should have been.

"Else you have the problem that Tremmert experiences: attempting to learn the IPA from your knowledge of English pronunciation, which will inevitably be influenced by your own accent."

Also look at paragraph four. Does it look familiar? I'm I quoting Mxsmanic without inverted commas? Not really: I had mean to delete that. I'd pasted it there to read it more easily but I hadn't meant to repaste it (as if you'd forgotten what Mxsmanic had written).
Bob   Wednesday, October 06, 2004, 04:20 GMT
''It seems that the RP's version is somewhere between [e] and [E] (if we trust Mac Uni). So is it incorrect to use one symbol over the other?''

Could we invent a whole different symbol for that RP sound that's somewhere between [e] and [E]. Supposed we used [e)] for that sound somewhere between [e] and [E].

Using Sampa, In my accent the ''e'' sound in ''bed'' is [E], [E:] is the vowel sound in ''bread'', [e] is the ''a'' sound in ''vane'' and [e:] is the vowel in ''steak'' in my accent. There is a phonemic distinction between length in my accent.

''Indeed, phonemic transcritions are better than phonetic ones. So, no, don't use some "English version" of the IPA that will only confuse everyone; ESL students, native speakers, those who have learnt the real IPA and those who haven't alike; use a phonemic alphabet.''

The question would be, which phonemic alphabet should be used though? There are many phonemic alphabets that will not work for my accent. In my accent ''whine'' and ''wine'' are distinct and ''qu'' and ''kw'' are distinct as well, but many phonemic alphabets ignore those two phonemes. Which phonemic alphabet is best to use?

If I were teaching a non-native speaker pronunciation, should I tell them about the ''whine'' vs. ''wine'' and ''qu'' vs. ''kw'' distinction in my accent.

''Oxford, Longman, Collins and Cambridge all use /I/ and /i:/. I suspect you are thinking about very old dictionaries or dictionaries from second-rate publishers.''

Why do very olde dictionaries use /i/ for the ''i'' in ''bit'' and /i:/ for the ''ee'' in ''reed''? Did people use to actually pronounce those sounds that way?


I have a dictionary that shews the distinction between ''wh'' and ''w'' by using [hw] for ''wh''. I don't think of ''wh'' as [hw] but as [W].

''Yes, it is silly. I agree. Of course the IPA does have diacritics to further narrow the range of the symbols but, as Tom says, it's still no infinite: there are only a finite number of diacritics too.''

Would the IPA be a little bit better if it included a few more symbols and diacritics?


How about saying that in the RP accent /e/ is pronounced [e)] rather than [e] or [E]. [e)] is my own invention.
Tom   Wednesday, October 06, 2004, 23:06 GMT
Why do very olde dictionaries use /i/ for the ''i'' in ''bit'' and /i:/ for the ''ee'' in ''reed''? Did people use to actually pronounce those sounds that way?

"The pronunciation scheme used in the first twelve editions of EPD was one that required rather few special symbols. It achieved this parsimony by transcribing the English vowels quantitatively. This meant writing /i:/ for the vowel of reed and /i/ for that of rid, using the same phonetic symbol with and without a length mark. The vowel of cord was written /O:/, that of cod /O/, and similarly for other pairs. Thus the difference in vowel quality (vowel timbre, vowel colour) between such pairs of vowels was not shown explicitly but had to be inferred from the presence or absence of the length mark"
Mxsmanic   Wednesday, October 06, 2004, 23:19 GMT
The vowel in "bed" is pronounced [E] in ALL standard pronunciations of English, be they GAE, RP, Australian, Scots, Irish, or even Indian. It's very misleading to represent the sound as [e].

The reason English publishers have traditionally used [e] (and other inappropriate symbols, no doubt) is simply that it's easier to typeset ... it doesn't require a special character. For relatively broad or phonemic transcriptions of English in isolation, without consideration of actual sounds or other languages, one can get away with this. But for ESL, it doesn't work, because it doesn't represent the real sounds and it can confuse students.

Note that IPA _can_ be used for phonemic transcriptions, Jim, and it is regularly used for that purpose. The IPA is suitable for all sorts of transcriptions, from very broad phonemic transcriptions to extremely detailed and narrow phonetic transcriptions. However you do it, though, there's no excuse for using [e] to represent [E] in any transcription intended for use with ESL students.

