So I'm pretty sure I understand labialized [ʷ] and some of the other superscripts, but I don't fully understand palatalized [ʲ]. An example of palatalized is Abkhaz, Selkup, Bulgarian, and Yanesha. The way I imagine the [ʲ] to work is you put the tip of your tongue on the hard palette while you make whatever sound, like the image below.

To start it looks like I got that wrong, it is the middle of the tongue that should touch the hard palette:

In technical terms, palatalization refers to the secondary articulation of consonants by which the body of the tongue is raised toward the hard palate and the alveolar ridge during the articulation of the consonant. Such consonants are phonetically palatalized.

"Pure" palatalization is a modification to the articulation of a consonant, where the middle of the tongue is raised, and nothing else. It may produce a laminal articulation of otherwise apical consonants such as /t/ and /s/.

But in those phonologies above, they palatalize these consonants:

  • [mʲ]
  • [nʲ]
  • [kʲ]
  • [tʲ]
  • [pʲ]
  • [βʲ]
  • [lʲ]
  • [ɡʲ]
  • [χʲ]
  • [ʁʲ]
  • [qʲ]
  • [zʲ]
  • [sʲ]
  • [fʲ]
  • [vʲ]

So there are nasals, stops, approximants, and fricatives in there. Placing the middle of my tongue on the hard palette while doing a [qʲ] or [ʁʲ] or [ɡʲ] feels difficult, basically I am just pressing my tongue from back all the way to the middle, but trying to avoid touching the velar area, but I don't see how sound can escape in both places the tongue is touching. Wondering if this is "double articulation" sort of thing, or how to do it better.

Then for the [lʲ], I can do the normal [l] and just press the middle of my tongue up to the hard palette, but it doesn't really change the sound.

The [tʲ] (when I do it) sounds sort of like either [tc] or [tɕ] or [tʃ]. Then [pʲ] sounds sort of like [pɕ], so I guess that is a different sound.

Finally, I don't see how [nʲ] can be any different than [ɲ]. And I don't see how [mʲ] will have any effect since the mouth is closed.

So basically, when I try to do it, only [tʲ] and [pʲ] seem (a) possible and (b) make any obvious sound difference.

Wondering what I am doing wrong or what I should be looking for, and if my assumptions and such are missing things. I'm also wondering if the "ʲ" aspect is occurring during the main sound being created, or before/after. I assume the effect it is trying to achieve is happening during the main sound. But I dunno, maybe you could write [pʲ] as in some orthography. Though that might work for stops, not sure what the non-stops would look like, maybe [s͡ç].

enter image description here

1 Answer 1


Palatalization is basically superimposing the tongue position of the vowel [i] on any other consonant, and is as you now understand. The actual timing of the i-like articulation with respect to the main consonant articulation varies from language to language and context to context; palatalization is most often realized and perceived as a feature of the consonant release and transition to the next vowel. If you reach satisfaction with [pʲ] but not [mʲ], that is probably because the nasal does not have the kind of noisy release that [p] has, so the difference will sound subtle (thus the perception that [pʲ] sounds like [pɕ]. Embedding the sound in the frame [a_a] maximizes the chances of hearing a difference.

You are right that [nʲ] (abstractly) sounds like [ɲ] – that is, some people may write [ɲ] for the same physical sound as others would use [nʲ] for. The same goes for [kʲ gʲ xʲ] versus [c ɟ ç]. Deciding between the two notations is usually more informed by phonological pattern rather than phonetic fact. For example, Russian has a lot of palatalized consonants, and general rules of palatalization which argue for writing [nʲ] as a palatalized [n], rather than [ɲ]. The choice of symbols may be influenced by a phonetic fact, that the exact timing and degree of palatal coarticulation can vary across languages, and writing [ɲ] rather that [nʲ] may indicate a greater overlap of palatality with the consonant. The same goes for [kʲ gʲ xʲ] versus [c ɟ ç].

Abkhaz is a special problem since [qʲ, ʁʲ, ɡʲ] there involve tongue raising and fronting simultaneous with tongue backing and lowering: a contradiction. I do not know what the argument is for such phonemes, and what supports that analysis over other analyses. One possibility is that this is related to the impoverished vowel inventory (only two vowels), and really there are more vowels and fewer consonants. Or, [Cʲ] may be a consonant cluster, [Cj]. I don't have access to Hewitt's grammar or Catford's unpublished book, so I don't know what the phonetic and phonological evidence is for that language.

(I should add that there is no IPA symbol for an alveo-palatal nasal, for which reason ɲ is pressed into service for a number of languages like Swahili or Saami).

  • 1
    Articulatorily, [nʲ] and [ɲ] are clearly different because the latter doesn't involve coronal occlusion. In practice, I suspect what is transcribed with [ɲ] and described as palatal may often be alveolo-palatal, though.
    – Nardog
    Oct 19, 2018 at 17:04
  • What are you referring to by "Catford's unpublished book"?
    – Nardog
    Oct 19, 2018 at 17:05
  • 1
    Note though that "coronal" in some theories includes palatals, see esp. Hume. I understand that Catford was working on a ms. The phonetics of Caucasian languages. I have no idea if anyone has a copy.
    – user6726
    Oct 19, 2018 at 19:30
  • You said that palatalization superimposes the tongue position of the vowel [i] on any consonant. Isn't that what is already done with any plosive consonant (eg P, B, T, D, K, G) followed by the vowel [i]? For instance, the [bi] in "beat" or [ti] in "tea". What are the differences between those and [bʲi] and [tʲi] ? Oct 31, 2019 at 16:56
  • It is a question of timing: in [bʲi] that tongue position is relatively earlier than it is in [bi]. Consequently, during the closure, the tongue will be less advanced in [bi].
    – user6726
    Oct 31, 2019 at 17:06

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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