An aim or a name?
« previous post | next post »
In a meeting the other day I heard a colleague say something that was either the first of these or the second:
A good test of whether a course is coherent in its content is whether we can give it an aim.
A good test of whether a course is coherent in its content is whether we can give it a name.
Either would make sense: it is surely reasonable to say that a course is coherent if and only if you can say what its goals are. To say that you want to give students an understanding of basic articulatory and acoustic phonetics seems like a goal, whereas to say you thought you might just expose them to a miscellaneous ragbag of bits and pieces about language, society, culture, and brain anatomy seems aimless. And it also seems reasonable to say that a course is coherent if and only if you can name it in a concise, appropriate, and memorable way: "Basic Phonetics" is a good name for a course, while "A miscellaneous ragbag of bits and pieces about language, society, culture, and brain anatomy" is surely not.
Phoneticians have long used the pronunciation contrast between an aim and a name to illustrate the puzzling phenomenon known as juncture. The sequence of segments involved in the two phrases would appear to be exactly the same: first the unstressed vowel known as schwa, then the alveolar nasal [n], then the diphthong heard in the word day, and finally the bilabial nasal [m]. So how come you can ever hear the difference (which it seems intuitively that you sometimes can)? Are word boundaries or syllable boundaries audible, in addition to segments? And if so, why is the difference fragile, so that in relaxed or rapid speech it can disappear? The puzzle is an interesting one, involving timing, and subtle matters of consonantal articulation and syllabification, and it has no clear and simple solution.
It was interesting, after encountering an aim and a name cited so often in elementary linguistics courses and texbooks, to be in a situation where an actual utterance demonstrated their near-homophony, and given the style of speech used, the nonlinguistic context did not suffice to resolve the ambiguity.
(It turned out that my colleague agreed that a course should have an aim, but had been intending to say that it should also be possible to give it a suitable name: it was a bad sign if given the content no suitable name suggested itself.)
Dan Lufkin said,
May 7, 2010 @ 7:41 am
Isn't the an/aim boundary audible as a glottal stop? Well, it wouldn't be audible at a distance, I guess.
[No, it certainly isn't; there's no reliable glottal stop occurrence in English, and certainly not after a consonant like [n].—GKP]
Carl said,
May 7, 2010 @ 7:48 am
This might come down to one's dialect/accent. I feel like I would pronounce "a name" as "uh name" vs. "an aim" as "æn aim", and I have a semi-standard American accent.
[I very much doubt this. The indefinite article before a vowel is sometimes pronounced as [æn], but not on any regular basis. You may think you always use it, but I'd bet you don't. Only recordings of your everyday speech when you didn't know you were being recorded could confirm my suspicions. If your phone is being tapped, that might be useful… if we could get hold of the recordings from the FBI.—GKP]
Ian said,
May 7, 2010 @ 8:04 am
There are glottal stops and vowel distinctions when pronouncing the words in isolation, but I doubt many people will retain those features when speaking normally.
D. Sky Onosson said,
May 7, 2010 @ 8:19 am
I would suspect there might be some slight differences in the duration of the two vowels involved and the /n/, for each phrase. My hunch is that "an aim" would have a longer /n/ and possibly /ei/, while "a name" might have a slightly longer schwa (or whichever vowel) before the /n/ – but we'd need to record and measure them to be sure!
fs said,
May 7, 2010 @ 8:21 am
I'm with Carl here. Pacific NW American accent. [And I bet you're both wrong.—GKP]
Mark P said,
May 7, 2010 @ 8:43 am
I doubt that most people actually say the two differently unless they are trying to distinguish between them.
language hat said,
May 7, 2010 @ 8:52 am
I agree with Mark P, but I can't help but feel the example is a little forced. It takes a considerable stretch of the imagination to find "whether we can give it an aim" a likely ending for that sentence — it just doesn't seem like a normal English utterance to me. (That's leaving aside the fact that the idea that being able to give something a name is a good index of whether it is coherent is a hoary one.)
