summary of counting clauses

[Terry Au]

Dear Colleagues:

Earlier I posted a question about how to count clauses in frog stories and
narratives in general.  Here are my original query message and a summary of
the responses.

Thanks,
Terry Au
************************************************************************
ORIGINAL QUERY:
My lab is trying to get a handle on counting sentential clauses in Frog
stories (and other narratives).  Our criteria for clauses are:

1. A subject and a tensed verb (e.g., "he ran" will count as 1 clause; "he
ran and she walked" will count as 2 clauses)
OR
2. Small clauses (e.g., "I wanted him to run" would count as 2 clauses
because the subject is different for the two lexical verbs: "I" being the
subject for "wanted;" "he" being the subject for "run")

We count the following as only one clause:
1. Simpe to-infinitives (e.g., "I want to run")
2. Conjoined verb (e.g., "I ate and drank;" "He cried and cried")
3. Conjoined subject (e.g., "John and Mary left")

**************************************************************************
SUMMARY OF RESPONSES:

1. Several colleagues noted that "I ate and drank" should count as 2
clauses, whereas "I ate and ate" should count as 1 (the repetition marks
aspect rather than a second situation. (See the next point.)

2. Most who replied said they have adopted Berman and Slobin's (1994)
definition: "a unit that contains a unified predicate.  By unified, we mean
a predicate that expresses a single situation (activity, event, state)."
Detailed criteria and examples can be found in Appendix II of "Relating
Events in Narrative: A crosslinguistic developmental study (Berman & Slobin
et al., 1994).

3. Different researchers' criteria vary somewhat on subtler points.
Disagreements center on "I want to go" (one clause according to Berman &
Slobin) versus "I want him to go" (two clauses according to B&S).  The
rationales for departing from B&S's criteria are:

Lowry Hemphill <lowry_hemphill at gse.harvard.edu>: We don't count
constructions like "I wanted him to run" as two clauses because only one
narrative unit is accomplished, an expression of intention (we don't view
this sort of clause as different from "I want to run" because the second
person's agency is only hypothetical in "I wanted him to run", not actual).
We do, however, consider something like "I ate and drank" to be two clauses
because it reports two potentially separate narrative events. Labov's work
with Waletzky has guided us here; we've also been influenced by Peterson and
McCabe's work on children's acquisition of narrative structure.

Judy Reilly <reilly1 at mail.sdsu.edu>: we have counted those verbs with
infinitival clomplements and same subject as 2 clauses, but only one
proposition as they reflect only one event.

"Daniela O'Neill" <doneill at watarts.uwaterloo.ca>: I have found a bit
problematic are cases such as the one you mention with respect to small
clauses such as "I wanted him to run". Breaking these up reduces the
readability of the transcript considerably. Although I think it is correct
to separate this into two clauses, I also wonder about whether this fits the
definition of Bermin and Slobin's "single situation". I am not sure whether
in an everyday sense we consider this utterance the expression of two
situations rather than one. However, Bermin and Slobin do mention that
utterances with subordinate complement clauses such as "he thought he could
get the bees" be considered two situations and be treated as two clauses.
And indeed, we have also treated instances of direct or indirect voicing as
two clauses, as in "he said he would find the frog". Bermin and Slobin also
note that when each verb has a different subject, then the utterance should
be separated in to two clauses and give the example "I want you to go" which
is almost identical to the one you gave.

Claudia L. Ordonnez <ordonecl at gse.harvard.edu>: I count 'I want to go...'
and any other expression with a state-of-mind-verb as 2 clauses to make it
easier to count evaluative features, which are quite important in my coding
syste. The 'to go' would be a predicted event (another evaluative feature)
or a story event, in case it really happens later in the story.

Martha Shiro <mshiro at reacciun.ve>: I was interested in evaluative language,
therefore it was important for me to mark the difference between 'I ran' and
'I wanted to run' and so, I divided the latter into 2 clauses. But in the
case of 'I started to run' I considered it as 1 clause because it had an
aspectual
reference (it referred to a phase of the action of 'running' and not to a
mental process as in the previous case).

Barbara Zurer Pearson <bpearson at comdis.umass.edu>: "I plan to go" would be 2
(in my opinion).  I'm not sure but I think "He tried to go" might also be 2,
whereas I thought of "wanted to go" as "modal-like" (as did B&S) [and hence
counts as one].  The more examples one gives, the slippery-er it feels.
Sometimes that's the best way to clarify one's own thinking.

4. Lowry Hemphill sent me his clausing manual--easy to understand and
similar to B & S on major points.  Interesting ideas on how to deal with
fragments and unfinished sentences.

5. Lois Bloom <lmb32 at columbia.edu> send me references for relevant articles
to read (which I did and found them useful):

Bloom, L., Lahey, M., Hood, L., Lifter, K., & Fiess, K. (1980). Complex
sentences: Acquisition of syntactic connectives and the semantic relations
they encode. Journal of Child Language, 7, 235-261.

We called "complex sentences" those that "include two verbs" or that
"theoretically combine the structures that underlie two simple sentences."
We subsequently did a number of studies of kinds of complex sentences with
different clausal relations, including  those with complements and
expressions of causality in particular.  All of these papers are reprinted
in
Bloom, L. (1991). Language development from two to three. New York:
Cambridge University Press.








*******************************
Terry Kit-fong Au
Professor
Department of Psychology, UCLA
Los Angeles, CA 90096-1563, U.S.A.

Office: (310)206-9186
Lab: (310)206-7840
FAX: (310)206-5895