Monday 15 November 2010


Implosives are sounds made with a glottalic ingressive airstream mechanism. Along with ejectives (glottalic egressive) and clicks (velaric ingressive), they are a sound-type taught to all serious students of phonetics. Every year at UCL I trained budding phoneticians not only to recognize them but also to produce them to order.

Most of the implosives found in the world’s languages are voiced. That means that the rarefaction in the pharynx and mouth (the glottalic ingressive mechanism) is combined with a simultaneous vibration of the vocal folds dependent on a pulmonic egressive airstream. Interestingly, in my experience learners generally find this combined airstream easier to produce than the purely glottalic one that is needed for voiceless implosives.

But does what we teach in the classroom agree with what we find in real languages?
Just over a year ago I wrote about the Zulu song Thula sizwe (blog, 8 Sep 2009).

Now Catherine Paver, in a very late comment on that posting, has drawn our attention to a website called “Sing Africa!”. Thank you! This site contains the words and music to a number of South African songs — Zulu, Xhosa, and Sotho — along with sound clips comprising not only the complete songs but also slow demonstrations of the authentic pronunciation of each word in turn.

Here you can confirm what I wrote last year about the Zulu implosives:
The implosives are only very mildly implosive and the main auditory difference between them and the plain voiced plosives is that plain [b] and [ɡ] are depressors, while weak-implosive [ɓ] and [ɠ] are not. In singing, though, such tonal subtleties are naturally lost.

In the word-by-word clip of Thula sizwe, listen to uŋgabokhala uŋɡaɓɔˈkʰala, the third word in the song. That’s what a bilabial implosive ɓ in this real language sounds like. The “implosiveness” is much less than in the exaggerated versions we tend to get in the classroom. Mea culpa, perhaps. We see the same thing in the last word, uzokuŋqobela uzɔɠuˈŋǃɔɓɛla, with both velar ɠ and bilabial ɓ.

The implosives of Sindhi seem to be similarly weak.

I wonder if there are any languages with really strong, noisy implosives.


  1. Same's often true for Semitic "emphatics" and pharyngeal fricatives.

  2. Mmm, I can only hear a normal [b]. But then again, I'm rather hearing impaired...

  3. Lipman,
    Same what's often true? I've always observed that Semitic "emphatics" are often implosive, but how can you have an implosive pharyngeal fricative? I do know dialects in which the allegedly pharyngeal alleged fricative is not fricative at all.

  4. much less than in the exaggerated versions we tend to get in the classroom.

  5. So you didn't mean Semitic "emphatics" are often implosive at all. Do you agree that in some Arabic dialects they often are? I already take your answer as agreement on the classroom "pharyngeals"!

  6. No, I wasn't talking about what "emphatic" is here. Where do you get that they often are implosive in Arabic dialects (other than in front of dentals maybe)?

    Concerning the question of fricativeness, do you mean they're rather approximants with native speakers in normal speech? I'd agree there.

  7. I followed the link to the Sindhi sound archive, in the main post, and listened to The North Wind And The Sun in Sindhi.

    I'm not a speaker of Sindhi, but as a speaker of two other Indo-Aryan languages (Marathi and Hindi), I got the impression that the narrator had his rhotic(s) influenced by British English. I say this because he pronounced certain R's as alveolar taps (as would be unremarkable in standard Sindhi), while others were pronounced as (post-)alveolar approximants.

    The approximant was most clearly audible in all occurrences of the word 'musafiru' [mɯsafɪrɯ],[mɯsafɪrə].

    It's become somewhat fashionable, in commercial Hindi songs, to replace the tapped or trilled rhotic with the approximant, ostensibly to 'smoothen' its sound. However, the approximant remains rare in colloquial Hindi (and Marathi). In fact, those speakers of these two languages who know little or no English (i.e. the majority by far) have a tough time pronouncing it. I'd be very surprised if the same weren't true for Sindhi speakers.

