Bill,
My impression is that "transitivity" is a term that was tied to a verb, and only later to prepositions as adverbs. I think it has been generalized to the notion of c-command.
The branching requirement of generative grammar by which the c-command is defined has been questioned because of word-order "scrambling" that occurs in the patterning in Latin or even more so in Finnish. But let's not go there.
It may be argued that probability distributions are a characteristic of language as produced and compared. When I measure a two-foot board, sometimes it comes out as longer or shorter by some small amount, because of the precision with which I am able to measure it. This need not have anything to do with the board itself, just the standard to which it is compared and the act of comparing it. Its manufacture might also have been faulty. There are several possible origins for the probability distributions we see. Language is changing. Different speakers have different notions about its proper or most appropriate structure. I would think that whether the explanation for this is formalized or not would be up to the person making the observations. Even language play could stand to be explained -- pun, spoonerism, riddle, etc. I'm not sure I would want to make any such phenomena off limits to investigation and explanation.
I think the best way to make the divide between syntax and semantics remains a matter of taste. (to be continued)
Bruce
-----Original Message-----
From: Assembly for the Teaching of English Grammar [mailto:[log in to unmask]] On Behalf Of Spruiell, William C
Sent: Friday, July 11, 2008 2:34 PM
To: [log in to unmask]
Subject: Re: Transitivity vs Intransitivity - The Linguists' Version of Hazing?
Bruce,
When you look at a large number of examples of sentences using a given verb, whan you find, in terms of "+/- use of a direct object," is a probability distribution. "Sleep" is rarely used with a direct object, *but it can be* (I've heard students say things like, "That lecture slept the whole class") -- the probability distribution is thus something like 99.99999% intrans. vs. 0.00001% trans, or even more assymmetric. With a verb like "run," the distribution will be more symmetric. Traditional grammar makes an argument move that says, roughly, "verbs don't _really_ having varying probabilities of use; instead, each verb has exactly one use, and what looks like variation is either due to homonymy, or the kind of language play that does not call for any formal explanation."
In short, the existance of probability distributions enables the "transitivity is tied to the verb" account -- but it doesn't *mandate* such an account. I typically put chocolate only in desserts, but that doesn't mean the chocolate I put in a Mexican mole sauce is somehow a chocolate homonym, or a mistake, or an outbreak of creativity. The assymmetry of chocolate use isn't inherent in the chocolate. The evidence only requires us to recognize the existence of either variation, or what looks like it. Our traditional system grew out of that for describing Latin, and (although I'd want to doublecheck this) my sense of Latin is that verbs were not used nearly as flexibly as in modern English. If *all* verbs fall into that "99.9999 vs. 0.0001" pattern, it makes sense to round off for simplicity's sake -- but that's not English. Had we started with Chinese as a model, we'd probably have less of an urge to pin things on specific words.
In the early to mid-60s, when the debate about whether terms like "subject" and "object" were enough to let you describe what went on with actives and passives, etc., two basic positions emerged early on. One was that grammar didn't need to deal with semantic roles -- "subject" (or "NP directly dominated by S") and "object" ("NP directly dominated by VP) were enough. The other position was that semantic roles had to be acknowledged, and there were three logical possibilities for this. You could say that the arrangement of roles in a clause limited what verb you could use (Fillmore's position, substituting "role" for "deep case"), you could say that clauses came with a process-type-and-role configuration (Halliday's position), or you could say that the verb determined the roles of the NPs that were complements to it (a version of the traditional system). Generative grammar went from denying the utility of semantic roles to accepting them, but when accepting them, it opted for the third approach as instantiated in the work of Gruber. It then quickly changed "semantic role" to "thematic role," then to "theta role," for reasons I'm unsure about (but which I suspect have a lot more to do with the sociology of linguists than with language). Chomsky was uncomfortable with the idea of such roles being part of "core" syntax, and an advantage of the verb-based approach is that it lets you move that type of information completely into the lexicon, which he did in "Remarks on Nominalization."
For the argument this thread is persuing, the key point is that pinning transitivity on the verb is ONE method of dealing with the observed distributions, and -- moreover -- it's a method that's required IF the theorist makes certain other assumptions (syntactic component not driven by semantics, etc.). Start from different assumptions, and you get different options. We can observe differing distributions, but we can't observe *why* there are different distributions -- we can only develop hypotheses and see how far they go. There's nothing wrong with internal consistency, but it can't be mistaken for empiricism.
Bill Spruiell
Dept. of English
Central Michigan University
________________________________
From: Assembly for the Teaching of English Grammar on behalf of Bruce Despain
Sent: Fri 7/11/2008 10:11 AM
To: [log in to unmask]
Subject: Re: Transitivity vs Intransitivity - The Linguists' Version of Hazing?