[The example is not forced; it occurred in unreflecting natural speech in a faculty meeting. And the colleague in question had to stop, and say the sentence again, and clarify. One reason you may think that giving something an aim doesn't seem plausible is that you don't have experience of the bureaucratization of British academia, which involves a quasi-governmental agency overseeing a process of making sure all courses in all universities have defined "learning outcomes" — goals with respect to what the students should gain from the course — and procedures for assessing whether the goals have been met.—GKP]
Spell Me Jeff said,
May 7, 2010 @ 9:11 am
The example might be forced, but history offers examples where the ambiguity clearly existed in the minds of speakers, in some cases actually changing the language. The classic instance is the development of the word nickname from ekename:
"an ekename" => "a nekename"
I'm surprised GP did not mention this, as the analogy is particularly apt. I think.
[Can't mention everything, SpellMe.—GKP]
Rodger C said,
May 7, 2010 @ 9:15 am
Perhaps we can give it an ewt or a nadder the color of a norange.
Ralph Hickok said,
May 7, 2010 @ 9:29 am
I also agree with Carl. I grew up in Wisconsin but I've lived in Massachusetts since my college years.
[Shut up, all of you; stop saying you agree with Carl.—GKP]
Franz Bebop said,
May 7, 2010 @ 9:45 am
Isn't this similar to polysemy? The same sound representing two distinct phrases with two distinct meanings?
There may be no phonetic difference between the two at all, and yet, the language is not in danger of falling into anarchy.
[We usually speak of polysemy when a given lexical item, with just one entry in the dictionary, has two related senses. Like cotton meaning (i) a certain plant and (ii) the fiber produced from that plant and (iii) the cloth produced from that fiber. The case I'm looking at is two distinct sequences of words that just happen to sound very similar. Like raise meat and rays meet.—GKP]
Army1987 said,
May 7, 2010 @ 10:03 am
And it also seems reasonable to say that a course is coherent if and only if you can name it in a concise, appropriate, and memorable way: "Basic Phonetics" is a good name for a course, while "A miscellaneous ragbag of bits and pieces about language, society, culture, and brain anatomy" is surely not.
I did have a course which was a miscellaneous ragbag of bits and pieces about special relativity, quantum mechanics, and atomic and molecular physics, and it was called "Introduction to modern physics". By the same token, you could name a course like that one "Introduction to cognitive sciences".
John Wells in Syllabification and allophony gives a longish discussion of the phonetic effects of syllable boundaries.
And "another" is quite definitely "a nother" rather than "an other" (my dictionary agrees by placing the stress mark before the /n/), and at 00:59 here you can also hear the strong form "eI n^ð@"
fs said,
May 7, 2010 @ 10:04 am
Well, not really. Polysemy is, as I understand it, the same word (or phrase) representing two distinct meanings. Here we handily recognize "a name" and "an aim" as two distinct phrases.
John Lawler said,
May 7, 2010 @ 10:06 am
Hoard, James. E. `Juncture and syllable structure', Phonetica 15, 1966, 96-109 contains a namusing little story that was the stimulus for the experiment. It contains many pairs like a nice man, an ice man, and even the triplet night rate, nitrate, Nye Trait; the subjects read it aloud.
Bob said,
May 7, 2010 @ 10:21 am
I was talking about this phenomenon with someone recently with regard to "a norange" becoming "an orange". We were wondering, are there particular phonetic characteristics that make a word vulnerable to this kind of re-interpretation? Obviously it must begin with an 'n', but there are plenty of words that start with 'n' and relatively few examples of this kind of word change.
Andrew (not the same one) said,
May 7, 2010 @ 10:27 am
Language Hat: I think 'give it an aim' would actually be quite plausible in the kind of academic English that Geoff is referring to. It means 'think up something that can be written in the box on the form headed 'Aim of this course'.
[Spot on, AndrewNotTheSame. If the form has a box labeled "Aim of the course", then you are required to give it an aim and write it in that box.—GKP]
Janne said,
May 7, 2010 @ 10:28 am
I would say that context would cleanly separate the two possible interpretations in most cases, and where context is of no help is precisely where you end up with this ambiguity.
Japanese has a lot of homophones – much more than English or other indoeuropean languages – and makes heavy use of ellipsis and yet there is little danger of speech being misunderstood. Context normally gives us all the clues we need to make a correct interpretation of any utterance.
