Monday, 4 April 2011

progressive assimilation?

Ernesto L .B Sastre wrote
I hope you can help me with this matter about phonetics. I couldn't prove that there actually existed progressive assimilation for /s/ to become /ʃ/. […]
I have been doing some online research about it, but didn't manage to find anything about it. I also asked around, but it was denied that there existed such assimilation. Your dictionary also states that there [is] only regressive assimilation.
Nevertheless , the notes I was given by my Phonetics teacher included this type of assimilation. Here is an example of it:

bookish style /bʊkɪʃ staɪl/. If there is assimilation: /bʊkɪʃ ʃtaɪl/

Have you ever heard of this before?

No, I haven’t. My immediate reaction was that this type of assimilation simply doesn’t happen. That’s why Ernesto can’t prove that it does, and also why he can find no reference to it in his on-line search.

However, phonetic research is not just a matter of finding out what published descriptions say about this or that phenomenon (in this case, they seem to say nothing). Genuine research involves making observations: observing and analysing what speakers actually say. I told Ernesto he should
Collect some evidence, and see what you find.

The phonetic context we are interested in is by no means unusual. Plenty of possible examples come to mind.
British citizen
cash some cheques
cash-strapped shoppers
crash site
wash six pairs of socks
horseradish sauce
fish soup
rush suddenly
push something


Does anyone know of any research into the phonetics of such ʃs sequences?

Introspecting, I feel pretty confident in saying that full-blown progressive (= perseverative) assimilation, thus ˈbrɪtɪʃ ˈsɪtɪzn̩ → ˈbrɪtɪʃ ˈʃɪtɪzn̩, just doesn’t happen. Nor does full-blown regressive assimilation, thus ˈbrɪtɪs ˈsɪtɪzn̩. However some kind of intermediate allophonic regressive assimilation, perhaps ˈbrɪtɪɕ ˈsɪtɪzn̩ or somethinɡ similar, seems possible.

I stick by my view that in English progressive assimilation is restricted to
(i) morphological assimilation of voicing, producing s in cats and t in kissed (compare dogs with z and raised with d); there are also possible one-off cases such as newspaper and (pre-decimal) fivepence; furthermore there is also allophonic assimilation of voicing, always in the direction of voicelessness; and
(ii) assimilation of syllabic to the place of a preceding obstruent, as in ribbon (ˈrɪbən →) ˈrɪbn̩ → ˈrɪbm̩.

I think Ernesto’s phonetics teacher was wrong. In English phonology, features typically spread leftwards, not rightwards.

37 comments:

  1. What about palatalisation of t, d, s, z before j, giving , , ʃ, ʒ?

    For example, the eye-dialect spelling gotcha for got you represents ['gɒtʃə] from /gɒt jə/. Or the word issue, which can be pronounced either 'ɪsju: or 'ɪʃu:.

    It's not full assimilation of the sort your correspondent's interested in, more a place-of-articulation assimilation. But it's definitely progressive.

    ReplyDelete
  2. 'gɒtʃə from gɒt jə also includes an assimilation of voice.

    ReplyDelete
  3. We usually categorize that as COALESCENT assimilation ("yod coalescence", LPD p. 52).

    ReplyDelete
  4. Surely 'newspaper', 'fivepence' (and also 'has to' etc) show regressive/anticipatory assimilation, not progressive?

    ReplyDelete
  5. And how do you call a change from ɔə to ɔː? Is that smoothing? How can one know whether both a diphthong and a single vowel are acceptable, correct pronunciation? Also, has the change from ˈlɑːndri to ˈlɔːndri and ˈkʌmptən toˈkɒmptən been named and documented?

    (Yesterday I bumped onto this:

    http://www.languagebits.com/phonetics-english/number-of-diphthongs-in-english/

    and couldn't quite figure out what kind of a disappearing vowel UNDERLINEaUNDERLINEə is.)

    ReplyDelete
  6. How about absurd æbˈzɜːd?

    ReplyDelete
  7. (Off-topic:) "ˈlɑːndri to ˈlɔːndri and ˈkʌmptən toˈkɒmptən" - known, documented and filed and "spelling pronunciation", I should think.

    ReplyDelete
  8. The æb- variant has been marked as regional in LPD, with an § (so-called section sign). And -z is nowhere to be found.

    ReplyDelete
  9. Oh, spelling pronunciation... Wow. Never would have said. LPD says "a pronunciation that, unlike the traditional pronunciation..." – wait a minute: isn't ɑː more traditional, older and waning?

    ReplyDelete
  10. When I try to say ˈbrɪtɪʃ ˈʃɪtɪzn̩ it comes out ˈbrɪtɪʃ ˈʃɪtɪʒn̩...

