WALL WOMEN.

Jim Quinn’s 1997 “Phillyspeak” is an amusing “guide to Philadelphia English” by the author of the immortal American Tongue and Cheek: A Populist Guide to Our Language, which is such a powerful and irrefutable blast at prescriptivist poppycock I once bought a bunch of copies for $1 each and handed them out to people who I felt might benefit (it actually turned jamessal into the doughty descriptivist he is today!). I was given the link by a proud Philadelphian who gave it her cautious stamp of approval. Here’s the start:

For some reason WHYY’s Morning Edition keeps changing the people who read the traffic tie-up reports on Shadow Traffic. And for some reason, none are native Philadelphians anymore. So I have to start my day without the hero who used to warn usabout gaper delays “caused by an overturned tractor trailer on the Wall Women Bridge.”
“Wall Women” — what a superb and bizarre way to pronounce the name of America’s greatest male poet of the 19th century, Walt Whitman! Nobody else says it that way, and no true Philadelphian can say it any other way.
No wonder I love, and proudly speak, Philly’s dialect. Where else can you tell somebody “I hate our winners,” and know they’ll understand you mean, not “I hate the Flyers” (the closest we come to a championship team), but “I hate the weather in January”? Where else — this is one of my favorites — can a man named Ian and a woman named Ann go through life hearing their names pronounced exactly the same way?

I like dialect pride, especially when expressed with such pungency. (Thanks, des!)

Comments

  1. I know someone named Anne from Massachusetts (different accent of course, but same phenomenon) who went to school with a boy named Ian. The kids called them “Boy Anne” and “Girl Anne.”

  2. Funny that even in standard AmE the difference between Walt Whitman and Wall Women is actually only two missing glottal stops.

  3. Do they really say “gaper delays” in Philly? I am from next door in Maryland, and the only people I’ve heard use that term are Chicagoans.

  4. Quinn and Philly? Oh, I must have this book.
    I’ve got a neighborhood buddy from Philly — from this part: http://www.dailyfinance.com/photos/most-dangerous-neighborhoods/ (#16) –and recently, complaing about what he considered a bullshit ticket, he used a word I wouldn’t have expected to be in his idiolect; he said, “I ain’t no scofflaw.” I loved it. I never thought that man would have used a double negative, let alone ain’t.

  5. That neighborhood looks a lot like the one in The Wire, quite solid in places.

  6. “scofflaw” is perhaps the finest word in Murkin.

  7. Paul (other Paul) says:

    For us across the Pond, what are “gaper delays” please.

  8. @Paul: I assume ‘gaper delays’ are caused by people who slow down to gape at an accident. We call it ‘rubbernecking’ hereabouts.

  9. Trond Engen says:

    Too bad I couldn’t find it with English subtitles, but for those who can follow:
    The Norwegian Federation of Gapers

  10. J.W. Brewer says:

    Is “scofflaw” really only an Americanism? It sounds positively Dickensian. There’s probably a minor character in, I don’t know, Martin Chuzzlewit or something named “Obadiah Scofflaw.”

  11. “Scofflaw”: invented in the USA.

  12. JWB: So much so that its creators are actually known. Quoth Etymonline:

    The winning entry in a national contest during Prohibition to coin a word to characterize a person who drinks illegally, chosen from more than 25,000 entries; the $200 winning prize was split between two contestants who sent in the word separately, Henry Irving Dale and Miss Kate L. Butler. Other similar attempts did not stick, cf. pitilacker (1926), the winning entry in a Pennsylvania Society for the Prevention of Cruelty to Animals contest to establish a scolding word for one who is cruel to animals (submitted by Mrs. M. McIlvaine Bready of Mickleton, N.J.).”

    Prohibition having ended some time ago, it’s now usually applied to people who ignore traffic tickets. The reason for the archaic feel, I think, is that verb-plus-object compounds (sometimes called “tosspot compounds” after a prominent example of the type) are no longer productive in English, except for rare conscious coinages like this one.

  13. I love the Philly dialect. I love how it sounds both Northeast urban and also southern. I want to start a water taxi service called the Philly Ferry. The slogan I made up is: “The Philly Ferry: On the go on the water.” But it has to be said by a real native: “The Philly Furry: Awn the gew awn the wudder.” All it’s missing is the long /I/ that’s super backed like a West Country accent. Maye it should be “I’m on the go on the water,” but somehow that’s less slogany.