Philip Spaelti said,
May 7, 2010 @ 11:06 am
Re: a(n)adder, a(n)ewt, a(n)apron, etc. [The loss of the n in orange certainly predates borrowing into English, so that case is different] and the characteristics such words must have. I don't think the characteristics are phonetic (apart from the [n]). The word must not be too common or familiar, and be predominately used in the indefinite singular. Adder and Newt are good examples because they commonly appear in phrases like "I saw a (n) adder", etc.
Skullturf Q. Beavispants said,
May 7, 2010 @ 11:15 am
When I was an undergraduate, I overheard a snippet of another student's conversation where they were either saying "I don't care about marks anymore" or "I don't care about Marx anymore".
My situation is a little different because I don't think anyone would claim there could be an audible difference.
George said,
May 7, 2010 @ 11:18 am
I think that, in rapid speech, for me the "an" in "an aim" is almost a syllabic "n", with no vowel before it, where the "a" in "a name" is definitely a schwa-ish vowel. I grew up in Detroit, Michigan.
Adam said,
May 7, 2010 @ 11:38 am
I was thinking about the changes of nadder, ewt, napron, norange, & ekename, and I wonder whether such changes would have been inhibited by collisions with existing words?
Apart from "person/thing that adds", which I suspect might have been more rarely discussed than the snake in the old days, I doubt there can't think of any other words very similar to the meta-analysed forms.
Mr Fnortner said,
May 7, 2010 @ 11:52 am
Is that a proscription against a glottal stop, GKP? If so, what then is the decided clenching of the throat, voiceless moment, and the subsequent expulsion of air that occurs in "an aim" between "n" and "a", I ask humbly? (Granted, all this happens in a flash, even to the point of vanishing in rapid or cavalier speech.)
Matthew Stuckwisch said,
May 7, 2010 @ 12:05 pm
I'm not with Carl, I'm actually with George, but I recognize we're probably in the minority of English speakers. For me, in unstressed situations like most, "a" reduces to schwa, but "an" reduces to a syllabic n. SAE (we like our syllabic consonants down here)
an aim = ṇ.eɪm
a name = ə.neɪm
Ellen K. said,
May 7, 2010 @ 12:08 pm
When I say them in my head, it's definitely "uh name" vs."æn aim". When I say them out loud, the difference disappears. I did have to say them out loud to hear that they really do sound alike.
notrequired said,
May 7, 2010 @ 12:15 pm
I think there's no glottal stop there, not even for a flash, unless you articulate the words really carefully so that the boundaries are clear, which isn't what is normally done in rapid, everyday conversation.
I'm with Geoffrey.
Mel Nicholson said,
May 7, 2010 @ 12:23 pm
I recall a study on distinguishing "ladder" from "latter" by one of John Ohala's grad students at Berkeley circa 1989. The indicators seemed to all be on the preceding vowel because swap-splicing the flap itself made no difference to perception, but playing with vowel duration did. It was around a 4:5 duration difference.
I'd be inclined to examine what goes on before the nasal rather than after it in this case also. Reliably marking the transition from vowel to nasal wouldn't be easy, but I'm guessing that transition is earlier in "an aim" than in "a name" with a similar total duration for the pair.
richard said,
May 7, 2010 @ 12:44 pm
I can't get over Geoffrey's description of the sentence occurring "in unreflecting natural speech in a faculty meeting." Unreflecting speech in a faculty meeting is a given, in my experience; I'm not so sure about the "natural" part….
FWIW, I too hear a difference in my mind between "an aim" and "a name," but not when I say it aloud, unless I try really hard.
(Yes, I know that's not what you meant by "natural").
Paul said,
May 7, 2010 @ 12:57 pm
This reminds me of the case of Поручик Киже (Lieutenant Kijé), where the phrase Подпоручики же (podporuchiki zhe, "the lieutenants"*) was mistaken for Подпоручик Киже (podporuchik Kijé, "Lieutenant Kijé").
* же can mean a number of things, depending on the context. I'm not familiar enough with the context to know what it means in this case.
Eric S said,
May 7, 2010 @ 12:58 pm
> Shut up, all of you; stop saying you agree with Carl.
I'm still laughing at this after 5 minutes.
Seriously, from now on I am purposefully replacing all my schwas with ashes in all environments. If I don't go completely insane I will expect GKP to apologize.
