Wikipedia:Reference desk/Archives/Language/2012 March 29

Welcome to the Wikipedia Language Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

March 29

Canadian vs. American accents

How reliably can an experienced judge distinguish between a Canadian and an American speaker of English, based on perhaps one minute of recorded speech? What features in one accent or the other are most useful for this? How frequently will a Canadian have an accent that might be American, or vice versa? (We can assume that the Canadian speaker is not from the Atlantic provinces and that the American speaks "General American," albeit with the cot-caught merger.) I'd be particularly interested in hearing whether there's been any research into this.

A personal note. I'm Canadian, but lived in northern California for a number of years. People there told me I didn't have what they thought of as a Canadian accent, and that I sounded American, something that surprised me. I estimate I'd have about an 80% success rate in the test I've proposed, but not better. Sometimes I can listen to someone for a while and still not know which country they're from. 64.140.121.160 (talk) 07:04, 29 March 2012 (UTC)

The give-away for me, other than talking about "The Crown", "MPs", "litres", "zed", etc., is if they pronounce "out" as "OWT" (American) or "OOT" (Canadian). I believe the American accent closest to Canadian is in the northern plains states (Minnesota, North Dakota, etc.). There are also some speech patterns, like ending a sentence in "aye ?", which are distinctly Canadian. If you lived in California for several years, you likely adapted your accent to match. This happens subconsciously. StuRat (talk) 07:27, 29 March 2012 (UTC)

Canadians do not in fact say "oot". Maybe Scots do, not sure.

The Canadian pronunciation of "out" is actually just like it's spelled: You start with an o sound, and dipthong a u sound onto the end. I'm not sure just which o and u sounds, but in any case it's closer to "oat" than "oot".

The American "out" on the other hand seems to start with the "a" vowel of cat, and transition into a "u" sound.

Either way, the differences between, say, a Toronto accent, and a General American accent, are pretty subtle, certainly less than between GA and strong regional American accents (rural South, traditional Boston or Brooklyn). --Trovatore (talk) 17:37, 29 March 2012 (UTC)

Unless we be out pirating, we prefer to spell it "eh". Clarityfiend (talk) 09:12, 29 March 2012 (UTC)

Is it true that for ay, in the sense of forever, is pronounced "eh" rather than "eye"? I think The King's Singers got that wrong. --Trovatore (talk) 00:21, 30 March 2012 (UTC)

Yes, I was spelling it the way it sounds. StuRat (talk) 17:05, 29 March 2012 (UTC)

An experienced judge wouldn't rely on accent so much as on word choice ("zed", "hydro" etc.), and in a 1-minute conversation may not have enough information to detect a difference. The accent border does not coincide with the political border. The accent of most of the Candadians I've met isn't readily distinguishable from General American, although I know a few that had a classic "Great White North" accent. But then, some Americans speak with a similar accent. Dominus Vobisdu (talk) 07:56, 29 March 2012 (UTC)

I'm excluding obvious giveaways like lexical items (zed, dreamt), matters of phonemic incidence (shone, bade, pasta) and topic of conversation (MP, litres), so we can assume nothing like this arises in the one-minute recording. I'm really only interested in accent. StuRat, I've always had the impression that it was closer to "oat" than to "oot." I know what you're talking aboat, though. 64.140.121.160 (talk) 08:20, 29 March 2012 (UTC)

The combination of Canadian raising and the absence of the Northern cities vowel shift is a good way to tell. But "Canadian raising" isn't really the same everywhere in Canada...personally I find it to be stronger rural parts of the country, or at least in rural parts of Ontario. And urban people (like me) will probably show some of the Northern Cities vowel shift, probably from watching too much American entertainment. A handy example is Coach's Corner. Can you tell the difference in the accents of Ron MacLean and Don Cherry? They are from different parts of Canada. Cherry's accent has some of that northern US vowel shift, while MacLean's has much stronger Canadian raising. Similarly in the US, rural and urban accents are different and the NC vowel shift and the raising aren't the same everywhere. It's definitely possible to distinguish accents - on one occasion I discerned that someone was from the same city as me, but that was based on listening to them speak for about an hour before introducing myself. But only one minute of speech probably isn't enough to pin it down exactly. Adam Bishop (talk) 08:33, 29 March 2012 (UTC)

