Snuckward Ho!

« previous post | next post »

According to John "Hindrocket" Hinderaker, "Snuck?", Powerline 11/27/2009:

Regular readers know that I have little regard for the New York Times. But I assumed that, no matter how misguided the paper's politics might be, it did have some standards relating to grammar and punctuation. So I was astonished to see this, on the front page of the Times' web site:

["The celebrity-seeking couple who snuck into a state dinner this week came face-to-face with President Obama and his wife, Michelle, the White House said Friday.]

My fifth-grade teacher, Miss Klock, would be spinning in her grave, except that she was a Republican and probably never had much faith in the Times in the first place. The reporters evidently knew better; here is how their piece begins:

The celebrity-seeking couple who sneaked into a state dinner this week came face-to-face with President Obama and his wife, Michelle, the White House said Friday in a disclosure that underscored the seriousness of the security breach and prompted an abject apology from the Secret Service.

The Oxford Dictionaries FAQ site has this to say on the subject ("Is 'snuck', as the past tense of 'sneak', a real word?"):

From the beginning, and still in standard British English, the past tense and past participle forms are sneaked. Just as mysteriously, in a little more than a century, a new past tense form, snuck, has crept and then rushed out of dialectal use in America, first into the areas of use that lexicographers label jocular or uneducated, and more recently, has reached the point where it is a virtual rival of sneaked in many parts of the English-speaking world. But not in Britain, where it is unmistakably taken to be a jocular or non-standard form.

Numbers from Mark Davies' Corpus of Contemporary American English confirm this judgment quantitatively (numbers are frequencies per million words):

Spoken Fiction Magazine Newspaper Academic
snuck 1.89 4.41 1.16 0.66 0.15
sneaked 0.66 5.42 1.65 1.76 0.30
% snuck 74.1% 44.9% 41.3% 27.3% 33.3%

And the same corpus shows a significant recent snuckward trend:

1990-1994 1995-1999 2000-2004 2005-2009
snuck 1.18 1.56 1.61 2.32
sneaked 2.00 1.84 2.16 1.81
% snuck 37.1% 45.9% 42.7% 56.2%

A similar trend is visible in Time Magazine (numbers are again frequencies per million words):

20s 30s 40s 50s 60s 70s 80s 90s 00s
snuck 0.00 0.00 0.06 0.12 0.12 0.15 0.19 0.21 1.09
sneaked 1.83 3.48 5.50 4.53 2.55 2.06 2.64 2.57 3.58
% snuck 0% 0% 1.1% 2.6% 4.5% 6.8% 6.7% 7.6% 23.3%

The pattern in the NYT is less clear — did a new copy-editor come on board in the mid-to-late 1990s? — but still, snuck has been getting about a fifth of the action over the past few years (numbers are counts of articles containing the word):

81-85 86-90 91-95 96-00 01-05 06-09
snuck 37 52 98 112 92 109
sneaked 229 259 294 507 550 412
% snuck 13.9% 16.7% 25.0% 18.1% 14.3% 20.9%

Mr. Hinderaker apparently believes that the evolution of species is "a rather obvious fraud, which cannot withstand the mildest scrutiny", and that "scientific support for [anthropogenic global warming] is weak", so the development of sneak as a strong verb is just one of several issues on which he can look forward to many years of increasingly enraptured outrage.

[Note: I was disappointed to find that John Hinderaker seems to have retired his nom de blog "Hindrocket". But please be assured that it's his own self-description, not a name-based insult created by others.]


  1. Lance said,

    November 29, 2009 @ 2:23 pm

    There's something particularly pleasant about your posting Mr. Hinderaker's grammar gripe, in which his second (and perhaps his third?) sentence begin with conjunctions/coordinators, immediately after Geoff Pullum posted an Economist reader's grammar gripe on that very topic. I wonder whether the Swiss grouch uses "snuck", and if so, whether he and Mr. Hinderaker could be pointed toward each other to direct their peevishness.

    [(myl) I can dimly perceive the outlines of a new field of international econo-peevology ("peevonomics"?). Division of labor, supply/demand curves, the dynamic stochastic equilibrium of gaffe/gripe exchange rates, etc.]

  2. Craig Russell said,

    November 29, 2009 @ 3:10 pm

    Wow, there's no missing the contempt for 'snuck' in the Oxford FAQ answer. What's funny is that I imagine for many young Americans 'snuck' is so well-established that you'd have trouble convincing them that 'sneaked' is a "real word". In fact, I seem to recall having that very realization during the two years I spent teaching high school and middle school (they don't believe 'dived' rather than 'dove' is a "real word" either).

    Other examples of strong verb forms replacing older weak -ed forms? It seems to be contrary to the expected pattern.

  3. Geoff Nunberg said,

    November 29, 2009 @ 3:34 pm

    The American Heritage usage note (which I think I wrote for the Third Edition in my former capacity as chair of the usage panel) says this:

    Snuck is an Americanism first introduced in the 19th century as a nonstandard regional variant of sneaked. Widespread use of snuck has become more common with every generation. It is now used by educated speakers in all regions. Formal written English is more conservative than other varieties, of course, and here snuck still meets with much resistance. Many writers and editors have a lingering unease about the form, particularly if they recall its nonstandard origins. And 67 percent of the Usage Panel disapproved of snuck in our 1988 survey. Nevertheless, an examination of recent sources shows that snuck is sneaking up on sneaked. Snuck was almost 20 percent more common in newspaper articles published in 1995 than it was in 1985. Snuck also appears in the work of many respected columnists and authors: "He ran up huge hotel bills and then snuck out without paying" (George Stade). "He had snuck away from camp with a cabinmate" (Anne Tyler). "I ducked down behind the paperbacks and snuck out" (Garrison Keillor).

    The high degree of preference for sneaked here isn't that surprising, bearing in mind that the panelists are for the most part mature writers, as befits their distinction, and that they generally feel themselves obliged to be on their best linguistic behavior when they give their responses. We asked the panel about this one again in 2005 and found the preference for sneaked was more-or-less unchanged, with 65 percent favoring sneaked in the sentence "The truant schoolboy ____ into the classroom while he thought the teacher wasn’t looking." The preference for sneaked was slightly higher when the verb was used transitively, and a lot higher (84 percent) when it was used transitively in the sense, "to do, take, or enjoy hurriedly or surreptitiously" in the example "The airline passenger broke the rules and _____ a smoke in the lavatory."

    MW's Dictionary of English Usage is very good on this one, with useful historical documentation. Though we didn't say so in the AHD note, my guess is that the spread of snuck, if not its inception, owed something to a perception of it as a humorous regionalism that was somehow morphologically apt: as Roy Blount said in a comment on his usage panel ballot: "I like 'snuck' — it's sneakier."

    [(myl) The 2009 edition has gone further down the primrose path:

    First recorded in writing toward the end of the 19th century in the United States, snuck has become in recent decades a standard variant past tense and past participle of the verb sneak: Bored by the lecture, he snuck out the side door. Snuck occurs frequently in fiction and in journalistic writing as well as on radio and television: In the darkness the sloop had snuck around the headland, out of firing range. It is not so common in highly formal or belletristic writing, where sneaked is more likely to occur. Snuck is the only spoken past tense and past participle for many younger and middle-aged persons of all educational levels in the U. S. and Canada. Snuck has occasionally been considered nonstandard, but it is so widely used by professional writers and educated speakers that it can no longer be so regarded.


  4. sollersuk said,

    November 29, 2009 @ 3:51 pm

    Brit here: I've seen "snuck" fairly often in American writing, and always assumed that it's one of the differences between usages when it comes to strong verbs, and that it's a dialect usage like "gotten" that didn't cross over into Standard English but did survive in American.

  5. MattF said,

    November 29, 2009 @ 4:00 pm

    But isn't it peculiar that the new form is irregular? I'm comfortable with irregular forms that are remnants of ancient declensions, now thankfully gone in the modern world of today– but 'snuck' doesn't fit into that category, I don't think. I'm not complaining, I'm just wondering where it came from.

