re: center embedding

Robert Luk (csrluk@csns02.comp.polyu.edu.hk)
Thu, 14 Mar 1996 10:15:03 +0800

I thought the point of using corpus to find real-life examples is to
get over with the problem to substantiate linguistic investigation using
sentences extracted from an accepted communicative context where decision
of grammaticality or acceptability may not be clear or commonly agreed
between expert or laymen. Effectively, we are trying not to obtain constructed
sentences from "vested interest" (devoid of academic linguistic motivation)
so as to remain more objective. So, the collection of evidence is not just one can construct the sentence to illustrate the linguistic point but something more (more than a philosopher who provide argument but find evidence of argument in a "neutral" setting or known "bias" setting). Perhaps this is not acceptable since interpretation of observation is linked to theories and "neutral" is a bit difficult to define. But, sentences constructed by linguists to falsify opponents argument is not what one feel to be "neutral". My slient applause
to those that check with their corpus and really show that those sentences occur in "neutral (perhaps uncontrolled) communicative context" (A prescritive or descriptive role of linguistic theory/model?).

Robert Luk
Dept. of Computing
Hong Kong PolyU

> This is the rat that the cat chased.
> This is the dog that bit the cat that chased the rat.
> This is the cow that tossed the dog that bit the cat that chased the rat.
> These sentences are the basis of the truncated version of "The
> House that Jack built" that has been under discussion these few days. The
> last one relativizes to, I think: The rat the cat the dog the cow tossed
> worried killed ate (NOT "the rat the cat the dog tossed bit chased",
> which is gibberish). So this is a grammatical sentence. It does not
> follow that it is easy to understand, or even understandable. Grammatical
> sentences which cannot be understood are legion, as I believe has been
> an established principle for forty years at least.
>
> As for whether this sentence "ever occurs", I am a native speaker
> of English and I just wrote it above, so it has indubitably occurred. As
> for the "BNC sampler" I don't know what it is, but give me a live speaker
> any day. Thanks. KVT
>
> If there is something else we are talking about and I am not
> catching on, I stand ready for enlightenment.
>
> Karl V. Teeter kvt@husc.harvard.edu
> Grammarian phone (617)495-8888
> Professor of Linguistics, Emeritus
> Harvard University
> Widener Library Room T
> Cambridge, MA, U.S.A. 02138
>