I always change the channel when Coach's Corner comes on. Honestly, I don't usually enjoy listening to a whole minute of Don Cherry. But in the interest of science, I did listen to this. I didn't catch any ou vowels that could potentially have been raised in the little Ron MacLean said, but in other respects he sounds very Canadian. There were quite a few raisable ous in what Don Cherry said, and they were raised. At 2:53 he says "that's" in a way that doesn't sound like the Northern Cities Chain Shift. (Elsewhere, he raises /æ/ befor /n/, but this is normal.) Immediately after that, he says "I could go on and on," and he doesn't use the almost [a] vowel you'd expect in Buffalo. I'd be interested in having somebody from Buffalo/Cleveland/Detroit/Chicago say what they think of the way he talks, because he doesn't sound anything like them to my ear. 64.140.121.160 (talk) 09:12, 29 March 2012 (UTC)

Something I forgot to mention. A Californian friend once told me the way I say o before l in worlds like cold or pole struck her as unusual. Is there a difference between Canadians and Americans on this point? 64.140.121.160 (talk) 09:26, 29 March 2012 (UTC)

Well, it's certainly not as strong as an accent from Buffalo etc, but he does sound different from MacLean, at least to my ears. (He sounds more like me, I mean.) Adam Bishop (talk) 09:36, 29 March 2012 (UTC)

It's hard for me to tell how much of the difference is due to his accent and how much is to do with the fact that he shouts 100% of the time. 64.140.121.160 (talk) 19:32, 29 March 2012 (UTC)

Adam, this paper offers support for the distinction you're making between the Prairies and Ontario. 64.140.121.160 (talk) 23:37, 29 March 2012 (UTC)

If someone sounds pretty much American, but says words like "process" and "progress" with a long-o, or pronounces "schedule" as if it were spelled "shedule", that's a giveaway. But those words might not turn up in a short recording. The "oot" and "aboot" for "out" and "about" are more likely to turn up. Or, if you can find an old Cheech and Chong bit called "Les Morpions". [1] Although Tommy Chong seems to be doing more of a Minnesotan than a Canadian accent. ←Baseball Bugs ^{What's up, Doc?} carrots→ 11:14, 29 March 2012 (UTC)

I once identified a Canadian solely by his pronunciation of sorry to rhyme with quarry rather than starry (and the fact that he was apologizing at all...) Pais (talk) 14:30, 29 March 2012 (UTC)

Schedule and sorry are examples of cases where the difference is in what I referred to as phonemic incidence. I'm interested in accent only, that is, in differences in how Canadians and Americans pronounce the same sound, not in when they choose completely different sounds. 64.140.121.160 (talk) 15:24, 29 March 2012 (UTC)

Back to the "oot and aboot" thing, then. ←Baseball Bugs ^{What's up, Doc?} carrots→ 00:07, 30 March 2012 (UTC)

Bugs, I understand what you're saying about progress and schedule, but the information I have is that Americans are divided on process. 64.140.121.160 (talk) 19:35, 29 March 2012 (UTC)

I've never heard an American say "process" with a long "o". That seems to be a British Commonwealth thing. Although we do say "proceed" with a long o, and also "progress" when used as a verb instead of a noun. ←Baseball Bugs ^{What's up, Doc?} carrots→ 00:07, 30 March 2012 (UTC)

All right, I've done some digging. Sources say about 88% of English Canadians use Canadian raising in /aʊ/-words, and 81% do in /aɪ/-words. A majority of /aʊ/-non-raisers are among Newfoundlanders (who never raise) and minorities of non-raisers in Montreal and Vancouver. /aɪ/-raising is too widespread in the U.S. to be a useful shibboleth. /aʊ/-raising is known to be found in parts of eastern New England and eastern Virginia, and may now be found sporadically in Michigan, under Canadian influence. (Hockey Night in Canada?) These are fairly limited areas, and they have other identifiably American characteristics, so I think /aʊ/-raising is an excellent indicator.

However, that leaves out 12% of Canadians, and other sources have said the Canadian Shift is the most systematic and reliable indicator. This concerns the parallel retraction of /æ/, /ɛ/ and /ɪ/. The trouble is I have no real intuitive idea of what I should be listening for. I'm not aware of the differences between Canadians and "General American"-speakers on these vowels. Do other people find the Canadian shift (or its absence) easy to perceive? 64.140.121.160 (talk) 15:12, 29 March 2012 (UTC)

I live in eastern New England where "Canadian raising" is part of many local speakers' pronunciation. I have lived in enough other places and have enough of an ear for that sort of thing that I do notice it if I'm paying attention. However, what gives away a typical Canadian accent is the combination of raising with the absence of other features of the eastern New England accent and the absence of the Northern cities vowel shift that is common in most other places in the United States where Canadian raising occurs. The one part of the U.S. whose regional accent I can't really distinguish from Canadian accents is a region encompassing parts of Minnesota and the Dakotas. I'm thinking it is mainly northern Minnesota and the eastern Dakotas that have what sounds to me like a "Canadian" accent. And, as you've said, many Canadians in the Maritimes, Newfoundland, and parts of British Columbia lack a typical "Canadian" accent. (As a side matter, Francophone Quebecois have a distinctly French Canadian accent when they speak English that does not sound much like a French European accent.) Marco polo (talk) 17:52, 29 March 2012 (UTC)

