Talk:Australian English phonology

From Wikipedia, the free encyclopedia

This article is part of WikiProject Phonetics, an attempt to build a comprehensive and detailed guide to phonetics and descriptive phonology on Wikipedia. If you would like to participate, you can edit the article attached to this page, or visit the project page, where you can join the project and see a list of open tasks.

This is the talk page for discussing improvements to the Australian English phonology article. This is not a forum for general discussion about the article's subject.
Please sign and date your posts by typing four tildes (~~~~). Put new text under old text. Click here to start a new topic. New to Wikipedia? Welcome! Ask questions, get answers.	Be polite Assume good faith No personal attacks Be welcoming	Article policies No original research Neutral point of view Verifiability

edit Archives and related discussion
Talk:Australian English
Archive I	Archive II	Archive III
Archive IV	Archive V	Archive VI
Talk:Australian English vocabulary
Archive I	Archive II	Archive III
Wiktionary:Appendix talk: Australian English vocabulary
Talk:Australian English phonology
Archive I	Archive II
Talk:Varieties of Australian English
Talk:Regional variation in Australian English

Contents 1 Variation between /aː/ and /æ/ 2 What about Perth? 3 Wrong Symbols? 3.1 /a/ & /aː/ 3.1.1 Good Points 3.2 Competing Orthographies 4 Feesh and Cheeps and Fush and Chups 5 Golf-gulf merger 6 Kambah sounds like Canberra? 7 Wrong symbols part two 8 Celery/salary merger 9 /ɐ:/ different before "r" 10 Pronunciation of "our", "ours" 11 Pronounciation of ʉː and iː 12 Madder

[edit] Variation between /aː/ and /æ/

I have changed the values of demand and grasp to 99% for Adelaide. I am Adelaide born - I and many of my friends pronouce demand and grasp with an /æ/.

That's all very well, but (a) adding information based on how you and your friends speak generally counts as unverified original research, (b) those figures are from a (fairly small, I think, looking at the numbers) sample of speakers in the different cities, and the quoted findings are 100% long A for both of those words in Adelaide - we can't change the findings of that study.--JHJ 11:10, 7 April 2006 (UTC)

[edit] What about Perth?

The table "Variation between /æ/ and /aː/", based on Crystal does not include Perth/WA. Does this reflect Crystal's work, or is it an accidental omission? Grant65 (Talk) 06:15, August 27, 2005 (UTC)

I'm not sure, but I think my mother has a copy of that Encyclopedia, so I might be able to find out. Other people might be able to find out sooner than I can though. (Given it's an encyclopedia, Crystal probably didn't do the original research, though ;) — Felix the Cassowary 12:03, 28 August 2005 (UTC)

I went to my mother's today, and had a look at it. Perth is indeed missing, including in the paragraph of discussion. It cites a chapter of a book as its source, and I think we should probably mention it—but I'm not sure how best to do secondary citing here. The original work was Bradley, D. (1991). ‘/æ/ and /aː/ in Australian English.’ In J. Cheshire (Ed.) English around the world: Sociolinguistic perspectives. Cambridge: Cambridge University Press. pp. 227–34. If anyone can find that source, it would probably pay to see if it mentions anything about Perth, including why it was omitted. —Felix the Cassowary ɑe hɪː jɐ 12:15, 4 September 2005 (UTC)

[edit] Wrong Symbols?

It's crept back in again. It's not /a/ & /aː/ but /ɐ/ & /ɐː/. Jimp 28Aug05

See discussion at Talk:Australian English#New_Australian_English_phonology_page. (Well, you already know it's there, but other people who come in later mightn't.) — Felix the Cassowary 12:03, 28 August 2005 (UTC)

I was going to put the link here to but hadn't got round to it. Then, on the other hand, why not move that discussion here? Jimp 29Aug05

Well now it's all here. Jimp 11oct05

[edit] /a/ & /aː/

G'day Felix,

I'm all for your splitting the article in two however I don't agree with your using /a/ and /aː/. It's all well and good to use /a/ and /aː/ amongst those who know that it really is /ɐ/ and /ɐː/ but do we assume this? What would the point be of writing the article if it's only for those who already know AusE phonology?

There's also the question of consistency. You haven't used /r/ but /ɹ/. You haven't used /uː/ but ʉː/. Isn't better to use the correct IPA transcription?

I'd already changed these back to /ɐ/ & /ɐː/ before I read your reasoning however, for the reasons I've set out here, I'm not going to change them back.

Jimp 28Aug05

I didn't use /r/ because I just copy-pasted from the RP English page and removed the unvoiced [w] sound, not used in Australian English. I didn't even know I hadn't used /r/! :)

As for not using /u:/, but /ʉː/, that's because, well, the correct sound is /ʉː/, according to the standard we're using.

