4

To me, it seems clear that there is a continuum between this group of sounds, as all of them (apart from ç, which I will touch on later in the post) are sibilants, and the only difference between them is in the place of articulation. However, when I try to analyse the differences in place of articulation, a one-dimensional continuum doesn't appear to form, as the position of both the blade and the middle of the tongue varies throughout the group. This is counterintuitive to me, as I can intuitively describe the progression s-sʲ-ç-ɕ as a simple continuous softening of the sound "s", with each subsequent element being a softer version of "s" than the last, and the progression ç-ɕ-ʃ-ʂ as a simple continuous hardening of "ʂ", with each subsequent element being a harder version of "ʂ" than the last. Furthermore, historical phonological changes such as s-->sʲ-->ɕ-->ʂ in Russian second-person singular ending -шь reinforce this intuition.

Similar progressions are z-zʲ-ʝ-ʑ-ʒ-ʐ, t(s)-t(s)ʲ-c-tɕ-tʃ-tʂ, d(z)-d(z)ʲ-ɟ-dʑ-dʒ-dʐ, and possibly even ɳ-n-nʲ-ɲ and ɫ-l-lʲ-ʎ.

I shall note that, in all of these progressions, the "bridge element" is always a palatal consonant which is functionally different from the rest of the group: ç is the only non-sibilant in the group and can be analysed as an advanced x; c is the only non-affricate and can be analysed as an advanced k; etc.

In light of all of that, here is my question: are the qualitative differences in the group s-sʲ-ç-ɕ-ʃ-ʂ and similar groups described above attributable to a single phonetic process? If so, what is that process? Additionally, why do the palatal consonants not fit in phonetically despite fitting in perfectly qualitatively?

10
  • 1
    I don't see the palatals ç & ʝ as fitting qualitatively at all tbh (the case with the stops/affricates is more complicated because you've got affrication going on as well as changing place of articulation), whereas all the others do form a continuum, first raising the dorsum of the tongue with the tip in place & then retracting (and eventually retroflexing) the tip leaving the dorsum largely in place
    – Tristan
    Jan 4 at 10:59
  • @Tristan I have thought about it a bit more, and isn't it that more of a 3-step process rather than a 2-step? 1) The body of the tongue is raised towards the hard palate to form sʲ 2) the tip of the tongue is then retracted past the alveolar ridge to form ɕ 3) the body of the tongue is lowered again to eventually form ʂ.
    – Max
    Jan 4 at 17:14
  • Also, whatever the process is, surely there must be some sound which is identical to s except that it's palatal instead of alveolar? To be honest, I don't really see in what way ç can't be described as that type of consonant, given that it's identical to s in every way (at least according to the Wikipedia article on it) except place of articulation. Still, extrapolating from the examples of ts and dz, it's reasonable to assume that such a sound, it's not ç, will qualitatively fit into the pattern (it's telling that many Russians can't tell c from tʲ), and I'm wondering how that's possible.
    – Max
    Jan 4 at 17:39
  • idk, to me going from s to ɕ via sʲ seems moreorless one continuous movement, but it's possible I'm doing something wrong. I'm not sure there is a palatal s, sibilants aren't really possible too far back in the mouth
    – Tristan
    Jan 5 at 10:28
  • 1
    Your description of the changes as "softening" or "hardening" of the sounds suggests that what you're noticing are acoustic differences. The relationship between articulation and acoustics is... complicated (to put it mildly). Perhaps examining some spectral slices of these fricatives will reveal a pattern that isn't obvious when thinking about them in terms of tongue position.
    – drammock
    Jan 7 at 22:59
2

From a diachronic perspective, this is simply retraction vs advancement. The place of articulation appears as the most "important" part of such a series, and so that's how the phenomenon is portrayed.

One commonly cited dramatic example is Spanish, e.g. Latin DIXIT becoming modern Castilian Spanish dijo /'dixo/. In this case, the consonant cluster with the velar + dental cluster /ks/ of Latin has become a velar fricative /x/, and in some varieties may uvularise to [χ] or become a glottal [h]. Of course, the /k/ did in some sense "promote" this change.

However, there is a difference between palatal [ç] and alveolo-palatal [ɕ]: sibilance, and that marks a "natural" break between the /s~ʂ/ axis and the /ç~χ/ axis (though there are those positing Swedish sj-sound as the sibilant version of /x/). The phonetic definition of sibilance includes the following (Ladefoged 1971):

having a comparatively large amount of acoustic energy at high frequencies

Of course, you can start adding further distinctions of retraction (dentalised vs alveolar vs "retracted" alveolar) - something that affects many varieties of English in consonant clusters.

One can also contrast laminal /s̺/ vs apical /s̻/, which are phonemically distinguished e.g. in Basque. Alveolar, postalveolar and retroflex consonants exhibit a lot of variation in the way the "active articulation" is done. Past a certain point in the mouth though, retraction causes the apical variant to no longer be possible.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.