    ReplyDelete
  11. I'm not sure if you're being sarcastic. The older pronunciation lɑːndrɪ is counterintuitive, as a written au usually corresponds to ɔː.

    ReplyDelete
  12. j.house - you're right, of course, Jill. Silly me.

    ReplyDelete
  13. No, I'm really not being sarcastic at all. I was reading this

    http://www.guardian.co.uk/theguardian/1999/jun/18/features11.g22

    on The Guardian's website and, clearly, the man says people traditionally used to say ˈlɑːndri and ˈlɑːnʧ for laundry and launch. Perhaps it's intuitive to you to say ɔː because you were taught to say it that way, but those people from the past had a different view of what's intuitive, no? Perhaps once, some hundreds of years ago people said bloːd for blood.

    ReplyDelete
  14. In fact, I say lɑːndrɪ and ˈlɑːnʃ most of the time. Still, to me and about everybody else today, the default pronunciation of au is ɔː, and it was when this spelling pronunciation took hold. (Always provided it's actually a spelling pronunciation; I understand this au issue is complex.)

    Those in the know - am I imagining that some people have progressive assimilation in absurd instead of the unassimilated or regressively assimilated versions?

    ReplyDelete
  15. And progressive/perseverative assimilation of place and manner before voiced dental fricatives?
    e.g. all they /ɔːl leɪ/, and they /ən neɪ/.

    ReplyDelete
  16. JWL asks me to post this on his behalf:
    I have given thaut to this matter and over very many years ’ve been in the habit of observing many diff·rent forecasters reading the early-morning BBC shipping bulletin, watching out especially for, amongst other things, the use of the term "Irish Sea". I cert·nly have a few times he·rd 'Iris Sea' but only so rarely out of hundreds of tokens, that I've never consider·d it worth noting the name of the speaker. Such a level of occurrence I shdnt be inclined to classify as better than that of tongue-slip, so I think these observations help to bear out those of JCW.

    ReplyDelete
  17. John,
    «some kind of intermediate allophonic regressive assimilation, perhaps ˈbrɪtɪɕ ˈsɪtɪzn̩ or somethinɡ similar, seems possible.»

    I can well believe that. Even though I have a retroflex ʃ, I'm not sure I can guarantee I wouldn't say ˈbrɪtɪɕ ˈsɪtɪzn̩ myself.

    But I think you're overstating the case against 'cash-strapped shoppers' and 'bookish style' with progressive assimilation, for the following reasons:

    In a comment on your blog entry for Tuesday, 22 March 2011, "how do we pronounce train?" vp says
    «On the basis of my own (originally near-RP, now somewhat influenced by AmE) speech, I would transcribe the clusters as /tʃr/, /dʒr/.»

    But later he adds
    «Actually I've just realized that I do produce phonetic [tr] after /s/. So for me "strain" is [stɻeɪn]»

    If the canonic realization for him of the clusters /tr/ and /dr/ is as he says, we could at a pinch call [stɻeɪn] a case of progressive assimilation in the context of his idiolect, but more to the point, in my comment on that comment, I say of the students who transcribe train as tʃreɪn

    «Well, for a lot of these people it's [ʃʧɻeɪn], to use the disliked combo of t and ʃ. As clear as day. This is pretty good evidence for the credibility of their transcription [tʃreɪn], is it not?»

    Even if this [ʃʧɻeɪn] alternates with [sʧɻeɪn] (unlikely) or [streɪn] in their speech, don't you think [ʃʧɻeɪn] is that much more likely in 'British strain' and [ʃʧɻæpt] in 'cash-strapped'?

    And I don't know whether it's because that establishes the allophone ʃ in st- in one context, making it more available in another, but I'm pretty damn sure I've heard I've heard it in expressions like "bookish style" too. The initial cluster seems to make it more prone to progressive assimilation than initial sV in your other examples.

    Does nobody remember that when Janet Street-Porter first appeared on the scene many years ago, she was pretty soon lampooned as "Janet Shtrai'-Paw'uh"?

    Lipman,
    Trust you to promptly come up with 'absurd', which obviously had a perfectly uncontentious variant with progressive assimilation. I don't like it,

    But Dinora, I was totally incredulous that in LPD that -ˈzɜːd variant "is nowhere to be found". And rightly so: it comes immediately after the non-RP variant you mention.

    So Lipman, you are certainly not "imagining that some people have progressive assimilation in absurd instead of the unassimilated or regressively assimilated versions".

    Paul,
    «Basically voiced, but becoming /s/ and /t/ when a voiceless consonant precedes.»
    You seem to be another member of the post-before-you-read fraternity. John said that at the start.