However, using /a/ and /a:/ for the open central vowels isn't wrong: In fact, it's what you're meant to do. They're fully open: opener than /æ/, so they aren't nearly-open central vowels, so [6] isn't the ‘correct IPA transcription’. Therefore, we have to pick one glyph over the other, and IPA principles state that if we have to pick two glyphs, we should pick the simpler one. Every other language which has only one fully open vowel uses /a/, in spite of the fact that the real sound is actually almost always [a_-] or [6_o], and not [a]. Why should we be different for AuE? (Hungarian, in fact, based on the graph at the Hungarian language page, uses /a(:)/ for a phoneme even further back and higher than Australian equivalent.)

Also, my reading indicates that the normal IPA transcriptions that are used in the literature are either Harrington-Cox-Evans with /a(:)/ (i.e. what I put up), or M.-D. (which is many times more common, anyway; even though it's less phonetically accurate).

I'm thinking of drawing a diagram like they have at Dutch language showing the positions of the vowels (after all, almost no language has the vowels in the cardinal positions—based on the Dutch language vowel chart, in fact, AuE /ʉː/ is further forward that Dutch /y/, and AuE /3:/ is nowhere near IPA [3:]). If we do that, would you object less?

— Felix the Cassowary 11:23, 28 August 2005 (UTC)

If this is what you're meant to do then why don't Harrington, Cox & Evans (1997)? Look at the graph: it's clearly /ɐ/ and /ɐː/ (and so it's labelled) it may not right on but it's closer than /a/ and /aː/. Jimp 28Aug05

The fact that it's closer to the [ɐ] isn't relevant to the intended use of IPA [a] vs IPA [ɐ]. IPA [a] is clearly intended for use as a low front or low central vowel, whereas IPA [ɐ] is intended for use as, essentially, a second lower schwa. Look e.g. at Template:Vowels or open front unrounded vowel, where it says:

This symbol (i.e. [a] —FC) very frequently used for an open central unrounded vowel, and this usage is accepted by the International Phonetic Association. Since no language distinguishes front from central open vowels, a separate symbol is not considered necessary. If required, the difference may be specified with the central diacritic, [ä].

I'm not Harrington, Cox, or Evans, so I don't know why they chose [6] and not [a]. Incidentally, their IPAification appears to have been based on Clark's (1989), and he used /ɑ/. Try finding that in an IPA chart. (Cox, actually, uses /ʌ/ and /a/ in what she's published since then.) And, like I've seen, all the printed stuff I've seen apart from HCE (1997) and the website has used either the M.-D. system (as Cox does) or HCE-but-with-/a(:)/.

(Btw, I fixed your links—for some reason, the syntax for links is different from the syntax for wikilinks, just to confuse everyone.)

— Felix the Cassowary 12:37, 28 August 2005 (UTC)

I'm not Harrington, Cox nor Evans either and I can't read their minds, however, if I might be allowed to hazzard a guess, I'd assume that they've used [ɐ] exactly because of this fact that it's closer to this than it is to [a].

Is this irrelevant? I'll go so far as to say that it is relevant ... at least for a phonetic transcription. Sure, if it's a phonemic transcription, then it doesn't matter. Perhaps the material you refer to is exactly that: phonemic transcriptions only.

But which are we using here: a phonemic or a phonetic transcription? Both, of course, but don't we want to tell readers how we actually pronounce things? We do, don't we? Thus, I argue, we should use a phonetic transcription based on the closest IPA symbols and, to avoid confusion, base our phonemic transcription on that.

Isn't this the reason that they used [ɹ] rather than [r] on the RP page? You speak of our using standard (with respect to /ʉː/). Are they using the same sandard on the RP page? Who are we? I don't know of this standard. The standard I'm using is the Mac. Uni. website the links to which you've fixed up for me (thanks for that).

Just an asside ... shouldn't we move this discussion to Talk:Australian English phonology?

Jimp 29Aug05

Felix,

You link to the Wikipedia vowel template and the page on the open front unrounded vowel where there is an a in brackets in the central position. This is interesting but why is it absent from the IPA vowel chart?

It seems as if we agree that the vowels in hut and heart are closer to X-SAMPA /6/ and /6:/ than they are to /a/ and /a:/ respectively, as the Mac. Uni. vowel chart indicates. You've refered to it as a fact.

What we disagree on is the relevance of this fact. Is this not the kind of fact a reader might be looking for in an encyclopædia? Where better in this encyclopædia to find this fact than Australian English phonology? I argue that this is a fact which should be included here.