  6. Andrew (not the same one) said,

    November 29, 2009 @ 4:27 pm

    Well, the Oxford article says it has 'crept out of dialectal use', so it is not wholly new.

    In my experience this form would not be used with what I think is the distinctively British sense of 'sneak', meaning 'inform' (as in 'I was planning to take a quick smoke in the lavatory, but someone snuck on me'.)

  7. Geoff Nunberg said,

    November 29, 2009 @ 4:28 pm

    There are a number of English weak verbs that have become strong over time, like wear, dig, spit, and more recently dive, for many speakers.

  8. Faldone said,

    November 29, 2009 @ 4:30 pm

    Not that I can think of any off-hand, but there are some weak verbs that have turned strong over the years. On the other hand, if sneak is from OE snīcan it could be argued that it should be a strong verb. However, unless some other forces were at work, the present tense should be snike and the past snoke.

  9. Sili said,

    November 29, 2009 @ 4:31 pm

    Interesting. From a purely euphonious point of view, "sneaked" grates on my ears. That implies to me, at least, that I've met the strong form most oftenoftenest.

  10. Sili said,

    November 29, 2009 @ 4:31 pm

    Bad preview! No cookie for you!

  11. mgh said,

    November 29, 2009 @ 4:43 pm

    I had not thought twice about "snuck" so this was news to me.

    My possibly incorrect impression is that I would never use "sneaked" without an object (*I sneaked into the movie theater vs I snuck in) but I would use them interchangeably with an object (I sneaked a look out the window vs I snuck a look).

    For what it's worth, google counts have the same trend:
    "snuck in" 784,000
    "sneaked in" 195,000

    "snuck a look" 165,000
    "sneaked a look" 209,000

    Mark, if you have time, perhaps you can find a clever way to refine your corpus analysis with this in mind?

  12. Bloix said,

    November 29, 2009 @ 5:07 pm

    I am a 50-year-old east coast over-educated urban white male without a shred of "nonstandard dialect" in my vocabulary, and I confess that until today I believed that "snuck" was standard English and that "sneaked" was an error.

    PS- how on earth does he know the party affiliation of his fifth-grade teacher?

    [(myl) Perhaps she lectured her pupils on politics. Or perhaps this is an instance of the high-velocity factoids emitted from a dorsal orifice that gave rise to Mr. Hinderaker's nickname.]

  13. Mags said,

    November 29, 2009 @ 5:07 pm

    Could "snuck" have been imported to the US from Ireland? "Sneaked" sounds weird to me (I'm a 30-something Dubliner). (We also use "gotten", btw).


    [(myl) Well, a Google search of nets 97 "snuck" to 133 "sneaked", or 42.2% "snuck". This is not very different from the 368-to-539 (40.6%) counts at Both are a fair bit greater than the 129-to-466 (21.7%) numbers at — but of course there's also at 894-to-3800 (19.0%).

    A few other ratios from U.S. papers: 110-300 25.0% 177-437 28.8% 84-100 45.7% 565:549 50.7%

    Irish papers: 158-99 61.5% 21-40 34.4%

    Australian papers: 1720-749 69.7%
    site: 98-53 64.9%

    And from U.K. papers: 46-175 20.8% 110-407 21.3% 1100-2180 33.5%

    So my current guess is that snuck is winning world-wide, with the UK apparently bringing up the rear, though still exhibiting a respectable frequency of snuck. (Of course, these numbers may tell us at least as much about the editing culture at specific publications, or about the ratio of blogs and reader comments to edited copy, as about regional variations in usage.)]

  14. Greg said,

    November 29, 2009 @ 5:37 pm

    I've heard about this "sneaked" thing before, but I never consciously "corrected" my speech (like I did with "unorganized" –> "disorganized", for instance) because it just sounds weird to me. Presumably "snuck" came about by way of analogy, but why "snuck" and not "snought" or "snake" or "snek"?

  15. Ben K said,

    November 29, 2009 @ 5:41 pm

    This issue was raised a while back on Late Night with Conan O'Brien (although it's not clear what dictionary he's holding):

  16. Chris H said,

    November 29, 2009 @ 6:00 pm

    I live in the UK — but I have to say that "snuck" seems most natural to me. I think (but may be wrong) that my friends would also agree. It may be a (inside the UK) dialect thing as well as a US-UK one, I guess, or have something to do with my age (18). "Sneaked" just sounds weird to me, like a child's over-generalisation.

    [(myl) More evidence that sneaked is toast: we can look forward to a century of increasingly shrill peeving on the subject, I expect.]

    I have a slightly different problem with "dived"/"dove" – they both sound unnatural to me, to the extent that I avoid saying the word altogether (I would spell it "dived" – if I had to).

  17. Tim K said,

    November 29, 2009 @ 6:06 pm

    I like to imagine that I'm at least a little bit erudite, and I would have told you 'sneaked' was wrong before I read this post. Good to know.

    I like 'snuck' better.

  18. Boris said,

    November 29, 2009 @ 7:01 pm

    I would imagine "sneaked" is harder to say out loud, but why "snuck" should be the past tense is unclear to me without a pre-existing usage or precendent. Also, why don't "peak", "reek", or "seek" have a similar past tense? They are just as difficult to pronounce.

    Also, I never realized "dived" was correct, though I'm sure I've seen (or heard) it used.

  19. Lazar said,

    November 29, 2009 @ 7:04 pm

    Does anyone else share the preference for "sneaked" in the transitive sentences that the AHD notes? Most of the time I use "snuck", but for something like "X a smoke", I think "sneaked" sounds more natural.

  20. David said,

    November 29, 2009 @ 7:09 pm

    Re Geoff Nunberg's comment, the occasional move from weak to strong is not confined to English: over the last 200 years, Swedish has seen the establishment of "dök" ('dove', instead of "dykte", cf. 'dived') and "dög" ('was good enough', instead of "dugde") as the standard forms. Guess it shows that language change never is as straightforward as one would like it, given that the overall long-term tendency in (I suppose most) Germanic languages is strong -> weak.

  21. Matthew Kehrt said,

    November 29, 2009 @ 7:10 pm

    I'm another American (25 years old, East Coast, college education) who finds both "sneaked" and "dived" to be utterly bizarre.

    Also, for all the comments pointing out that verbs turning strong is unusual, I think this is not true in at least nonstandard American English. "Drug" for dragged comes immediately to mind.

  22. Tim said,

    November 29, 2009 @ 7:51 pm

    As a 28-year-old native speaker from the American Mid-West, snuck is absolutely standard for me. If someone said "sneaked" to me, I would definitely balk at it.

    The problem I have with dove/dived is the past participle. I think I probably tend to use dove as the past tense, but, whenever I consciously think about it, I want the accompanying past participle to be diven, which is obviously wrong.

    "Also, why don't 'peak', 'reek', or 'seek' have a similar past tense?"

    The past tense of seek is sought. Unless you're asking why it isn't suck?

  23. Shannon said,

    November 29, 2009 @ 7:54 pm

    Like Bloix and others above, if asked I would have said that "sneaked" was an error, not "snuck." Ditto with "dived" vs. "dove." As they say, you learn something new every day! To place myself in context, I'm a 30-year-old college-educated Californian.

  24. John Cowan said,

    November 29, 2009 @ 8:04 pm

    There are two kinds of irregular verbs in English: the irregular weak verbs, which end in -d (but not -ed) or -t in the preterite and participle, and the strong verbs, which change their vowels in the preterite or the participle or both (a subset of these also add -en or just -n to the participle). Normally, verbs tend move from the strong to the regular weak (henceforth just "weak") category in proportion to their rarity, only the commonest strong verbs surviving to the present.

