2

On English allophones on Wikipedia, there is an example of the pronunciation differences between "night rate" and "nitrate",

Night rate: unreleased [ˈnʌɪt̚.ɹʷeɪt̚] (without a word space between [ . ] and [ɹ])

Nitrate: aspirated [ˈnaɪ.tʰɹ̥eɪt̚] or retracted [ˈnaɪ.t̠ɹ̠̊˔ʷeɪt̚]

What exactly are the differences? I don't understand the diacritics and the subtle differences between the similar consonants.

Qian
  • 121
  • Wait, why the [ʌɪ] in "night rate"? I'd assume this was related to Canadian raising, but that would happen in both versions. – alphabet Jun 16 '23 at 02:49
  • I'll leave this one for someone else who can explain what's going on with the /tr/. – alphabet Jun 16 '23 at 03:00
  • 1
    Speaking as someone from the US Midwest, "night rate" does have a bit of a separation between the words, while "nitrate" pretty much runs the syllables together. The difference is not dramatic, but someone from the Midwest would generally be able to hear the difference. – Hot Licks Jun 16 '23 at 03:13
  • @alphabet It actually makes sense to me that the first one is raised but the second is not. BTW, this is one of those "chrain" and "chree" cases, but this time with "chrate" for the second but not the first. Somewhere we have a question about traitor or trader that talks about this affrication. See here for how trade becomes something like [t͡ʂɻʷeɪd]. Also see these answers. – tchrist Jun 16 '23 at 03:43
  • @HotLicks people don’t have gaps between the individual words when they speak. That’s your mind playing tricks on you! – Araucaria - Him Jun 16 '23 at 08:14
  • That is one of the worst Wikipedia phonetics pages I’ve ever seen. Full of basic errors, contradictions and omissions :-( – Araucaria - Him Jun 16 '23 at 08:24
  • 3
    @Araucaria-Nothereanymore. - It's subtle, but it's there. Perhaps a better way to express it would be to say that with "night rate" one perceives the "T" as ending the word "night", while with "nitrate" the first syllable is "nigh" while the second is "trait". – Hot Licks Jun 16 '23 at 12:24
  • @HotLicks Yes, and also, because of that, you might hear a glottal stop used for the /t/ in night rate as well. This could easily give the impression off a hiatus. (There'd be a hiatus in the sound in the middle of a regular [t] too, but your brain would rightly tell you that that was part of the [t]!) – Araucaria - Him Jun 16 '23 at 12:27
  • The Wikipedia page on IPA explains all the symbols: here's the section on diacritics. – Stuart F Jun 16 '23 at 13:24
  • @StuartF No point using that for that wikipedia page. for example, the transcription of night rate with an aspirated /t/ is phonetically impossible. They've also used the labialisation diacritic on the diphthong instead of the [r] in the last transcription. And the [t] there is going to be retracted in all the examples. That page is a cluster-bundle of errors. – Araucaria - Him Jun 16 '23 at 14:53
  • @Araucaria-Nothereanymore. Is the article right that the /t/ in "night rate" is never retracted, and that there's a difference in the voicing of /r/? – alphabet Jun 16 '23 at 19:04
  • @alphabet I don't think it quite says that (that it's always true - although it does imply it). But to answer that indirectly, night rate will have a retracted [t] if the /t/ is realised as a [t] and not as a glottal stop. And in that case, the [ɹ] may indeed be partially or fully devoiced. [And the two together may still form an affricate, albeit a less forceful one than in nitrate.] – Araucaria - Him Jun 16 '23 at 23:49
  • @alphabet Having said that, a glottal stop (as opposed to the [t] with no audible release), will be extremely common here. – Araucaria - Him Jun 17 '23 at 09:28
  • @Araucaria-Nothereanymore. I think their source might be here. That paper claims: "For nitrate, the first /t/ is aspirated, released, and retroflexed, whereas the /r/ is devoiced, suggesting that it is part of a cluster. By comparison, the first /t/ in night rate is unaspirated and unreleased, suggesting that it is syllable final, whereas the following /r/ is voiced, suggesting that it is syllable initial." – alphabet Jun 17 '23 at 17:47
  • They claim to have acoustic measurements supporting the existence of an aspirated /t/ in "nitrates," namely a longer "burst" in the /t/ sound, but I suspect that they might be wrong on this point; they may have actually been measuring the /t͡ʂ/ many speakers have. – alphabet Jun 17 '23 at 17:53
  • Though that paper cites a number of others with the same example, so maybe this error goes back even further. – alphabet Jun 17 '23 at 17:56
  • @alphabet Hmm. Hope not! That paper is by psychophysicists, not linguists. And the people they cite are ... themselves! And the journal is second of the only two psycholinguistics journals in existence and towards the very bottom end of the 117 experimental psychology journals. Yes, the article has a lot of citations, but that's because the lead author cites themselves over 64 times in that one paper, and presumably has self-cited along with the other authors continuously for the last quarter of a century since that was written. – Araucaria - Him Jun 17 '23 at 23:09
  • @alphabet From a linguistics point of view the phonetic analysis is bizarre. But perhaps that's best discussed in chat instead of having a lengthy exposition of the idiocies of Wiki and the authors of that paper under this fine and well-intended question. – Araucaria - Him Jun 17 '23 at 23:24
  • Let us continue this discussion in chat. – alphabet Jun 17 '23 at 23:34
  • 'Nigh-trate is the chemical : 'nigh' as in 'drew nigh to'. 'Night rate' is the payment : 'night' as in 'sight'. – Nigel J Jun 18 '23 at 22:18
  • Night rate, nigh trate; at all, a tall. It's a late T. – HippoSawrUs Jun 22 '23 at 01:05
  • This question got partially answered in the chat with @Araucaria-Nothereanymore. See here. Part of the issue is that this Wiki edit screwed the page up. I haven't had a chance to write up the details. – alphabet Jul 10 '23 at 21:42

