Ambisyllabicity

« previous post | next post »

The comments on "Hypertonal conlang" (12/8/2024) include a lengthy back-and-forth about where the syllable break should be located in English words like "Cheryl". I was surprised to see that no one brought up the concept of ambisyllabicity, which has been a standard and well-accepted idea in phonology and phonetics for more than 50 years. It continues to be widely referenced in the scholarly literature — Google Scholar lists about 2,170 papers citing the term, and 260  since 2020.

The most influential source is Dan Kahn's 1976 MIT thesis, Syllable-based Generalizations in English Phonology”.  There's more to say about the 1970s' introduction into formal phonology of structures beyond phoneme strings (or distinctive feature matrices), but that's a topic for another time.

Dan's 1976 thesis introduces the concept on pp. 33-35, under the heading "Section 3 Ambisyllabicity". You can read the whole thesis here, but for convenience, here's the text of that section:

In all traditional treatments of English syllabication, a word like atlas would consist of two syllables, [at] and [las]. Since each syllable is well-defined, it makes sense to speak of a "syllable boundary" as occurring between the [t] and [1] of atlas. This phenomenon of well-defined boundary is observed in a large class of cases in English, leading to the general assumption on the part of many phonologists that it is always possible to segment an English utterance into n well-defined syllables, i.e., to choose (n-1) intersegmental positions as syllable-boundary locations.

However, this conclusion is not a logical necessity. There need not correspond to every pair of adjacent syllables a well-defined syllable boundary. For example, as opposed to a word like atlas, where the boundary between syllables is uncontroversial, it would seem completely arbitrary to insist that hammer contains a syllable boundary either before or after the [m].

In the past this fact has been typically either ignored (but see below), in which case one arbitrarily assigns a syllable boundary in a word like hammer, or else taken as evidence that the concept of the syllable is an untenable one. The position taken here is a middle one between these two extremes: it makes sense to speak of hammer as consisting of two syllables even though there is no neat break in the segment string that will serve to define independent first and second syllables.

Using Pike's term "sonority" (each syllable contains exactly one "peak of sonority"), there appears to be a sonority trough at the [m] in hammer, as opposed to a complete break in sonority between the [t] and (1] of atlas. It would seem reasonable to maintain, then, that while hammer is bisyllabic, there is no internal syllable boundary associated with the word. As an analogy to this view of syllabic structure, one might consider mountain ranges; the claim that a given range consists of, say, five mountains loses none of its validity on the basis of one's inability to say where one mountain ends and the next begins.

The observation that polysyllabic words in English need not have well-defined syllable boundaries has in fact been made before. Careful phoneticians not committed to a theory of well-defined syllabication have suggested that intervocalic consonants in English may belong simultaneously to a preceding and a following vowel's syllable.

For example, in discussing words like being, booing, Trager & Smith (1941:233) say, "…in cases like these, the intersyllabic glide is ambisyllabic (i.e., forms phonetically the end of the first and the beginning of the second syllable), so that these words exhibit a syllabic structure exactly parallel to that of such words as bidding…"

Smalley (1968:154) points out that it is easy to identify the "crests" of syllables but notes that "it is not always possible to determine an exact syllable boundary. A consonant between two syllables may belong phonetically to both." He gives the English word money as an example of this phenomenon.

The difficulty speakers of English experience in saying, in many cases, just where one syllable ends and the next begins, referred to by Abercrombie (see quote above), is doubtless due to their uncertainty about arbitrary syllabication conventions in these ambisyllabic cases. The only phonologists who to my knowledge try to deal formally with the phenomenon of ambisyllabicity in English are Anderson & Jones (1974). For them also, words like hammer, being, booing, bidding, and money would involve ambisyllabic segments. I will have more to say about their proposals below and in Chapter II.

One reason that Dan's work was so effective was the notation he suggested. Here's his Figure 4, illustrating the syllable structure of atlas and pony:

His ideas was to treat syllables as "suprasegmental" or "autosegmental" entities, analogous to tones, listed on a separate tier and connected to the segmental sequence by links that obey order allow segments to be shared by two adjacent syllables.

There have been many additional ideas about syllabic (and sub-syllabic and super-syllabic) structure, the relationship of those structures to phonological features, and the place of the traditional concept of "segment" in all that. But essentially all of these proposals agree on the idea that there are some phonological entities (whether segments or features or whatever) that belong to two adjacent syllables (or syllable-like entities).

Update — the earlier works cited in the quoted section from Kahn 1976 include:

Trager & Smith, "An Outline of English Structure", 1957.
Smalley, "Manual of Articulatory Phonetics", 1963.
Anderson & Jones, "Three theses concerning phonological representations", 1974.