    In the case of stick, dive, and hang there were originally two verbs, a primary strong verb and a weak derived verb, which partly or completely merged phonologically and have been competing. Strong dive was lost in English by the 13th century, but was retained in Scots until at least the 16th century, whence it passed to America and began to return to England in the 20th century.

    On the other hand, a few weak verbs became strong by pure analogy: this is the case of ring (late OE), string (17th century), sneak (19th century in the U.S.), and probably strive (a French borrowing, but probably of Germanic origin, so it may belong to the previous group).

    (A)bide is in a third category: originally strong, it became weak by early Middle English times, but the strong forms had been revived by the 19th century, whether from another dialect or from analogy is not known, and with somewhat different timing for bide and abide.

    So that's a total of only eight or nine weak-to-strong conversions (plus their derivatives, like overhang) versus hundreds in the other direction. Note also that some strong verbs have weak homonyms made by zero derivation: shine 'reflect light' is old, but shine 'cause to shine, polish' is a 19th-century U.S. innovation and as such weak.

  25. J. Goard said,

    November 29, 2009 @ 8:24 pm

    I think I probably tend to use dove as the past tense, but, whenever I consciously think about it, I want the accompanying past participle to be diven, which is obviously wrong.

    Interesting. I have a similar reaction, but very much want it to be "doven".

  26. mollymooly said,

    November 29, 2009 @ 8:35 pm

    I think I use both "sneaked" and "snuck"; I have found myself struck by reading one where I would have used the other. I endorse the idea that "snuck" is "morphologically apt" in many instances. The verb "sneak" may not strictly be inherently jocular, but I suspect it gets used disproportionately often in informal contexts.

    @Andrew (not the same one):

    In my experience this form would not be used with what I think is the distinctively British sense of 'sneak', meaning 'inform'

    I've never heard this sense, but I imagine it's a verbing of the noun "sneak" rather than an extension of the pre-existing verb "sneak". As a newly-coined verb, it ought to be regular.

  27. mollymooly said,

    November 29, 2009 @ 8:45 pm

    Whatever about Britain, I don't think "dove" has any currency at all in Ireland. The first time I read the word "swandove" I stared a good two minutes before working out what it meant. (A cygnine-columbine hybrid? A yoga position?)

  28. Trond Engen said,

    November 29, 2009 @ 8:48 pm

    Many posters say they would use the weak sneaked transitively. John Cowan cites a new-formed weak causative to shine. As he also says, this is an inherited pattern, and there are quite a few inherited verb pairs of the kind. It's tempting to suggest that new strong forms hit verbs with a clearly active meaning in opposition to an imagined causative, or to reserve the old weak verb for a needed causative. Is there a way to tell? The opposite tendency of merging these pairs may blurr the picture, but I'd imagine that at least shortly after such a split first occured one might be able to see a clear differentiation in meaning among the innovators.

  29. Simon Cauchi said,

    November 29, 2009 @ 9:26 pm

    @mollymooly. According to the New Oxford Dictionary of English, the verb — "sneak", meaning tell tales — came first, and it's regular:

    "2 (no obj.) Brit. informal (especially in children's use) inform an adult or person in authority of a companion's misdeeds; tell tales: _she sneaked on us_"

    The corresponding noun is likewise said to be Brit., informal, and especially used by children.

  30. Andrew (not the same one) said,

    November 29, 2009 @ 9:40 pm

    Simon Cauchi: yes, I had always assumed that the verb came first, and the noun meant 'one who sneaks'; thanks for confirming that.

    Regarding 'dove'; my grandmother (born early 1900s) was Canadian and always said 'dove'; to me (British, born 1960s), when I first heard this, it seemed totally alien.

    I think some people are still missing the 'crept out of dialect' bit. The answer to the question why people chose 'snuck' rather than 'snaught' or whatever is that the form already existed, though not in the standard version of the language.

  31. Micaya said,

    November 29, 2009 @ 10:02 pm

    Snuck sits better than sneaked for me, actually. What does not sit well is snucked. For example, heard at 30th Street Station in Philadelphia: "So I had a hamburger snucked in one pocket and a cheeseburger snucked in the other . . . "

  32. Mark S. said,

    November 29, 2009 @ 10:17 pm

    It's interesting to see how well "snuck" is doing in the press despite warnings against it. (Those instances of "snuck" can't all be in direct quotes.)

    The New York Times Manual of Style and Usage gives "sneak, sneaked, sneaking. Never snuck. "

    The Associated Press Stylebook, which is the standard at many other U.S. newspapers, gives "sneaked Preferred as the past tense of sneak. Do not use the colloquial snuck."

  33. Ethan said,

    November 29, 2009 @ 10:19 pm

    Micaya: "a hamburger snucked in one pocket…"

    Are you sure this wasn't actually "a hamberger snugged in one pocket and a cheeseburger snugged in the other"? "snug" and "snugged" are both idiomatic to me, and the quoted sentence would be a typical use of "snugged".

  34. dogmother said,

    November 29, 2009 @ 10:27 pm

    The past tense of seek is sought

    Except in HTML5, where it's seeked, because the correct form "would confuse the heck out of people".

  35. Aviatrix said,

    November 29, 2009 @ 10:48 pm

    I was an A+ English student in Canada and to me snuck is standard, "sneaked" is jocular, and dived and dove are equally acceptable. I reread the "celebrity-seeking couple" sentence a couple of times trying to find a problem with the comma splicing before I read on to find the real gripe.

  36. Lazar said,

    November 29, 2009 @ 11:26 pm

    "Note also that some strong verbs have weak homonyms made by zero derivation: shine 'reflect light' is old, but shine 'cause to shine, polish' is a 19th-century U.S. innovation and as such weak."

    I've noticed this in baseball with the phrasal verb "fly out", which takes the preterite/particple form "flied out".

  37. Kenny V said,

    November 29, 2009 @ 11:39 pm

    Just for anecdotal evidence of the trend toward "snuck":

    I'm a graduate student in Classics whose BA was in linguistics, and I consider myself very knowledgeable about English, both prescriptively and descriptively, compared to the common person and the common pontificant peever; but "sneaked" doesn't even sound like a word to me. In my idiolect, "snuck" is correct.

  38. Boris said,

    November 29, 2009 @ 11:42 pm

    @Tim, yes, of course it is. I just reread what I wrote and have no idea what I was thinking. Still, I would not necessarily expect "suck" because "seek" is spelled differently. Maybe an analogy with "read" or "lead" is more appropriate for "sneak" (pronounce like "snake" maybe? or "snack?"),.

  39. David Green said,

    November 29, 2009 @ 11:54 pm

    Is "sneak" = "inform" related to "snitch"?

  40. Nathan said,

    November 30, 2009 @ 12:09 am

    In my sophomore year of college (1996) I took an introductory anthropology course in which the professor, making some point about language change, raised the subject of the past tense of to speed. Most of my classmates insisted on sped as the only correct form, but he claimed (I don't remember any evidence being presented) that speeded was much older.

  41. Jerry Friedman said,

    November 30, 2009 @ 12:27 am

    A new sometimes irregular verb is text, same in past tense and past participle.

    Here's a debate.


  42. Simon Cauchi said,

    November 30, 2009 @ 12:33 am

    Is "sneak" = "inform" related to "snitch"?

    Off the top of my head: yes, of course it is. A straightforward case of palatalization (of the final consonant) and centering (of the vowel). Now I'll consult my reference books and will eat my words if I'm wrong.

    PS All my books say the origin of "snitch" is unknown. But "tweak" and "twitch" are indeed related, and I'll believe the same of "sneak" and "snitch" until someone proves otherwise.

  43. Julie said,

    November 30, 2009 @ 12:47 am

    You don't have to be young….I'm 50, and I don't think I'd heard the word "sneaked" spoken out loud until I was in college. (Of course I'd seen it in writing.) I had a small-town northern California upbringing. I do not think either of my parents (both born in California) would ever say "sneaked."

    My husband, on the other hand, is also a California native, but has a strong negative reaction to "snuck."