  14. There is currently a spam comment above that I love; it reads: “When you are embarrassed and don’t really know the best way to write the term paper services essays, you would to buy research paper from the professional research papers writing service. That will really save valuable time and money.”
    Where do I to sign up? In addition to the artful prose, I like the counterintuitive assertion that paying someone to do something instead of doing it yourself will save money.

  15. J.W. Brewer says:

    Quinn’s piece is very good, and seems like it might hold the record for having the most technical phonological description of any article ever published in a modern US alt-weekly. I grew up about 25 miles SW of South Street (just on the Delaware side of the Del./Pa. line), natively speaking a pretty “high” form of formally-edumacated SAE with only a little regional substrate, but I definitely have some of these features (the very fronted or tensed or whatever the right jargon is O, and the mayor/mare merger, for instance).

  16. My limited personal experience suggests that a more “reduced” pronunciation is often interpreted as authentic and micro-local, even though it may be very widespread geographically and has more to do with age or race or class or whatever. I know that after I moved to Bawlmore, Merlin, I was sometimes praised for sounding like a native Bawlmorean.

  17. I’m surprised to see “winner” for “winter” classed as a Phillyism: that’s how I say it, and I learned my English in California.

  18. I echo Tom Recht.

  19. Well, I don’t think the term necessarily means “phenomena found only in Philly”; it also covers “nonstandard features prominent in Philly even if other folks have them too.”

  20. “Funny that even in standard AmE the difference between Walt Whitman and Wall Women is actually only two missing glottal stops.”
    Marc, those aren’t glottal stops, those are unreleased t’s. The p-t-k finals are alive and well in AmE.
    “I’m surprised to see “winner” for “winter” classed as a Phillyism: that’s how I say it, and I learned my English in California.”
    Where in California? Concord here – Bay Area – those are distinct in my variety.

  21. Jim, if you pronounce “winner” and “winter” differently and “Walt Whitman” without glottal stops, your realization of /t/ is very different from most Americans’, I think.

  22. Jim, aren’t unreleased t’s realized as glottal stops?
    It’s the same phenomenon that makes “ship, man” and “shit, man” homonymous.
    As for the finals, I think ‘t’ is rapidly fading into a glottal stop. Pronouncing the final t in ‘cat’ with the same realization as the final ‘p’ in ‘cap’ or the final ‘k’ in ‘took’ almost sounds foreign to me.

  23. Ben: If time is money (and it is for most of us) then you can certainly save money by buying things. To grind coffee by hand would take me approximately forever, so I own a grinder and sometimes buy pre-ground coffee, so that I can spend time on my own money-making pursuits.

  24. marie-lucie says:

    winner/winter
    In the Canadian regions that I know, a winter is not a winner.
    The first people I met who surprised me by saying inneresting and innerested (in 4 syllables) were Californians.

  25. Where in California? Concord here – Bay Area – those are distinct in my variety.
    LA and later Oakland/Berkeley. I think I do distinguish winter/winner sometimes, but not always (depending maybe on the formality of the occasion, or maybe just on how much sleep I’ve had).
    Glottal stops and unreleased [t]s aren’t the same: with the latter, you make the alveolar closure for the [t] but don’t release it, with the former there’s only a closure in the glottis. That said, in a word like Walt where there’s a preceding alveolar, I don’t think there’s any difference.
    Speaking of word-final [p t k], I noticed on a recent visit to London that hip young people there seemed to be pronouncing these with a very audible ‘pop’-sounding release, exactly the opposite of the unreleased/glottal-stop realization common on this side of the water. UK correspondents, please report: is this a London thing? A middle-class thing? An urban yuppie thing?

