Phenonemon?
« previous post | next post »
In a comment about the video lecture in yesterday's post about David Lodge, JPL asked:
Why does he say "phenonemon" [sic] (purposefully enunciated) at 4:42?
Let's start with the question of what Lodge actually says. Here's the relevant passage, in which Lodge describes some of the real-life background for his novel Small World:
Indeed
it was this intense double exposure to the international conference
which first gave me the idea for that novel.
I felt I'd stumbled on a new cultural phenomenon
a product of what I called
or had my character Morris Zapp call
the global campus.
(For the video as well as the audio, here's a link to the immediate context.)
Zeroing in a bit:
I felt I'd stumbled on a new cultural phenomenon
And examining the word itself (and side-stepping the question of how to IPA-ize the vowels), my judgement by both ear and eye is that it's
[fɛˈnʌ.nʌˌnvn]
rather than
[fɛˈnʌ.nʌˌmvn]
That is, Lodge has not swapped the /m/ with the following /n/, but rather has replaced the /m/ with an /n/:
You can made your own judgement by listening — and verify visually that the nasal consonant in question lacks the formant transitions (fore and aft) that would be expected if it were labial.
For those of you who've missed the opportunity to be inducted into the mysteries of spectrogram reading, here's a fragment from earlier in the same lecture (around 14:19) that illustrates the point:
These were hosted by a different university each year,
and were accommodated in a college or hall of residence
whose student occupants were absent during the Easter vacation.
Zeroing in a bit, we can see the typical labial transition following the /m/ of accommodated:
So Lodge's performance of "phenomenon" involves a perseveration/anticipation of the phoneme /n/, not an exchange of /n/ and /m/. And if by "purposefully enunciated" JPL means that Lodge pronounced the word that way on purpose, I'm pretty sure that he's wrong — this is a typical kind of speech error.
Speech errors occur on average about once per thousand words, and the Lodge lecture comprises a bit more than 2,000 words, so it's not a surprise to find a speech error in it. (Of course, different people, contexts, and texts yield different rates…)
Victor Mair said,
January 5, 2025 @ 8:11 am
I've listened to that word spoken by Lodge, both in isolation and in sentence context, and each time it sounds like "phenomenon".
Mark Liberman said,
January 5, 2025 @ 8:25 am
@Victor Mair: "I've listened to that word spoken by Lodge, both in isolation and in sentence context, and each time it sounds like "phenomenon"."
Thereby demonstrating that you are subject to the well-known effect of top-down expectations on speech perception, as illustrated long ago in the "phoneme restoration" effect, where "When part of an utterance is replaced by another sound (e.g. white noise), listeners report that the utterance sounds intact-they perceptually restore the missing speech".
See the discussion of 比较 bi3jiao4 in "How they say 'Beijing' in Beijing", which mentions a case where
a native speaker of Chinese … transcribed bi3jiao4 as [bidʒau] and refused to believe, even after repeated careful listening, that the medial consonant was [pronounced as] a glide, not an affricate. So I had him make a spectrogram, in which the lack of any stop gap or fricative noise was visually plain. His response was to conjecture that I had somehow hacked the computer program, in order to play a joke on him for pedagogical effect, because it was so completely obvious to his ear that the non-existent affricate was actually there.
Philip Taylor said,
January 5, 2025 @ 8:44 am
Unlikle Victor, and despite (probably) sharing his "top-down expections", I found the "phenomenon" in "I felt I'd stumbled on a new cultural phenomenon" intensely jarring — it sounded to me (at second hearing — I had already listened to the whole thing yesterday) like "phenonenen", with all /n/s and one clear /ɒ/, the final vowel being a schwa.
Jason M said,
January 5, 2025 @ 10:52 am
Tally me for the “wait, what’d he just say?” camp when I first heard it. I wasn’t sure what it was but seemed mangled somehow.
