Swam = Swawm?

Anatoly Vorobey wrote me as follows (I’ve added italics and a link):

I was looking up a sound change in Jespersen’s A Modern English Grammar, specifically the rounding of a after w: the way words like swap, war, watch, etc. switched from the vowel of bat to the vowel of bot. Apparently happened in the 17th-18th centuries post Shakespeare (he rhymes watch/match etc.) Before [k],[g] the change didn’t happen (whack etc.), and also before [m], but here Jespersen says:

In swam the only pronunciation known in England is [swæm], but in America [swɔm] may also be heard; this is the regular phonetic development, while [swæm] must be due to the analogy of other preterites: began, drank, etc.

I don’t think I’ve ever heard swam pronounced as “swom” by Americans or otherwise – have you ever heard any such thing?

I most certainly have not, and I was intrigued enough to post the question: have you heard [swɔm] for swam, or do you know of its existence?

Comments

  1. I’m pretty sure I heard “swum” in small- town Texas when I was a kid in the 50s and 60s– surely by analogy with “swung”. But I don’t remember it quite sounding like “swom”.

  2. Yeah, “swum” is familiar, but (as you say) a different thing.

  3. I don’t think I’ve ever heard that pronunciation either, but then I wasn’t around in 1928, so maybe things have changed since then.

  4. Never heard it that I can recall. In particular, I can’t say I have any recollection of hearing my maternal grandfather (born 1903) saying it, yet somehow if he *had* said it it would sort of feel like it fit in plausibly with various other now-archaic features of his idiolect.

  5. Apparently happened in the 17th-18th centuries post Shakespeare

    Or even later, as Tony Harrison recently taught us.

  6. alternative link for unAmerican readers

  7. It was published in 1909, not 1928.

  8. Swum is the past participle of swim, as in “I’ve swum there before.” I often see the simple past form used in place of the past participle, e.g. “They’ve went home” instead of “They’ve gone home,” but I don’t think it’s common for the past participle to be used where a simple past is intended.

    I’ve never heard a rounded pronunciation for swam myself.

  9. David Eddyshaw says

    The only similar word that comes to mind is dwam [dwæm], which is certainly part of my active vocabulary, but may not be all that English.

    [They’re always “wee.” A muckle dwam would probably prove fatal. I hope never to find out.]

  10. i’ve never heard it, or heard of it before.

  11. Strange that many people now pronounce wort (isolated or in ragwort, etc.) /wɔːt/ (or /wɔːrt in AmE/) as if it were wart, rather than /wɜːt/ (/wɜːt/) to match every other wor* word – apart from worry and verb forms like wore, worn. Almost universal here in Ozland, where “St John’s wort” sounds like a holy skin problem. Same elsewhere?

    Then there’s /dɔːl/ for dhal, which I’ve noticed in much Anglo-Indian English speech. A hyper-British shift, perhaps defensively “non-native”?

  12. Swom was already obsolete for Noah Webster in 1828. Compare Johnson in the second edition of 1773, with no label such as Obſolete (which he uses elsewhere, although I don’t know what his criteria or cut-off points for Obſolete actually were).

  13. My comment didn’t appear for a long time, giving me no opportunity to make this fix: “rather than /wɜːt/ (/wɜːrt/)”

  14. My preceding comment didn’t appear for a long time, giving me no opportunity to make this fix: “rather than /wɜːt/ (/wɜːrt/)”. (And is this one now a near-duplicate?)

  15. The vowel in “war” doesn’t sound the same as the vowel in “watch” to me. Is the r-coloring misleading me?

  16. Craig, “war” does have the same vowel as “swarm”, “warn”, “warp”, “wart”, “thwart”, “quart”, “dwarf”, etc. Essentially “a” after “w” in many cases makes the sound you’d expect if it were spelled “o”, and “o” makes a different sound before “r” from what it makes in other contexts.

  17. Akismet seems to be hungry tonight.

  18. I was thinking about what this obsolete spelling swom was meant to represent. I suppose it may have been sounded the same as swum and used the same spelling convention as come and love and wonder, etc.

  19. @Jongseong Park: Seen serves as a simple past of see for a lot of dialectal AmEng speakers. Sink is also notorious for confusion between sank and sunk – you’ll hear both used in either role.

  20. I wonder if what they heard was the attempt to adopt a Mid-Atlantic accent by people who didn’t actually grow up speaking that kind of English. I can certainly imagine plenty of Americans in 1928 thinking that [swɔm] would sound more like RP.

  21. David Marjanović says

    Swamp certainly argues that the /æ/ in swam is analogical, but of course that says nothing about when the analogy happened.

    Then there’s /dɔːl/ for dhal, which I’ve noticed in much Anglo-Indian English speech. A hyper-British shift, perhaps defensively “non-native”?

    I’d say it’s simply Beng[ɔː]li. The h is what’s hypercorrect.

  22. i’ve never heard it, or heard of it before.

    Same here. Born in the U.S. mid-west, lived in most of the states from Maryland to Maine,
    except Mass., where they talk funny.