14 Comments »

  1. Chris Button said,

    December 14, 2024 @ 7:16 am

    I'm surprised the post does not cite the work of John Wells.

    For example, his 1990 article "Syllabification and allophony" says the following:

    Ambisyllabicity?

    So strong is the presumption among linguists for CV.CV structure as universally preferred that many writers assume it to be true for English even in the face of strong counter-evidence such as is discussed here. Fudge (1984: 21) asserts, on no substantial evidence that I can detect, that competitive, for example, has a stressed syllable /pe/; but American writers, alert to the implications for /t/ allophones, correctly insist on /.ˈpet./ (see, for example, Webster’s Ninth New Collegiate Dictionary). Grunwell (1982) assumes, equally without justification, that a word such as better is /ˈbe.tə/.

    A more sophisticated idea is that a ‘left-captured’ consonant such as the /t/ in better is ambisyllabic, belonging to both syllables simultaneously (Kahn 1976: 33; Gussenhoven 1986). This notion has a respectable origin in the phonetic approach to the syllable in terms of sonority: the intervocalic consonant represents a trough of sonority and ‘belongs’ to neither peak. In modern terms, ambisyllabicity may be felt to allow us to satisfy at the same time both the putative universal preference for CV.CV and the overwhelming allophonic arguments in favour of CVC.V. The principle of Occam’s razor, though, shows that ambisyllabicity is not a useful concept. Those who believe in an absolute universal preference for unchecked (open) syllables must, I believe, accept that in English this can at best be true only of deeply abstract representations, and that by the level at which allophonic conditioning becomes relevant a resyllabification rule must have come into operation, namely the principle I propose. And this is uneconomical, since a word such as additive, morphologically /æd+ɪt+ɪv/, would have to have been switched to phonological /ˈæ.dɪ.tɪv/ before surfacing again as [ˈæd.ɪt.ɪv]. There may be occasions when the Duke of York gambit is necessary (Pullum 1976), but I do not believe this is one of them.

  2. Mark Liberman said,

    December 14, 2024 @ 8:08 am

    @Chris Button: "I'm surprised the post does not cite the work of John Wells."

    Um, because the 1990 Wells citation (though certainly relevant) is 14 years after the clear presentation in Kahn 1976, and is one of several thousand post-1976 discussions that cite Kahn 1976?

  3. Andrew Usher said,

    December 14, 2024 @ 8:59 am

    I didn't bring it up, though I believe in it, because I thought it was already being effectively granted. It was being asked which was the better syllabification, not assuming that only one could be correct. Only the number of syllables and their nuclei have any real existence, the boundaries between them are otherwise arbitrary; ambisyllabicity simply emerges from that.

    In the Kahn quote, the mountain-range analogy is wrong – there's a clear definition of the boundary between mountains in that case, the low points on the ridge-line between peaks. Sonority doesn't work like that, even if you believe it's an absolute standard, the low points will be found during one sound, bot between them. Hence ambisyllabicity. But for the same reason I'd also disagree that there's one correct syllabification for 'atlas' – as the /t/ is the sonority minimum, why is it not also ambisyllabic? Why not a-tlas? Of course, we know that would be odd in English phonology and that at-las is more _useful_ in terms of a model of how English pronunciation works. So for all words, and it may and does happen that the best syllabification of the same phonetic sequence differs between languages or dialects.

    In the particular case of intervocalic /r/, I gave my opinion that for Brits and others with the traditional pre-R vowel system, it is probably best to throw the R to the second always, while in General American, it's definitely better with the first vowel if that is stressed and not a diphthong.

    I don't follow John Wells here, and it seems his dismissal of ambisyllabicity is just begging the question. In 'additive', if you accept ambisyllabic consonants, both intervocalic stops are, so it makes no sense to move them around as he shows. I seem to remember that his syllabifications were eccentric in maximising codas.

    k_over_hbarc at yahoo.com

  4. Chris Button said,

    December 14, 2024 @ 9:05 am

    @ Mark Liberman

    My point is rather that Wells makes a very effective point. Unfortunately, it's a point that makes the institution uncomfortable. The most powerful ones usually do. Lumping the point in with "several thousand" other points, really just misses the point.

  5. Andrew Usher said,

    December 14, 2024 @ 9:09 am

    'The institution'? Really, do you see Wells as a rebel?

  6. Chris Button said,

    December 14, 2024 @ 9:19 am

    @ Andrew Usher

    Yes, like all the best academics.

    He very effectively challenges the orthodoxy. Here that orthodoxy goes CV.CV.