Mark Liberman said,
January 5, 2025 @ 10:56 am
@Philip Taylor: "…despite (probably) sharing his "top-down expections"…"
You and every other speaker of English, at least those who are familiar with the words and concepts involved. And the same for all speech-to-text systems in the past few decades.
There are individual (and circumstantial) differences in the relative strength of the various relevant sources of information about what someone said, and their mode of combination — what the words are, how they're combined, what the meaning of the passage is, what the audio signal-to-noise is like, what the listener's hearing is like, whether visual cues are also present, and so on and on…. There's a large scientific and technical literature on this, but suffice it to say that for some people in some circumstances the various top-down effects are strong enough to override what would otherwise be clear acoustic evidence.
Bloix said,
January 5, 2025 @ 11:57 am
Might be moderately relevant to note that Lodge's final academic novel, about a recently retired linguist, is titled Deaf Sentence.
Gregory Kusnick said,
January 5, 2025 @ 12:30 pm
If you watch the BBC's Gardeners' World, you'll hear "anemone" pronounced as "an enemy" with some frequency. These don't seem to be one-off production errors; rather, the presenters seem about evenly split between those who consistently (mis)pronounce it this way and those who put the m and n where they belong.
JPL said,
January 5, 2025 @ 6:35 pm
In the sentence from my comment that you quoted, the parenthetical "purposefully enunciated" meant something like, "enunciated with full and conscious attention", as opposed to being the result of an articulatory "misstep". (Not "on purpose", really, since that expression seems to indicate that he said it that way in spite of the fact that he knew it was wrong, perhaps to provoke a response.) But I should have expressed my thought more accurately if I had included "apparently" before the "enunciated", since I "couldn't believe what I was hearing", and that's why I was asking a question. So I'm happy to be corrected; except that I was led to the alveolar rather than labial interpretation partly by looking at his lips when he was saying the word, and he appeared to be closing his lips on that last consonant. (Although I do now see his tongue coming into position as well.) So what about that aspect? How does what a lip-reader might identify as a labial have the spectrogram evidence of a non-labial? Is interchanging the two nasal consonants in a context like that also a common speech error?
Andrew Usher said,
January 5, 2025 @ 6:40 pm
When knowing already of the error, I can't possibly hear the correct 'phenomenon' but I _can_ hear 'phenonemon' rather than the more accurate (according to your analysis) 'phenonenon'. A form of auditory dissimilation that also affected the commenter that noted it? Or is it more prompted by a belief that the first type of speech error is more likely than the second, and is that merited?
k_over_hbarc at yahoo.com
(Please do something about the kind of spam seen at https://languagelog.ldc.upenn.edu/nll/?p=67692#comment-1626266 which should be preventable. I've long learned to ignore spam, but I can't ignore my words being used in that way and being recorded for posterity on the Internet!)
John Swindle said,
January 5, 2025 @ 6:49 pm
After listening to it enough times I was eventually able to hear the m->n switch you've described. But why do I hear "I felt I'd stumbled er a new cultural phenonenon"?
/df said,
January 5, 2025 @ 9:59 pm
In "sea anemone", a phrase that I would have learned well before the horticultural word itself, I find it quite difficult to avoid m n switching ("-nome"). The plain word is more likely to come out accurately, especially when concentrating, but of course that's not what's interesting. Yet the name "Nemone" seems to be unaffected.
JPL said,
January 5, 2025 @ 11:10 pm
Correction: I think I meant "… led to the labial rather than the alveolar interpretation …" above, rather than what's there. (Another switcheroo(ni).)
Philip Taylor said,
January 6, 2025 @ 5:57 am
"I felt I'd stumbled er a new cultural phenonenon" — try as I might, John, I cannot persuade myself to hear an " er " in this sentence … (two commas, one em-dash, one ellipsis in a total of 90 characters — see thread "The cost of commas")
John Swindle said,
January 6, 2025 @ 8:00 pm
@Philip Taylor: Intrigued, but not good at reading spectrograms, I took the YouTube version Mark linked and listened to it at 0.25 speed. I agree with you — "on a new" is distinguishable.
