Talk:American and British English pronunciation differences

From Wikipedia, the free encyclopedia

In refactoring this out of American and British English differences, I've added a lot of new individual words and rearranged the material. Most of the additions I made were from Wells, especially words he's marked with an asterisk as being different. I cross-referenced against various dictionaries, and generally where there was disagreement I cut the Gordian knot by just removing the word. No doubt I've added many shiny new mistakes. Joestynes 02:53, 22 September 2005 (UTC)

I've removed raspberry, because I'm pretty sure it is in fact a trap/bath word. (I haven't checked this, but it's not easy to check; British dictionaries don't generally show the short vowel pronunciations of BATH words, in spite of how common they are in the UK.) I think we should use phonemic transcriptions in many places, partly because it's what dictionaries (in Britain) generally do, partly because then we don't need to mention irrelevant detail like dark L, and partly because it allows for regional variation a bit better (though not perfectly): for example the short A /æ/ is actually [a] in much of the UK. But I should say that this page looks a lot better than the pronunciation section of American and British English differences used to look.--JHJ 17:34, 2 October 2005 (UTC)
The trap-bath split page mentions the phonetic context as preceding [f, s, θ, ns, nt, ntʃ, nd, mpl], whereas raspberry has following [z]. Joestynes 11:30, 4 October 2005 (UTC)
Some people say it with /s/ (well, I do, anyway). But regardless of that, the vowel variation in raspberry is presumably inherited from that in rasp, which is a straightforward BATH word. Maybe there should be some mention of raspberry (and also Glasgow, which also has BATH-type vowel variation before /z/) on the trap-bath split page, but I don't think it's needed here.--JHJ 12:19, 5 October 2005 (UTC)

Contents

[edit] FACE and GOAT vowels

I don't really think these can be thought of as British/American differences; there's too much variation on both sides of the Atlantic for that. I've left the GOAT one in with a comment on the variation because of the difference in the commonly used symbols, but I've deleted FACE, where /eɪ/ is commonly used for both RP and GA.--JHJ 19:42, 5 October 2005 (UTC)

[edit] Schwa / wedge

This

In GAm, realisations of [ʌ] and [ə] are more similar than in RP, and are sometimes viewed as stressed and unstressed variants of a single /ə/ phoneme. (This is reflected in the common American use of uh as a pronunciation spelling for schwa.)

was replaced with this

In RP, realisations of [ʌ] and [ə] are more similar to each other than in GAm; RP [ʌ] is closer to [ɐ].

Since the page really only details phonemic differences (ie that might be reflected in broad transcriptions) rather than phonetic differences, I have deleted the corrected version altogether. Joestynes 20:42, 20 December 2005 (UTC)

