March 19, 2005

Saying more with less

The students in my Advanced Introduction to Linguistics course (for grad students in other departments) had a textbook exercise (from Akmajian et al.) on subject omission in informal English -- specifically, about the omission of dummy IT plus following form of BE in declarative sentences:

Odd that Mary never showed up.
Too bad (that) she had to leave town so soon.
Amazing that he didn't spot the error.

One student, Tyler Schnoebelen, added a footnote to his answer:

Interestingly, when you drop it and be from the beginning of these sentences, the resulting sentence feels like much more of a glimpse into the thoughts and feelings of the speaker.  "Odd that Mary never showed up" is in some ways closer to "I think that it is odd that Mary never showed up."

This struck me as a perceptive observation: the Subject Omission and the I think versions are more expressive/subjective, while It is odd that Mary never showed up is more reportive/objective.  This is an old observation for I think, but I don't recall having seen it for omission of dummy subject IT (which is usually lumped together with other types of Subject Omission, as in Taught three classes yesterday and See any penguins yesterday?, though I'm coming to think that there are several different constructions here).

I was moved, in fact. to think about a possibly parallel difference in the Article Omission cases, which also came up in the homework exercise:

    The problem is, we have to leave now.
    Problem is, we have to leave now.

These do seem subtler to me; still, I suspect that more than simple economy of expression is going on in the Article Omission cases.

Indeed, though economy of expression, or brevity, was surely the original motive for people's omitting highly predictable sentence-initial material, in a kind of strategic reshaping of linguistic material, the conventionalization of the results seems to have followed quickly on this bit of creative language use, and rather "specific contributions to the pragmatically conveyed meaning" (quoting Geoff Pullum's words to me about these examples) came to be associated with what started as phonological elision in a weak position.

In other correspondence on this phenomenon, Elizabeth Traugott noted that this case "fits in with various well-known phenomena like absence of 'that' after verbs like 'see'. Generally, less structure seems to be equated with more subjective [my emphasis], except in the domain of discourse markers, where the more one has the more (inter)subjective things get!"  That gets things started; then they become conventional, and in fact grammatical.

Meanwhile, Eve Clark noted,  "I think it's something of a convention in writing that one uses a lot more ellipsis to convey 'internal feelings, attitudes'.  Does this spill over to actual spoken usage?"  Good question. 

And Bruno Estigarribia, our local expert on Initially Reduced Questions, tells me that "Haegeman finds a lot of subject deletion in root and in embedded clauses (the latter being usually considered not acceptable) in English diary registers.  That also suggests a correlation between 'internal discourse' and subject deletion (which reminds me slightly of Bakhtin 1981)." His references:

Bakhtin, M. M.  1981.  The Dialogic Imagination: Four Essays.  Trans. Caryl Emerson and Michael Holquist.  Ed. Michael Holquist. Austin: Univ. of Texas Press.

Haegeman, Liliane. 1997. Register variation, truncation and subject omission in English and in French.  English Language and Linguistics 1.233-70.

Haegeman, Liliane & Tabea Ihsane. 1999. Subject ellipsis in embedded clauses in English. English Language and Linguistics 3.117-145.

So, lots of things to think about.

zwicky at-sign csli period stanford period edu

Posted by Arnold Zwicky at March 19, 2005 11:38 PM