    ReplyDelete
  18. The American comedy show South Park used progressive assimilation of s to esh to great effect in the phrase "fish sticks". Here's a clip (with mild cursing and an off-color joke): http://www.youtube.com/watch?v=D3u5qkqCTDQ

    ReplyDelete
  19. @Lipman:

    I'm not sure if you're being sarcastic. The older pronunciation lɑːndrɪ is counterintuitive, as a written au usually corresponds to ɔː.

    /lɑːndrɪ/ isn't "older" than /lɔːndrɪ/, and /lɔːndrɪ/ isn't a spelling pronunciation. Dobson actually claims the reverse: that the spelling of such words with "au" or "a" is derived from the predominant pronounciation. (The one exception is "aunt", which we might expect to be spelled "ant", on account of its being in the BATH set).

    Generally, the class of BATH, PALM and THOUGHT words where the stressed vowel precedes a nasal derives from words with variation in Middle English between /au/ and /a/. In some styles of speech the /au/ pronunciation took its regular development into the THOUGHT vowel: in others it had the less regular development into the PALM vowel (as in the word "palm" itself). The Middle English /a/ variant explains the pronunciations with the TRAP vowel in flat-BATH varieties of English.

    ReplyDelete
  20. Is it possible that the use of /ʒ/ by some speakers, rather than the historically expected /ʃ/, in words such as "version" and "incursion" is due to progressive assimilation from /r/? (This could apply synchronically only to rhotic speakers, obviously).

    ReplyDelete
  21. Yes. As I said, it's more complex.

    ReplyDelete
  22. I can think you can do this for:

    /...ʃ#st.../

    It's probably not progressive assimilation, but rather palatalization (or regressive assimilation) of /s/'s place to /t/.
    E.g.:

    /...ʃ#st.../
    /...ʃ#ʃt.../
    [...ʃt...]

    We've all heard /st/ clusters pronounced as [ʃt], right?

    ReplyDelete
  23. Doesn't Sean Connery do something like this?

    ReplyDelete
  24. Oops. Of course æbˈzɜːd exists. My mistake. Plus the Guardian link is not the right one.

    http://www.guardian.co.uk/uk/2001/dec/23/monarchy.comment

    ReplyDelete
  25. The only given example I can actually see this happening with is "cash-strapped shoppers". It may not be with everybody, but I find it a waste of energy to try and say /kæʃ stʃræpt/, and I wind up naturally eliding the word to /kæʃ tʃræpt/.

    ReplyDelete
  26. For some reason, [bʊkɪʃ ʃtaɪl] doesn't sound *that* weird to me, even if I can't specifically remember anyone doing something like that. On the other hand, if there was no /t/ after the /s/ it would sound a whole lotta weirder.

    @Dinora: I guess a-underline-schwa would be ire as in Ireland or sapphire, the “fuller form” being bisyllabic PRICE + LETTER. I guess dictionaries don't bother with that distinction because for most speakers today higher and hire have the same range of possible realizations.

    ReplyDelete
  27. I'm actually finding it hard to pronounce 'bookish style' with [ʃs] - whenever I try it comes out as [ʃʃ] unless I concentrate. Of the other ones, it only seems to happen in 'cash some cheques' and 'cash-strapped', but it's definitely happening. Maybe it only happens before consonants or unstressed vowels? The others seem to end up with something like [ɕs] as you suggested. I dunno, it just seems a bit presumptuous to say that this would never happen in English.

    Anyway, I'm from Edinburgh, if that helps.

    Also, this blogger comment form is an absolute nightmare to use - I've had to go back and rewrite the entire thing because I wasn't logged in to my wordpress account and it got deleted when I went back to this page.

    ReplyDelete
  28. I said
    «Even if this [ʃʧɻeɪn] alternates with [sʧɻeɪn] (unlikely) or [streɪn] in their speech, don't you think [ʃʧɻeɪn] is that much more likely in 'British strain' and [ʃʧɻæpt] in 'cash-strapped'?

    And I don't know whether it's because that establishes the allophone ʃ in st- in one context, making it more available in another, but I'm pretty damn sure I've heard I've heard it in expressions like "bookish style" too. The initial cluster seems to make it more prone to progressive assimilation than initial sV in your other examples.»

    Thanks for the confirmations of those observations, if not the reasons behind their predictability. But that would involve reading my post, and nobody reads anything much on here.

    Note that I said "more prone to progressive assimilation than initial sV in your other examples". I do see what prompted the original question about progressive assimilation, and I am now drawing attention to the "more".