Bloix said,
May 7, 2010 @ 1:08 pm
Re juncture: My father used to say "oran juice." He ran the two "j" sounds together. He would look in the refrigerator and ask, "Are we out of oran juice?" (Of course we weren't. My mother would get up and move the milk aside for him.)
jammy said,
May 7, 2010 @ 1:57 pm
I always like the concept of misinterpreting word like this. I'm sure a lot of
people have heard some variant of this joke:
'A string walks into a bar and asks the bartender for a Singapore Sling. The bartender regards him with a scornful eye and says, "We don't serve strings here!" Downtrodden but resilient, the string leaves the bar with a theory of how to get back in. He messes up his hair, ties himself up and walks right back in. The string orders another Singapore Sling. The bartender leers at him and says, "Aren't you that string that was in here just a second ago?"
"Nope," the string says, "I'm afraid not." '
Instead of course, "a frayed knot".
unekdoud said,
May 7, 2010 @ 2:32 pm
For me, it feels like I pronounce the "day" dipthong further in front in "a name" then "an aim". (But I'm not an experienced listener, so maybe that's just an illusion.) The other possibility is that I'm stressing the "aim" just a bit harder than the "an", and the "a" a bit harder than the "name".
Stephen Nicholson said,
May 7, 2010 @ 2:35 pm
1) I agree with your colleague.
2) On a related note, I've been in a few college classes where the name didn't seem to match-up with what the professor was teaching.
3) If's Carl's phone is being tapped, he might be able to get copies via the freedom of information act. But unless the case is closed, I wouldn't wager even a penny on it being successful. And even then, only a dime.
J. Goard said,
May 7, 2010 @ 3:46 pm
@George and Matthew:
I study English articles in L2, and also teach English to Koreans, which has made me very sensitive to my pronunciations, and caused me to test them countless times. One of the big perceptual problems for L2 learners is that "a" before any sonorant is often realized as syllabification. While I agree with you that presence of a schwa in fast speech is usually only compatible with a name, the syllabic [n] is fully compatible with either (and probably my typical realization).
(I also use n-less a pretty often before vowels too, BTW, especially diphthongs (get a ice cream) where the [n] even feels awkward.
D. Sky Onosson said,
May 7, 2010 @ 4:45 pm
@ J. Goard
I do the same with the n-less a. For me, a apple is pretty natural.
Rubrick said,
May 7, 2010 @ 5:06 pm
The puzzle is an interesting one, involving timing, and subtle matters of consonantal articulation and syllabification
This I can well believe. If imagine suitable contexts and say the two phrases, I'm convinced that I can feel something different, but it's very hard to describe what that difference is. If I were to try, I'd say that for "a name" I push straight through with a single breath, including the "n", but for "an aim" there's a tiny interruption in exhalation during the "n", so that when the tongue releases from the palate at the end there's less pressure built up; the "a" is "starting from scratch", as it were.
Even if this self-impression is accurate, of course, I have no idea how this difference shows up acoustically.
The obvious question is whether a suitably trained neural net can reliably tell the difference.
dazeystarr said,
May 7, 2010 @ 6:18 pm
I'm reminded of an episode of Family Ties in which Alex, uncharacteristically, receives a failing grade on a paper and is trying to tell his parents. Struggling to articulate the despised letter, he stretches out the "n" in "an": "I got an…annnnnn…nnnF." Both parents respond in unison, "A neff?" Cue laughs from studio audience.
I'm trying to work out why the bit worked–why it was clear that they were saying "a neff" versus "an F". As Prof. Pullum notes, there's no clear or simple solution.
As far as "an aim" versus "a name", I swear the vowel's different for me. In "a name" it's definitely a schwa; in "an aim" it's /æ/ in careful speech and appears to be /ε/ in rapid speech.
Hugo Quené said,
May 7, 2010 @ 6:29 pm
Once upon a time, decades ago, I investigated the timing and acoustic cues involved in juncture, and their perceptual relevance, in Dutch. It's still an interesting puzzle indeed.