  44. David Costa said,

    November 30, 2009 @ 12:47 am

    What pre-existing strong verb is the alternation in 'sneak'/'snuck' based on? I can't think of any other verbs that match it.

  45. alex said,

    November 30, 2009 @ 2:34 am

    It's interesting that the vowel-changing verbs are "strong." How does this relate to perfect tenses? I'm an American midwesterner, and I definitely use "snuck," but only in perfect tenses. That is, "They sneaked into the cottage" but "They have snuck into the cottage." Kind of like "The boat sank." "The boat has sunk."

  46. Chris said,

    November 30, 2009 @ 4:55 am

    Seems to me (as a non-linguist) that these strong forms of traditionally weak verbs are usually introduced as a purely facetious invention. Sometimes they catch on, sometimes they don't.

    "The manner of his death was thus:
    He was druv over by a bus."

  47. phspaelti said,

    November 30, 2009 @ 5:14 am

    Note that not all vowel-changing verbs are "strong". Verbs like "bring/brought, catch/caught, sleep/slept, etc." which add a "t" are old weak verbs that have aquired a vowel change through historical forces. One give-away for a weak verb is that the past and the past-participle are the same.
    True strong verbs have 3 forms "sing/sang/sung, ride/rode/ridden". This is in fact often how one can spot newly formed "strong" verbs, since their past participle will often be either a "t/ed" form or the same as the past tense. So "dive/dove/dived" and "sneak/snuck(or sneaked)/sneaked (or snuck = past)".

  48. Dionne said,

    November 30, 2009 @ 5:14 am

    I have the same problem with shined/shone. Of late, people have taken to talking about "She shined in the stage production of so-and-so" and it really grates. I have no problem with "she shined her shoes" though.

  49. Kellen Parker said,

    November 30, 2009 @ 5:57 am

    seek > sought, sneak > snought.

  50. Colin John said,

    November 30, 2009 @ 6:35 am

    @Kellen Parker
    So break > brought!

    Im a BrE speaker in my 50s and was really surprised by the proportion of 'snuck' in UK sources. I think I've hardly ever heard anyone who wasn't American or US-influenced use it in speech, although TV & films (or would you prefer me to say 'movies') mean that it's very familiar as a usage. 'Dove' for 'dived' is less common in my experience and 'drug' for 'dragged' would cause a real double-take if I heard it used seriously by any BrE speaker.

    [(myl) Many of the examples in the Guardian seem to be in blogs or opinion pieces, but on the other hand, the authors seem quite thoroughly British, e.g. an Old Etonian here, and other apparent natives here and here. And here's snuck in a regular 2004 news story about Southampton tying Middlebrough, written by Amy Lawrence, who

    is deputy football correspondent for the Observer. She lives in London, round the corner from the former site of the stadium she fell in love with aged six

    So it seems that snuck is now firmly rooted in British soil.]

  51. John Atkinson said,

    November 30, 2009 @ 6:53 am

    Simon Cauchi: 'Is "sneak" = "inform" related to "snitch"? Off the top of my head: yes, of course it is. A straightforward case of palatalization (of the final consonant)…'

    One would hypothesise that "sneak/snitch" is one of those doublets derived from Norse and Old English respectively. Like "skirt/shirt", "kirk/church" and so on. Note that modern Norwegian has the verb "snike", sneak, steal.

  52. Plegmund said,

    November 30, 2009 @ 7:07 am

    Not something I would have fruck out over.

    [(myl) Nice. Then again, you're not the designated ranter at a wingnut opinion site.]

  53. Lazar said,

    November 30, 2009 @ 7:37 am

    @Nathan: That reminds me of the situation with "plead", with "pled" raising a great deal of ire.

  54. jim said,

    November 30, 2009 @ 8:09 am

    I'm English, and I'm about 50-50 sneaked/snuck. I think I would generally prefer a specific one in a specific instance, but after a few minutes' thought I've not been able to come up with any simple rule as to which is which. I'd agree that 'sneak' in the schoolboy sense of 'inform', 'snitch' is always weak, though.

    Perhaps I'm not a typical case, in that I tend to jocularly make strong verbs out of weak ones on the spot when speaking colloquially. In my experience, though, it's more common for weak verbs to be stronged than for strong ones to be wuck.

  55. Matthew Kehrt said,

    November 30, 2009 @ 8:12 am


    Oh I never realized that about strong verbs having three forms. It does indicate to me that my usage of spell/spelled/spellt is entirely idiosyncratic, something I have long suspected but never bothered to look up.

  56. Aaron Davies said,

    November 30, 2009 @ 9:54 am

    @dogmother: regularizing irregular verbs is pretty common in computer science, mostly due to the tendency to have to refer to computer language elements as literally as possible. similarly, i'd be very surprised to find anyone talking about a hard drive's having "sought" to a specific point while finding a data block. perhaps if more people programmed in Perligata…

  57. Dan Lufkin said,

    November 30, 2009 @ 9:54 am

    Then we have the distinction between "hung a picture" and "hanged a horse-thief."

    And Red Grange's famous "He slud into third base."

    Steven Pinker's The Language Instinct has some excellent stuff in Ch. 5.

    Including "Nothing could have subdone him the way her violet eyes subdid him."

  58. jsg said,

    November 30, 2009 @ 10:32 am

    "What pre-existing strong verb is the alternation in 'sneak'/'snuck' based on? I can't think of any other verbs that match it."

    Think about stick/stuck and strike/struck. We find the "short U" vowel in many past tense forms, regardless of the vowel in the present (also, for example, sing/sung and hang/hung). That may be enough to establish an irregular form for the past tense of "sneak" even though nothing matches its vowel in the present tense. The semantically similar slink/slunk might have some pull as well.

  59. Spell Me Jeff said,

    November 30, 2009 @ 10:48 am

    To make plain what a number of the more playful posts have suggested: The argument for snuck on the basis of euphony doesn't seem to wash. There are many analogies indicating that Americans are quite comfortable with the eek + t pattern, e.g., peeked, reeked, freaked, leaked, freaked being the closest, as it also begins with a voiceless fricative + voiced x combination (if that even matters).

    If sneaked is dissonant, I would argue that usage (or lack of it) makes it so.

    Peak-ed (meaning sickly) is an interesting variation. Imagine saying, "I sneak-ed into the theater."

  60. Ellen said,

    November 30, 2009 @ 11:12 am

    We also wouldn't say "I peak-ed" into the theatre". Only adjectives get the two syllable pronunication in modern English. Blessed is another one, with blest for the one-syllable pronunciation as an adjective. Whereas as a verb, blessed and blest are both one syllable and nearly homophonous.

    Also, regarding "seek to a point", I'm not at all in the computer field and sought sounds wrong there. I think because it's a different meaning of "seek" than the usual. It comes across as jargon. So, more a vocabulary difference than a grammar difference.

  61. Nick Z said,

    November 30, 2009 @ 11:23 am

    @Spell Me Jeff: I'm not sure I understand what you mean by 'peak-ed'. Do you mean that the final -ed is pronounced as a full syllable, i.e. (in pseudo-IPA) [pi:ked] rather than [pi:kt]?

    I don't know this usage of 'peaked' anyway – although I see it is in the OED (orig. colloq. and regional.), where the pronunciation given is [pi:kt]. I do know 'peaky' = 'a bit poorly'. Otherwise only 'piqued' (of interest).

    On the general case of 'snuck' vs 'sneaked' – as a British English speaker 'snuck' isn't non-standard enough to pull me up when reading it, but (especially in speech) it would definitely strike me as jocular/American/informal.

    'Dove' = 'dived' for me is completely impossible, but is understandable as an Americanism.