  26. mayor/mare merger
    Are these pronounced differently in (Standard) English, and what is the difference?
    aren’t unreleased t’s realized as glottal stops?…the same phenomenon that makes “ship, man” and “shit, man” homonymous.
    Are there really dialects where they are homonymous? I would have thought that even in the unreleased articulation there is a difference. For non-native speakers it can be very difficult to hear the difference, but for a native speaker the subtle difference is normally audible.
    I’m surprised to see “winner” for “winter” classed as a Phillyism: that’s how I say it, and I learned my English in California.
    When people say that ‘winter’ is pronounced as ‘winner’, are they saying that the two words sound very similar, or that they are not conscious of any difference when pronouncing them. IMHO there is a difference.
    When I say ‘winter’, I’m aware that the pronunciation can be very close to ‘winner’. And yet I’m conscious of pronouncing them differently. In my most casual pronunciation of ‘winter’, there is still a very slight difference in the pronunciation, a slight catch or very lightly articulated /d/. Some people might be forgiven for thinking that they are pronounced the same, but as long as the intent to pronounce them differently is there it’s not correct to say they are pronounced ‘the same’.
    So my question is: Do those people who pronounce ‘winner’ and ‘winter’ the same really make no distinction between the two, or are they just saying that the pronunciation is very close?

  27. When speaking rapidly I’m sure I pronounce “winner” and “winter” identically. But I pronounce “mare” with one syllable and “mayor” with two, at least in careful speech. I don’t have IPA handy at the moment, but “mare” has an open e and “mayor” a close e and an r-colored schwa.

  28. Bathrobe: as long as the intent to pronounce them differently is there it’s not correct to say they are pronounced ‘the same’
    I don’t think that’s true. If the spectrograms are identical, or if listeners can’t tell which token is which out of context, then the pronunciations are the same, whatever the “intent” (I’m not actually sure what you mean by that word). It sounds like you do make a very slight phonetic distinction between “winner” and “winter”; I’m sure that I don’t, at least in some tokens.

  29. Like Marie-Lucie my experience of Canadian English (in Ontario and Quebec, in my case) is that WINNER and WINTER are quite distinct, but when I taught in the Southern United States they were (in casual speech) homophonous, with earlier intervocalic /nt/ and /n/ having merged as /n/. I distinctly remember a misspelling “inter city” for “inner city” in an essay: it’s a kind of error which I had never encountered North of the border.

  30. Darn, it’s no book! At least I finally read and enjoyed the essay — thanks, LH (and once again, for old times’ sake, for saving me from the dark side).
    That neighborhood looks a lot like the one in The Wire, quite solid in places.
    Yeah, it’s a scary place. More than a person per day is killed in Philly, and a lot of that killing takes place right there, blocks from where my friend lives. Of course, a lot of really good people live there, too — most of the residents are good people — but there are enough hardcore criminals and misguided kids for taking a stroll to be dangerous. Even on my friend’s block, where a lot of kids play outside, a crew started selling crack on the corner, and a few months ago a body turned up. The drug war really is a travesty.

  31. Marc,
    “Jim, aren’t unreleased t’s realized as glottal stops?
    See Tom Recht’s comment about t’s and glottal stops. They sound alike to you because in your variety they are phonemically identical. Here on the West Coast if anything they sound Polynesian. No, not really.
    “It’s the same phenomenon that makes “ship, man” and “shit, man” homonymous.”
    In your particular dialect of English. These are definitely not homophonous in most varieties of English.
    “As for the finals, I think ‘t’ is rapidly fading into a glottal stop.”
    They are where they are, but not where they aren’t. It is not happening in North America, where intervocalic t’s are beingcoming flap r’s – soft but still voiceless and distinguishable from d’s.
    It’s not just a London thing either; a woman from Aberdeen, who had been living in the US for many years BTW, had to repeat the word “thirty” to me twice before I caught what she meant. That’s a pretty good distance from London, but it’s still on the east coast. I wonder if the same thing is happening is Liverpool or Bristol. Or in Danish, speaking of the east coast.

  32. I don’t think that’s true. If the spectrograms are identical, or if listeners can’t tell which token is which out of context, then the pronunciations are the same, whatever the “intent” (I’m not actually sure what you mean by that word).
    I tend to disagree here. If the speaker is consciously pronouncing a /t/, even if it’s not very distinct, there is a /t/ in there. If his children can’t hear the difference and decide that ‘winter’ and ‘winner’ are pronounced exactly the same, and proceed to pronounce it that way, then there is no /t/. There has also been a linguistic change. Presumably a very slight difference will come up on a spectrogram.

