7

Suppose that, in some hypothetical language, there were two different words:

  • /tump/
  • /tump/

What's the difference?, you might ask. In the first one, the word is one syllable long. In the second one, the word is two syllables long because the /m/ is syllabic.

How do you express this difference with IPA?

curiousdannii
  • 6,193
  • 5
  • 26
  • 48
Peter Olson
  • 1,412
  • 1
  • 13
  • 23
  • 2
    Good question but I think you might have chosen bad examples. – hippietrail Sep 16 '11 at 00:13
  • @hippietrail I feel like I'm missing something important, but you have my permission to edit the example however you like. I just used a random phonemic sequence that seemed to work. – Peter Olson Sep 16 '11 at 03:15
  • 1
    I don't know if there are any minimal pairs in English to use a side-by-side illustration like you've chosen. But English words like "button" and "bottle" are often described as having syllablic "n" and "l". Not in all transcriptions but I'm not sure about all dialects. If not perhaps we can compare dialects rather than minimal pairs. – hippietrail Sep 16 '11 at 07:50
  • 1
    Importantly, for me it doesn't ring true to have a syllabic consonant beside a vowel, but perhaps that is indeed frequent but just outside my experience. – hippietrail Sep 16 '11 at 07:51
  • @hippitrail IMO, button and bottle, do have syllabic consonants, but it doesn't seem like the number of syllables in them could be ambiguous. That is, it seems to me that it would be impossible to pronounce /bʌʔn/ or /badl/ without the /n/ or /l/ being syllabic. – Peter Olson Sep 16 '11 at 13:21
  • I don't really know what you're getting at with syllabic consonants changing the number of syllables in a word. You might ask about it another question. "Button" and "bottle" indeed have two syllables whether they're analysed as having syllablic consonants, a schwa as the second vowel, /t/, /d/, or glottal stop etc. – hippietrail Sep 16 '11 at 15:10
  • 1
    @hippietrail, I think what the asker is saying is that the syllable count in those hypothetical forms is dependent on whether or not the nasal is syllabic. In his hypothetical language, the word in which the [m] is not syllabic contains one syllable, but the one in which the [m] is syllabic contains two syllables. – musicallinguist Oct 11 '11 at 15:14
  • 1
    Incidentally, I think part of the confusion here is also stemming from everyone's use of diagonal slashes in transcriptions. Normally we would notate the pronunciations (i.e. phonetic transcriptions) of words with brackets, so the pronunciation of button would be notated as[bʌʔn̩]. Few linguists would analyze that word as having the underlying form /bʌʔn̩/; rather, they would transcribe it with a vowel (maybe a schwa) in the second syllable and posit a rule whereby the vowel-nasal sequence in the underlying form gets realized as a syllabic nasal in the surface form. – musicallinguist Oct 11 '11 at 15:25
  • 1
    There are languages that have been analyzed as realizing underlying nasals that aren't accompanied by vowels as syllabic on the surface (Swahili being a well-known example), so not all syllabic nasals have to stem from vowel-nasal sequences underlyingly. But usually syllabic nasals are in complementary distribution with non-syllabic nasals, so within a single language you wouldn't expect to find a minimal pair that differs only in the syllabicity of the nasal. It follows that you wouldn't expect two underlying forms (which we transcribe with diagonal slashes) that differ in this regard. – musicallinguist Oct 11 '11 at 15:47
  • @musicallinguist: My examples are actually from lexicography where I once compared a range of dictionaries for how they transcribe the pronunciations of various phenomena, so not exactly linguists. But I wouldn't be surprised to see as much variation in how linguists would represent these as in how lexicographers do. Lexicographers can make it messier than necessary by often using [] where linguists would use //. – hippietrail Oct 12 '11 at 08:27
  • @hippietrail, just to clarify, I think the examples you provided were quite reasonable and many linguists would agree with them as possible pronunciations. I just meant that linguists don't use // for pronunciations (actually the comment was directed more toward @Peter). – musicallinguist Oct 12 '11 at 14:52
  • Just happened to stumble upon this old question now, randomly. An example of an existing language that actually does this is Cantonese, which has phonemic syllabic nasals as well as rimes that end in nasals. For example (ignoring tones, since my knowledge of Cantonese tones is zilch), 衫溼 ‘the shirt is wet’ is [saːm sap], while 沙唔溼 ‘the sand is not wet’ is [saː m̩ sap]. (Silly example, but you get the picture.) – Janus Bahs Jacquet Mar 09 '15 at 14:13

2 Answers2

12

I've always seen a dot below the syllabic consonant. This is common in British dictionaries which indicate syllabic consonants - the various dictionaries vary quite a bit in this regard as it happens. But the majority of British dictionaries do use IPA which is still quite uncommon in their American counterparts.

Then again Unicode provides COMBINING VERTICAL LINE BELOW for this purpose:

/bʌtn̩/, /bɒtl̩/

hippietrail
  • 14,687
  • 7
  • 61
  • 146
  • +1 - By the way, I wanted to try looking up one of your examples in two dictionaries, so I've searched button in the NOAD (American) and the OALD (British), neither had the dot; I don't understand why, but anyway, I agree with you: it's the one adopted by the IPA. :D – Alenanno Sep 16 '11 at 11:37
  • Online dictionaries might not do it for technical reasons like not enough fonts, web browsers, operating systems, or some of their software working with the Unicode symbols. You should find it in print dictionaries because the publishers have full control over their typesetting and are used to weird symbols. Also you can Google directly for the IPA! – hippietrail Sep 16 '11 at 11:50
  • I guess you're right, not all computer might be able to view those symbols, although it might be pretty standard by now. :D I didn't check my paper dicts, I was just trying to see if it was present in the dictionaries I knew, because I never paid attention to that. :) – Alenanno Sep 16 '11 at 22:58
  • 3
    Nitpicky correction: the IPA diacritic for a syllabic consonant is NOT a dot; it's an "understroke", as shown in the IPA diacritic chart: http://upload.wikimedia.org/wikipedia/en/e/e3/IPA_diacritics_2005.png. Strictly speaking there is no IPA diacritic that consists of a single dot under the symbol, although a small circle represents voicelessness and a pair of dots represents breathy voice. – musicallinguist Oct 11 '11 at 14:49
  • @musicallinguist: I'm sure I'd seen a dot in British dictionaries but I don't have access to my dictionary collection right now as I'm hitchhiking around the world. Anyway I covered both my memory and Unicode versions. – hippietrail Oct 12 '11 at 08:24
6

You can use [ . ] such as /tu.mp/ as a symbol for a syllable break (often left-off in normal usage). But I don't think your specific case (tump) is possible in a language a syllabic consonant would appear after a syllable ending with a vowel.

Edit: there is also a syllabic consonant diacritic. For your examples the transcriptions would look like this [tump] [tum̩p].

Louis Rhys
  • 8,501
  • 6
  • 45
  • 71
  • Out of curiosity, why don't you think it's possible? Unless I'm missing something, it seems to be easily pronounceable. – Peter Olson Sep 17 '11 at 00:43
  • @peter sorry, more correctly what I meant was my intuition said nobody would pronounce them like that. On second thought I might be wrong. – Louis Rhys Sep 18 '11 at 15:37