  62. Acilius said,

    November 30, 2009 @ 11:25 am

    In the 1980s, I took a first-year course on ancient Greek at a North American university. One of my fellow students used the word "snuck." Another objected to this form. When we thought about it, "snuck" didn't sound right to us, and so we asked the corrector what form she would prefer. "Sneaked," she said. That definitely didn't sound right, so we spent a minute or two trying to figure out what the past tense of "to sneak" should be. I proposed "snook"; that met with a strong objection from a classmate named Mr Snook. Then the professor came in, and the discussion ended.

  63. Spell Me Jeff said,

    November 30, 2009 @ 11:48 am

    Nick Z: I'm not sure I understand what you mean by 'peak-ed'. Do you mean that the final -ed is pronounced as a full syllable, i.e. (in pseudo-IPA) [pi:ked] rather than [pi:kt]?"

    I don't know this usage of 'peaked' anyway – although I see it is in the OED (orig. colloq. and regional.), where the pronunciation given is [pi:kt]. I do know 'peaky' = 'a bit poorly'. . .

    Yes, two syllables, [pi:ked]. My 1982 American Heritage lists it as a separate entity, an adjective apparently pp. of peak, "to become sickly," origin unknown. My gut feeling is that forms such as peaky are derivations.

    I've only ever heard the word used jocular-pretentiously, in the spirit of, "[As grandmother would say] you look absolutely peaked!"

  64. J. W. Brewer said,

    November 30, 2009 @ 11:49 am

    phspaelti's point is interesting, w/o necessarily getting into how the terms (e.g. "true strong") should be used. It is perhaps to be regretted that those innovators who, uh, unweakened the verb did not have the foresight to innovate a full set of strong forms. "Have snuck" seems to be, as it were, the non-standard standard (and are there people out there who would consistently use snuck for the imperfect but "have sneaked" for perfect?). Googling reveals some instances of "have snucken," but I expect they're mostly jocular, and in any event you don't have the nice pattern of getting yet another vowel in play. If I were trying to analogize from, e.g., sink/sank/have sunk or ride/rode/have ridden (where the vowel seems to be steadily moving from high to low), I would have come up with something like sneak/snuck/have snocken (more the "talk" vowel than the "rock" vowel), or perhaps sneak/snuck/have snook(en), in tribute to the apparently etymologically mysterious snook in the phrase "cock a snook." Maybe the German should be "hat geschnoechen"?

    On shone v. shined, one advantage of "shined" is minimizing "nervous cluelessness" by avoiding the need to choose between the variant pronuncitions of "shone": e.g. sounds-like-Sean v. rhymes-with-stone. Wiktionary says this is a BrE v. AmE issue, but I think in practice it may be more complicated than that.

  65. Spell Me Jeff said,

    November 30, 2009 @ 12:12 pm

    J. W. Brewer, I'd only read your first sentences when "have snucken" leapt immediately to mind, and then I laughed when I came upon it several sentences later.

    I wonder if the "u" sound was originally "oo," falsely analogous to take, took, taken. It just seems odd that a quasi-strong form should bootstrap itself. Not that I'm seriously proposing this . . .

    On the pronunciation of "shone" as "sean". I'd only ever heard that in Pink Floyd's "Shine on You Crazy Diamond," where it always struck me as odd. Why it never struck me as a Britishism, I have no idea, so much of their work being (or so it seems to me) quintessentially British. Now I can check that off the list of mysteries.

  66. Ellen said,

    November 30, 2009 @ 12:16 pm

    Spell me Jeff wrote:

    On the pronunciation of "shone" as "sean".

    By "sean" do you mean "Shawn" or "seen"?

  67. Ellen said,

    November 30, 2009 @ 12:19 pm

    Oops… didn't see the post above yours with "Sean" capitalized. Nevermind, that clarifies.

  68. Bill Walderman said,

    November 30, 2009 @ 12:46 pm

    I wonder whether anyone has heard "crunk" as a strong past/past participle of "crank," meaning "to start" (as a car or other motor vehicle). I heard this form used entirely unselfconsciously by a group of men in rural southeastern Alabama about 30 years ago. It seems to be another instance of an originally weak verb developing strong forms, in defiance of the principle of levelling.

  69. Skullturf Q. Beavispants said,

    November 30, 2009 @ 12:54 pm

    I've noticed that a friend of mine uses "dug" for the past tense when it comes to digging holes, but he uses "digged" when he's talking about really getting what you're saying there, man.

  70. JimG said,

    November 30, 2009 @ 12:56 pm

    I'd bet that when the snake first made his surreptitious approach to Eve, he never imagined that people would fook up his verb.

    Prof. Liberman started this thread by writing: "Regular readers know that I have little regard for the New York Times."

    [(amz) NO NO NO! MYL didn't write this. He quoted it, mockingly.]

    I'd quibble…

    [(amz) Well, then, go and quibble with Hinderaker, who did write this.]

  71. J. W. Brewer said,

    November 30, 2009 @ 12:58 pm

    SpMJ: I had actually considered mentioning in my earlier comment that if only due to Pink Floyd some fairly massive demographic subsets of the US population (you, me, and a couple tens of millions of others — maybe including an absolute majority of white males between the ages of 30 and 55?) have been repeatedly exposed to the "shawn" pronunciation variant. Now, there are certain variant pronunciations (or lexical items or syntactic forms) that you experience in British-origin rock music that are somehow in context clearly Anglicisms, so they wouldn't influence the usage of American listeners unless they were trying to be Anglophilic/affected. (Of course, the fact that many British rock singers were themselves trying to imitate what they took to be an American style, and sometimes even a BVE style, further complicates matters.) But that particular Pink Floyd one in context doesn't feel marked-as-Anglicism to me. Also, that the Brits themselves supposedly lack a consensus on how to pronounce "scone" further increases, at least for me, the puzzlement associated with "shone."

  72. dwmacg said,

    November 30, 2009 @ 1:34 pm

    @ JimG:

    The "Regular readers know…" quote was from Hindrocket, not Liberman.

    Regarding the pronunciation of "shone" my daughter used to have a tape of "Pippi Longstocking" spoken by what appeared to be a speaker of American English who pronounced it as "Sean"; it struck me as odd coming from an American English speaker, but maybe she'd been listening to Pink Floyd.

    Oh, and "snuck" sounds better to me, and I have a doctorate, so that really should settle matters.

  73. Lugubert said,

    November 30, 2009 @ 1:39 pm

    david wrote "Guess it shows that language change never is as straightforward as one would like it, given that the overall long-term tendency in (I suppose most) Germanic languages is strong -> weak."
    My personal tendency in Swedish is to strong the weaks when I can get away with it.

  74. Sal said,

    November 30, 2009 @ 2:29 pm

    I always say "sneaked," thanks to my fifth grade teacher.

    To me "snuck" sounds like a slimy body fluid, which is enough reason to not use it.

  75. JB said,

    November 30, 2009 @ 3:22 pm

    I think in British English, "shone" is pronounced "shon," not like "Sean," which is pronounced "shawn." The first rhymes with "don," the second with "dawn." -JB, 25, NY

  76. Mark F said,

    November 30, 2009 @ 3:43 pm

    Andrew (not the same one) wrote

    I think some people are still missing the 'crept out of dialect' bit. The answer to the question why people chose 'snuck' rather than 'snaught' or whatever is that the form already existed, though not in the standard version of the language.

    I think you're misinterpreting what the OED people meant. At some point the strong verb form had to arise. There can't be a dialect of American English where 'snuck' has "always been" the past tense of 'sneak', unless there is or was a dialect in Britain where that's true. According to John Cowan's comment which at least seems pretty authoritative, whatever dialect it crept out of actually created that past tense form, turning a regular verb into an irregular one. That's also how I read the original post.

  77. Dan T. said,

    November 30, 2009 @ 4:01 pm

    I pronounce "shone" like "shown", rhyming with "own", "throne", "thrown", "bone", or "hone".

  78. Andrew (not the same one) said,

    November 30, 2009 @ 4:13 pm

    Mark F: Ah, OK. I still think that the oddness of everyone going over to a strong form is reduced if it didn't happen all at once, but began in a dialect and then spread.