Vowel length is not phonemic in standard English, and pretending otherwise is also very misleading to ESL students. The distinction ESL students must make is in vowel position, not in vowel length. If the position is wrong, the vowel will be misunderstood, no matter what it's length. But if the position is right, the vowel will generally be misunderstood, even if the length is unconventional. So the position is phonemic; the length is not.
ScotsJim   Thursday, October 07, 2004, 00:31 GMT
''Vowel length is not phonemic in standard English, and pretending otherwise is also very misleading to ESL students.''

Vowel length is phonemic in English. I pronounce the ''ee'' sound in ''feet'' as [i] and the ''ie'' sound in ''relief'' or the ''ei'' sound in ''receive'' as [i:]. So, how can you say that it's not phonemic.

Mxsmanic writes ''standard English''. What do you mean by ''standard English''? I (along with many Scots) distinguish the ''ee'' in ''feet'' [i] from the ''ie'' in ''believe'' [i:]. The difference between the two sounds is only in length. Does that mean that my accent is nonstandard?
Mi5 Mick   Thursday, October 07, 2004, 02:16 GMT
I think it's unique to Mxsmanic's accent. Most English speakers pronounce "feet" with a long vowel, and "fit" with a short vowel.
Mi5 Mick   Thursday, October 07, 2004, 02:32 GMT
It's true; there's no difference between these vowel sounds, at least phonemically for me. SFSwin confirms this.

What's "vowel position"?
Mi5 Mick   Thursday, October 07, 2004, 02:33 GMT
It's true; there's no difference _in length_ between these vowel sounds, at least phonemically for me. SFSwin confirms this.

What's "vowel position"?
Mxsmanic   Thursday, October 07, 2004, 04:32 GMT
I'm surprised by how poorly understood the term "phonemic" is here.

Vowel length is phonemic if there exist pairs or sets of words with different meanings that are distinguished from each other exclusively by a difference in vowel length. No such minimal pairs or sets exist in standard English. Therefore vowel length is not phonemic.

In every case in which there is a difference in vowel length between two words, there is also a difference in vowel position. Most people (including myself) pronounce "feet" with a long vowel, and "fit" with a short vowel, but these words are also distinguished by a difference in the vowel positions: feet is pronounced /fi:t/ and fit is pronounced /fIt/. You can pronounce feet as /fit/ and fit as /fI:t/ and you will still be understood, but you cannot pronounce feet as /fIt/ or fit as /fit/ and expect to be understood. The position of the vowel is phonemic; the length of the vowel is not.

Vowel position is the position of the tongue. Vowels are pronounced by producing sound with the vowel folds (which naturally has a kind of buzzing quality), and then modifying that sound by passing it through the resonant cavity of the mouth. The mouth attenuates certain frequency bands and enhances others, and the exact frequencies affected depend primarily on the precise position of the tongue. Thus, different vowels are formed by placing the tongue in different positions, and the resulting vowel sounds can be identified acoustically by noting the frequency distribution of the sound they contain.

Conventionally vowels are often plotted on a trapezoid that schematically represents tongue positions. The vowel [i] is pronounced with the tongue high and forward, and is called a close front vowel; this is the upper left corner of the trapezoid. The vowel [u] is close (high) and back. The vowel [a] is open (low) and front. The vowel [A] is open (low) and back. Other vowels are in intermediate positions within the trapezoid. The schwa, for example, is right in the center, and is thus the "central vowel" of all central vowels, and often is the easiest and most natural vowel sound to produce.
Jim   Thursday, October 07, 2004, 04:44 GMT

Have you even bothered to look at the links I'd posted? Who am I to trust the Linguistics department at Macquarie University or you? Who'd be a better authority on the Australian accent? ... or the Kiwi accent? ... or any accent for that matter?
Mi5 Mick   Thursday, October 07, 2004, 06:22 GMT
>>"... vowel length is not phonemic in English." Bollocks. It may not be phonemic in Mxsmanic's dialect but it certainly is in mine. Here's a perfect example: consider the minimal pair "some" vs. "psalm"; in my accent the difference is the length of the vowel.<<

I don't know anymore Jim and Mxsmanic.

Maybe it's just that a syllable comprising several "sound elements" is perceived as longer compared to a simpler syllable of a single sound element. So take a diphthong: it's perceived as having a longer sound over that of a monophthong but both a diphthong and monophthong can be sounded within the same time frame -- Mxsmanic should have communicated this idea better to avoid the misunderstandings.

The "vowel position" also affects our perception, somehow.