  33. Incidentally, I’m not talking about people consciously adopting spelling pronunciations, I’m talking about the existence of the phoneme in the speaker’s mind when the word is being articulated. Perhaps I’m old-fashioned in clinging to the psychological reality of the phoneme, but I’m still doubtful about purely phonetic approaches that look solely at the phonetic matter of language.

  34. But there are two different issues here. One is the psychological reality of the phoneme to the speaker, and I don’t think many people would deny that if a speaker things they’re pronouncing two words differently, the difference is real to them. But if nobody else can hear the difference, it doesn’t make sense to treat it as an objective fact.

  35. Bathrobe says:

    speaker things they’re pronouncing two words differently, the difference is real to them. But if nobody else can hear the difference, it doesn’t make sense to treat it as an objective fact.
    The question is, is this possible? If a speaker is consciously pronouncing words differently according to the phonology of their language, how can the difference be completely inaudible?
    On the other hand, I do think there might be cases where some people think they are pronouncing things when they are not. I can’t think of any specific examples, but in something akin to ‘prescriptive grammar’ there are people, I suspect, who are capable of thinking that they are pronouncing something according to the spelling when they are actually pronouncing it according to the innate phonology of the language.
    But other than this, and the case of people with speech difficulties, I would be curious to know of any cases where speakers are consciously producing a sound that is completely inaudible to anyone else.

  36. If a speaker is consciously pronouncing words differently according to the phonology of their language, how can the difference be completely inaudible?
    I think I’m saying -ible and -able differently, but I’m pretty sure I’m really not—that even I wouldn’t be able to tell the difference if you played examples back to me. I think that sort of thing is fairly common, and I’m surprised you find it so inexplicable. Another example: I’ve met Spanish speakers who refuse to accept that b and v are just two ways of spelling the same phoneme; they’re convinced they say them differently.

  37. Tom Recht says:

    I would be curious to know of any cases where speakers are consciously producing a sound that is completely inaudible to anyone else
    There’s the case of gesture overlap. A speaker may actually make the closure for the [t] in a phrase like imperfect learning, but if that gesture is masked by the preceding velar closure for the [k], it’ll come out as [ɪmpɚfɛk lɚnɪŋ] nevertheless. Different motor plans, same phonetic output.

  38. I think I’m saying -ible and -able differently, but I’m pretty sure I’m really not
    I would call that a spelling pronunciation (or spelling-influenced perception of pronunciation). I know they are spelt differently, but I personally wouldn’t regard the two as being pronounced differently. For instance, I suspect Daniel Jones would represent them both as something like /əbl/. (I know DJ is a bit old, but I cite him here as a linguist who represents pronunciation norms as distinct from spelling norms.)
    A speaker may actually make the closure for the [t] in a phrase like imperfect learning, but if that gesture is masked by the preceding velar closure for the [k], it’ll come out as [ɪmpɚfɛk lɚnɪŋ] nevertheless.
    I guess that is possible. Personally, even if I say ‘imperfect learning’ quickly, there’s normally still an audible /t/ in there — maybe just a slight catch before the /l/ (that is, the /t/ isn’t released) — but still a slightly different pronunciation. But I could imagine that in really rapid speech the /t/ could be completely swallowed up.
    What bothered me with the earlier comments was that it wasn’t clear whether people were saying “I pronounce ‘winter’ and ‘winner’ identically” or “when I pronounce ‘winter’ and ‘winner’ they come out sounding the same’. The first would be equivalent to saying “I always pronounce ‘imperfect’ as ‘imperfec’ (because for me the ‘t’ no longer exists in that word)”; the second would be equivalent to saying “When I say ‘imperfect learning’, it comes out as ‘imperfec learning’ because the ‘t’ basically becomes inaudible in that particular situation”. It seems to me there is a rather large difference between the two.
    Tom, since you say that ‘winter’ is pronounced ‘winner’ in your speech, does that mean that Tom Recht’s Pronouncing Dictionary would represent them both the same, as /wɪnɚ/?