    JB: As you say, in British English 'shone' is pronounced 'shon', with the same vowel as in 'don'. But as I understand it, in standard American English that vowel sound doesn't exist – a word spelt 'shon' would be pronounced in a way that a Brit would hear as 'shahn'. It may be that 'shaun' is the nearest approximation to British 'shone' available in American speech.

  79. Lazar said,

    November 30, 2009 @ 4:55 pm

    @Andrew: Right, the phoneme /Q/ doesn't exist in General American. All instances of it have evolved into /A:/ (as in "cot") or /O:/ (as in "cloth"), the latter tending to become opener as it moves into the vocalic space vacated by /Q/.

  80. J. W. Brewer said,

    November 30, 2009 @ 5:02 pm

    In some dialects of AmE, don and dawn have the same vowel. For me they're close but distinguishable (which doesn't necessarily mean my vowel in "don" is what a Brit would expect). You would think the name Sean would rhyme with the name Dawn rather than the name Don, but when I just tried to be introspective about that, self-consciousness destroyed any confidence as to accurate perception of my own "natural" pronunciation. But I will accept that in my dialect the Pink Floyd version of "shone" is closer to rhyming with Don than with Dawn and if that slightly-but-perceptibly distinguishes it from Sean then I apologize for having led others astray.

    Returning to the original subject of the thread, "snook[en]," as one of my proposed past participles for sneak, was intended to rhyme with look rather than kook, but your dialect may of course vary.

  81. Lazar said,

    November 30, 2009 @ 5:15 pm

    Just for an anecdote: there's a forum I post on where posts are numbered, and there's a certain weekly community thread where (among other things) people compete to get posts ending in hundreds – this accomplishment is called a "sneck", because the first time it happened, a fellow poster remarked facetiously, "You sneck up on me."

  82. Ellen K. said,

    November 30, 2009 @ 5:16 pm

    I too, distinguish Don/Dawn (and don/dawn, but there context pretty much will make things clear, plus don is rare). Curiously, I perceive the two vowels as being variants of the same phoneme except where there is a minimal pair that are pronounced differently. My sense is Sean can be pronounced either way, though I would expect Shawn to be pronounced like Dawn. I'm an American from the middle of the country.

  83. Ellen K. said,

    November 30, 2009 @ 5:23 pm

    Lazar, what do you mean by /Q/? Is that supposed to be a look alike for an actual IPA symbol for a vowel? Do you mean the /ɑ/ (lower case A, but not a)?

  84. empty said,

    November 30, 2009 @ 5:46 pm

    As you say, in British English 'shone' is pronounced 'shon', with the same vowel as in 'don'. But as I understand it, in standard American English that vowel sound doesn't exist – a word spelt 'shon' would be pronounced in a way that a Brit would hear as 'shahn'. It may be that 'shaun' is the nearest approximation to British 'shone' available in American speech.

    This is the story I have told myself about "scone". Americans switch from saying "scoan" to saying "scahn" (or "scawn") because they think that's what Brits say, while in fact Brits say "shon", using a vowel that Americans tend to approximate by "ah" (or "aw").

    But, British people, it's still not clear to me about "shone": do (most of) you say "shoan" or "shon"? I think I know that you don't say "shahn" (or "shawn").

  85. Simon Cauchi said,

    November 30, 2009 @ 6:14 pm

    Lazar's /Q/, I'm pretty sure, represents not /ɑ/ but /ɔ/. That's to say, an open-mid back rounded vowel, not the open one. In other another system of phonetic transcription it's represented by a lower-case o with the mirror-image of a cedilla under it.

  86. Peter Taylor said,

    November 30, 2009 @ 7:46 pm 1100-2180 33.5%

    Bearing in mind the Guardian's reputation you have to wonder how many of those 1100 were meant to say "snack"…

  87. Nick Z said,

    November 30, 2009 @ 9:54 pm

    I suspect that we are getting a bit off-topic here, but, for what it's worth, for me (BrEng) the following words rhyme. I will attempt to provide IPA vowels, but a) I've no idea if they will come out right, b) I'm rotten at doing vowels in IPA.

    shone (past tense of shine): gone : on : con : Don [ɒ]

    Sean : dawn : torn (non-rhotic, remember) [ɔː]


    shown (past tense of show) : tone : sewn : stone : moan [əʊ]

    Both scone [skɒn] (=shone) and scone [skəʊn] (=shown) are acceptable (so long as you don't mind 50% of people telling you whichever you say is wrong). The Stone of Scone [sku:n] is pronounced to rhyme with moon.

  88. Mabon said,

    November 30, 2009 @ 11:15 pm

    Re: Don/Dawn:
    My wife's cousins, grandfather and granddaughter, are Don and Dawn.
    All of us being Bostonians, it is often impossible to differentiate which of the two the speaker is referring to.

  89. dwmacg said,

    November 30, 2009 @ 11:33 pm


    Growing up outside of Boston, we often had a similar confusion with our brother Bob and our sister Barb(ara).

  90. empty said,

    December 1, 2009 @ 12:36 am

    When a boy named Ian moved to (I think) the Boston area, he was teased for having a girl's name. his new neighbors had never heard the name before, and to them it sounded like (their pronunciation of) "Ann". True story.

  91. Mark F. said,

    December 1, 2009 @ 12:41 am

    I'm sorry to stay off topic, but this has been bothering me. Is there a site that has audio for minimal pairs with [ɒ] and [ɔ]? The distinction is completely opaque to me. The isolated sound samples in Wikipedia do sound different, but I can't tell what aspect of the difference really counts.

  92. slobone said,

    December 1, 2009 @ 12:49 am

    I wonder whether strong forms that are originally jocular or from dialect are more likely to gain ascendancy if the verb itself is one that tends to be used in an informal or humorous context? "Sneak" certainly seems to qualify — the quote from Roy Blount made me think of that. It's hard to imagine the verb being used in any form in a legal brief or a presidential address.

    Would JFK have said "The Russians have sneaked missiles into Cuba"? Or even "snuck"?

  93. Jerry Friedman said,

    December 1, 2009 @ 1:37 am

    @Mark F.: It's not exactly what you want, but maybe the AUE audio archive will help. Particularly the various samples of people saying, "Bother, father caught hot coffee in the car park." (Though some Americans, like me, say "coffee" with an /ɔ/, I think most British people use /ɒ/.)

  94. Jerry Friedman said,

    December 1, 2009 @ 1:38 am

    , damn it. Sorry if anyone has to go any trouble over my mistake.

  95. Lazar said,

    December 1, 2009 @ 2:36 am

    @Ellen K: I was using X-SAMPA, which is the ASCII text conversion for IPA (mainly because I was too lazy to find the characters). /Q/ is the X-SAMPA version of /ɒ/, and /A/ is the X-SAMPA version of /ɑ/.

  96. Cecily said,

    December 1, 2009 @ 6:34 am

    Re certain accents not having an audible difference between Don and Dawn and between Bob and Barb, what's the etiquette for pronouncing a person's name THIER way?

    I'm English and I knew an Australian living over here called Meg, which she pronounced it Mig. I found myself going to great lengths to avoid using her name because if I pronounced in my way, it wasn't really her name, but if I pronounced it her way, it might have sounded as if I was doing a sarcastic impression of her speech.

    We've drifted out of touch, but I'm still not sure what I would do in a similar situation.

  97. Graham said,

    December 1, 2009 @ 6:41 am

    For me (British, in my 30s) I'd use both sneaked and snuck. However, thinking about it, I don't think they're exactly interchangable. I would use sneaked where the subject is trying to be 'sneaky' – i.e. being deliberately quiet or stealthy to avoid attracting attention. "The cat burglar sneaked into the room, keeping to the shadows." However I'd use snuck where the subject is using an window of opportunity to engage in an activity without attracting attention. "Bob snuck out for a crafty smoke when his boss went into the meeting." So snuck implies speed for me, while sneaked implies stealth. Maybe this could be because sneaked has a longer vowel for me than snuck?

