"tsk tsk"

[Beli Orao]

Where does this phrase come from? Does it go back to PIE? There is a cognate in South Slavic languages (c c [ts ts]).

[Jetboy]

I'm pretty sure it's paralinguistic, not a phrase, so it likely doesn't go back to PIE. It actually originally represented a velaric ingressive click, I believe, though I didn't make the connection between the orthographical representation and the actual sound until it was mentioned in Catford's Practical Introduction to Phonetics, mainly because /tɪsk tɪsk/ is so often used sarcastically as opposed to the actual click. But anyways, the Slavic word is probably convergent, something along the lines of onomatopoeia, since PIE having clicks with only these few identical survivors seems quite unlikely. Apparently tut-tut is supposed to represent the same thing; see here

[tiramisu]

Arabic has [!\_<_w]. I thought this would be useful information to provide, but I have no interpretation to offer.

[Salmoneus]

Jetboy: 'paralinguistic' features are still linguistic features. Why would it be odd for PIE to have such restricted clicks - when English has just as restricted clicks!?

[Jetboy]

I'm not saying it would be odd for it to have one, just that it would be odd for it to survive for four-thousand years as the sole click. Going mainly off John Wells, I was under the impression that, since paralinguistic features can vary so much geographically, they tend to do so temporally as well. Also, according to etymonline, tsk originates in the twentieth century.

[Soap]

That might be just for the spelled out form. Still, I doubt that /!\/ or whatever it is actually goes back anywhere near 4000 years since we'd see more of it in other languages for sure if it did.

[zompist]

It may just be underinvestigated. Dictionaries generally don't have entries for "tsk tsk", interjections, special pragmatic words, etc.

[Radius Solis]

FWIW, paralinguistic clicks are very common around the world, and the feature of having them may easily go back to PIE, who knows. But I doubt the individual pairings of clicks with their uses derives in any traceable way from anything millennia old, just because the pairings seem unstable. The POA of my own various paralinguistic clicks is not highly specified, in that I could register disgust with a dental click or a lateral one, but the lateral one could also register satisfaction or a piquing of interest in something, either of which I could also express with an alveolar click (but never dental). And even the exact number of clicks per usage is not set in stone - it can be one or sometimes two, one of mine consists usually of three (for filling space while thinking), and for at least one purpose (calling pets) there may be an extended string of them. And then other people around me employ clicks in ways I wouldn't and sometimes do not catch the meaning of, and vice versa, and I've seen others misunderstand each others' clicks too. My experience is a limited data sample to be sure, but if it's at all representative, the pairings of pronunciation with meaning for paralinguistic clicks may not be stable enough for us to draw many conclusions about their history. I mean I could be wrong here, it just seems like an awful mess to try to reconstruct any coherent historical picture from.

It's also possible that paralinguistic sound effects like clicks are not really features of particular languages, i.e. that which ones you have might depend much more on your culture, or exactly where you were raised, than it depends on the language you were raised to speak. I have no data either way, I'm just pointing out that this is a factor that we need to examine too before we can really talk about tracing a linguistic genealogy of paralinguistic clicks.

I'm not saying it would be odd for it to have one, just that it would be odd for it to survive for four-thousand years as the sole click. Going mainly off John Wells, I was under the impression that, since paralinguistic features can vary so much geographically, they tend to do so temporally as well. Also, according to etymonline, tsk originates in the twentieth century.

For one thing, we have more than just one click. I employ at least three distinct ones (see above) used in at least half a dozen different ways, and I would expect most people to have at least the two basic POAs, alveolar and dental. Maybe everyone has the lateral one too, I don't know. It can take some paying attention for a while before you find out how many clicks you really do use - snap attempts to inventory them are highly prone to undercounting, as I keep finding when I discover yet another. (I've been paying attention ever since I did a lot of reading up on clicks a couple years ago.)

For another, etymonline can only tell you when "tsk" is recorded that way in text, it cannot tell you how far back the sound goes.

[Gulliver]

The OED puts the earliest instance of tut as c 1529. I'm taking this as basically a variant spelling of tsk, which the OED dates back to 1947 in a book called Phonemics, which I think is good evidence that it pre-dates that in spelling at least as this source lists both tsk and tut as variant spellings of one thing. The earlier tut is listed as /tʌt/, with the later tsk-or-tut as an alveolar click (actually, I think it's more dental, but that's beside the point). Hmm.

This suggests that there are two 'words' - tut /tʌt/ and tsk (or tut) as the click. But we already know that. The trouble is that it's not the sort of thing that was written down very often - writing materials being expensive and literacy being a prestige activity - so sources like the OED are of limited usefulness.

The World Atlas of Language Structures has a page on the paralinguistic use of clicks, complete with a map. There does appear to be some pattern in how clicks are used (contrasting "logical" and "affective"), but there is no breakdown of the phonetics (which clicks mean what). Europe is overwhelmingly affective (expressing a positive or negative effect), but the Mediterranean and wider Indo-European area is overwhelmingly logical ("yes" and "no").

However, languages of the second type, in which para-linguistic clicks have affective usages but not logical ones, are widespread in many different parts of the world. From a geographical perspective, they would seem to constitute a kind of default background upon which other isoglosses may be superimposed in particular regions. (WALS)

It is entirely possible that clicks shifted meaning from affective to logical in some languages (or vice versa), but the geographic distribution is too widespread and too inconclusive to say anything much about where it came from.

Basically - no one knows and it's hard to tell.

[sirdanilot]

As for Dutch, I know these paralinguistic clicks:

[ǀ] (dental click): slight annoyance
[ǀχ:] or [ǀh:]: stronger annoyance. the more annoyed, the longer the χ/h
[ǁ]: when someone states a problem or something, I sometimes do this, followed by "tsja" (which means something like 'well yeah...'). I also know a very, very annoying guy who does this whenever you say something, or whenever he wants to say something. When he does the click, I usually say "oh you want to say something" just to annoy the guy.

[finlay]

Europe is overwhelmingly affective (expressing a positive or negative effect)

[Thomas Winwood]

It actually originally represented a velaric ingressive click

Clicks are lingual ingressive because the rear articulation isn't significant, and varies depending on the shape of the front of the tongue.

[Jetboy]

It actually originally represented a velaric ingressive click

Clicks are lingual ingressive because the rear articulation isn't significant, and varies depending on the shape of the front of the tongue.

Ah, apologies, I was going off Catford– hence the "I believe" immediately following that statement.

[Magb]

This discussion reminds me of how paralinguistic use of pulmonic ingressive sounds, which most people probably think of as being a quirky thing found in a few languages, turns out to be quite common cross-linguistically: http://ingressivespeech.info/.

[blank stare II]

Apparently paralinguistic sounds suffer from regional variation as well:

tsk-or-tut as an alveolar click (actually, I think it's more dental, but that's beside the point)

IMD it is alveolar

[Skomakar'n]

Usually tss in Swedish (often actually pronounced like this, though, rather than as a click), but in Donald Duck comics, it seems to be t-t-t... Another variation is pft, which is the sound of blowing a little air through the lips, in a generally disapproving manner, but with a different intonation and POA (more like [ɸ]?), it can be equivalent to a very small giggle.

[Cathbad]

Where does this phrase come from? Does it go back to PIE? There is a cognate in South Slavic languages (c c [ts ts]).

But that's a click too. Actual [ts ts] sounds to me like someone trying to imitate bird chirping, very badly.