  39. I would call that a spelling pronunciation (or spelling-influenced perception of pronunciation). I know they are spelt differently, but I personally wouldn’t regard the two as being pronounced differently.
    This… doesn’t make much sense. You can’t just wave away the question of pronunciation with “I personally wouldn’t regard”; the whole point at issue is what “pronounced differently” means, and my position is that you have to distinguish the speaker’s psychological perception from differences that can be considered objective in that a recording device shows them and people can consistently distinguish them when played back in isolation. You seem to be saying that there is no such difference, that psychological perception = physical reality, and you’re not really responding to what I think are clear counterexamples.

  40. Hat, I think you are misreading Tom and Bathrobe. A distinction in articulation may be perfectly objective (showing up on an MRI of the vocal organs, for example) and yet produce no detectable auditory difference. For example, I articulate /r/ with an alveolo-palatal gesture rather than a retroflex one. It makes no difference to the sound stream (as far as I can tell — a spectrogram might say otherwise) whether the tip of my tongue is turned up or down, but I’m sure it would show up in the pictures. This is not at all the same as asserting that if a North American thinks they say Peter with a voiceless consonant, then they actually do.

  41. There are three separate situations:
    1) Clearly audible phonetic distinction; the speaker and all hearers agree.
    2) Distinction detectable to MRI but not to hearers, who mistakenly claim the speaker is not making a distinction they cannot perceive.
    3) Distinction exists only in the head of the speaker, being imperceptible to both hearers and recording devices.
    I’m not seeing Bathrobe acknowledging the last of these. (Not sure why you’re lumping Tom in there; as far as I can tell, he’s making the same argument I am: If the spectrograms are identical, or if listeners can’t tell which token is which out of context, then the pronunciations are the same, whatever the “intent”.)

  42. Hat, I thought you would dive for your copy of Daniel Jones to prove me wrong about -ible/-able.
    I checked the Internet and found that:
    (1) Merriam Webster treats ‘audible’ and ‘laudable’ as homophonous (apart from the extra ‘l’ in ‘laudable’), i.e., \ˈȯ-də-bəl\ vs \ˈlȯ-də-bəl\
    (2) Macmillan gives /ˈɔːdəb(ə)l/ and /ˈlɔːdəb(ə)l/ for British English, but /ˈɔdɪb(ə)l/ and /ˈlɔdəb(ə)l/ for American English. Despite this, they both sound the same to me in the recordings given. Perhaps Americans can hear a difference within their own phonology; I can’t since I have a different phonology.
    I agree that If the spectrograms are identical, or if listeners can’t tell which token is which out of context, then the pronunciations are the same. My point is that, if such is the case, then apart from certain purely physical / phonetic phenomena such as the one that Tom pointed out, then I suspect the speaker actually is pronouncing them the same. That is, both ‘winter’ and ‘winner’ are, in fact, /wɪnɚ/ in the speaker’s ‘phonological representation’ of the word (the speaker’s ‘pronouncing dictionary’). It seems to me that a speaker who consistently pronounces ‘winter’ as /wɪnɚ/ isn’t actually pronouncing it as /wɪndɚ/ at all! He/she’s pronouncing it as it sounds: /wɪnɚ/. The speaker may be aware of ‘winter’ as the spelling form (‘canonical form’), seen as the ‘correct’ form of the word, but is not his/her actual pronunciation.
    I don’t agree that you and Tom are on the same side in this. Tom gave an example where a sound ‘disappears’ in a specific articulatory environment, i.e., there is a gap between the phonological representation and the way it is phonetically realised due to articulatory restraints. You gave an example where there is a gap between some kind of ‘canonical form’ and the actual phonological form. That is, before ‘audible’ ever reaches the stage of taking a specific phonetic shape, it has already been degraded by phonological (not phonetic) considerations from /ˈɔdɪb(ə)l/ to /ˈɔdəb(ə)l/, whereby (to simplify) unstressed vowels become schwa.