  98. Stuart Martin said,

    December 1, 2009 @ 7:12 am

    "I confess that until today I believed that "snuck" was standard English and that "sneaked" was an error."

    I'm a native NZE speaker in my early 40s, and this was pretty much my reaction to the "gripe" too. "Sneaked" just feels really, really strange and unnatural.

  99. Mr Fnortner said,

    December 1, 2009 @ 8:48 am

    Who would've thunk it. My neighbor's daughter brung home an A in English. She must've snuck around and studied some.

    Seems "received" grammar has little stomach for regression toward non-standard false strong verbs. Yet those above and more crop up with regularity away from the mainstream. I wonder if the early Angles and Saxons liked the aural variety that came with the verb change, and would have done more of it had the language been insulated from outside influence.

  100. mollymooly said,

    December 1, 2009 @ 10:15 am

    Re: Don/Dawn
    Some Cork people confuse Carl Lewis with his sister Carol.

  101. Bloix said,

    December 1, 2009 @ 10:35 am

    "which she pronounced it Mig."
    As Killick would say.

  102. Cecily said,

    December 1, 2009 @ 11:19 am

    Sorry about that, though it was arguably the least serious typo in my post.

  103. Dan Lufkin said,

    December 1, 2009 @ 11:24 am

    In my idiolect the distinction between "Don" and "Dawn" is vowel length: [dɔn] vs. [dɔ:n]. They sound entirely different.

  104. Ellen said,

    December 1, 2009 @ 11:31 am

    Cecily, what typo? I see odd grammar, but nothing that looks like a typo.

  105. Jerry Friedman said,

    December 1, 2009 @ 11:48 am

    Proving yet again that the hardest typos to see are the ones in big type, boldface, or (in this case) all caps. Also, Cecily's "in" for "it" ("pronounced in my way") looks like a typo, since she used "it" in the parallel clause of that sentence.

    Anyway, Cecily, I pronounce my name "Jairy", but if you use /E/ as in "jet", I'd say that's how you should pronounce it. You wouldn't making it any less my name; it's just an accent difference.

  106. John Lawler said,

    December 1, 2009 @ 1:18 pm

    I notice nobody yet has mentioned the shwaful preterite of shrink, immortalized in the title "Honey, I Shrunk the Kids". I learned shrank, but it's true that shrunk sounds natural; certainly it's frequent enough in the preterite. This is a shifting from the 3-vowel ring, rang, rung pattern to the two-vowel wring, wrung, wrung pattern. The new pattern is more regular, in that the preterite and the past participle forms are identical, the same way they are in regular {-ed} verbs. It is also simpler, with 2 instead of 3 forms; this is not a coincidence.

    Opposed to that, I think, is the disapproving flavor that seems to taint preterite ablaut forms with shwa (especially when followed by eng: shrunk, thunk, sunk, drunk, brung, snuck), whether 'legitimate' or not.

  107. Jarek Weckwerth said,

    December 1, 2009 @ 1:19 pm

    @Mark F: These two places may be useful if you want to hear the difference:
    Howjsay, one of the very few places online that has a sizeable free British pronunciation dictionary;
    Soundcomparisons, which compares accents using individual words, not texts.

  108. Ellen said,

    December 1, 2009 @ 2:22 pm

    Jerry, I was referring to the part of Cecily's post that Bloix quotes.

  109. Ken Brown said,

    December 1, 2009 @ 3:06 pm

    empty said: "But, British people, it's still not clear to me about "shone": do (most of) you say "shoan" or "shon"? I think I know that you don't say "shahn" (or "shawn")."

    I think we all say "shon" to rhyme with " "con" or "gone". I have no memory of anyone saying "shoan" (to rhyme with "cone" or "bone") let alone "Sean/Shawn" (which for me rhymes with "corn" as well as "fawn")

    How do cot/caught-merged Americans say "con" as in "con trick" or "con artist"? Does it rhyme with ""lawn" and "fawn"? Or with "lan" and "fan"? (Which I suppose would make it a homophone of (tin) "can")

    And, me being a True Brit, the thing that brought me up short in the opening post wasn't "snuck" (which I'm perfectly happy with but I doubt if I'd be as likely to use as "sneaked") but "…the White House said Friday…" I deeply, deeply, want to put an "on" or a "last" in that.

  110. Scott said,

    December 1, 2009 @ 3:36 pm

    "How do cot/caught-merged Americans say "con" as in "con trick" or "con artist"? Does it rhyme with ""lawn" and "fawn"? Or with "lan" and "fan"? (Which I suppose would make it a homophone of (tin) "can")"

    For cot/caught-merged Americans "con" rhymes with "lawn" and "fawn", words like "can" and "fan" have a completely different sound.

  111. Simon Cauchi said,

    December 1, 2009 @ 4:13 pm

    ". . . the White House said Friday . . ."

    An American Germanism, I believe.

  112. Julie said,

    December 1, 2009 @ 4:29 pm

    In my cot-caught merged American accent, I use the same back vowel for fawn, lawn, con, dawn, Don and Sean. The vowel can vary considerably, depending on context, but does not approach the one in can.

  113. Bloix said,

    December 1, 2009 @ 4:38 pm

    "the least serious typo in my post"
    But the most amusing!

  114. Faldone said,

    December 1, 2009 @ 5:24 pm

    I think we all say "shon" to rhyme with " "con" or "gone".

    Classic example of why this method of describing pronunciations doesn't work. For me, an American with no cot/caught merger, 'con' and 'gone' don't have the same vowel sound.

  115. Thomas Matthews said,

    December 1, 2009 @ 5:25 pm

    Baseball fans of my generation will recognize this as Dizzy Dean English. The great pitcher-turned-announcer could easily have said something like, "They almost throwed him out, podner, but he snuck back to first."

  116. Graeme said,

    December 1, 2009 @ 5:51 pm

    Is it possible that 'snuck' is gaining favour as to the ear it sounds less 'sneaky'?

    Compare the implications of:
    'I snuck into the theatre after the show started.'
    'I sneaked in when no one was looking"

  117. Sneaking suspicions … | Agência de Notícias da Livraria 30PorCento (Blog) said,

    December 1, 2009 @ 11:06 pm

    […] Fonte: Books, Inq. — The Epilogue … Snuckward Ho! […]

  118. Jerry Friedman said,

    December 2, 2009 @ 12:55 am

    @Ellen: Sorry about the misunderstanding.

    @Simon Cauchi: Why shouldn't "Friday" for "on Friday" be a gallicism or a hispanicism?

    @John Lawler: Elsewhere in the multiverse, I saw "That's what sprung to my mind" today. But what flavor of disapproval did you mean—the speaker's disapproval of whatever stunk or stung, or us prescriptivists' disapproval of preterite "sunk", "shrunk", "rung", etc.?

  119. Mark F. said,

    December 2, 2009 @ 1:27 am

    Cecily —

    I don't think people necessarily expect you to pronounce their name in their accent. I don't expect British people to pronounce my name rhotically.

  120. Paul said,

    December 2, 2009 @ 8:14 am

    That's me telt. I once amazed a class of (British) students by having hidden a piece of paper on which I had written the word "posh". I then asked the class how they would pronounce "scone". There was a pretty even split between [skɒn] and [skəʊn]. Then I asked what they thought of the other pronunciation, whichever was the one they personally used. Enough of them (from both sides) shouted out the same word for them to be impressed when I produced the piece of paper.

  121. Ken Brown said,

    December 2, 2009 @ 9:42 am

    Faldone replied to my "I think we all say "shon" to rhyme with " "con" or "gone"." with Classic example of why this method of describing pronunciations doesn't work.

    It's a fair cop, guv. But I was just copying the examples used by someone else before.