"Phenomenon," however, definitely becomes "phenonemon." At 0.25 speed, with video, the closure of the speaker's lips at the beginning of the final syllable is obvious and heard accordingly.
My next step, in the name of science of course, will be to enjoy some of Lodge's fiction.
JPL said,
January 8, 2025 @ 6:20 pm
Side note:
Articulation of a sequence of three apico-alveolar consonants clearly in quick succession is not that easy to produce. That's why trumpet players learning the technique of "triple-tonguing" are instructed to "say tu-ti-ka, tu-ti-ka" into the mouthpiece. (Not "tu-tu-tu", or even "tu-ki-tu".)
@John Swindle:
If that's where we stand on the question of what was uttered (and I agree that, judged by what I can hear, it seems to be "phenonemon"), how do we explain the spectrographic evidence? (BTW, I mentioned the tongue movement: it's probably getting into position for the word-final "n".)
Also, if that's the case, my question is still out there.
Jonathan Smith said,
January 8, 2025 @ 8:24 pm
McGurk effect or something, as both John Swindle and JPL were watching the video. Speaker's lips kinda close at this point, part of the production error.
John Swindle said,
January 9, 2025 @ 12:55 am
I think Jonathan Smith has nailed it. I'm increasingly hard of hearing and can't hear speech very well unless I can see the speaker. The "phenonemon" pronunciation appears when I watch and listen to the sample at 0.25 speed and disappears when I look away.
JPL said,
January 9, 2025 @ 1:41 am
@Jonathan Smith:
Except that I was not looking at the screen when I first heard it, definitely and jarringly. After letting it go on a little bit the puzzle bothered me and I went back to check, multiple times. I know enough to separate the two modes of evidence and not to automatically assume they are causally related; in fact my usual practice, especially with video, but also in person, is that when I want to listen closely I avert my eyes from the speaker. That principle is evident in the way I expressed my observations above. Even now I asked how we can explain the spectrographic evidence. In trying to arrive at an accurate description of the sound that was actually produced that is the more important question. (BTW, what would your own judgment be wrt what sound was uttered?)
Jonathan Smith said,
January 9, 2025 @ 3:07 pm
@JPL I mean, I heard pheNoNeNoN in the audio. Then I watched the video and heard your version (pheNoNeMoN) — so concluded that my shifted perception was McGurk-adjacent in the sense that, while audio and video are not delinked, sound and mouth shape are incongruous. My judgment re: sound is "n" with lips closed-ish :D
JPL said,
January 9, 2025 @ 4:32 pm
@Jonathan Smith:
OK, fair enough. So actually we still have not reached a consensus, as I continue to hear a difference in audible properties between the second consonant in the series and the third, describable as alveolar vs labial. So I was wondering if this could be reconciled with the spectrographic data. (Normally (as I would do in a field-work context) I like to work with the audio only. I'm pretty good at imitating and producing sounds, and demanding with a native speaker about whether I've got it right. When I listen to my "n-n-m" sequence, it sounds closer to what I hear Lodge enunciating than my "n-n-n" sequence.) Actually, I didn't look closely at the visual clues until just before my second comment, just as a check. I was unsure about the tongue movement, as that could either be for the third nasal in the sequence or the word-final nasal. So for me the question "What's going on here?" is still open, but I'm not adamant about it.
Andrew Usher said,
January 9, 2025 @ 9:58 pm
I have not watched the video so can't be affected by McGurk, but on repeatedly playing the audio I can still hear it either way. There's certainly some difference between the second and third nasals but I can't localise it – it might even be just the spreading of nasality into the surrounding schwas.
In such doubt I defer to she spectrographic evidence.
JPL said,
January 17, 2025 @ 7:11 pm
I learned today from Adam Roberts's Substack that David Lodge was, effectively, deaf.
https://profadamroberts.substack.com/p/david-lodge-rip