I think there may be something worth mentioning here. Some Americans seem to perceive the vowel of cut and schwa as allophones of one phoneme, but in an RP context they're usually treated as separate phonemes /ʌ/ and /ə/. For example, the OED 3rd edition transcription scheme [1] uses different symbols for the vowel of strut and the first vowel of another in its "British" transcriptions, but the same symbol /ə/ in its "American" transcriptions. The OED even has a minimal pair: muzz /mʌz/ and one of the pronunciations of Ms /məz/. --JHJ 21:38, 20 December 2005 (UTC)
That was the basis for the original para (which I wrote). My instinct was to revert the edit; however I cross-checked the replacement para against other Wikipedia articles and it seems to check out. It's possible that GAm "allophones" are phonetically more distinct than RP "distinct phonemes"; that would be a statement about phonetic tradition rather than objective reality, but still worth mentioning. I don't believe the new version's info merits inclusion, even if it is accurate. Joestynes 11:08, 21 December 2005 (UTC)
I'm the one who did the recent change but I agree with Joestynes because I had never noticed the distinction between the two until I read about it. Since the article is about the actual pronunciation differences (presumably noticeable differences) and not how dictionaries consider the phonemes, the deletion seems appropriate. AEuSoes1 06:20, 22 December 2005 (UTC)
Well how about replacing with something like this:
In GAm, realisations of [ʌ] and [ə] are often viewed as stressed and unstressed variants of a single /ə/ phoneme. (This is reflected in the common American use of uh as a pronunciation spelling for schwa.) RP transcriptions traditionally retain two distinct symbols.
which makes no claim that the underlying phones are more similar in GAm. Joestynes 11:52, 22 December 2005 (UTC)
That seems like a good idea, but I'd suggest dropping the phonetic symbols [ʌ] and [ə], which are a bit confusing here, and using example words instead. After all, the vowels may not be [ʌ] and [ə]; it's just that /ʌ/ and /ə/ are the usual transcriptions of the two RP phonemes.--JHJ 12:17, 22 December 2005 (UTC)
I don't like it. It's taking a pronunciation distinction and changing it to, basically, a spelling one. "uh" is simply the closest thing in GAm spelling that comes to indicating a schwa; whereas in RP, they have "er." It also seems to imply that the two sounds are closer in GAm without stating it.AEuSoes1 20:44, 22 December 2005 (UTC)
However, there is some evidence (e.g. the OED transcriptions) that there is a tendency to regard the strut vowel and schwa as the same phoneme in GAm but not in RP and other British accents.--JHJ 21:05, 22 December 2005 (UTC)
True, but that is not a pronunciation difference; it's a transcription one. OED only shows the ideology of a British editor, grammarian or linguist who noticed the difference or considered them different and I'd hardly consider the view of experts trained to notice the difference between schwa and wedge as indicative of regular speakers' ability to do so.
This can probably go in another article about RP and GAm differences but using only the OED reference to come to the conclusion that we're making seems a bit of a stretch.
I knew a fellow who had trouble telling the difference between scwha and wedge until he was told that schwa only appeared in unstressed syllables. Maybe a non-linguist RP-speaking editor could verify the other half of what we're talking about. AEuSoes1 21:33, 22 December 2005 (UTC)
According to http://www.oed.com/about/oed3-preface/pronunciation.html the OED's American pronunciations are "based on a model devised by Professor William Kretzschmar of the University of Georgia". I'd like to know how standard his analysis is, though, so I'm not suggesting basing anything entirely on the OED pronunciations.--JHJ 21:51, 22 December 2005 (UTC)

First of all, for both RP and GA the stressed vowel is a low mid central vowel, though lower in RP than in GA, that is more accurately transcribed [ɐ]. (In RP it needs to be lower to remain distinct from [ɜ], while in GA these vowels are distinguished by rhoticity, and there is greater leeway in height.)

I (GA speaker) generally hear schwa as being either [ɐ] or [ɪ] (or maybe [ɨ]). I believe the difference is allophonic, being decided by environment, but I understand this difference is phonemic in RP. Perhaps this is part of the reason for the different transcriptions? Either way, if we don't use distinct symbols for obscure vowels, then they will need to be transcribed as two vowels, [ɐ] or [ɪ]. kwami 22:51, 22 December 2005 (UTC)

Here's what Wells (1982: 132) has to say about it: "GenAm characteristically has a somewhat backer quality of /ʌ/ than present-day RP, [ʌ+] as against [ɐ]; Scottish and Canadian speech are usually like GenAm in this respect, but southern-hemisphere accents like RP. ... Even in GenAm it may well be considered that stressed [ʌ] and unstressed [ə] are co-allophones of one phoneme." --Angr (t·c) 17:33, 23 December 2005 (UTC)

[edit] Issue with issyu vs ishu

The article presently states "In some words where [j] has been coalesced in GAm, it may be retained in RP: e.g. issue is RP ['ɪsjuː] or (as GAm) ['ɪʃuː]" ... I must take ['ɪʃ·juː] with this assertion... Anyone who said ['ɪʃuː] when speaking GAm would raise eyebrows and cause people to wonder what kind of bizarre accent they were trying to cover up with such an affectation. I can't think of an alternative example ATM...perhaps someone else can? Kwami? Tomertalk 05:43, 15 March 2006 (UTC)

