Beware of intrusive stops

A message I got yesterday (noted more fully here) reminded me of something from the 2002 documentary Spellbound, which I saw earlier this year. The film follows 8 kids as they make their way to the National Spelling Bee championship in 1999. One of these kids is Harry Altman, a really smart, sweet, and precocious 11-year-old who unfortunately loses on the word “banns“. An article in The Jewish Journal of Greater Los Angeles notes that this word “refers to a Christian marriage notice” and that Harry had never heard of it before because he’s Jewish. But in the film, Harry himself offers a completely different reason for getting the spelling wrong: the word was incorrectly pronounced by the spelling bee announcer as “bands”.

For me and (I believe) for most speakers of English, these two words are simply homophonous. The technical reason that “banns” sounds like “bands” has to do with the timing of articulatory gestures: the transition from the velum-down (nasal) gesture of the [n] to the velum-up (oral) gesture of the [z] does not line up perfectly with the transition from the closed (stop) gesture of the [n] to the critical (fricative) gesture of the [z]. The velum moves up before the closure becomes critical, which leaves you with an oral stop: [d]. This sort of thing is called an intrusive stop, but it sounds no different than a “real” stop, at least not at any consciously perceptible level. It’s the same phenomenon that makes “prince” sound like “prints” — though there’s also a transition in voicing going on there, so I suspect that makes some difference. Another interesting example is “hamster”, which is often misspelled as “hampster” (as of this writing, there are 2,710,000 ghits for the former, 160,000 for the latter), and in this case there’s both a voicing transition and a transition in place of articulation from labial [m] to coronal [st] (note that the intrusive [p] is also labial). A similar thing, but involving lateral-to-central rather than nasal-to-oral transitioning, happens for speakers who pronounce “false” something like “faults”.

Poor Harry deliberated an awfully long time, asked for the word to be repeated several times, pronounced the word to himself many times … it was sad to see him go down like that. But I sincerely doubt that he’s right in his assertion that the announcer mispronounced the word (and I also doubt that it would have made much difference had Harry been raised a Christian).

3 thoughts on “Beware of intrusive stops

  1. Kie Zuraw

    It’s been a while since I rented that movie, but I recall that the announcer made a big effort to not produce a [d] (though that may have been after the contestant asked “bands?”).

  2. Eric Bakovic

    Interesting. I wonder if the audio’s good enough to verify that instrumentally … if it can in fact be verified based on the acoustic record. FWIW, in order to produce this word without the intrusive stop, I feel like I have to avoid making a full closure during the entire length of the nasal, so that the result is more like a nasalized fricative. (Coincidentally, I made nasal place assimilation to fricatives (and the blocking thereof) the central case in my contribution on local assimilation to Paul de Lacy‘s forthcoming edited volume The Cambridge Handbook of Phonology.)

  3. Pingback: There’s a pattern here to see » How well do you spell?

Leave a Reply

Your email address will not be published. Required fields are marked *