  23. David Eddyshaw says

    I suspect dhal has become [dɔ:l] for these speakers on the analogy of tall, fall, ball, call, gall …

  24. David Eddyshaw says

    I suspect dhal has become [dɔ:l] for these speakers on the analogy of tall, fall, ball, call, gall …y

  25. David Eddyshaw says

    I suspect that dhal has become /dɔ:l/ for these speakers on the analogy of tall, fall, ball, call, gall …

  26. David Eddyshaw says

    Akismet is eating my comments too.

  27. David Eddyshaw says

    I suspect this is one of those cases we’ve encountered before, where there is a symbol in the post heading that triggers the Wrath of Akismet and causes her to work to rule.

  28. Well, shit. Does Akismet not like ɔ? Because I’m not sure what to do about it. I suppose I could change it to o, but That Would Be Wrong!

  29. David Eddyshaw says

    Akismet would do well to remember that I was only narrowly dissuaded from exorcising her. My patience is not inexhaustible.

  30. Well, shit. Does Akismet not like ɔ? Because I’m not sure what to do about it.

    “SWAM rhyming with BOMB”? “SWAM with the LOT vowel”?

  31. OK, I’ve changed the post title. We’ll see if that appeases Akismet.

  32. Yes, that’s what I should have written. Jespersen should have written [swɑm], unless American pronunciation changed in that regard between 1928 and the 1960s.

    On this subject, I just asked my Punjabi friend how to pronounce that Indian food made of lentils or similar legumes, and he used [ɑ], as far as I can tell. There also seemed to be an [h] after the [d]. (When I asked him about it, he said 26 was a “miserly” number of letters for an alphabet.) However, he pronounces “Bengal(i)” and “Nepal(i)” with an “aw” sound similar to mine (not the RP vel sim “aw” that’s a non-rhotic “or”).

  33. I think he’s just using it (dia)phonemically for LOT.

  34. Hm, then what IPA symbol might he have used for THOUGHT?

  35. In the linked work he uses /ɔ/ for LOT, /ɔˑ/ for THOUGHT. I’ve seen the same (except with the modern length mark [ː]) in some mid-20th-century British works.

  36. Yes, the IPA transcription for English introduced by Daniel Jones in the English Pronouncing Dictionary (first edition 1917) used /ɔ/ for LOT as opposed to /ɔˑ/ for THOUGHT, cutting down on the number of distinct vowel symbols used by marking only the quantitative difference between pairs of similar but not identical qualities. It also had /i/ for KIT and /iː/ for FLEECE, /u/ for FOOT and /uː/ for GOOSE, and /ə/ for schwa and /əː/ for NURSE.

    A.C. Gimson took over the editorship of the English Pronouncing Dictionary in 1977 and introduced the now familiar practice of explicitly marking the qualitative difference between these pairs as well, i.e. as /ɒ/ vs /ɔː/, /ɪ/ vs /iː/, /ʊ/ vs /uː/, and /ə/ vs /ɜː/. But English dictionaries in Korea kept using Jones’s older system for a while longer, which is why I am familiar with /ɔ/ representing the LOT vowel.

  37. Thank you, adam, mollymooly, and Jongseong Park!

  38. I’m still not seeing my comment immediately after posting it, so I missed the chance to edit the errors. The symbol used by Daniel Jones for THOUGHT should of course be /ɔː/ instead of /ɔˑ/.

    One thing I might add is that as you can see from Mollymooly’s link (which I did not see at the time I posted my comment), Clive Upton’s 1995 edition of the Concise Oxford English Dictionary again uses /əː/ for NURSE, since for many speakers there is no appreciable difference in quality between this and the schwa vowel.

  39. The problem isn’t the post title, it’s the URL — or at least that was the case the last time there was a cursed thread: The Ḥatäta, or The Hatäta. Apparently, the blog platform tries to sanitize the URL, but it fails for some characters.

  40. Well, there always are some characters you can’t sanitize.

  41. Well, there always are some characters you can’t sanitize, try as hard as you want.

  42. Well, there always are some characters you can’t sanitize.

    According to the Unicode specs for converting byte streams into character streams, an unrecognizable byte sequence for a given encoding is replaced by a standard “replacment character” (mojibake, tofu). That’s one procedure that could be called sanitizing. Another would be to reject the entire sequence, which is more like garbage disposal than sanitizing.

  43. The problem isn’t the post title, it’s the URL

    I’d forgotten that. I’ve changed the URL now; we’ll see if it does the trick…

  44. PlasticPaddy says

    @stu
    I took Hans to be playing with other meanings for “character” and “sanitise”. However, maybe your response, despite its impressive factivity, is also meant playfully.

  45. Playfully ? Factivity ?? I rarely venture beyond the facetious and factitious here, since only they are free of charge.

  46. Yes, I charge Stu a hefty fee every time he makes a serious point.

  47. Strange that many people now pronounce wort … as if it were wart

    cf. “whorl”, which in BrE usually sounds like its etymon and near-synonym “whirl”, but for AmE and me* makes a Useful Distinction in pronunciation.