Bob Ladd said,
May 7, 2010 @ 6:41 pm
The commenters who are convinced that they pronounce these differently are probably right, but so are those who are convinced that it's nevertheless difficult or impossible to tell the difference. In a paper published in Journal of Phonetics in 2003, Astrid Schepman and I created dozens of pairs of names like Joe Neeson – Joan Eason which we got people to read in neutral contexts (i.e. the speakers were not deliberately trying to signal the contrast between the two members of the pair). The biggest acoustic difference we found was also the most important difference identified by Ilse Lehiste in her 1960 monograph on "juncture", namely that word-final consonants are shorter, on average, than word-initial consonants (in our data, 47 msec vs. 67 msec in the case of /n/). (We were actually interested in other differences, not relevant here.) There were glottal stops (at the beginning of e.g. Eason) in only 10-15% of cases. In a subsequent perception experiment, we found that people, given a forced choice, could correctly identify the name intended by the speaker only about two-thirds of the time.
Scriptor Ignotior said,
May 7, 2010 @ 9:37 pm
For me, a apple is pretty natural.
I have noticed that many Americans accept a apple if there is a hesitation after a. Elsewhere speakers tend to "correct" the article (I'll have a … an apple), even though the n is not needed for juncture in a … apple.
For me, this thread is an inventory of the hazards of what Rubrick calls "self-impression", above. In my experience, people are usually sure that they say tacked and tact differently. And it is striking how speakers confined to the standard phonemic repertoire of English do not "hear" aspiration, and cannot say what it is about a wrongly aspirated labial in asparagus, or a non-standardly unaspirated labial in parent. They just sound suspect, like mixing up Hilbert space and Hilbert's pace.
Ellen K. said,
May 7, 2010 @ 10:57 pm
Dazeystarr, seems to me in that context, if the parents were saying "an F" they would pronounce "an" with the /æ/ vowel, and there would be a pause between the words, so it would be easy to tell "a neff" from "an F". Which is different than using one of these sort of word pairs in the middle of a longer utterance.
blahedo said,
May 7, 2010 @ 11:33 pm
I am *able* to pronounce "a name/an aim" differently, in roughly the way others here have reported (schwa in the former, [ε] in the latter), but I'm also sure that in normal running speech, they're not easily distinguishable.
However, it does suggest a solution to the "a neff" dilemma—if you want the audience to clearly hear that (rather than "an F"), push the vowel schwaward. It doesn't have to be very far to be clear.
D. Sky Onosson said,
May 8, 2010 @ 3:54 am
@ Scriptor Ignotior
I can definitely produce a apple without any substantial interval between the two words, and have certainly noticed myself doing so on occasion (I'm Canadian btw, though I wouldn't doubt this occurs in the U.S. as well). In fact, I'm not even sure there needs to be a glottal stop between the vowels.
I like your point about "self-impression", though; we need real data, not what everyone thinks they hear.
Simon Cauchi said,
May 8, 2010 @ 6:06 am
GKP: "there's no reliable glottal stop occurrence in English, and certainly not after a consonant like [n]"
I'm not sure what you mean by "reliable", but I can think of one speaker of the Cockney variety of BrE who instructed a class I once attended (in NZ) on bidding in Contract Bridge. One of her favourite, much-repeated expressions was "If you've got it, flaunt it", which she pronounced as "If you've go' i', flaun' i'". Those last four apostrophes are meant to represent audible glottal stops. If I remember rightly, the n of "flaun'" was heard as a nasal vowel, not as a consonant at all: a nasalized open o.
Daniel Klein said,
May 8, 2010 @ 6:14 am
I've always heard this sort of thing described as "form masking" (for masking, form asking)
language hat said,
May 8, 2010 @ 9:16 am
Ah, well then I withdraw my remark. I am, happily, unfamiliar with such boxes and the accompanying gobbledygook.
No they wouldn't. Did you learn nothing from your saying-it-out-loud experiment farther up the thread?
Ellen K. said,
May 8, 2010 @ 12:00 pm
Yes, Language Hat, I did. That it's good to say things out loud. And when I say "An F" and "A Neff" out loud, as I would in the context of the parents who were responding back, they are very distinctly different. Remember, don't just say the words. Act them. Be those parents asking a surprised question. For me, when I say it, it's as I said and you quoted, the /æ/ vowel in "an F", and, more importantly (since that's the one they didn't say), as I also said and you didn't choose to quote, a pause between the words. A pause that doesn't happen in a sentence like the one that started this whole thread, and that doesn't happen when just saying the words plainly, but that does happen when asking them as a surprised 2 word question. (Actually in "a neff" is more a lengthening of the uh and the n sound, with a quieting, rather that a full stop.)