You mention the fact that other languages with only one fully open vowel use /a/. You ask why AusE should be different. My answer is that, no, it should be no different ... if it's a phonemic transcription only. You mention Harrington-Cox-Evans' use of /a(:)/ in the literature. All I can guess is that they're giving only a phonemic transcription in this literature.

What I argue is that we should be giving a phonetic transcription. Don't we want to give the whole story? Using /a(:)/ doen't give the whole story. This goes for /ɹ/ ʉ:/. But I'm just repeating myself now. What we need is a show of hands. I wonder who with me & who's with you.

Jimp 31Aug05

I'm replying on Talk:Australian English phonology; you're right, that's the more appropriate palec. — Felix the Cassowary 11:25, 31 August 2005 (UTC)

and the expression "good on you" or "although both of these are - something seems to be missing between "or" and "although". --Oop 23:52, September 1, 2005 (UTC)

Sorry about not responding to your other comment there, I'll do them both now (I must've missed it in my watchlist).

A central [a] is missing from the IPA vowel chart because that's not its cardinal position. Its cardinal position is indeed front. That doesn't mean it's incorrect to use it in non-cardinal positions. The IPA doesn't contain symbols for every possible vowel; nor even every vowel used in a language; only for those which contrast. No language contrasts a front open from a mid open from a back open vowel, so there's no need to include three fully open vowels; languages can therefore use /a/ to phonemically represent a sound that is phonetically in a centralised position (and indeed, the IPAssociation does so).

If we're going for a phonetic transcription, we're going to have to mark a lot more—the [uː] allophone of /ʉː/; the [ɔʊ] allophone of /əʉ/; the fact that /əʉ/ is more like (ahem) [ɐʉ] or [ɜʉ] depending on how liberally we wish to interpret the symbols; the fact that /ɜː/ is more like [øː] or [ɵː]. I don't think this level of detail is necessary, and in any case it's misleading as you'd have to either still include the discussion, or diacritics, because the only Australian vowel that's on (or rather, has a mean value that's basically on) the cardinal pronunciation is /oː/.

Secondly, the purpose of the symbols, there at least, are more to serve as a guide to interpreting pronunciations; that's why the section on vowels begins by discussing the fact that there's two ways of writing Australian phonemes. It's not meant to be, and I don't think it should be, a guide to the phonetics of Australian English. Firstly, because to do that with symbols would not do justice to the phonetics, even dressed up in diacritics. Secondly because we can do that more comprehensibly using graphics and textual descriptions, as I've already included.* Thirdly because if we write /ɪə/ as [iː] as would be phonetically justifiable/justified, we'd serve only to confuse people when they actually do read phonemic transcriptions written in Australian English. And finally because if people do want to know the actual phonetic pronunciations of Australian English rather than graphical, textual or symbolic approximations, they're better served by travelling to our fair country, or failing that going to the primary literature.

* Regarding the graphic, I've also included the GIMP source file on the Wikimedia Commons site—follow the links from it—so if you do want to change the character I've used for /a(ː)/, you can without too much difficulty, assuming you have the GIMP. The font I've used is Gentium.

When you say that perhaps the material I've been reading that uses /a/ rather than /ɐ/ is phonemic material only, to some extent you're right. It includes stuff that's been using broad phonetic transcriptions because the precise phonetics of /a/ isn't the matter. But stuff that is concerned with the exact phonetics of /a/ has either used diacritics to symbolise the low central nature of the vowel or, more often, has been using textual and graphical descriptions, including references to the frequencies of the formants of the vowel. Because that's the only way you can capture the phonetics of a vowel on paper.

Regarding standards, I don't mean that there's some standards body that dictates or even suggests appropriate IPA characters for Australian English. I mean a standard as in the normal way of doing something. In Australian English, when length is marked, the normal way of referring to the vowel phoneme in ‘food’ is with /ʉː/, so therefore, /ʉː/ is the ‘standard’ glyph.

— Felix the Cassowary 11:24, 31 August 2005 (UTC)

[edit] Good Points

You put forth some very good points, Felix. Also, the vowel chart you've created serves to negate most of my concerns. I'd been worrying that the use of /a(:)/ would indicate quite fronted vowels but the graph makes the fact that these are central clear enough. I'm not about to go changing things back to /6(:)/ (in fact I've just done the opposite). At least not until I have a bunch of others (who seem to know what they're talking about) who also would prefer /6(:)/. Thanks for the graph. It would be nice to have one for diphthongs I'd do it myself if only I had time to figure out how. Anyway, all in all, I think you've done a good job on this page. Jimp 1Sep05