  43. Bathrobe says:

    I think my point is that there are possibly three different levels here:
    1. Canonical form, which is largely represented by the spelling. This is the form regarded as the ultimate basis of pronunciation, used when spelling words out with extra clarity. ‘I want you to say it in an audible voice! Did you hear me? AU-DI-BLE! So stop mumbling in your beard!’
    With English spelling, the problem usually arises with words like ‘debt’ or ‘bough’, where the spelling doesn’t give an acceptable ‘canonical form’.
    2. Phonological representation: The form that people have in their minds when actually pronouncing a word. Having studied linguistics, for me this is in IPA. But even without IPA I believe that most people have a specific target pronunciation in their own mind. In a case like ‘audible’, the target pronunciation for me is /ˈɔːdəb(ə)l/, not /ˈɔːdɪb(ə)l/. Consistently coming out with /ˈɔːdɪb(ə)l/ would sound strange and affected to me, although not for Hat, so this may be a dialect difference.
    For a speaker who says ‘coughing’ as ‘coughin”, the ‘canonical form’ may be \ˈkȯfɪŋ\ or /kɒfɪŋ/, but the ‘phonological representation’ (the way the speaker is intending to pronounce it) is definitely \ˈkȯfən\ (Merriam Webster) or /kɒfən/ (Macmillan British)! It’s not a case of ‘falling short’, it’s a case of actually intending to pronounce it that way, although some speakers might deny it or admit that they are ‘wrong’.
    3. Phonetic realisation: The way the word actually emerges and is recorded by physical means or perceived by native speakers who are listening. Sounds may, as Tom pointed out, disappear in certain circumstances.
    My question then is: When people say that they pronounce ‘winter’ as ‘winner’, are they talking about 2 or 3? If ‘winter’ is invariably pronounced the same as ‘winner’, then surely the two words have been merged at level 2, although they may still be separate at level 1, mainly due to the influence of spelling.

  44. Hmm… With your clarification, I’m not sure we have any real disagreement. Your last question is a good one.

  45. The case of -able/-ible is particularly hard, because unstressed vowels in English are normally out of conscious attention. It was only a few years ago that I convinced myself that I have the weak vowel (abbot-rabbit) merger, which means that unstressed /ɪ/ has merged with /ə/. The word that finally persuaded me was chicken, where if the two vowels have the same quality, you don’t have a WVM, and if they are different, you do. Note that the spelling -en doesn’t bias you one way or another, which is a great help. (Not everybody is fully merged or fully unmerged: see this Antimoon page for some of the variants.)

  46. Are all vowels in English going to merge into schwa one day? Speakers in other European languages are so particular about pronouncing them properly, and it’s typical of English (or US or Australian etc.) speakers of German & Norwegian to mix up Es and As as schwas.

  47. enurlinimiter says:

    The next time I read a blog, I hope that it doesnt disappoint me as significantly as this 1. I mean, I know it was my choice to read, but I really thought youd have something intriguing to say. All I hear is usually a bunch of whining about something that you could fix if you happen to werent too busy looking for attention.
    [Interesting spam left for context --LH.]

  48. Tom Recht says:

    Tom, since you say that ‘winter’ is pronounced ‘winner’ in your speech, does that mean that Tom Recht’s Pronouncing Dictionary would represent them both the same, as /wɪnɚ/?
    Well, as I said above, in my speech I actually have two pronunciations, [wɪnɚ] and [wɪnthɚ]. (I can’t figure out how to make that h superscript in HTML.) I don’t think I have anything like [wɪndɚ] or [wɪnɾɚ], though. I don’t know if this means that in my lexicon there are two phonological representations for the word (the choice between them being situation-sensitive or random), or that there’s one representation and a postnasal-t-omission rule which is only sometimes applied. I’m not even sure if that’s an empirically meaningful question.
    I agree with Hat’s “three separate situations”, except that I think there’s also a
    4) An articulatory distinction exists, but no corresponding auditory distinction (as in my gesture-overlap example or John Cowan’s alveolar vs. retroflex [r]).

  49. David Marjanović says:

    Interesting spam!
    From a recently closed thread, because it fits here:

    Mongolian manages to do without diacritics using clever orthography while Kazakh which is quite similar phonologically uses almost half a dozen diacritical letters:
    ң
    ғ
    қ
    ұ
    ә
    і
    and this is on top of Mongol ө, ү and 33 letters of Russian Cyrillic.

