A few days after I got done [saying](http://noncompositional.com/2008/07/as-you-leave-i-was-wondering/) that _as_ used in the “speech act” sense was nearly impossible, though not uncommon in play-by-play commentary, I encountered an example.
At a dinner I was at recently, one participant remarked that a roommate would continually make requests indirectly, e.g. “Do you think you’ll do the dishes?” “I wonder if we should do some cleaning this weekend.” [language changed to protect the innocent]. She expressed some frustration with that sort of talk, wishing that the roommate would “be direct” and just say
> Can you please do the dishes?
In case anyone was wondering if, maybe, somehow, _can you X_ was still only indirectly a request.
A topic of continued interest (primarily by(?) Geoff Pullum) is [linguification](http://www.google.com/cse?cx=001269089414569134552%3Aqvjtfauf7ou&ie=UTF-8&q=linguification&sa=Search), or the expression of a particular idea or argument in terms of language. A couple years ago I [wrote about](http://noncompositional.com/2006/10/when-folk-linguify/) one particular type of linguification that often takes the form of
> You rarely hear X and Y in the same sentence
and which expresses, basically, X and Y are really different and totally unrelated. In case X is a modifier of Y, then it means Y is anything but properly described by X (e.g., you never hear “politician” and “honest” within three words of each other).
In the June 2008 issue of _Language_, Frederick (Fritz) Newmeyer writes, in a footnote:
> The late Jim McCawley wrote somewhere that he can always pick out theoretical linguists at academic cocktail parties. We are the ones who talk about the Fibonacci sequence, the laws of thermodynamics, and Romance clitic climbing, all in the same sentence.
This reflects what I guessed was a folk theory of discourse, one part of which is the idea that sentences have only a single topic, and that nothing is present in the scope of negation (e.g., “studying Romance clitic climbing clearly has nothing to do with the laws of thermodynamics, which is just as well considering that I know nothing about the latter”). This particular form of linguification relies on the idea that if any concepts are mentioned in the same sentence, this indicates at least that the speaker is interested in all three topics, and perhaps even that the speaker is somehow arguing for a significant relation between them. (Or perhaps just that they are [concept-dropping](http://hayesdavis.net/2008/05/11/concept-dropping/) in order to impress their colleagues).
I was surprised and very happy about the sudden interest on LL in “yeah no” ([initial post](http://itre.cis.upenn.edu/~myl/languagelog/archives/005523.html) and the [aftermath](http://itre.cis.upenn.edu/~myl/languagelog/archives/005525.html))since I’ve also been paying attention to the little guy in some recent research (following up, in a sense, from two earlier posts of mine [here](http://noncompositional.com/2006/09/no-yeah/) and [here](http://noncompositional.com/2006/09/no-yeah-i-really-said-it/). I won’t reveal too much now (have to keep you in suspense!) but there’s one use of it which is really cool, and which is illustrated by LL’s own Geoff Nunberg on NPR.
Get yourself an archived recording of Talk of the Nation on June 2nd, 2004. It’s a discussion between Neal Conan and Geoff Nunberg on his then-new book _Going Nucular_. At about 10m30s into the session, they’re taking calls, and you’ll hear this:
> JIM: Well, three categories here. I’ll do them real quickly. First I guess I’ll call it the category of the cachet of erudition. ((dozens of seconds omitted; some talk of the word _robust_ included near the beginning))
> CONAN: Any thoughts on those, Geoffrey?
> Mr. NUNBERG: Yeah. No. I think ‘robust,’ for example, is an instance of one of those vogue words that for one reason or another is just picked up and people like the sound of it. You’re right. I don’t know if it’s erudition but there’s a kind of pleasure in saying a word like that and everybody plugs into it.
This sort of thing happens not infrequently on NPR (and other radio talk shows, I presume). It might even be getting to the point where “yeah no” (and sometimes just “no”) is almost just a question-uptake marker more than anything else. Listen for it. Other good places to listen for it: asymmetric social contexts, especially where one person constantly feels the need to maintain a non-disagreement or positive-alignment stance with other people there, either just in order to be nice (think job interviews) or to prevent themselves from being misunderstood (think any academic context).
It’s too bad that neither Burridge and Florey, nor Moore’s thesis from what I saw, take up the task of comparing “yeah-no” and “no-yeah” to the use and distribution of _yeah_ and _no_ individually. Maybe I’m just too much of a lexical semanticist. But, if they had, they might have found what I did in going through a bunch of LDC mixer corpora (as well as some more natural conversations): namely, that in most but not all cases there’s not a whole lot special about the combination in and of itself. It doesn’t seem to fulfil any core function that either “yeah” or “no” aren’t observed to do on their own (assuming you assign some combination of propositional and discursive meaning to each _yeah_ and _no_ when they do appear on their own, and also when they appear in combination). There may be cases where it is appropriate or even ideal to use both, as many of Mark’s readers pointed out (like in responses to negative or leading questions), but the resulting “meaning” is not really anything beyond what you might expect by putting the two together. I think that’s the case with most of the examples in the two LL posts. However, Australia could be a completely different story. In particular, if you have access to the Burridge and Florey article, check out the “athletic” use, which to my ears is actually rather strange.
Finally, it may interest my more sociologically- or CA-minded readers (all none of them?) that this sequence of words was analyzed by Schegloff in a 1992 paper, “Repair After Next Turn: The Last Structurally Provided Defense of Intersubjectivity in Conversation”, Am. J of Soc. He also looked at just “no” in “Getting Serious: Joke->Serious ‘No'” (2001, J of Prag.), but this “no” can also be combined with “yeah” resulting in a very similar effect. This is the use that I don’t think any of the LL-responders has mentioned, where you say something joking as a sort of ice-breaker, and then “(yeah) no I just wanted to ask you about….” This is as opposed to Mark’s (right-on, by the way) characterization of the “no” as indicating “divergence from (perhaps shared) presuppositions or expectations” — note the similarity to “intersubjectivity” in Schegloff’s title.