Thanks. I'd like to do one for the diphthongs too, but it's also a case of figuring out how... I don't really know if the Gimp can draw likes with arrows (I've copy-and-pasted the one for /i:/ from the Dutch diphthongs chart, but I don't know that that's suitable in general). I might see what I can do tonight though. — Felix the Cassowary 03:43, 1 September 2005 (UTC)

And voilà! (The arrows look a little wonky, but it’s still good, I insist) — Felix the Cassowary 10:32, 1 September 2005 (UTC)

Nice graph. Don't worry about the wonkyness of the arrows: we've got to live with pixels. Is /i:/ really a diphthong though? It certainly has a significant on glide but I though diphthongs had to have two targets. I'll see if I can't dig up that article I read ... here it is. Cox found this long on glide but only one target. Also the Mac. Uni. site I've refered to lists /i:/ as a monophthong. Jimp 2Sep05

I don’t know whether an onset is enough to make it phonetically a diphthong, but this is suggestive:

Two vowels classified as monophthongs may exhibit diphthongal qualities. /i/ is typically diphthongal in character in that it has a variable onglide for most Australian speakers. /u/ can sometimes be realised with onglide but is more typically monophthongal.

Phonemically, I think it only makes sense. In many (but not all) contexts the only distinctive feature between /ɪ/ and /ɪə/ is length, so length has to be the thing that distinguishes them (in fact, in almost all the contexts where there's another feature that would distinguish them—e.g. word finally—/ɪ/ isn't permitted). So phonemically the length contrast is between /ɪ/ and /ɪə/. Note also from what you cited Short/long pairs of vowels, which are similar in phonetic quality, retain comparable onglide propotions and compare and contrast /iː/ with an onglide of almost 50 per cent length, and /ɪ/ and /ɪə/ with around 10 per cent each.

If /iː/ is another phonemic long vowel, then we need to create an extra category. Because the targets of /iː/, /ɪ/ and /ɪə/ are close enough to be identical, it'd need to create a category of long vowel + onset or long vowel - onset, but diphthong seems to be suitable.

Anyway, it could be best not to attempt to answer the question. This version doesn’t, but it’s a lot busier. If you like it better, you can add it to the article and I’ll make one that shows only the short vowels. The article you gave indicates that this is theoretically valid (e.g. The results of the duration data confirm the traditional analysis of Asutralian English vowels that there are essentially two length types: long, which includes diphthongs, and short. A three-way length distinction as proposed by Bernard (1967b) is not support by the present data.)

(In any case, I did include /i:/ on the other chart as a way to avoid answering the question.)

—Felix the Cassowary ɑe hɪː jɐ 04:20, 2 September 2005 (UTC)

The /ɜ:/ is more like /ɘ:/ in my mouth and in that vowel chart it's also appears closer.

No it's not, though it might be more like [ɘ:]. But that isn't the basis on which we alter the page; we're following established conventions, not making knew ones. Also, the exact pronunciation varies; for me it's a slightly retracted [øː] (which looks almost the same as a slightly advanced [ɘ:] when mapping only F1 and F2, as that chart is ... sounds almost the same, too). Let's follow the convention. —Felix the Cassowary 10:27, 21 May 2006 (UTC)

/ɪː/ and /ɪə/ are allophones and that should be noted.

It is, although without specifying the exact qualities (because [ɪɐ] is also used, and probably others). —Felix the Cassowary 12:50, 12 June 2006 (UTC)

[edit] Competing Orthographies

The table below is of various competing orthographies for Australian English vowel phonemes:

Competing Orthographies for Australian English Vowel Phonemes
Mitchell & Delbridge (1965)	Durie & Hajek (1994)	Ingram (1995)	Harrington, Cox & Evans (1997)	Examples
ɪ	ɪ	ɪ	ɪ	hid
i	iː	i	iː	heed, happy
ɛ	e	e	e	head
ɛə	eː	eə	eː	hair, yeah
æ	æ	æ	æ	hat, had
æ	æː	æ	æ	bad
ʌ	a	ɐ	ɐ	hut
a	aː	a	ɐː	hard, fast
ɒ	ɔ	ɔ	ɔ	hot
ɔ	oː	o	oː	horde, dawn
ʊ	ʊ	ʊ	ʊ	hood
u	ʉː	ʉ	ʉː	hoon
ə	ə	ə	ə	banana, centre
ɜ	ɜː	ɵ	ɜː	heard
eɪ	æɪ	æɪ	æɪ	hate
aɪ	aɪ	aɪ	ɑe	hide
aʊ	æʊ	æʉ	æɔ	howl
oʊ	əʉ	əʉ	əʉ	hope
ɔɪ	oɪ	oɪ	ɔɪ	hoist
ɪə	ɪː	ɪə	ɪə	beard, hear, idea
ʊə	ʊː	ʉə	ʊə	tour, pure

