Bob,
Most extant theories of language provide an explanation of why that sentence is ambiguous (I'm not sure whether to read your "A formal theory of language..." as restrictive or not). I'm sure there is a theory somewhere that completely rejects the notion of hierarchical element grouping, but I doubt it gets out much. There are a number of theories that reject the idea that hierarchical element grouping is specific to a language faculty, arguing instead that humans do this kind of thing in a lot of domains, but that's a different issue.
So, yes, there are theories that try to account for exactly what you're talking about -- they do so by discussing grammatical/rhetorical relationships above the sentence level, and by not assuming an *essential* distinction between grammar and pragmatics. I don't know of any of these that would claim a single sentence, without any context, is effective, however. What looks like an effective one-liner is effective because of its context, not because of the absence of such. A deliberately contextually-weird one-liner used as a comedic opening (e.g., starting a routine with the single sentence, "However, no" followed by a silence) operates by taking as context people's expectations that the speaker will follow standard maxims of conversation, a la Grice.
For your specific example, we just need to start by positing that, when speakers encounter a sentence that can be read multiple ways, at some level they assign interpretations (including structural info) for all the possible readings, with the readings being in competition. In many, if not most cases, one reading will be strongly supported by previous context/text, while other readings will be hampered; the reading with the most support and least "suppression" will "win" (network theories, for example, usually discuss things functioning in this way). Our conscious perception of ambiguity is usually triggered by a context in which two or more of the interpretations are running neck-and-neck.
In your pair of examples, the first sentence has multiple readings, at least two of which remain in contention. Absent strongly-predispositive previous context (and this kind of joke tends to rely, as part of its setup, on an immediately previous topic shift), the weirdness of elephant-in-pajamas imagery suppresses the adj-PP reading. The second sentence strongly supports the adj-PP reading, though, reversing the original weighting. This is how a lot of humor works, and as a prerequisite for dealing with it analytically, we need to be able to talk about how a sentence can be "about" previously-existing expectations, be they from immediately prior linguistic input (as in your example), or from social norms ("So, I found out that when you're taking a test on classroom management, saying that you'd put the kids in Velcro vests so you could stick them to carpet patches on the wall probably isn't a good idea"). In short, you need a view in which it's elephants all the way up.
--- Bill Spruiell
-----Original Message-----
From: Assembly for the Teaching of English Grammar [mailto:[log in to unmask]] On Behalf Of Robert Yates
Sent: Thursday, May 12, 2011 7:50 PM
To: [log in to unmask]
Subject: Re: study shows effectiveness of grammar: was transformational versus traditional
Bill,
To the degree that a text can be misunderstood, we could consider such a text not nearly as effective as one that is less likely to be misunderstood. I'm not quite sure that any theory of language can be founded on that principle.
A formal theory of language provides an explanation for why the following sentence is ambiguous.
1) One morning I shot an elephant in my pajamas.
So, does this ambiguity mean this is not an effective text?
Yet, if this sentence is followed by (2), is it now effective?
(2) How he got into my pajamas I'll never know.
As a text, those two sentences are relatively famous. I have no idea what theory of language tries to explain why those two sentences together appear to be an effective text, but sentence (1) is not effective.
Bob Yates, University of Central Missouri
>>> "Spruiell, William C" <[log in to unmask]> 05/12/11 5:14 PM >>>
[Advance warning: this is just the usual me-arguing-with-Bob thing]
Bob,
The point I was making was that verb tense ties in to coherence; I was using the fact that randomization ruins coherence as *evidence* of that connection (which it is). You had opened your post with the statement, "I know of no theory of language which lays out the principles of an 'effective text'." I considered as my rhetorical task to establish that (a) you don't need a definition of ultimate effectiveness as a prerequisite to saying that some texts are more effective than others, (b) grammatical choices can move things on the effectiveness scale, and (c) plenty of theories do talk about how such choices affect effectiveness.
The argument is only a red herring if (1) saying that effectiveness and coherence are related has no bearing on this argument, or (2) you think I was saying that beginning writers randomize their choices. As for (1), do you think that? As for (2), where did I say that? I said randomization would hurt coherence, which it will. It's a basic transitive chain -- if verb tense choice affects coherence (and randomization effects act as proof of that), and coherence affects effectiveness (I'm relying on general consensus for that one), then verb tense affects effectiveness, at least to some extent.
In short, I wasn't saying that beginning writers randomize anything. And I wasn't saying that you don't talk about verb tense selection. What I do think is that when *you* talk about tense selection across a paragraph, you probably believe you're talking about a fundamentally different domain than when you talk about subject-verb agreement in a clause, and that the phrase "grammatical theory" is something you believe applies to the latter domain only. That's the part I think is a definition issue.
--- Bill Spruiell
-----Original Message-----
From: Assembly for the Teaching of English Grammar [mailto:[log in to unmask]] On Behalf Of Robert Yates
Sent: Thursday, May 12, 2011 4:41 PM
To: [log in to unmask]
Subject: Re: study shows effectiveness of grammar: was transformational versus traditional
I love red-herrings in arguments about what developing writers do. Bill writes:
Randomizing verb tense choice in an essay will make it ineffective. Ignoring context established earlier in the text will quite likely result in bad pronoun choices, and hence an ineffective text, etc. The work on genre that Craig mentions has extensively developed descriptions of correlations between particular linguistic choices at specific points in a text and whether the text is considered a good example of science writing, or a good example of a narrative. You may or may not view that kind of correlation as being within the domain of "grammar" -- but that's a definitional issue, and one that I doubt educators are quite so invested in.
* * *
We have to be very careful about the implications of correlations of particular grammar forms in a particular genre. As someone who teaches ESL such correlations help make decisions about what needs to be taught if students are expected to write such texts.
However, let's use an example from Biber et al. They found in fiction that passive with the by-preposition phrase occurred about 500 times for every one million words while in academic writing such passives occurred 1500 times for every one million words. (p. 938)
Does this mean that if I write fiction (their category) and have NO passives with by-prepositional phrases I have composed an ineffective text? Likewise if my academic text has only 850 passives for every 1 million words, is my text ineffective?
***
Here is the red-herring:
Randomizing verb tense choice in an essay will make it ineffective.
Perhaps, I'm fortunate, but my students don't randomize verb tense choices. Perhaps from a mature writing perspective, the non-target-like verb choices in my students' texts may appear random, but I don't think they are. To consider them as random is committing what Robert Bleyv-Roman calls the "comparative fallacy."
As their instructor, I think is my responsibility is to figure out what their principle(s) of tense choice is and show why that principle is inappropriate for the kind of academic text they are trying to write.
Again, Jim Kenkel and I have several papers looking at both native and non-native speaker writing from this perspective.
Bob Yates, University of Central Missouri
To join or leave this LISTSERV list, please visit the list's web interface at:
http://listserv.muohio.edu/archives/ateg.html
and select "Join or leave the list"
Visit ATEG's web site at http://ateg.org/
To join or leave this LISTSERV list, please visit the list's web interface at:
http://listserv.muohio.edu/archives/ateg.html
and select "Join or leave the list"
Visit ATEG's web site at http://ateg.org/
To join or leave this LISTSERV list, please visit the list's web interface at:
http://listserv.muohio.edu/archives/ateg.html
and select "Join or leave the list"
Visit ATEG's web site at http://ateg.org/
To join or leave this LISTSERV list, please visit the list's web interface at:
http://listserv.muohio.edu/archives/ateg.html
and select "Join or leave the list"
Visit ATEG's web site at http://ateg.org/
|
