It’s back in business with elision, folks…

I think it was John Wells who first came up with the term pseudo-elision, though I can’t quote chapter and verse.  It’s a useful term, I think, but it seems to me that there are several variations and each could do with its own name.

The term refers to the situation where the target sound actually disappears, but leaves some trace of its former presence behind.  An often quoted example from English is [ju:nɪvɜ:s:ti] for university. The vowel of the fourth syllable has been elided, but the preceding fricative has been lengthened to compensate for its absence.  This means that the end of the word is distinct from, for example, thirsty, which never had a vowel following the fricative in the first place.  This kind of pseudo-elision we could call compensatory-lengthening pseudo-elision.

Another variety involves the loss of a sound but the retention of the co-articuatory effects of that sound on a neighbour.  A good example is the word can’t pronounced as [kɑ:ʔ] (with a nasalised vowel – I can’t get that to display properly).   Although the [n] has disappeared, the nasalisation of the preceding vowel remains, making the word distinct from cart = [kɑ:ʔ].  The [ʔ] could also be [t] in both words, of course.  How about co-articulatory pseudo-elision for this one?

There is at least one other kind I can think of for English.  An example of what I am thinking of is what can happen to words like support and suppose when they lose the [ə] of the first syllable.  [spʰɔ:t] and [spʰəʊz] are possible pronunciations of the words.  However, these appear to contravene the normal distributional characteristics for strongly aspirated voiceless plosives in English, which do not normally occur in initial s+C clusters. I spose (ho, ho) we could call this anomalous distribution pseudo-elision..

I notice that pseudo-elision does not appear in SID. I must do something about that. There. That’s my first New Year’s resolution.

7 thoughts on “Pseudo-elision

  1. In one of his recent blogs Jack Windsor Lewis mused on elisions as well. One of his examples was the word ‘co-operate’ with the elided forms /kwɔːpəreɪt/ and /kɔːpəreɪt/. Let’s concentrate on the former of the two. Where does the /w/ come from? My suggestion is that it’s the remainder of two processes: 1. glide insertion leading to /kəʊ`wɔːpəreɪt/ and 2. deletion of the initial diphthong resulting in /kwɔːpəreɪt/. Does this fit into your pseudo-elision system somehow?

  2. Petr,

    That’s a possibility, I suppose. However, as the diphthong əʊ already has a hiɡh back rounded vowel quality as its final tarɡet, it would seem uneconomical to posit the insertion of a w and then the elision of the whole diphthong. Simply, deleting the ə portion of the diphthong would give us what we need. This would mean recognising what one might call intra-segment elision. Watch this space for some more musings on that topic.

  3. John,
    your explanation is definitely more elegant in descriptive terms than mine. But I assume you will not deny the fact that at the boundary between the 1st and 2nd syllable of co-operate in other than a very careful pronunciation one can and most probably will insert a wyn as a consonantal bridge over the vocalic hiatus resulting in /kəʊ`wɔːpəreɪt/. Following your line of arguments would mean that in allegro speech the wyn glide insertion is suppressed and the the back vowel mutates to a wyn. Hmm!

  4. When Petr sez “at the boundary between the 1st and 2nd syllable of co-operate in other than a very careful pronunciation one can and most probably will insert a wyn as a consonantal bridge over the vocalic hiatus resulting in /kəʊ`wɔːpəreɪt/” I’m afraid I can’t really agree with “most probably will”. I feel especially vigorous articulation is the likely pre-requisite for such an occurrence. I think a version very often occurs whose transcription cd be equally reasonably either /kəʊ`ɔːpəreɪt/ or /kə`wɔːpəreɪt/. The latter notation wd of course be needed to represent Petr’s relatively vigorous articulation. I think it shd be remembered that the second element of the GB ‘goat’ diphthong is a wyn (as I call it) or a ‘w’ if you prefer. Various people have transcribed that diphthong as /ow/ including John Wells for his own speech (in maître phonétique in 1962). By the way I hope no EFL student will run away with the idea that there’s anything undesirable about the classic pronunciation /kəʊ`ɒpəreɪt/ even tho I do claim that it’s probably no more commonly he’rd a version than the one we’ve been quoting for ‘co-operate’.

  5. First, I should like to add that the wyn-symbol I used in my transcription should have come out as a superscript symbol in this blog and thus signal a linking glide (similar to the /j/-glide in [weɪjɔːl] transcribed with a superscript j). I’m sorry I forgot to mention this in a follow-up comment. When one adds stress marks ‘co-operate’ should be transcribed as [kəʊw`ɔːpəreɪt] with superscript wyn. So no vigour is needed for pronouncing it this way, is there.
    Second, I’m afraid I’m not yet warm to the idea that the second element of the GB GOAT diphthong is a labio-velar approximant per se, unless you convince me of the fact that you always pronounce the second element of this diphthong with a double stricture of the vocal tract – a bilabial and a velar one.

  6. Petr,

    A close back rounded vowel, which is what ʊ is, by definition must have a double stricture, one labial and the other velar. In fact, all rounded vowels are double articulations.

  7. If all rounded vowels have a double stricture, then all spread vowels likewise have a double articulation: one is labial and the other palatal or central.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.