    Well. The differences in the phonology are indeed small, but they explain these letters:
    ң – /n/ and /ŋ/ have merged in Mongolian, [ŋ] is now the syllable-final allophone; thus, the letter н suffices for both.
    қ and ғ – Чингиз Хаан pronounced his title Қаған, and it’s spelled accordingly in Mongolian script, but since then, /q/ has become [x], and /ʁ/ mostly disappeared, creating the long vowels (elsewhere it merged back into /g/).
    ә – /æ/ is a Turkic thing that Mongolian never had.
    ұ and і – these are the normal /ʊ/ and /ɪ/; the letters у and и for [u] and [i] are used for what is phonologically /ʊw/ and /ɪj/, of no concern to Mongolian.

    Are all vowels in English going to merge into schwa one day?

    For the people of Missoura, I hear, all unstressed vowels have already merged.
    And in German, all unstressed vowels* merged early enough that they’re all spelled e!
    * Except in unstressed morphemes that consist of a single syllable, like -nis and -ung, and in more recent loans, most from Classical or French. All these keep their distinct pronunciations, but all are short; long vowels cannot exist without at least secondary stress, in stark contrast to neighboring Czech or Hungarian or not neighboring Finnish or Mongolian.

    1. Canonical form, which is largely represented by the spelling. This is the form regarded as the ultimate basis of pronunciation, used when spelling words out with extra clarity. ‘I want you to say it in an audible voice! Did you hear me? AU-DI-BLE! So stop mumbling in your beard!’

    This only exists for written languages, I guess. In my dialect, we actually switch to Standard German for such extra clarity – it tends to have more phonemes per word, too.

    3. Phonetic realisation: The way the word actually emerges and is recorded by physical means or perceived by native speakers who are listening. Sounds may, as Tom pointed out, disappear in certain circumstances.

    And sounds may approach each other without completely merging. Have a look at Exhibit A – I concur: many on the other side of the pond turn -nt- between vowels into a nasalized flap that sounds much like [n] if you’re not used to it, but it’s not quite the same (my occasional claims of Trahno in Canada are exaggerated). Now, I’m sure some do take the next step and merge it completely into [n], turning the tubes of the Internet into innertubes – but most don’t.
    Exhibit B:

    Jim, aren’t unreleased t’s realized as glottal stops?

    In environments (other than preceding /s/) where English /p t k/ are not aspirated, the aspiration isn’t simply left off, it’s actively taken away by holding one’s breath – by articulating a glottal stop. Before pauses or in consonant clusters, they are often additionally unreleased; when they are released, some people release the glottal closure later, producing ejective consonants – I’ve heard both RP speakers and Americans end like with an ejective; it’s quite an experience. :-)
    In front of /n/ and /m/, and in front of pauses, many Americans regularly turn /t/ into an unreleased glottal stop. Lots of accents in England, maybe all except RP by now, turn every unaspirated /t/ into a glottal stop unless there’s a /s/ in front of it (…in which case it may not actually be /t/, but that’s another story).

  50. David Marjanović says:

    I can’t figure out how to make that h superscript in HTML.

    You’re not supposed to (well, the tag is <sup>, but it probably doesn’t work here anyway). It’s a separate IPA character: ʰ, called “small modification sign H” by Unicode.

    John Cowan’s alveolar vs. retroflex [r]

    These do actually sound slightly different, and I bet they have nonrandom geographical distributions. But dental and interdental fricatives can be made to sound exactly the same. As soon as I found out, I stopped doing interdental ones – too much work for someone who isn’t used to move his jaw to speak.

    Except in unstressed morphemes that consist of a single syllable, like -nis and -ung

    Or indeed -land and -mann: the parts of compounds stay separate phonological words, unlike in English.

  51. seo link building says:

    Incredible points. Solid arguments. Keep up the amazing effort.
    [Defanged but kept for context —LH.]

  52. For the people of Missoura, I hear, all unstressed vowels have already merged.
    Well, no. All stressed reduced vowels, yes. There remain a vast number of unstressed and unreduced vowels, what are marked in dictionaries as secondary stresses; they are only predictable on diachronic grounds, and not always then. Indeed, in the accents that show happY-tensing (probably a majority of all speakers by now), unstressed final /ɪ/ has actually gone from reduced to unreduced. I don’t expect to see reduction of all unstressed vowels a la Russian or Portuguese any time soon, if ever.
    Incredible points. Solid arguments. Keep up the amazing effort.
    For once a spammer, detestable as he is, has told neither more nor less than the exact truth. Perhaps there is hope for his soul after all.