Hmm. Maybe I'm wrong. I say "ishyu", but according to the [American Heritage] dictionary (which, as everyone knows, is the only correct source for pronunciations), the only pronunciation given is "ishu". Tomertalk 05:43, 15 March 2006 (UTC)
TShilo, you must be putting me on. The only real pronunciation reference is Webster's Third. Period. AHD can't hold a candle to W3. Get real, man. --JackLumber 22:43, 15 March 2006 (UTC)
Yeah :-p, but >:-o ...<back to my point? :-\> Tomertalk 01:37, 16 March 2006 (UTC)
Why don't you just speak English instead of jotting down a haphazard hodgepodge of weird punctuation marks? Half-kidding. Anyways, that's what Webster's third has to say about the issue pronunciation issue:
/'i(,)shü, 'ish(,)yü, 'i(,)shůə, 'ish(,)yů, before a vowel often -_sh(y)əw; chiefly in the southern US -_sh(y)ə before a consonant or pause or before a vowel in a following word; chiefly Brit 'i(,)syü or 'i(,)syů or 'isyəw/.
' and , are primary & secondary stress; ü is as in "moon," ů as in "foot," other symbols are self-explanatory. Hope this helps.--JackLumber 13:03, 16 March 2006 (UTC)
So I'm not a freak then...well, at least not bcz of how I pronounce "issue". My point stands then, that we need to find a better example, if, indeed, one exists. Tomertalk 00:35, 17 March 2006 (UTC)
Hmm... Based on a cursory investigation, I'd say that the commonest General American pronunciation is the "coalescing" one. Assuming there is such a thing as General American. I personally would table the ishu until someone comes up with a better solution.--JackLumber 19:09, 17 March 2006 (UTC)
Just to add to the discussion, my oxford electronic dictionary has ['ɪsjuː] and ['ɪʃuː] as equal transcriptions of issue (no pronunciation is associated with American and British speech as this dictionary tends to do). However, tissue is ['tɪʃuː]. I pronounce tissue as a perfect rhyme of issue but I honestly wouldn't notice if someone epenthisized a [j] right after the /ʃ/. I'm not saying that GA does one or the other, I just don't think that it's all that marked or even noticeable. AEuSoes1 20:13, 17 March 2006 (UTC)
The somewhat cumbersome transcription of Webster's 3rd actually highlights your point, AE—some nuances are really hard to catch. "Tissue" & "issue" rhyme in my own speech too. Not surprisingly, W3 transcription of "tissue" goes exactly like "issue" (except for allowing for a dialectal "tishee" in the phrase "tissue paper.")--JackLumber 18:28, 18 March 2006 (UTC)

[edit] Is it just me...

Or are a lot of these entries seemingly written for linguists? In order to effectively read this entry, chances are, you are already aware of the content. It's fairly incomprehensible to someone not already well versed in linguistics. I think more examples should be given to make this article useful for laypersons.

I agree completely. Whereas the other articles on differences between British and American English (e.g. the one on spelling differences) are extremely clear, this one is incomprehensible to the non-linguist. I think this article should be a candidate for a complete re-write to make it useful for non-specialists. ~~

[edit] Loanword 'o'

Can't see this mentioned anywhere. There's a marked difference between RP and GAm of loanwords with the letter 'o'. For example the penultimate vowels in risotto and gnocchi are pronounced with ɒ in RP but oʊ in GAm. Also applies to French and Spanish loanwords.

(Curiously, both RP and GAm agree on the use oʊ of for the final vowel.)

Anyway, I'm not a trained linguist so maybe someone else could write this up properly if it's not buried in there already.

That sounds fairly accurate, but I don't know if it's OK with others yet. Your observation on the "different" British vowel is simply to do with the normal pronunciation of "o" before two or more consonants, eg. doting (oʊ) vs. dotting (ɒ). -- THE GREAT GAVINI {T|C|#} 19:10, 23 August 2006 (UTC)
Some examples are already in the Miscellaneous pronunciation differences table; there is a remark "(And A in many other foreign names and loanwords)" in the chianti-gulag-pasta row in the table. Something similar for O would I think suffice; it applies to some such words but by no means all. Add any you can think of to the table (check dictionaries first though: dictionary.com disagrees with you about the American pronunciation of risotto ) jnestorius(talk) 18:03, 30 August 2006 (UTC)

[edit] The pronunciation of harass

I am trying to determine which pronunciation of 'harass' is favoured by whom - American vs British, or do both countries allow both variants? I should appreciate any help you could give me.