I have put this comparative table in this talk page to show some of the competing orthographies for Australian vowel phonemes that have been created by various Australian Linguists (see here for references). – AxSkov (☏) 16:45, 31 August 2005 (UTC)

M.-D. don't use lenght symbols, do they? At least, the Macq. doesn't, nor do modern publications that use them. — Felix the Cassowary 03:46, 1 September 2005 (UTC)

No, M-D doesn't use lengths that I know of, it was a mistake. It was an oversight, somehow I put those length symbols in the table when I was compiling it. – AxSkov (☏) 04:38, 1 September 2005 (UTC)

Good-good, I haven't read anything by them so I didn't want to edit it if I was wrong. — Felix the Cassowary 04:41, 1 September 2005 (UTC)

[edit] Feesh and Cheeps and Fush and Chups

I presently have this in the section of the kit vowel in Australian English including the common stereotypes. ; /ɪ/ : for example kit, bid, hid. (M.-D. /ɪ/.) The target for this vowel tends to be tenser than in other varieties of English. Because of this, some New Zealanders often claim that Australians say feesh and cheeps for fish and chips while some Australians conversely claim that New Zealanders say fush and chups.

Does this include unnecessary information? Cassowary thinks that it does. But those stereotypes are very real, and so should perhaps be mentioned somewhere in the article. 64.194.44.220 02:44, 19 December 2005 (UTC)

Should we include all stereotypes? (well, except for the fake ones, like the way Americans say "sprint" as "flicrk", that's a very common fake stereotype). Should we include that some Australians will stereotype Australian Wogs as doing the same? the fact that Americans will stereotype Australians as saying "buy" for "bay" or "boy" for "buy"? How about the "g'doy moyt" stereotype of Australian wogs, by non-wog Australians? Anyway, how is New Zealand English relevant to a discussion of Australian English? It's not, but in order to include that statement as relevant you'd need to explain why Aussies say Kiwis say "fush and chups" (by the way, the kit-bit split is the wrong place to link to, because there is no split in New Zealand English).

I'll agree that it might perhaps be relevant in a discussion of the differences between Australian and New Zealand English, but that's not what that is. That's a discussion of Australian English, without regard for others.

—Felix the Cassowary (ɑe hɪː jɐ) 03:01, 19 December 2005 (UTC)

[edit] Golf-gulf merger

The golf-gulf merger article is in danger of being deleted. 64.194.44.220 22:57, 19 December 2005 (UTC)

And, User:64.194.44.220, you can save it. Just find us a reliable source. Jimp 06:40, 21 December 2005 (UTC)

[edit] Kambah sounds like Canberra?

Hi, at present the article on the Canberra suburb of Kambah, Australian Capital Territory contains the assertion "Kambah sounds like Canberra." It doesn't really the way I hear most people say the two words, maybe my hearing is not good or I move in the wrong circles. They quite possibly both derive from Ngambri. Perhaps, as to pronounciation, a fellow-editor, more skilled in phonics than I am (not at all!) can help. To provide meaningful input, perhaps you would have needed to hear the words pronounced by Canberrans, but maybe not.--A Y Arktos 18:59, 12 January 2006 (UTC)

I expect you would. Can you make a recording of both words? I would guess however that "Kambah" was pronounced /kæːmbə/ (i.e. first syllable like "ram", second like "slumber") and "Canberra" is pronounced /kæːmbrə/, so if my guess is right the only difference is the missing "r" in "Kambah". I've never heard the word "Kambah", so I could be way off though. Maybe it rhymes with "Rumba"? —Felix the Cassowary 02:33, 13 January 2006 (UTC)

I don't have the means (or nous) to make recordings and upload them. In non-phonetic terms I believe Kambah is pronounced that the first syllable rhymes with cam as in 'cam shaft', the second syllable is bah as in 'bah humbug'. The first syllable of Canberra would rhyme with a 'can of drink'. Your IPA symbals show the pronounciation being similar but would these rhymes make a difference? Thanks --A Y Arktos 21:05, 23 January 2006 (UTC)