1 Answers1

-1

To the extent this is about phonetics, I think it's not about English Language & Usage; that would deny dialects, if not idioms.

Ignoring any technicalities from phonetics, the active point is the space in 'night rate', which is pronounced.

To make that sound like 'nitrate' would mean eliding the sound, in pretty-much the same way that 'nitrate' elides the spelling.

Robbie Goodwin
  • 2,564
  • Nobody really "pronounces" spaces in normal speech, though your brain may trick you into thinking that you do; that said, the word boundary can creates other phonetic changes that are pronounced. – alphabet Jul 10 '23 at 21:40
  • Alphabet, I'm sorry to be the one to disillusion you and if you see any useful, or even technical difference between 'pronouncing' and 'leaving a pause for' something, please explain it. If the pause is not pronounced how, in ordinary English, does it exist? – Robbie Goodwin Jul 10 '23 at 21:48
  • There are no actual pauses between words in typical speech. It just sounds like there are. If you listen to (say) a video of ordinary speech at a sufficiently slow speed, you'll be able to hear this. – alphabet Jul 10 '23 at 21:50
  • The "pause" you're hearing is typically a sort of illusion. In this case, the word boundary might be realized as the glottalizing (or glottal reinforcement) of the first /t/. But it's not an actual delay between sounds. – alphabet Jul 10 '23 at 21:51
  • Sorry and no, the pause is no kind of illusion. It clearly is an actual delay between sounds. – Robbie Goodwin Jul 10 '23 at 21:55
  • Nope. There is no gap in the articulation between words in everyday speech. Again, listen to (say) videos of people talking at a slow playback speed on YouTube. Pauses usually occur on word boundaries, but the vast majority of word boundaries are not marked by actual delays between sounds. – alphabet Jul 10 '23 at 22:04
  • Just like how the /k/ sound in keep is different from the one in cool, even though they sound the same to English speakers. Your brain just makes them seem equivalent. – alphabet Jul 10 '23 at 22:06
  • Alphabet, if you really believe 'night rate' and 'nitrate' have identical sounds, why not say so?

    Either way, where did you get the idea that 'cool' or 'keep' had any bearing on the differences between 'night rate' and 'nitrate'.

    In more than 60 years of listening I've never noticed anyone unable to hear what I guess you must mean by the 'gap in the articulation between words in everyday speech' and if you yourself can't hear it, why can you not explain how 'can't hear' doesn't sound the same as 'caantear'?

     – Robbie Goodwin Jul 12 '23 at 22:30
  • 1
    They do not have identical sounds--not even close! But the difference is not an actual pause, it's a set of phonetic phenomena that occur on word boundaries and would appear in a narrow phonetic transcription. – alphabet Jul 12 '23 at 22:32
  • 1
    These include devoicing, glottalization, changes in vowel length, prefortis clipping, etc. For example, the vowel in "night" is significantly shorter than the vowel in the first syllable of "nitrate," and the first /t/ is usually pronounced quite differently. This is what lets you hear word boundaries when there's no actual pause. – alphabet Jul 12 '23 at 22:35
  • Thanks and while I do think this is going too far, still the simple fact is that 'word boundaries' are different from whatever you'd like to term the boundaries within words, and one difference is pauses – Robbie Goodwin Jul 14 '23 at 19:52