Lou Hevly said,
May 8, 2010 @ 2:47 pm
I can make "an F" and "a neff" sound different by making the "uh" sound of the latter last longer before pronouncing the "n":
"a neff" = uh.nef
"an F" = uhn.eff.
However, I have to do it on purpose, which is perhaps what the actors did to get their laugh.
I think a similar thing happens with Americans who pronounce their intervocalic "t"s as "d"s. When we want to say something like "the injured sheep was bleating", we mentally "hear" that our natural "d" pronunciation would sound wrong ("bleeding") and consciously pronounce a "t".
ACW said,
May 8, 2010 @ 8:16 pm
I'm commenting here late, but nobody seems to have brought up the following theory. I'm not sure I believe it, but I'm going to state it as if I did.
1. There is no difference in relaxed, informal pronunciation between "an aim" and "a name".
2. In most situations, all ambiguity is resolved by context.
3. In the rare actually ambiguous cases, the listener settles quickly on one interpretation, and an inhibitory process suppresses competing readings in a process similar to the one involved in Necker cube perception.
4. This interpretation selection is so definite that there is a powerful sense that evidence for it was present in the stimulus. Other ambiguity experiments (more easily controlled, to ensure that the stimulus really is ambiguous) often elicit confabulated "explanations" of what triggered the selection.
5. The illusory distinction in the stimulus transfers to a conviction that one makes a corresponding distinction in speech. These convictions can be completely convincing, and can't be tested without copious candid recordings. They can't be tested by speakers, who will adjust their own practice to correspond to expectations. All these purported glottal stops, syllabifications, and vowel adjustments are exactly this sort of confabulation.
ACW said,
May 8, 2010 @ 8:18 pm
Sorry to double-comment, but I neglected to add that I constantly run into people who insist that they distinguish absolute homophones in pronunciation, with a similar confabulation based on spelling. I had a long argument once with someone who claimed to pronounce "fair" and "fare" differently.
Scriptor Ignotior said,
May 8, 2010 @ 8:47 pm
Onosson:
To be clear: I was making a related point, not reinterpreting or misinterpreting your own.
ACW:
I tacked that way with my point about tact. Spellings, morphologies, syntactic and semantic roles – the more these differ the less likely speakers are to discern homophones.
Sparky said,
May 9, 2010 @ 2:57 am
I would assume that some kind of soundwave analysis of recordings could show whether there is, in fact, a difference.
But I'm going to predict that there is none. We rely on context to determine which is intended. And if the context is ambiguous, the meaning remains ambiguous.
Incidentally, I don't believe "an apple" sounds any different from the first three syllables of "Annapolis."
Onosson said,
May 9, 2010 @ 3:33 am
@ Scriptir Ignotior:
Not to worry – that was perfectly clear to me, and I was only doing the same.
@ Sparky:
You might be right, but you might be in for a surprise too. In the research I've been doing for my M.A. I've recorded allophones of an English vowel where one is almost 200% the duration of the other – yet I'm sure few people are aware of any difference at all.
Scriptor Ignotior said,
May 9, 2010 @ 4:15 am
Onosson:
We could start an infinite regress. ☺
Anthea Fleming said,
May 9, 2010 @ 6:30 am
Can't resist oldish joke.
Two stout elderly nudists sitting on veranda.
First:"Have you read Marx?"
Second: "Yes, it's these damned cane chairs."
Peter said,
May 10, 2010 @ 1:54 am
It's quite easy to specify which wording was intended, by using /ei/ versus /æ/ at the start. I would of course just use a schwa unless I was intending to stress the difference.
There's also an old joke that my grandfather was fond of: What's the difference between a prince, a bald man, a monkey and a lake?
The prince in the heir apparent, the bald man has no hair apparent and a monkey has a hairy parent.
richard said,
May 10, 2010 @ 2:41 am
I agree with those who agree with Carl .. I have a New Zealand accent salted with a smidgeon of Australian. Both in my head or out in the real world I hear myself say "uh name" and "an aim".
Anthea Fleming said,
May 24, 2010 @ 2:22 am
I just heard a woman talking on the radio (here in Melbourne, Australia) claiming to be "a cannibal".
Then it appeared that she believed she had been saying "accountable".