Yes, it's clear that there are many characteristics of eastern New England speech that make it easy to recognize as being from the U.S. rather than Canada. Apparently, Maritimers (but not Newfoundlanders) do exhibit Canadian raising, according to what I've read. In Montreal and Vancouver, only a minority of speakers are non-raisers. In Montreal, native English-speakers of Italian ethnic origin are much less likely to raise than others. Most British-origin and Jewish English-speakers do raise in Montreal, but I've been unable to confirm whether they do so at the same rate as Canadians elsewhere, or if the discrepancy there is due solely to non-British-origin ethnolects. British-ancestry and Jewish Montrealers raise just as much as other Canadians. I found some information about sporadic Canadian raising in Detroit under recent Canadian influence, but you seem to suggesting that it occurs more broadly in the northern U.S. and is of long standing there. There are other aspects of Minnesota and North Dakota accents that are similar to Canadian ones, particularly as regards peripheral realizations of /eɪ/ and /oʊ/. Are you saying that Canadian raising is among the features that make these accents Canadian-like? 64.140.121.160 (talk) 19:27, 29 March 2012 (UTC)

I'd be very interested in seeing the source for those percentages of English Canadians having Canadian raising. Such exact percentages are usually pretty suspicious when talking about language. You definitely hear Canadian raising frequently, but my personal impression is that it's in the minority (and I have seen sources say that it's receding), at least among people in the GTA and in general among younger people. Of the numerous roommates I had during my university years in Canada, only two had Canadian raising (plus me, after I started adopting it half on purpose, half involuntarily). It definitely isn't a reliable indicator of a Canadian accent, because so many Canadians don't do it. --Terfili (talk) 22:11, 29 March 2012 (UTC)

Terfili, I took the 88% figure from The English Language in Canada by Charles Boberg, on page 151. But I've now found what I think is the original paper. It's based on a geographically representative sample of 86 students from across Canada studying at McGill, with raising defined as a difference of at least 50 Hz in the F1 of raised and unraised /aʊ/. The paper has additional details on the regional distribution of Canadian raising. 64.140.121.160 (talk) 23:23, 29 March 2012 (UTC)

Thanks for the link to that interesting paper! But keep in mind that those results are valid only for that particular sample, and can't be generalized to the general population (even though it is a reasonably large sample compared to many other studies). --Terfili (talk) 13:12, 30 March 2012 (UTC)

Terfili, my original question was about successful perception of the differences between Canadian and American English. Do you have any information about that? More specifically, is the "Canadian Shift" or its absence easy to perceive? 64.140.121.160 (talk) 16:17, 30 March 2012 (UTC)

I don't know how noticeable the Canadian shift really is (I have to admit I never noticed anything either before or after reading about it, but that could be because I was living in Canada and already used to it). A similar shift seems to be taking place among some speakers in the US too though (must be something triggered by the caught-cot merger), so in any case it probably isn't a very reliable indicator of Canadian English either. Overall, I doubt that it is possible to reliably distinguish English Canadians from GA-speakers based only on their vowels, unless of course the Canadian shows Canadian raising or the American doesn't have the caught-cot merger. It probably comes down to individual words, of which several people here have given examples (another one that comes to mind is [s] vs. [z] in "resources"). What no one has brought up are differences in intonation, which I think are often much stronger clues to where someone is from than what their vowels are like. But I don't know about any studies dealing with that. --Terfili (talk) 09:46, 31 March 2012 (UTC)

In resource, I thought that /s/ and /z/ were both possible in both countries. I thought the difference was that the pronunciation with the first syllable stressed as /ˈriː/ was possible in the U.S., but not in Canada. In addition to the cot-caught merger, the Canadian shift and Canadian raising, Boberg's paper quotes the ANAE as identifying the following features as being characteristic of Canadian English and not being "found to the same degree or with the same regularity in neighboring American varieties":

• "the articulation of the long vowels /ey/ and /ow/ (day and go) with tense, peripheral nuclei that approach monophthongal forms";
• "the articulation of /aw/ (cow) in relatively back position, a feature shared with some neighboring parts of northern United States."

The Boberg paper, published in 2008, describes the Canadian Shift as being "apparently mostly confined to Canada."