  53. Bathrobe says:

    in my speech I actually have two pronunciations, [wɪnɚ] and [wɪnthɚ]. … I don’t know if this means that in my lexicon there are two phonological representations for the word (the choice between them being situation-sensitive or random), or that there’s one representation and a postnasal-t-omission rule which is only sometimes applied.
    In structuralist days would they have spoken of two allomorphs?
    But a rule-based approach may be better. How do you pronounce ‘splinter’, ‘sprinter’, and ‘sinter’?

  54. How do you pronounce ‘splinter’, ‘sprinter’, and ‘sinter’?
    Or how about “the second Mrs de Winter”?

  55. marie-lucie says:

    Sinter ??
    I think that center (the American spelling of centre) is part of the same series.

  56. Tom Recht says:

    Sinter isn’t in my lexicon, but for sprinter, splinter, de Winter I think I’d be increasingly likely to pronounce the t in that order.

  57. And centre (being a more common word than sinter)?

  58. Bathrobe says:

    Or banter, canter, ranter, enter etc., maybe Fanta, Santa, shanty, etc.

  59. marie-lucie says:

    LH, thanks for the sinter page. I had never heard of this process.

  60. Tom Recht says:

    Bathrobe, I don’t think anyone would omit the [t] in shanty since there’s a secondary stress following. For the other words you ask about I think it’s a question of frequency: for me, [t] is optional is center, enter, Santa, but not in banter, canter, ranter.

  61. Tom, thanks for the answers to the persistent questions. That’s an interesting ‘gradient’ of pronunciation you have there. I’m not sure how you would explain it theoretically, but the phenomenon has been outlined much more clearly now. Not an automatic omission or assimilation of the ‘t’, instead an omission/assimilation in frequently-used word.
    Incidentally, I would probably use the same nasalised vowel followed by an alveolar tap in both ‘winter’ and ‘shanty’.

  62. I think I mostly agree with Tom: the t can disappear in very common words like center, enter, inter- (net, national…), but not less frequent* words like banter, canter, splinter, and definitely not words i’ve never had to use before like sinter.
    *less frequent than very frequent

  63. Maybe less frequent than very frequent, but surely ‘I’ve got a splinter in my finger’ (or ‘I have a splinter in my finger’) is everyday enough to warrant the dropping of the ‘t’…

  64. marie-lucie says:

    -nt- > -nn-
    I wonder if the pronunciation differences even in the speech of a single person also have to do with the quality of the stressed vowel before the consonant cluster.
    As I (and Etienne) mentioned before, in Canada winter and winner are always clearly differentiated, but it seems to me that I have heard both Santa Claus and Sanna Claus (versus Santa alone). In this region there is Saint Francis Xavier University, popularly known as Saint FX, pronounced both Saint EvEx and SainnEvEx. Similarly there is a francophone institution called Université Sainte-Anne, known in English as SainnAnne.

  65. David Marjanović says:

    Hat, have you considered switching to a blogging platform that lets less spam through? WordPress appears to be good, and while Blogspot is too stupid to allow <blockquote>, it does have Captcha.

    Indeed, in the accents that show happY-tensing (probably a majority of all speakers by now), unstressed final /ɪ/ has actually gone from reduced to unreduced.

    Yes; I’m talking about “Missoura”, where unstressed final /ɪ/ has instead merged into [ə].
    There are even intermediates out there. There’s a song that contains “and make me happyyyyyyy”, with [ɘː] (as it happens, that’s the sound of the Polish y). Isn’t that what the spellings happeh and kitteh are trying to convey?

  66. Hat, have you considered switching to a blogging platform that lets less spam through? WordPress appears to be good
    Yes, many times. There are several reasons I’m still using this ancient platform: 1) laziness and conservatism on my part, 2) a reluctance to make it harder for people to comment (though I realize everybody’s used to captcha by now), and 3) (probably most important) my stepson/tech guru, who would have to do the actual work of making the switch, is ready, willing, and eager to do so but so swamped with his day job and two young kids he barely has time to sleep. But it will happen, sooner or later. For a while I thought the spammers had settled down to a manageable trickle, but now they’ve come roaring back.
    [And immediately after I posted that, 34 spam comments appeared. Bah.]

Speak Your Mind

*