Ah well, I do think some people pronounce "Canberra" the way it's written (in IPA as /kæːnbərə/) but it always strikes me as odd and unnatural, like the way Americans say "Melbourne" as Mel-BORN instead of MEL-ben. Presumably the person who made the assertion says "Canberra" like I do, whereas you say it like it's written. Now, as for the second syllable of "Kambah", if you say it's pronounced as "bah humbug", then in the IPA that'd make it /kæːmbaː/, but the difference in pronunciation between /aː/ and /ə/ at the end of a word is very slight—mostly just the length of time you say it for. My guess—as someone who's only spent about an hour in Canberra, and most of it in a car—is that if you listen carefully you'll find some people pronounce it with the longer /aː/ and some with the shorter /ə/. Maybe the person who wrote it used the shorter sound. Or maybe the person who wrote it was just alluding to the fact that the long and short sounds sound similar. (PS: I hope the IPA symbols are showing properly for you. I think I've explained them well enough to understand them, but if I haven't I'll try again. It's hard explaning things you know like the back of your hand ;) —Felix the Cassowary | toːk 10:28, 24 January 2006 (UTC)

I live in canberra and we pronounce canberra as /kænbra/ and kambah as /kæːmba/

[edit] Wrong symbols part two

graph, chance, demand, dance, castle, grasp, to contrast.

This time the question is not really about the symbols used so much as the phoneme. For some of these words I use /aː/ but I don't think I ever use /æ/. I'm pretty sure it's always /æː/ for me whenever it's not /aː/. Then, on the other hand; whereas I say /graːf, graːsp, kaːsl̩/ & /kəntraːst/; /græːf, græːsp, kæːsl̩/ & /kəntræːst/ all sound definitely unAussie as opposed to /græf, græsp, kæsl̩/ & /kəntræst/. Is there any pattern as to which side of the BAD-LAD split these TRAP-BATH split wavering words fall? Jimp 16:22, 22 January 2006 (UTC)

If you're using a flat-a in these words, the division between long and short flat-a is exactly the same here as it is anywhere: short before fricatives, long before /n, m/ unless certain other criteria are met (which they never are for these words­). —Felix the Cassowary 03:48, 23 January 2006 (UTC)

Do Australians actually use the term "flat a"? In Britain, I usually hear "short a" for /a/ and "long a" for /ɑː/, though that wouldn't make as much sense in the Aussie context where what we call the "short" one has sometimes become long.--JHJ 08:53, 23 January 2006 (UTC)

/aː/ is usually called "broad a" here when it comes from the TRAP-BATH split, or just "the /aː/ sound" otherwise*, with "long a" meaning /æɪ/ (taken from the Middle English value). I have no idea where I picked up the term "flat a", but it's less common and only ever used in contrast to broad a. In contrast to long a, and normally, you get "short a" meaning both /æ/ and /æː/.

I think part of the reason for this is because the broad-a/flat-a split is finished and the differences are socially conditioned, whereas the long/short flat-a split still seems to be spreading to new environments (e.g. by the time English was exported to Australia (cf. the Philidelphia and Estuary English systems), it seems it hadn't yet spread to before /ɡ/ whereas it is now there with exceptions (either way) exclusively neologisms; in 1985^†, it hadn't yet spread to before /dʒ/ whereas by now (IME & unsourced) I'd say /bæːdʒ/ was the normal pronunciation of "badge").

• * Orthographic a represents so many sounds /a aː æ æː æɪ e eː ɔ oː ə/ that it's hardly worth having names for them all.
• † Blake, B. J., (1985). ‘Short a’ in Melbourne English. Journal of the International Phonetic Association, 15: 6–20.

—Felix the Cassowary 11:16, 23 January 2006 (UTC)

[edit] Celery/salary merger

Altered the reference from 'In Victoria..' to 'for some Victorian speakers...' since this merger is far from universal. 61.69.244.153 03:15, 12 March 2006 (UTC)Oscar (12th March 2006)

[edit] /ɐ:/ different before "r"

Australian English /ɐ:/ sounds more fronted to me (closer to /æ:/) in syllables ending with a (theoretical, unpronounced) "r". Any thoughts on this from native speakers of AuE? Richwales 05:47, 18 September 2006 (UTC)

Do you mean that the vowels of ‘farce’/‘farm’ and ‘pass’/‘palm’ in AusE sound different to you? They’re not. The ‘r’ is just one orthographic notation of the long vowel. For instance, when you form colloquial words by appending one of the affixes -ie/-y, -er/-a or -o after the first syllable (or some other opportune place), newly intervocallic fricatives may be voiced. Hence, ‘afternoon’ /aːftənʉːn/ becomes /aːvəʉ/, spelt ‘arvo’.

Do you also mean you’re not a native AusE speaker? Is your own accent rhotic? You could be hearing a difference you’re used to hearing as significant, when really, it’s just because the AusE vowel is central vs your own which might be back; or you might have misunderstood an allophone (I’ve heard it as a side-remark that in Melbourne, /a(:)/ is fronted after /k/).