As a total non-native to any north-American dialect, I'd like to confirm what others have said above: if someone pronounces about as "aboot", while otherwise having what to me sounds like an American variety of English, that someone is Canadian, until otherwise proved. --NorwegianBlue ^talk 21:02, 29 March 2012 (UTC)

It's not "aboot" -- see my remarks above. Closer to "abote", but not that either. --Trovatore (talk) 21:07, 29 March 2012 (UTC)

Well yes, I agree, sort of. I certainly agree that the "oo" sound in Canadian about/abote/aboot is a diphtong. I think we'll have to resort to IPA to sort this out completely, but regrettably I'm not very good at IPA. The IPA transcription of this diphtong may have been given by other editors above, but not explicitly for the word "about". I also checked the article Canadian raising, but it did not contain enough detail for me to try to reproduce the diphtong that I have in mind. Anyway, I'm pretty sure we'd agree on what would be a Canadian "about", given sound samples. --NorwegianBlue ^talk 21:33, 29 March 2012 (UTC)

I think that there are degrees of raising. What I hear in Eastern New England (and pretty widely in Toronto) is something like [ʌʊ]. To most Canadians, that is perceived as "no accent", I think. I think that there is a variant with a higher first element something like [ɘʊ] (note that the first element is not a schwa), which is caricatured as "oo". This higher-starting diphthong may be the one that is more common in rural Ontario and that is really marked as Canadian. By contrast, the corresponding diphthong in the New York City area, where I grew up, is more like [aʊ]. Marco polo (talk) 01:14, 30 March 2012 (UTC)

What I heard in Toronto could maybe be [ʌʊ] but I heard it more as [ɔʊ]. My personal pronunciation (California with some Southern influence) is something like [æʊ]. --Trovatore (talk) 04:03, 30 March 2012 (UTC)

Trovatore, [æʊ] would be a good transcription of the unraised allophone of /aʊ/ in Canada, particularly before a nasal consonant. There is considerable variation in the raised variant. The further east you go, the more front it is. Quebec, which resembles western Canada more than it does Ontario, is the only exception to this. (However, it is not clear from the sources I have if this would be the case in Quebec if only British-ancestry speakers were taken into account, as opposed for example to native English speakers of Italian or Jewish descent, who present significant differences.) In the paper I linked to above, Charles Boberg writes "The data [...] suggest that raised /awT/ is produced about 150 Hz further forward in southern and eastern Ontario than in western Canada, corresponding to phonetic values approximating [ε^>ʊ] and [∧^<ʊ], respectively." 64.140.121.160 (talk) 16:42, 30 March 2012 (UTC)

Americans seem to pronounce "toque" with a silent "que", while Canadians pronounce it "too-k". ~AH1 ^(discuss!) 15:23, 1 April 2012 (UTC)

Canadians pronounce the word /tjuːk/ (or /tuːk/ by those with yod-dropping, as you say) and usually spell it tuque. The word toque is from European French and tuque is from Canadian French. 64.140.121.160 (talk) 22:42, 1 April 2012 (UTC)

Pronunciation of /oʊl/ in North American English

In Canadian English, as in many varieties of North American English, /oʊ/ is centralized to a moderate extent. However, in The English Language in Canada, Charles Boberg writes that "centralization of /ow/ in [Standard Canadian English] is blocked by a following /l/." (That is, the vowel remains back.) Although the vowels in go, stone and toe have an average F2 of 1291 Hz, in words like cold the vowel averages an F2 of 932 Hz. To what extent is this characteristic of Canadian English shared with various dialects in the U.S.? 64.140.121.160 (talk) 15:12, 29 March 2012 (UTC)

I'm Canadian, and I can't tell any difference between the /oʊ/'s in "tone" and "toll". However, you may want to take my recommendations with a grain of salt. Interchangeable|talk to me 20:41, 29 March 2012 (UTC)

California, not Canadian, but for me tone has a diphthong while toll has a monophthong. In fact, the only place I have s.t. approaching cardinal [o] or [ɔ] is in toll, tore, toil (≈ /ton, tor, tojl/); everwhere else I have ≈ /ʌw/. I suspect this is because the velarization of the dark el absorbed the offglide of the diphthong, so I wouldn't be surprised if tone and toll developed differently in Canadian or other English varieties. — kwami (talk) 21:02, 29 March 2012 (UTC)

Kwami, I'm Canadian, and it sounds like we're basically in agreement. My toll is a back monophthong or something close, and my tone vowel is a diphthong whose nucleus is somewhat centralized, though less than yours. I think what you're saying implies that your speech has the characteristic I described above. 64.140.121.160 (talk) 22:57, 29 March 2012 (UTC)