Other vowel sounds to compare in terms of length, particularly in non-rhotic English: seed and seared; tied and tired; balm and bum. Strictly speaking, the length of these syllables doesn't carry any phonemic weight; each word can be sounded within the same time frame.

However, certain vowels have to be sounded _relatively_ longer than other ones, in natural speech, to make phonemic distinctions and/or to make natives intelligible to one another. And this is what Jim is driving at.
Jim   Thursday, October 07, 2004, 07:38 GMT
What do you know, Mxsmanic, about the Australian accent that I, an Aussie, don't? What do you know that those linguists at Macquarie University, an Australian university, don't. The vowel charts on their website directly conflict with your claim that "The vowel in 'bed' is pronounced [E] in ALL standard pronunciations of English," You might as well say that Aussie English is substandard.

Be that as it may, is <e> really such a bad symbol for use in phonemic transcriptions? I wonder, Mxsmanic, whether you insist on using the "correct" symbol for /r/. Do you? Would a phonemic transcription be wrong if it used <r> instead of the up-side down <r> of the IPA? I say it's fine because it fits English orthography. It's the same case with <e>. This is the letter we normally use to write this sound when we write English words, so why not use it in phonemic (not phonetic) transcriptions?

I say you might as well feel free to base your phonemic transcriptions on English orthography. You don't need to doggedly follow the IPA. Sure; if you like, you can; of course. However, whether we're writing for ESL students or native English speakers they're more likely to be familiar with English orthography than with the IPA and it's likely that learning English orthography will be more important and/or useful to them too.

Is vowel lenght really not phonemic? In my accent the difference in pronunciation between "psalm" and "some" and that between "balm" and "bum" is primarily one of vowel lenght. Any difference in position is negligible. If someone were to mispronounce "balm" or "psalm" by shortening the vowel, I'd think they'd said "bum" or "some"/"sum" (unless context made it clear). Vice versa if those short vowels were lenghtened.

In general it is both lenght and position which distinguishes monophthongs in English. There are long vowels and there are short vowels. How do you get about saying that vowel lenght is not phonemic? You wanted a minimal pair ... I've given you one (two if you count "balm" vs. "bum" which Mick mentioned ... though not as a minimal pair). Here's another: "dead" vs. "dared", pronounced by many Aussies as [ded] vs. [de:d] (IPA transcription).

By the way the "position" of a vowel also refers to how wide open your mouth is and the roundness of your lips.
Mi5 Mick   Friday, October 08, 2004, 02:29 GMT
Actually, "seed" and "seared" have different vowel positions; "seared" is like a long "sid".

"Tired" ~ "ti-y'd" contains a semi-vowel; "tied" does not.

But other examples where length distinguishes monophthongs in Australian English: can't and cunt, shan't and shunt, harm and hum, part and putt, calm and come, pull and pool (regional).
Mxsmanic   Friday, October 08, 2004, 07:20 GMT
There aren't any vowels in standard English that have to be pronounced long or short in order to carry meaning. Vowels tend to be long or short in the language on a fairly consistent basis, but that's an effect, not a cause. The meaning is carried by the position of the vowels.

As for university linguists, they are welcome to come here and discuss the matter if they so choose.

It's very misleading and risky to tell ESL students that vowel length is phonemic; it's not. They should concentrate on getting the vowel position correct; if they can do that, they'll be understood. They can learn the conventions of vowel length over time through imitation, if they want to diminish their accents. They don't have to worry about vowel length in order to make themselves understood, nor do they have to recognize vowel length in order to understand.
Mi5 Mick   Friday, October 08, 2004, 08:20 GMT
Well you're wrong Mxsmanic and you're obviously incapable of assimilating our ideas which are probably new to you. You repeat the same vague arguments without progressing and you're unable to argue directly against any of our points, which include real-life demonstrations. Maybe this is because you refuse to accept anything different or because you have a narrow view of standard English.

Vowel length does carry meaning in non-rhotic English and to some extent, rhotic English -- whether it's less critical or not, but valid. We've just explained it to you with some simple examples, how vowels of the same position are contrasted by length, but somehow you just can't register anything. If you still don't get it, go away and read up on it, or listen to how it works in the rest of English speaking world. Then write something here with more oomph than just the same old generalisations and vague ideas which are misleading.

By the way, this area is a problem for ESL students and it impacts on their abilities to communicate. Even if vowel length isn't a technical requirement in some rhotic forms of English, it's important enough because native speakers conform to its application, systematically.