  8. Just listened to the same story in Urdu, on the same site. The transcription is scandalously erroneous. I'd better write to them.

  9. How does one pronounce a vowel that immediately follows an implosive? Is the airstream for the vowel also ingressive, or does it suddenly change direction to be egressive?

  10. vp: it changes. NB the airstream for an implosive is GLOTTALIC ingressive, but that for a vowel is PULMONIC egressive. Voiced implosives actually combine the two (see my second para).

  11. Lipman,
    Where I get that they often are implosive in Arabic dialects is from the impressions of my own ears, I'm afraid. I see from Google that there's a promising-looking article On two Cases of Consonant Change in Modern Arabic Dialects by MM Bravmann - 1960 mentioning "the implosive character of the Arabic t", which I assume must be ṭā, and I'm curious to know whether the dialects in question are the ones I have heard on a spectrum from Libya to Algiers, where it's tā that's funny, varying to tsɛ:. There seems to be a parallel with the abovementioned development of implosives in Sindhi here. Shall we club together and buy the article?

    It's quite startling with the ṭā (Standard Arabic /tˤ/), but I have no idea about its distribution within that spectrum, or indeed quite what it is. The funny thing is that ḍād (Standard Arabic /dˤ/) doesn't necessarily seem to go with it, which is the opposite of what you would expect with an implosive. But I do believe in the unvoiced implosive variants, though I remember thinking they varied to ejective with some sort of epiglottal involvement. I realize I shouldn’t have adopted your word "often", having misunderstood the reference of it. I still don't know what you mean by "in front of dentals".

    Concerning the question of fricativeness, yes I do mean varieties of ‎ʕ (though not ħ) are rather approximants with native speakers in normal speech, but including glottal stop as well as glottal approximation, in all these cases likely to be associated with creak or epiglottal involvement.

  12. I know ṭā and ḍād are traditionally transcribed as /tˤ/ and /dˤ/, but learning Arabic I felt that there was a lot more going on here than just a simple voice contrast. Did my lying ears deceive me?

  13. I think we should avoid being blinded by these discrete words ("implosive", "ejective" and "pharyngeal", for example) which attempt to compartmentalize features of a language that are much fuzzier in practice.

    Ejectives can evolve into implosives or pharyngeals and back again, can't they? So it's not shocking that such a widely spoken language as Arabic should show variety among speakers.

  14. mallamb, they probably really mean t.

    Anyway, I merely wanted to give another example of NNSs who often have a hard time not to exaggerate. There are so many more, of course, eg finding the right sound between [h] and [x] in, say, Russian or (different forms of) Spanish.

  15. Re-Semitic emphatics. First, in Arabic dialects, they are most often primarily realised with secondary pharyngealisation (/velarisation) and are often much tenser than their 'plain' counterparts. In many dialect types emphatic /T/ and uvular /q/ are plain voiceless where, e.g. /t/ and /k/ are voiceless aspirated (see ongoing work, incl. acoustic analyses, by Bellem and by Bellem and Watson). This seems connected with dialectal sound-system typology involving also retention of interdentals.

    Second, Semitic emphatics are now widely believed to have originally been ejectives. They are still ejectives in Ethio-Semitic, and Modern South Arabian languages seem to be in a state of flux - mostly there is a velar /k'/ corresponding to Arabic /q/ but the other emphatics are 'backed' and either variably ejective or only ejective / glottalised in certain contexts (see ongoing work by J.C.E. Watson and by Watson & Bellem).

    There is a strong correlation between ejective consonants and creaky vowels (think glottalisation / laryngealisation) - I have a number of clear examples of significantly creaky vowel following ejective in Tigrinya and Mehri. It has been suggested (see e.g. Heselwood 1996) that due to some re-aligning in the sound system, laryngealisation may become associated with pharyngealisation, leading to Semitic ejectives becoming in some language groups pharyngealised obstruents (sorry, that's in a nutshell).