    * My accent 1. is rhotic, 2. preserves voiceless wh, and 3. distinguishes horse from hoarse, which makes “whorl” hard to pronounce but easy to spell.

  48. Thank-you, mollymooly. I am happy to hear about your accent. Very helpful.

  49. David Eddyshaw says

    maybe your response, despite its impressive factivity, is also meant playfully.

    The distinction is not a meaningful one.

    https://www.smbc-comics.com/comic/p-bot

  50. Further non-NURSE -w[h]or- words:

    LOT: Worrall

    THOUGHT:* Worple, Worplesdon

    FOOT worsted [sometimes**], Worcester

    ??: wor [Geordie]

    *plus exotic names like Dworak, Dworkin, Jaworski, Woronoff. Is Whorf English or exotic? (Worf is Klingon.)

    **are there any speakers with rhotic non-NURSE “worsted”? Noah Webster spelled it “wusted”.

  51. Explaining Worcester here in the US is notoriously tricky – there’s the fact that the canonical ENE pronunciation elides both r‘s but that the “correct” rhotic pronunciation still elides the first, then having to explain (especially if through text) that the stressed vowel isn’t /ʌ/ or /uː/ but rather /ʊ/. Then on top of all that, the fact that there’s a bona fide alternate pronunciation Wister which is used by about a quarter to a third of the city’s residents – and which is sometimes wrongly touted as the correct one just because it’s the most distinctive.

  52. Yes to all of that — it took me a while after I moved to Massachusetts to figure it out.

  53. There are other American toponyms that seem to be deliberate respellings of “Worcester” to accommodate the non-non-rhotic, e.g. the Wooster Square neighborhood in New Haven, Wooster Street in Manhattan, and the entire city of Wooster, Ohio. All of which are FOOT rather than GOOSE, I believe.

  54. Sadly not named after Bertie Wooster.

  55. Indeed, it appears from further googling that the instances I mentioned are all named not for Bertie but instead for David Wooster (1710/11 – 1777), who was the major general in command of the Connecticut state militia during the Revolution and died of wounds sustained in combat against the British. Still possible that Gen. Wooster’s surname reflected in its etymology a simplifying respelling of the toponym “Worcester,” of course.

  56. there’s a bona fide alternate pronunciation Wister

    Isn’t that just a strong New England accent, with a vowel so short and back and it sounds like an ‘i’ to those from away?

    Having recently moved to Maine I am trying to master some approximation of correct pronunciations, but my Massachusetts-born BFF still mocks my attempts at “Market Basket.” I’m non-rhotic but haven’t quite got the vowels right. In British English, the first vowels of both words are identical, but up here there’s a subtle difference.

  57. PlasticPaddy says

    @david l
    How do you say the a in Basil? That could be closer (just add a rasping quality) to Bostonian Basket than “Barsket”.

  58. David Eddyshaw says

    Isn’t that just a strong New England accent

    It’s the alternative in the place itself, in not-New England.

  59. @PP: I say ‘basil’ with a short a, British-style, but it’s not quite as, um, sharp as the New England ‘a’ in basket.

    oh, I see the ambiguity in what I wrote. I meant that the vowels in Market Basket are identical to each other in Britspeak, but in Massspeak they differ slightly from each other. They are very different from southern Brit (my native language), closer to northern British, as in ‘bath,’ but not quite that either.

  60. In my accent, PALM is distinguished from TRAP by quantity but from START by quality. There are some PALM words where similar-accented Irish people often insert /r/ because the vowel we hear from furriners has the quality of our START rather than our PALM. “Chica[r]go” is one example, where the existence of “cargo” abets the illusion.

  61. (Worf is Klingon.)

    but also russian*, which may or may not be relevant to his vowels.

    .
    * belarussian, according to Memory Alpha; russophone jewish**, according to the most committed trekkies i know. regardless, he describes himself as a member of the house of rozhenko as well as of the house of mogh, and later the house of martok.

    ** which would presumably place his roots in raysn, which unfortunately does not help with the vowels at all, since if he had any familial yiddish to draw on it could be from a wide variety of topolects.

  62. That infamous West Virginian, Hasil Adkins, had a brother named Basil, both pronounced with an /æ/.

  63. Isn’t that just a strong New England accent, with a vowel so short and back and it sounds like an ‘i’ to those from away?

    No, I’m well familiar with the accent here and people do use the one or the other. I once even heard “Wister” used in person by the then-mayor, Joseph O’Brien.

    Market Basket

    Agreed, ENE uses a front-leaning [a:] for PALM/START/(BATH) which isn’t too far off from its /æ/. Of note the dialect has the TRAP/BATH split as a traditional feature, but it’s spotty and recessive, and I can’t recall ever hearing it as deep into the wordstock as basket (usually something very basic like bath or laughaunt being the only one with real staying power), so you’re right that the two here wouldn’t be identical.

    basil

    I picked up /æ/ from watching British TV as a kid; I’d still definitely use that for the name, although on the herb I’m probably 50/50.

Speak Your Mind

*