Craig,
This idea that a sentence can be "transitive" is very irksome to me and I wonder if others have the same difficulty. It seems very elementary that the presence of the direct object in a sentence tells us that the finite verb of the sentence is transitive. I suppose that you want to throw out all the traditional terminology or abrogate it for the non-traditional view of a functional grammar. I think that it would be more productive to build on functional terminology and stop confusing us (just me?) with what looks like a straw man constructed to look very much like a chip on the shoulder.
I have worked a bit with the functional concepts and believe that they can be very fruitful in the improvement of writing and effective expression. I am looking forward to seeing more and more come from that area of analysis. I must say that in my work in knowledge representation and semantic networking, what I have attempted from the functional point of view has not been helpful. I think that functional analysis, as with the relatively new branch of cognitive science, does seem to lack many of the formal elements needed to make it compatible with the forms of logic and mathematical modeling that have been built up from traditional apparatus. Some work is being done to fill this gap, but it has not interfered with or abrogated the formal terms of traditional grammar. I read yesterday [http://linguistlist.org/issues/19/19-2183.html] of the book, Gerard J. Steen, Finding Metaphor in Grammar and Usage (John Benjamins, 2007), where the author proposes methodological strategies to ferret out a formalization of this complex phenomenon into at least eight different relationships or mappings between grammar and usage. (The formalizations are not explicated, but at least a heuristics is made clear showing how to proceed to such a goal.)
Bruce
From: Assembly for the Teaching of English Grammar [mailto:[log in to unmask]] On Behalf Of Craig Hancock
Sent: Friday, July 11, 2008 6:41 AM
To: [log in to unmask]
Subject: Re: Transitivity vs Intransitivity - The Linguists' Version of Hazing?
Bill,
I think John's post is very thoughtful, but I want to go ahead anyway with a sort of contrary, maybe too theoretical position. I have been working on two writing projects that have forced me to think about these issues more than is probably healthy. Now I'm stuck with all these thoughts that need purging.
I suspect you're right, that much of what animated our teaching historically is giving students what they need in order to handle the "shibboleths" of usage. So we decide that students need to master the tricky concept of transitivity so that they, too, can handle sit and set "properly" and look down on those who don't. In the language of the blue collar friends I grew up with, this is stupid. We also have studies that show that grammar doesn't improve reading and writing, and is in fact harmful precisely because it pulls us away from more productive activities, like reading real books and doing real writing, and if this is what grammar is all about, then the studies are absolutely right. It's all trivial.
What's more difficult, I think, is that we are also heirs to a structural grammar tradition that attempted to put grammar study on a more formal, more scientific basis.This meant making observations on the basis of what we could observe about forms and the "rules" that seem to govern them. This leads naturally to classification: on the basis of form, how can we classify a sentence as transitive? Because this is a difficult enterprise, we try to make it easy by focusing on the prototypes. But there is nothing to stop someone from asking the "so what" question and no easy answer to it. If the only thing we are worried about in relation to grammar is error, then it seems natural to dismiss this as a distraction and to reduce grammar down to "error in context" with as little metalanguage as possible.
We can say that grammar has a huge role in the making of meaning, but our central traditions, in both traditional grammar and in linguistics, at least in this country, don't acknowledge that.
To make grammar useful, we need to reconceive it in a very fundamental way.
I don't like to make such a contrary blanket statement, and I hope my many friends in the group understand that this is said in deep respect, but at this point in time we (ATEG) are part of the problem.
Craig
Spruiell, William C wrote:
Craig, Diane, et al.:
I suspect one of the main reasons that both the classification and the amount of flexibility with it are big deals in grammar textbooks is simply the existence of verb sets like "sit vs. set," "lie vs. lay" and the like. Those seem to have turned into shibboleths fairly early, and stayed that way. One of my clearest memories of third grade Englsih was filling out pages of exercises in which I had to figure out whether to put "lie" or "lay" in blanks. We have to talk about transitivity to deal with those sets (and to talk in an organized way about why proto-geezers like me balk at statements like "I graduated high school"). What I think gets left out of the discussion, frequently, is the possibility that it's the existence of "set" that keeps us from using "sit" transitively, and vice versa - in other words, verbs can "block" the flexibility options of other verbs. Where there is no opposing verb, we're much freer to play with the transitivity of the one we have.
I usually start class discussions of transitivity with that "sit vs. set" pair (and "rise/raise"), since students here in Michigan typically use those the same way formal usage demands (unlike the case, of course, with "lie vs. lay"). Those work well as prototypes for comparison, and I do something similar with the transitive and linking interpretations of verbs like "taste" ("The chef tasted the soup" / "The soup tasted too salty"). Having concrete models like that, and then discussing how particular examples fade off into grey areas, seems to work fairly well. What would be a problem is giving them a set of verbs, out of context, and asking them to label the transitivity of each; it's only with the relatively rare "sit/set" type that that's easily possible, I think.
Bill Spruiell
Dept. of English
Central Michigan University.
From: Assembly for the Teaching of English Grammar [mailto:[log in to unmask]] On Behalf Of Craig Hancock
Sent: Thursday, July 10, 2008 12:43 PM
To: [log in to unmask]
Subject: Re: Transitivity vs Intransitivity - The Linguists' Version of Hazing?