    Further, there is some association between pharyngeals and epiglottals - some dialects of Arabic have a clear epiglottal realisation of the voiced pharyngeal (Iraqi Arabic /?/ is often an epiglottal stop in final position). In particular, see John Esling's work on epiglottals. In NE Caucasian langs, pharyngeals are often better described as 'epiglotto-pharyngeals' and are clearly different from e.g. the Arabic pharyngeals. They behave differently phonologically, too, as they pattern with palatalisation. (See e.g. Sandro Kodzasov.)

    Going back to implosives, the only case where I think it is possible that there may be variable implosivisation of an Arabic emphatic is in dialects where the 'Daad' is a stop (not a fricative or lateral), and ('partially'?) implosive realisation of 'Daa' may have the effect of increasing the voicing duration and thus perceptually what we may call sonorance. Certainly there is a lowering of the larynx (I'm sure this is mentioned in some of the literature, but I can't for now recall where and don't have my papers to hand). My untested hunch is that this would be the case mostly in careful speech or recitation, and I'd be very interested to see further comments on it.

  16. luke said...
    I know ṭā and ḍād are traditionally transcribed as /tˤ/ and /dˤ/, but learning Arabic I felt that there was a lot more going on here than just a simple voice contrast. Did my lying ears deceive me?

    There seems to be quite a bit of agreement on here that your ears are neither lying nor deceiving. I think phoneticians should give their ears the benefit of the doubt and then proceed to attempted refutations, and I don't see that the statement your ears have prompted you to make is falsified. Not by a long chalk!

    Do you mean they probably really mean tā [tʰɛː, tsɛː] etc rather than ṭā [tˤɑː] or whatever? Have you any evidence that they are even more insane than me?

    I'm sorry to say I got carried away by the possible relevance of implosive or ejective Daad versus an "emphatic" fricative of various desccriptions to John's reference to the "regular correspondence between the z of (isi)Zulu and the ejective t̕ of the closely related (si)Swati" in his next blog entry, and made what might with generosity be described as further comments on your discussion of this Daad in connection with that perhaps parallel correspondence.

    I was struck by the implosivization I was so sure I was hearing in the ṭā (Standard Arabic /tˤ/), because I didn't think then that there was such a thing as an invoiced implosive, but I was also sure I was hearing implosive /b/ in Algerian, and I did think implosive /dˤ/ where it wasn't /zˤ/ or something.

  17. Lipman,
    That was still too cryptic for me, but I've found it in Google Books now. What a disappointment. It was just about knocking down the straw man W. Marçais. Again, I gave the benefit of the doubt, and considered the possibility that in 1922 "l'implosion de ت" may have meant Saussurean-style implosive function. But nah, the context is "l'explosion formidable" of q. And we may be sure it doesn't mean that with reference to some intervening vocoid. It's just anecdotal 1922 rubbish, and it's not even appropriate from the perspective of 1960's phonetic terminology, never mind hypothetico-deductive method or common sense, to say "there is no proof" of either, when what would be needed is disproof of a descriptive hypothesis, and neither attains to the status of any such thing. And then Bravmann goes on to say "a k- sound in a group like this is usually articulated as an implosive – the very opposite of what Marçais assumes for the articulation of q."

    How naïve of me to imagine this article could have had anything to do with the sort of articulatory implosion we have been talking about. I won't bore you with further quotations, but it becomes obvious that "implosive" is being used by Bravmann at least to mean "unexploded".

    And the worst of it is that I now seem to remember that that use of it was the first I came across in those days, and that I thought nothing of it until I got the Saussurean and Sweetian traditions sorted out in my head.

  18. This is from quite a while ago, but I found this post when googling.. a language I work on, Uduk, has differences between voiced, voiceless, aspirated, ejective, and implosive in at least 2 places of articulation.. and the implosives in careful speech are quite distinct. Does anyone have any references regarding implosives in Arabic, btw? I'd be curious to read about that.


Note: only a member of this blog may post a comment.