—Felix the Cassowary 08:00, 18 September 2006 (UTC)

No, I'm not a native AusE speaker. I grew up in California, but then lived for several years in Ontario, Canada before moving back to California. My everyday speech is that of Ontario (I've been mistaken many times, by Canadians, for a native speaker) — but as a result of having become "bi-dialectal", I've become consciously aware of many speech distinctions that are allophones in one dialect or another.

I just finished watching the public memorial for Steve Irwin. The service opened with a speech by John Howard, in which he clearly exhibited the distinction I'm talking about here in words such as "remarkable", "hearts", and "arms" on the one hand, and "quality" on the other.

On the other hand, most other speakers at the Steve Irwin memorial — such as Anthony Field and Wes Mannion — did not make this distinction. I wonder if it might be a regional variation. I'm pretty sure that, when these vowel sounds are different, they are treated as allophones and the difference is not noticed by most Australians.

Richwales 03:03, 20 September 2006 (UTC)

In Britain, and I believe Australia too, the vowel in quality is different from that in father and calm, and the non-rhotic vowel written ar is the latter. E.g. in RP barmy and balmy are both /ˈbɑːmɪ/ but quality is /ˈkwɒlɪtɪ/. So you'd need a pair like balmy/barmy to check whether the ghost of the /r/ has any effect on the vowel quality.--JHJ 07:40, 20 September 2006 (UTC)

As JHJ says, the vowel in ‘quality’ is different from the vowel in ‘remarkable’. ‘Quality’ is /kwɔləti/ and has the same vowel as ‘cot’, ‘hot’ or ‘what’/‘watt’. (I don’t know how much the American balm/bomb, cot/caught and very/vary mergers have affected all of this, so I’ll explain it basically all: In a word spelt with a ‘w’, ‘wh’ or ‘qu’ followed by a short ‘a’ not followed by a velar consonant, the ‘a’ is pronounced as if it were a short ‘o’. So the A’s in was, what, watt, waffle, quantity, quality, Warrigul (a town) or warm, quarter, war are all treated as short o’s, and the words are pronounced as /wɔz wɔt wɔt wɔfəl, kwɔntəti kwɔləti, wɔrigəl/, /woːm, k(w)oːtə, woː/ respectively. On the other hand, wary, wafer, whale or wack, wag, wang are /weːri, wæɪfə, wæɪəl/, /wæk wæːg wæŋ/. The word ‘water’ is /woːtə/ but I have no idea why; forms of strong verbs like ‘swam’ /swæm/ are also excepted.)

I’m afraid if you didn’t hear a distinction in other speakers of AuE, it would be because you missed a distinction that was made.(*) I doubt any Australians would consider the sounds /ɔ/ and /aː/ to be alike; the difference of length (if nothing else) is very salient. Even /a/ and /ɔ/ (cut vs cot) aren’t all that alike; /æ/ and /a/ (cat vs cut) are nearer.

(*) To my (native AuE) ear and lips, the Australian English vowel /ɔ/ is probably less rounded than its symbol suggests. And this might be age-related, so John Howard’s /ɔ/ might be rounder than the younger speakers, making it easier for you to hear the difference. But that’s just a guess...

—Felix the Cassowary 08:48, 20 September 2006 (UTC)

I may be getting misled here by the North American father-bother and cot-caught mergers here, so OK, I'll concede that the vowel in "quality" is a different phoneme from the vowel in "arm". Let me ask you (and/or any other native AusE speaker) a different (but possibly related) question: Do you perceive the vowel in "arm" to be the same as, or different from, the initial portion of the diphthong in "life"? Richwales 16:45, 20 September 2006 (UTC)

Similar. Whether it’s the same or not depends on how broad the speaker’s accent is. Personally I’ve never been very sure there’s much point in using a different symbol (/a(:)/ or /ɐ(:)/ vs /ɑ(e)/, rather than something more like /a(:)/ vs /a(i)/), but that seems to be the lie of the land. —Felix the Cassowary 22:44, 20 September 2006 (UTC)

[edit] Pronunciation of "our", "ours"

I am aware that in AusEng, "our" is often pronounced as [æɔ], eg "our car" [æɔ kɐː]. But there are two things I'd really like to know:

1) Is this pronunciation used before vowels, eg "our aunt" [æɔ ɐːnt]?