Dinae,
Thanks for the response. I'm not at all surprised; that's a pretty good description of what we can expect to find in most grammar texts. Students certainly have difficulty in applying these classifications, so it makes absolute sense to present clear examples of the central prototypes. I just worry that classification becomes an end in itself, and it's hard to fault those who say that studying grammar (learning how to classify sentences this way) doesn't carry over into reading or writing. In fact, when we being to look at the kinds of nuanced choices available to us as language users, the patterns, as defined, aren't practical. The fact that so many sentences aren't easy to classify is a plus for a language user, but not for someone trying to use these patterns as some sort of guiding light. In the attempt to make it simple, we rob it of nuance and of life.
Craig
diane skinner wrote:
Craig,
Without question, the discourse context must be considered,
nonetheless, classifications remain. Klammer et.al. do not state a
"rule" for any conditionals concerning implied direct objects. So, to
this regard, your conclusion would be accurate; however, they define
sentence types "on the basis of prototype sentences, those that
clearly fit a pattern without complication or ambiguity" (209). They
also do not include other considerations, such as the derived
intransitive, an intransitive member of an ergative pair:
The sun melted the ice sculpture.
The ice sculpture melted.
Diane
On Tue, Jul 8, 2008 at 5:32 AM, Craig Hancock <[log in to unmask]> <mailto:[log in to unmask]> wrote:
Diane,
Just to bring this back to the previous discussion, it seems clear that
Klammer et. al.'s approach is to look at the sentence (or clause, since
transitivity happens in all clauses, even downranked ones) rather than the
meaning of the verb apart from that context.
In other words, they don't have a rule that says "If a direct object is
implied, the sentence is transitive."
Would that be accurate?
Craig
diane skinner wrote:
Hi all,
Klammer, Schulz, and Volpe, in Analyzing English Grammar, 4th ed. use
the following definitions for intransitive verbs:
"You can test whether a verb is intransitive by dividing the predicate
into phrases. If all the phrases except the main verb phase are
optional adverbial modifiers, then the verb is intransitive. If you
can substitute a prototypical adverb (like here, then, or slowly) for
the phrase, it is an adverbial phrase.
Ex: The bus stops here each Monday" (p.203).
To explain transitive verbs, they write: (1) "Are the subject noun
phrase (NP) the actor, the verb and action, and the object NP the
'receiver' of the action? if the answer is yes (as in John hit Bill,
where hitting is an action, John performs the action, and Bill
receives the action), then the sentence is probably Type V [i.e.,
contains a transitive verb].
(2) "To find the direct object, ask who? or what? after the subject
noun phrase and verb: John saw who/what? If the answer does not rename
the subject, it should be the direct object" (p. 221).
(3) "Is the verb followed by a direct object, a noun phrase that has a
referent different from that of the subject noun phrase? if so, the
sentence [contains a transitive verb]" (227).
Additionally, they include examples of transitive verbs with reflexive
and reciprocal direct objects: "Elmer cut himself with a Swiss Army
knife" (222),
and transitive verbs with object complements: "Cheryl considered
Carl's bean soup salty" (223).
Grammar is so much fun!!
Diane
On Sun, Jul 6, 2008 at 7:17 PM, Veit, Richard <[log in to unmask]> <mailto:[log in to unmask]> wrote:
Dee, Craig, Tabetha,
Whether we regard a sentence as transitive depends on how we choose to
define the term, and we are choosing different definitions. Our diverse
definitions cover a lot of territory. The options seen in the four postings
(copied below) range from the narrowest to the broadest:
1. To be transitive, the verb must be immediately followed by a direct
object. [Tabetha's definition?]
2. To be transitive, the verb must have an overt direct object (in any
position). [Craig's definition?]
3. To be transitive, the verb must have a deep-structure direct
object, whether or not it is overt in the spoken sentence. [My definition]
4. To be transitive, one must be able to imagine a direct object for
the verb. [Dee's definition?]
These definitions yield different results:
* "He remembered my birthday." Transitive for all four
definitions.
* "He remembered fondly the old days." Transitive for definitions
2, 3, and 4; intransitive for definition 1.
* "Did he remember your birthday?" "He remembered." Transitive for
definitions 3 and 4; intransitive for definitions 1 and 2.
* "He sings in the shower." Transitive for definition 4;
intransitive for definitions 1, 2, and 3.
* "He coughed loudly." Intransitive for all four definitions.
My apologies if I mischaracterized your definition, but we certainly are not
all in accord, and the differences are interesting.
Dick Veit
-----Original Message-----
From: Assembly for the Teaching of English Grammar
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/
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/
________________________________
NOTICE: This email message is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply email and destroy all copies of the original message.
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/
----------------------------------------------------------------------
NOTICE: This email message is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply email and destroy all copies of the original message.
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/
|
