I don't know if grunts/growls are considered a legitimate part of a linguistic system, but they get used from time to time to convey things like, "I'm gonna kill you" or an angry "Yes!" so I thought I'd ask. As for place, /ɦ/ seems to be the lowest down the vocal tract, but some grunts/growls are made below that. So are these still glottal, using the vocal folds to produce a fricative, or are they sub-glottal, whatever that might be...?

2 Answers 2


First, such sounds may be a part of communication, but so is a slap on the head or a snort. If you define language as any audible means of communicating, then (1) it would be and (2) ASL is not language. So it kind of depends on what you mean by "linguistic".

There are very many kinds of noises that humans can make that could be called "grunts" and "growls". One version of a growl is approximated by the relatively non-exotic IPA transcription [ɹ̰ɑ̰ɹ̰ʷ]. You'd really need a set of recorded samples to match against supposed IPA transcriptions. I don't think that any of the IPA extension symbols are particularly applicable in describing such noises.

Whether or not such sounds have a place of articulation at all depends on what you mean by "place of articulation". A traditional phonetic view is expressed by Ladefoged and Maddieson The sounds of the world's languages, that POA is a property of consonants (not vowels). But of course you are free to define POA however you want, and phonologists generally assign a place of articulation (or, more accurately, a "place") to all segment except those occasionally claimed to be placeless. A grunt is like schwa, so the POA of a grunt would be whatever you say about schwa.

A sub-glottal POA is an impossibility -- there's nothing to articulate, below the glottis. All sounds somehow involve the larynx and lungs, but that does not give sounds a laryngeal or pulmonic POA.


First, you will be surprised how many sounds IPA (the International Phonetic Alphabet) can describe when you look at the full charts.

There are additional IPA symbols for "disordered speech" (see, e.g., http://teaching.ncl.ac.uk/ipa/consonants-extra.html ) extending the repertoire.

It anything else fails, you can resort to higher level markup (e.g., XML tags like <grunt/> or <grmbl/>).

And yes, they are legitimate parts of a linguistic system.

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.