Charl Meyer (South Africa)

In American English, stress is usually on the second syllable; in British English traditionally on first. That's pretty much it. -- THE GREAT GAVINI {T|C|#} 19:04, 23 August 2006 (UTC)
I've heard both ['hærəs] and [hə'ræs] (or [hɑ'ræs]) in native American English speakers. I'd argue that the latter is more common nowadays amongst the younger generation with the former being preferred by older generations, at least in Southern speakers. --Alai 03:26, 24 August 2006 (UTC)
Check out AHD's usage note. Second-syllable stress does appear to be commoner, however. My pronunciation is kinda like [hə'ɹe(ː)əs]. JackLumber. 15:35, 11 September 2006 (UTC)

[edit] -atory

From the article:

... regulatory with AmE /'ɹɛgjʊlɪˌtɔɹi/ and BrE /ˌɹɛgjʊ'leɪtəɹiː/...

But the OED Online clearly gives first-syllable stress as the only acceptable pronunciation, and I as a native Brit agree with it. So something's amiss. —Blotwell 04:19, 11 April 2007 (UTC)

I suspect this is a disputed change in progress in BrE. Cambridge Advanced Learner's Dictionary gives only the third-syllable stress, marking the first-syllable stress as US. Many OED pronunciations are decades old; the CALD is highly non-prescriptive. However, the general claim made seems plain wrong in any case ("The ending -atory is similarly different: in BrE primary stress moves from the root word to the first syllable of the suffix, while in AmE the stress of the root is unchanged"). In mandatory, retaliatory, conciliatory, discriminatory, there is no shift in BrE; in oratory, there is a BrE shift backwards; in compensatory, there is an AmE shift, though not the same as BrE. So I've expunged the offending sentence, and just enumerated the individual differences. The moral is: if you're looking for a pattern, don't just make one up, find a reliable source. jnestorius(talk) 22:27, 12 April 2007 (UTC)

[edit] Clippings and Hackish

I'm wondering whether/how to incorporate this from The New Hacker's Dictionary, s.v. Commonwealth Hackish:

It is reported that Commonwealth speakers are more likely to pronounce truncations like `char' and `soc', etc., as spelled (/char/, /sok/), as opposed to American /keir/ and /sohsh/. Dots in newsgroup names (especially two-component names) tend to be pronounced more often (so soc.wibble is /sok dot wib'l/ rather than /sohsh wib'l/).
  • I wonder if the dots in newsgroups difference is still current; in Web URLs don't the dots always get pronounced (as in Oracle's blurb "We're the dot in dot com")? In which case, has this affected Newsgroup usage?
  • Do Americans pronounce SOC in one consistent way? If it's newsgroup "soc.foo.bar" then [soʊʃ] would correspond to "social" as stated. If you have a SOC101 sociology course, it should be [soʊs] for the majority who say [ˌsoʊsiˈɑlədʒi] rather than [ˌsoʊʃiˈɑlədʒi]. For "Law Soc." I surmise one simply says "Law Society" rather than "Law Suss". In BrE in all these, SOC is [sɒk] "Sock".
  • The [keɹ] given for char presumably corresponds to [ˈkeɹəktɚ] for character; for those without the Marry-Mary merger the full word is [ˈkæɹəktɚ], which can't make [kæɹ] but might make [kɑɹ].

Most pertinently, if there is as stated a greater tendency for BrE to pronounce truncations without regard for the pronunciation of the source word, this should be reflected in other words, and indeed in non-geek words. But I'm having trouble finding further examples. One candidate is cos for cosine (still geeky, I know): BrE [kɒs] or [kɒz] (Concise OED); is the AmE [koʊs], [ˈkoʊsaɪn], or [kɑs]? (cite?) OTOH, sin is always pronounced like sine. OTOH there are cases where AmE changes the pronunciation as BrE. This guy lists these (not all of which I agree with):