  7. Dick Margulis said,

    December 14, 2024 @ 10:40 am

    Of course, you guys are talking about syllables as sound, not syllables as written word breaks, so this is just a side note. But in written {En·​glish (Merriam-Webster) | Eng·lish (American Heritage)}, the main concern of non-linguists is where to put the damn hyphen when a word breaks at the end of a line of type. And for that purpose, the conventions, even though there are two conflicting conventions, agree that doubled consonants are always* split, so it's ham·mer, which neatly resolves the ambiguity for my purposes but probably just kicks the ambiguity can down the road for yours.

    * There may be exceptions I'm not thinking of.

  8. Mark Liberman said,

    December 14, 2024 @ 10:44 am

    @Chris Button:

    I must be missing your point. Who exactly are these orthodox CV.CV-ers? According to Wells, their position is just that CV is a "natural" or "unmarked" pattern, since it's trivially obvious that there are other phonotactic patterns. What implications this really has for e.g. the medial /r/ in "Cheryl" is not clear, though Wells feels that it implies onset-maximization. But in any case, the only "orthodox" folk that Wells cites are Fudge's 1984 work English Word Stress, which recapitulates the ideas in Fudge 1969 that were explicitly (and I think wrongly) ignored or rejected by most orthodox phonologists, and Grunwell's 1982 book "Clinical Phonology", which looks interesting but I've frankly never heard of, and is not cited by any "orthodox" phonologists whose work I've seen.

    Wells 1990 argues that "not only onsets but also codas are maximized in stressed syllables", and that "ambisyllabicity is not a useful concept". That might be true, but it doesn't seem relevant to the observation that the earlier discussion failed even to note the possible existence of ambisyllabicity as a position. And I remain somewhat puzzled about which sect views onset-maximization über alles as orthodox doctrine…

  9. Jonathan Smith said,

    December 14, 2024 @ 1:35 pm

    Agree "cher.yl or che.ryl" etc. is not meaningful. The mountain analogy above is good — but IMO why even bother claiming explicitly that the trough belongs simultaneously to both mountains? Just count peaks, right? On which framing my interest over there was in situations where the same "mountain gestalt" may be construed by different viewers as consisting of definitely 1!! or definitely 2!! peaks. Whereas an opposing view was that you just have to look hard enough and you'll figure out Reality.

  10. Chris Button said,

    December 14, 2024 @ 3:41 pm

    @ Mark Liberman

    Is John Wells' proposal mainstream/orthodox? Not as far as I'm aware

    Is John Wells' proposal preferable to more mainstream/orthodox proposals? Well, it seems to me to be grounded in solid phonetic principles based on how people actually speak.

    Is John Wells' proposal relevant here? Even if John Wells weren't one of the world's most eminent phoneticians, I still think his proposal merits a mention whenever ambisyllabicity rears its (superficially attractive) head

    At least, that's how I see it …

  11. Jerry Packard said,

    December 14, 2024 @ 4:39 pm

    It seems to me that ambisyllabicity is an eminently useful concept both substantively and theoretically; substantively because many if not most speakers would admit that, e.g., a nasal on a syllable boundary like ‘pony’ can be thought to belong phonetically to either or both sides. Theoretically because, e.g., the theoretical apparatus that allows either one tree branch or two to attach to the nasal [n] in ‘pony’ gives more explanatory possibilities than an apparatus that doesn’t. And frankly it fits more nicely with the notion that any definition of ‘syllable’ can give you an accurate syllable count, but won’t be able to tell you what phonetic stuff belongs to each syllable in a non-arbitrary way.

  12. Yves Rehbein said,

    December 14, 2024 @ 5:34 pm

    I addopt Steve's approach but instead of using two syllables and 175 tones, I have to make do with one syllable and let context do the heavy lifting.

    Mi mi mi. Mi mi mi, mi mi mi. Mamma mia!

    [That's nonsense. Syncrony equals diachrony and ontogeny recapitulates phylogeny. Not!]

    Meme meemi [yours truly]

  13. Yves Rehbein said,

    December 14, 2024 @ 5:51 pm

    *adopt

    Which is almost on-topic. I do not either agree with the statement of "additive, morphologically /æd+ɪt+ɪv/," indeed it would be ad + , *dheH- as in fact, if Wiktionary is to be believed. That is is different from aditus, from adeō. But I do not see how that's relevant to Cheryl vis-a-vis churl, Carl, Carolus, Carola, Carulla (?), and I do not really understand the argument beginning "The principle of Occam’s razor". So mi mi mi it is.

  14. Peter Cyrus said,

    December 15, 2024 @ 5:49 am

    Without implication for the broader question of ambisyllabicity, English seems to have lots of suffixes that begin with a vowel, and most of the examples cited involve them. I would break the syllables at the morpheme boundary: ham|er.

RSS feed for comments on this post · TrackBack URI

Leave a Comment