2) Can the adjective "ours" be pronounced as [æɔz]? --The Lazar 03:35, 30 September 2006 (UTC)

Yes (well, I have [æɔwaːnt] in that context) and yes; these are usual. Some people do say [æɔəz] tho, particularly if ‘speaking properly’. If this is a perculiar feature of AusE and doesn’t constitute original research, then it should be mentioned, perhaps along with Australian-specific pronounciations like auction with a short vowel. The pronunciation [aː] and [aːz] like they use in England are unlikely to be heard here, but I have a friend from NZ at Uni who does precisely that. —Felix the Cassowary 08:11, 30 September 2006 (UTC)

I can second everything Felix said (except for the bit about having a NZ friend at Uni). Jimp 07:00, 29 January 2007 (UTC)

[edit] Pronounciation of ʉː and iː

Would there be any other studies conducted that include ʉː and iː? I believe that the NSW pronouciation is different, eg. "pewl" vs "pool" and "oils" vs "eels". - Diceman 13:10, 24 October 2006 (UTC)

There is some variation between those two pairs (I’m not sure what you mean by ‘oils’, I suppose the vowel of ‘fee’ followed be a the last syllable of ‘royals’). I don’t think I’ve seen any proper regional studies on the topic tho. My impression is that it’s a change that’s probably affecting the whole country, but has spread from Adelaide to Melbourne to Sydney over a period of at least a generation so younger speakers in Sydney might still say ‘pewl’ [pʉːəɫ] whereas older speakers in Adelaide probably normally say ‘pool’ [pʊːɫ].

The only study other than the ones I’ve cited in the article that I can remember reading that discuss the issue, was one about whether a given vowel + l combination was pronounced monosyllabically or disyllabically. It didn’t actually discuss the phonetics of the vowel, but I presume that the pronunciations of /ʉːl/ were monos. [ʊːɫ] (your ‘pool’) vs dis. [ʉːəɫ] (‘pewl’) and of /iːl/ were monos. [ɪːɫ] (‘eel’?) vs dis [iːəɫ] (where [iː] contains the onset standard for AusE /iː/; ‘oil’?). It was limited to Adelaidean English, and found that there the monosyllabic pronunciation of ‘pool’ was all but mandatory for the speakers involved, but there was some amount of variation for ‘eel’.

—Felix the Cassowary 23:55, 24 October 2006 (UTC)

The classic (exaggerated) example would be the Kath and Kim pronounciation of "Look at me". The other times I've heard it are on Toyota World Sport (some Rugby commentator) and the Patrick Stevedore dockworkers (their pronounciation of "defeated").

Doesn't sound like there's enough to mention these two additional vowles the article yet. - Diceman 06:24, 28 October 2006 (UTC)

[edit] Madder

Here's a potential minimal pair: madder and madder. The only problem is that I don't think I've heard the latter ever pronounced so I'm only guessing that it would have /æ/ as opposed to the former's /æː/. Jimp 01:32 (UTC) P.S. Though it be only a guess as I've mentioned this is perhaps interesting in itself: why would I guess so? Indeed if I read the article on the plant using the pronunciation of the surperlative in my head, it does sound off, so what's the go here? Jimp 01:37, 29 March 2007 (UTC)

Two sides of the same coin:

That pair alone isn’t enough to prove that /æː/ is a phoneme in Australian English; the syllable boundary could be after the /d/ in “more mad” (keeping with the morpheme boundary) but before it in the plant’s name (for a basic CV.CV word). Better evidence for the difference is pairs like ran~Wran or mad~lad. Even ran~Wran could be described with “ran” being faithful to the length of the base form, “run”, but “Wran” following the regular course.

As for the way you’re reading it; that’s simple: lengthening doesn’t (never) happens if the consonant is followed by a vowel or approximant, unless a morpheme boundary intervenes; the contrast is restricted to closed syllables. You’re treating the plant’s name as monomorphemic, so lengthening would be anomalous.

—Felix the Cassowary 12:56, 29 March 2007 (UTC)

That's some interesting stuff there Felix. However, wouldn't /mæ.də/ break the rule (with its short stressed vowel in the open syllable)? Jimp 07:10, 2 April 2007 (UTC)

No; that’s precisely what you expect. English /æ/ lengthening is not Swedish vowel lengethening. The plant is /mæ.də/; the superlative is in some (possibly historical) way /mæd.ə/ [mæːd.ə]. Today, it might be the case that the superlative is /mæːd-ə/ [mæː.də] where the length is retained from the underlying specification, rather than added.

Also, don’t forget that /d/ doesn’t/didn’t generally cause lengthening; it’s just the case that four words ending in -d happen to have long vowels: mad, glad, sad and bad. All other words ending in -ad have a short /æd/: had, fad,[*] lad, dad, pad, ... and I also read neologisms like spad, blad with short vowels.

[*] I’ve always pronounced the “politically correct” name for a certain brand of fake cigarette lollies I liked as a kid as [fæːdz], presumably because the old name for them was [fæːgz]. But a passing superficial fashion is [fæd].

—Felix the Cassowary 15:12, 2 April 2007 (UTC)