  • spec (short for specification)
  • mayo (short for mayonnaise, where the "y" is never pronounced)
  • bro (short for brother; it's a completely different vowel)
  • expat (short for expatriate; the vowel in the short form is short, but it corresponds to a long vowel in the original)
  • deli (last vowel is longer)
  • bio

There is not enough evidence to support the general claim IMO, so maybe just the specific examples should be added. jnestorius(talk) 23:19, 12 April 2007 (UTC)

[edit] Translation to English

Can someone translate the phonology so that someone besides a phonologist can understand this? Militärschokolade 23:10, 5 September 2007 (UTC)

[edit] Pronunciation of "calm" and "palm"

In England, these words are pronounced with the "ah" vowel, but I often (but not always) hear Americans pronouncing them with the "aw" vowel instead. Also, some Americans pronounce the L, which doesn't happen in England. Is this worth integrating into the article? I think it is a valid difference.

Incidentally, why do some Americans use the "aw" vowel instead of the "ah" vowel used in normal English? Avengah (talk) 01:47, 11 April 2008 (UTC)

British English hasn't been peculiarly normal in a long time. And, in various dialects, “aw” and “ah” may not have quite the values that you presume, leading to misinterpretation of your question. I suggest that you use the IPA in your queries. —SlamDiego←T 04:55, 11 April 2008 (UTC)

As an American, I am also quite curious why our pronunciations differ since our way isn’t normal English.--DavidD4scnrt (talk) 04:21, 11 April 2008 (UTC)

OK, well I'm trying to help the article, and if you can't get past my use of "normal English" to actually help the article, it's Wikipedia's loss, not mine. Anyway, I don't know IPA, not to mention it doesn't display properly on my computer anyway (using IE6, Win XP Pro, and I can't upgrade to IE7). I'll try to explain it more clearly.
In England, the words "talk", "walk" etc. use the "aw" vowel, which is the same as "or" for England. However, the words "calm", "palm", "psalm" etc. use the "ah" vowel.
In America, the words "talk" and "walk" also use the "aw" vowel (which is a different "aw" vowel than the English "aw", but nonetheless it is still "aw"). Some Americans merge "aw" and "ah"; for these people this is irrelevant, but for the others, there are some Americans who use their "ah" vowel in "calm" and "palm" etc. but some who use their "aw" vowel in "calm" and "palm" etc. I hope I've made it a bit clearer! I'm trying to write this from a NPOV, so please give me a little bit of credit! Avengah (talk) 16:10, 11 April 2008 (UTC)
Wikipedia certainly wants your positive contributions, but it has limited tolerance for arbitrarily labelling other cultures as abnormal or improper. (Unless, perhaps, you could show them to be objectively such, but that won't be possible in the case of American English.)
Wikipedia is a pretty good place to learn the IPA, and it has long been a popular system in British dictionaries. You can look up the words of interest in recent editions of the OED and of the SOED; the IPA is used therein. And, below, I've supplied a link to decent TTFs of the IPA.
When it comes to the words that you list, there is more variation within American English (and within British English) than is captured by your description. I'm not sure that there is much hope for a practical discussion unless someone can dig-up a statistical study to which to refer. —SlamDiego←T 04:58, 12 April 2008 (UTC)
I agree with SlamDiego - I suspect there is more variation in the range of pronunciations of palm/calm within BrE and within AmE than there is between BrE and AmE, and that there is therefore nothing for this particular article here. (And, as something of an aside, but to illustrate such variability, the suggestion made by Avengah that in BrE "the aw vowel ... is the same as or" is complete nonsense, in the variety of BrE that I speak!) Snalwibma (talk) 11:42, 12 April 2008 (UTC)

[edit] I can't view IPA properly!

Some symbols, such as the symbol for lengthening a vowel (similar to :) and the symbol that represents the "bit" vowel, do not appear properly - they instead display as a square. I'm using IE6 on WinXP Pro. Does anyone know why this could be, and is there a font I can download to fix the problem? I can't upgrade to IE7. Thanks! Avengah (talk) 17:01, 11 April 2008 (UTC)

Try SIL's IPA fonts. —SlamDiego←T 04:46, 12 April 2008 (UTC)