    Ryan's "fish sticks" corroborates my hypothesis, but even the sV "fish soup" is not immune: compare it with "fish pie". With people who do the progressive assimilation, I hear the ʃ as perceptibly vastly longer in "fish soup" than in "fish pie", and the s as vastly shorter than it should be in an unadorned (non-cluster) onset, and although pace JWL I have admitted to the possibility of ˈbrɪtɪɕ ˈsɪtɪzn̩ in my own speech, I would not admit to the possibility of fiɕ sʊu̯p. I think I'm more likely to have the partial progressive assimilation fɪʃ ʃ͡sʊu̯p that I have described for that. I don't know why that is, but I'm betting on it being because the tɪɕ of ˈbrɪtɪɕ has no stress at all.

    ReplyDelete
  29. For absurd, I note that the pronunciation with -z- was unknown to the editors of the 13th and 14th editions of the 'Daniel Jones' dictionary, but became the primary pronunciation in the 15th edition. LPD's poll found 23% of British speakers using it. Was there a Madhur Jaffrey effect - a single TV personality using an odd pronunciation which caused that form to become established almost overnight among a section of the population?

    Alternatively, we might consider German influence, or perhaps cross-contamination from the word observe.

    ReplyDelete
  30. I agree that a likely alternative explanation for the new pronunciation of absurd is analogy. See the OED, s.v. "S": "Between vowels, and when phonetically final, a single s is mostly /z/ . But there are many anomalies and uncertainties, especially in classical derivatives: cf., e.g., absurd /æbˈsɜːd/ , observe /əbz-/ ; with regard to some words usage is divided, as in absolve /æbs-/ , /æbz-/ [...]"

    ReplyDelete
  31. Er, on the other hand there's the word absolve, in which the -z- pronunciation predominates. Taking absV- words as a whole, and ignoring absurd itself, one would say that -z- occurs in stressed syllables and -s- in unstressed, a situation which parallels the ex- prefix (compare exit ˈɛksɪt vs. exist ɪgˈzɪst).

    ReplyDelete
  32. mypresentandthoughts said...

    > See the OED, s.v. "S":

    Thanks for that: I hadn't seen your post when I posted my own last one (http://phonetic-blog.blogspot.com/2011/04/progressive-assimilation.html?showComment=1301953233815#c9041369307986419484).

    > "[...]cf., e.g., absurd /æbˈsɜːd/ , observe /əbz-/ ; with regard to some words usage is divided, as in absolve /æbs-/ , /æbz-/ [...]"

    On the (off-topic) subject of the quality of the initial vowel: note that æ in these quotes from the OED corresponds to æ̆ in the original (NED aka OED1), which should have been translated into IPA ə in OED2 (but apparently wasn't).

    ReplyDelete
  33. Paul Carley,
    «And progressive/perseverative assimilation of place and manner before voiced dental fricatives?
    e.g. all they /ɔːl leɪ/, and they /ən neɪ/.»

    Does it seem to you that cases like that are more likely to be further reduced to single l and n than say 'all the'?

    I was replying above to a Paul who said of the morphological assimilation of voicing, producing s in cats and t in kissed (compare dogs with z and raised with d) which John had already referred to in his blog entry at (i),
    «Basically voiced, but becoming /s/ and /t/ when a voiceless consonant precedes.»

    That wasn't you, but someone who posted as plain "Paul". This post has disappeared without trace, so I'm afraid you may have thought it was you I was accusing of being of the post-before-you-read persuasion. Please forgive any such impression I gave.

    ReplyDelete
  34. I still haven't made it clear. In the first part of his post, which I didn't copy, it was obvious that this other Paul thought he was introducing the progressive assimilation producing s and t in the plural and past tense morphemes for the first time, so he wasn't just appending to John's remarks the tendentious specification that the forms in question were "basically voiced".

    Perhaps I should explain that that was a further annoyance to me, because although I understand why people argue that sort of thing from the z and d in forms like 'ties' and 'tied', it's quite arbitrary to say those are the default forms. No allomorph is basically anything. All forms are equally context-determined phonological forms of higher-order entities.

    ReplyDelete
  35. Going back to Dr. Wells' motive for the blog post in question, I might say that I found only one reference to such assimilation type. Finch and Ortiz Lira (1982, p. 82) give the same example (bookish style) as a case of progressive assimilation. I think that this type of adjustment may only be possible when a word ends in /ʃ/ and the next word starts with a consonant cluster such as /st/ and problably with /sp/ too, as the title of the book these authors wrote some 30 years ago: A course in English Phonetics for Spanish Speakers. So, in my view and from my experience as an Non-native English (speaker-)listener, this kind of assimilation is possible but not in every context.

    ReplyDelete
  36. Hi I was surfing and found your blog. its very nice to have such topics in this blog.
    Thanks.

    ReplyDelete
  37. Hi,there!

    What is assimilation for:/dʒ/+j=
    Exampe: Judge your action.

    What is assimilation for:/g/+j=
    Example:Beg your pardon.

    ReplyDelete