Most sound changes that involve consonantal place of articulation are gradual changes between two POAs that are contiguous: for example, a velar gets gradually fronted until it becomes a palatal. What are some examples of changes where a sound skips from one POA to another that is not contiguous to it?

I can think of these:

  • labiovelar stop to labial (very common; at least 5 independent examples in IE alone)
  • interdental fricative to labiodental (also common)
  • debuccalization changes



I don't have enough reputation to put this as a comment, so I'll just leave it here: one way that a sound change can occur between two PoA without ever moving through all of the intermediate positions is through secondary articulations that gradually become more prominent, while the original primary articulation is lenited or lost. This is in fact present in your example of "labiovelar stop to labial", and is also the reason for phenomena like r-labialization (many english varieties have actual phonetic labialization of /r/, without losing the rhotic articulation.) While these sort of changes appear to "jump" from one articulation to another, the process can still be a gradual one, since both articulations can be present at the same time. Besides labialization, some other secondary articulations that can become primary are palatalization and glottalization.

Changes like debuccalization are probably a true example of "skipping" discontinuously from one articulation to another. Another example would be the change t > k commonly seem in Polynesian languages, which seems to occur without passing through all the "intermediate" articulations that are physically between them in the mouth.

| improve this answer | |
  • Excellent point about gradualness, thanks. For the Polynesian t > k change I was always under the impression that it was a gradual one (though I never actually looked into it); how is it known that this isn't the case? – TKR Nov 22 '14 at 19:15
  • What I meant is that the sound doesn't appear to pass through the palatal place of articulation, which is physically intermediate between the dental/alveolar and velar positions. I'm drawing that conclusion from this paper, which presents some analysis of the phonological development of the change, but has no attested example of a palatal intermediate step. home.ccil.org/~cowan/temp/43.2blust01.pdf – brass tacks Nov 22 '14 at 23:43

I can think of a few sound changes in natural languages that skip over another phoneme's place of articulation. Many involve /r/:

  • British English: R-labialization is realization of /r/, normally a alveolar approximant [ɹ ~ ɻ], as a labiodental approximant ([ʋ]) in dialects of southeastern England. Non-Cockney tend to hear this sound as /w/. This appears to have happened in Haitian Creole as well.
  • American English (possibly related): Realization of /r/, normally a pharyngealized velar approximant (I think), as a velarized bilabial approximant [w] in Elmer Fudd's derhotic idiolect (and to a lesser extent that of Barbara Walters).
  • French, German, and Hebrew: Guttural <r> is realization of /r/, originally an alveolar tap or trill [r ~ ɾ], as uvular fricative or trill [ʀ ~ ʁ].
  • Portuguese /r/ is all over the place.

And a couple that do not involve /r/ phonemes:

  • Taiwanese (source): /f/ becomes realized as [x] or /x/ as [f] in some sociolinguistic contexts
  • The Spanish sound spelled soft g and j used to be a postalveolar fricative [ʃ], but in the seventeenth century, it moved back past the velar place of hard c and qu and became an uvular fricative [χ].
| improve this answer | |
  • I think it's common for "r" to be "all over the place". It might even be "normal". f~h is probably pretty common though I'm not sure I've noticed f~x before. – hippietrail Oct 19 '14 at 5:52
  • ..bearing in mind that "discrete" is a phonemic notion, not a phonetic one. Discretion is the better part of phonemicization. – jlawler Nov 18 '14 at 2:36

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.