    Its hard to do this by introspection because I can't hear myself speak(presumably linguists and actors pick up the knack)

    I'm not sure what the IPA for what I think of as a short "o" in words like "lot" and "shone" and "gone" and "cot" and "rock" is. A number of websites describe the "lot" vowel in RP (which isn't my accent!) as /ɒ/ and others have /o/ – if I can trust the IPA sound samples in Wikipedia, neither is quite what I think I (or most of my neighbours and family) say when we say those those words. And I don't have the technical vocabulary to describe what I think the difference is.

    (Is this because the transcriptions are phonetic rather than phonemic, and the people saying them are perhaps American and so don't in fact use the same vowel as me? Or am I just mishearing myself?)

    But in words like caught/court and fought/fort (two pairs of homophones) I think I have an "open-mid back rounded vowel" maybe IPA /ɔ/ – though in some context its perhaps more of a dipthong, gliding from that into a similar unrounded vowel or maybe a schwa.

    For me, an American with no cot/caught merger, 'con' and 'gone' don't have the same vowel sound.

    ???? Presumably you have a partial merger if you use a vowel other than the "lot" one for one of those words but not the other?

  122. Peter Taylor said,

    December 2, 2009 @ 1:09 pm

    @Jerry Friedman: the hispanicism would be "the White House said the Friday" ("la Casa Blanca dijo el viernes").

    @Paul. Interesting. I say [skɒn], but my immediate instinct on [skəʊn] is "Northern". My slightly more considered response is "regional". If it is more a class thing than a regional thing then I think it's likely to be lower to lower-mid and upper-mid to upper vs lower-mid to middle-mid.

  123. Ellen said,

    December 2, 2009 @ 1:28 pm

    @Ken Brown: For American's who don't have the cot/caught merger, con has the vowel of cot (and lot), and gone has the vowel of caught.

  124. Faldone said,

    December 2, 2009 @ 2:57 pm

    Ken Brown: ???? Presumably you have a partial merger if you use a vowel other than the "lot" one for one of those words but not the other?

    You pronounce 'lot' and 'cot' with different vowels? If so, I might have a merger but I can't see calling it a cot/caught merger, partial or otherwise. For me cot and caught are/kɑt/ and /kɔt/. And lot and cot both have the /ɑ/.

  125. Jerry Friedman said,

    December 3, 2009 @ 1:37 am

    @Peter Taylor: Thanks for the Spanish correction.

  126. Andrew (not the same one) said,

    December 3, 2009 @ 10:30 am

    Ellen: that's interesting. I think a similar distinction may have existed in (some dialects of) British English once; certainly there are songs in which 'gone' has to be sounded as 'gorn' (non-rhotic), and is sometimes written that way. (How about 'off'? And 'often'? I think Fowler discusses whether 'off' should be pronounced 'awf'; it was clearly a live issue in his time.)

  127. Ken Brown said,

    December 3, 2009 @ 12:01 pm

    Faldone said You pronounce 'lot' and 'cot' with different vowels? no of course not – but you said you used a different vowel for "gone"

    Ellen said: For American's who don't have the cot/caught merger, con has the vowel of cot (and lot), and gone has the vowel of caught.

    Thats sort of what I meant. So some of the "cot/lot" words have moved over and others not, which makes sense. Presumably a shift like this doesn't happen all at once.

    Andrew (not the same one) said, …certainly there are songs in which 'gone' has to be sounded as 'gorn' (non-rhotic), and is sometimes written that way. (How about 'off'? And 'often'? I think Fowler discusses whether 'off' should be pronounced 'awf'; it was clearly a live issue in his time.)

    My copy of Fowler doesn't seem to mention "off" but he says that "aw" for "o" in "gone" and "soft" is southern rather than northern. He rants pleasingly about "often": "The sounding of the t … is practised by two oddly consorted classes – the academic speakers who affect a more precise enunciation than their neighbours … & the uneasy half-literates who like to prove that they can spell…." To me "awf" and "awfen" for "off" and "often" sound old-fashioned and affected.

    Its in Pirates of Penzance:

    General: Tell me, have you ever known what it is to be an orphan?
    Pirates: Oh, dash it all!
    Pirate King: Here we are again!
    General: I ask you, have you ever known what it is to be an orphan?
    Pirate King: Often!
    General: Yes, orphan. Have you ever known what it is to be one?
    Pirate King: I say, often.
    All: Often, often, often.
    General: I don't think we quite understand one another. I ask you, have you ever known what it is to be an orphan, and you say "orphan". As I understand you, you are merely repeating the word "orphan" to show that you understand me.
    Pirate King: I didn't repeat the word often.
    General: Pardon me, you did indeed.
    Pirate King: I only repeated it once.
    General: True, but you repeated it.
    Pirate King: But not often.
    General: Stop! I think I see where we are getting confused. When you said "orphan", did you mean "orphan",a person who has lost his parents, or "often", frequently?
    Pirate King: Ah! I beg pardon – I see what you mean – frequently.
    General: Ah! you said "often", frequently.
    Pirate King: No, only once.
    General: Exactly – you said "often", frequently, only once.

    And the best-known "gorn" song is:

    A mother was barfing 'er baby one night,
    Youngest of ten and a poor little mite.
    the mother was poor, and the baby was fin,
    only a skelington wrapped up in skin

    She only turned round for a minute,
    To get the soap off of the rack,
    She only turned round for a minute,
    But the baby had vanished away

    'er baby 'ad vanished completely,
    vanished completely away,
    'er baby 'ad vanished completely,
    but she 'eard, an angel say….

    Your baby has gone darn the plug'ole
    Your baby has gone darn the plug
    The poor little fing was so skinny and fin
    He should 'ave been washed in a jug

    Your baby is perfickly happy
    'E won't need no barfin no more,
    Your baby is perfectly 'appy,
    Not lost but gorn before.

    (Loads of very different versions. To the tune, roughly, of My bonny lies over the ocean)

  128. Jerry Friedman said,

    December 3, 2009 @ 12:02 pm

    We caught-is-not-cot (or CINC, as we say at alt.usage.english) Americans typically pronounce "off" and "often" with the "caught" vowel. Likewise many other words with "o" followed by a continuant. This is apparently a survival of British pronunciation that just barely hangs on there. My data-free feeling, though is that the American pronunciation of "on" as "awn" is increasing and is an invention of ours.

    Another possible invention of ours is "og" as "awg", as in the well-known "dawg". Some people limit this to "dog" and its derivatives and maybe "dogma"; others, including me, use it in most of the -og monosyllables.

  129. Lazar said,

    December 3, 2009 @ 4:07 pm

    @Jerry Friedman: What you're referencing is called the lot-cloth split, and it's a nearly universal corrolary of the cot-caught distinction in American English. Certain words switched from /Q/ to /O:/ before /Q/ collapsed into /A:/:

    short "o" followed by a voiceless fricative: off, cloth, broth, loss, lost, etc.
    short "o" followed by "g" or "ng": dog, log, long, song
    specific cases: "gone", in some dialects "on"

    As for Britain, the lot-cloth split had some currency in old-fashioned RP (it manifested in the pre-fricative cases and in "gone", but not in the pre-velar cases, which were an American innovation), but is now pretty much extinct, surviving as a caricature of very posh speech. For example, lamenting "the lorst pah of the British empah".

  130. “Snuck” sneaked in « Sentence first said,

    June 18, 2010 @ 1:11 pm

    […] brings us nicely to Language Log, where late last year Mark Liberman posted an example of extreme snuck-peeving complete with a strict-teacher reference (“My fifth-grade […]

  131. Bobo Linq said,

    August 16, 2010 @ 1:51 pm

    Perhaps the weak-to-strong metamorphosis is driven by playfulness. My five-and-a-half-year-old daughter, in making up a story involving a contest in which the Norse god Loki competed, said:

    "Loki chat — that means he cheated — in the contest."

    I can't think where she got the "cheat -> chat" idea ("shit -> shat" is not part of our lexicon).

RSS feed for comments on this post