I think Johanna is referring to cohesion in the following passage. > I have said over and over that sentence-level grammatical choices (such as > subject, coding of various referents via grammatical relations like > possessive or object of preposition, and other listers have reminded us > of things like sentence-initial adverbials) are responsible for text > coherence. It seems difficult to me to maintain that "sentence-level grammatical choices" are responsible for "the underlying semantic relations that allow a text to be understood." By definition, "sentence-level grammatical choices" are cohesive ties. > Aspects of text coherence that sentence structure determines > are things like salience of a topic, connection between main and > subtopics, and general information flow, such that the reader can track > referents of the text and connect new information to the correct > referent. Witte and Faigely observe "there is no evidence to suggest that a large number (or a small number) of cohesive ties of a particular type will positively affect writing quality." This passage by Johanna demonstrates an interesting problem of coherence as I have defined above. > One well-known example of a violation of such coherence is > dangling modifiers, or modifiers placed such that the identity of the > head is ambiguous. I'm not yet far enough into this field to know > whether the research is there to nail down other violations. But a > little work I have done myself and the little I have read in > functionalist discourse analysis points to some robust correlations > between grammatical structure and text purpose. Note how the passage begins with a claim about dangling modifiers and ends by making a claim about correlations between grammatical structure and text purpose. The writer of the passage connects those two claims by citing the amount of research she has done on the two topics. I don't think that is an obvious connection to the general reader. The observation I have just made has nothing to do with the grammatical choices the writer has made. In fact, the but is a cohesive tie and appears appropriate as a signaling of the contrast between the research on dangling modifiers with robust correlations on grammar and text. Of course, email posts are not polished pieces of writing. I know that I have typos in my own posts and leaps just as egregious. One aside about this passage. I would never classify dangling modifiers as a problem of the underlying ideas in a text being confused. My experience is that students know exactly what their string means but are not aware of an unintended meaning. Another aside about this passage. I am familiar with Biber et al, A grammar of spoken and written English and have used their descriptions for some of my own work, but I have never figured out what they mean for teaching. Perhaps, Johanna could share with us an example or so of a robust correlation which should inform the teaching of writing. For example, a while back on this list, someone claimed that the past perfect in English is disappearing. Biber et al (p. 461) show that in every million words of fiction in their corpus the past perfect occurs approximately 5,000 times as compared to only 3,500 times for the present perfect. This is the only register ( the others are conversation, news, and academic writing) in which occurrences of the past perfect are more than the present perfect. Does this mean that a student should track her use of the past perfect in fiction writing and make sure it is used more times than the present perfect? Does this mean that if one teaches fiction writing, then the past perfect should be taught? Johanna, asks some very interesting questions about grammar choices. I wonder if she is talking about the grammar we find in polished writing or the writing of our students. > Grammatical choices have a lot of control in this area as well. Choice > of pronouns and other point-of-view establishers are strong indicators > of how close a social relationship to the reader the author wishes to > convey. For instance, are 1st- and 2nd-person pronouns used, instead of > more objective-sounding and exclusive (of writer and reader) 3rd-person > pronouns? Is information stated in the form of questions, giving the > illusion of direct communication between writer and reader, or are > declaratives used, creating greater distance between writer and reader? > Are imperatives used, inviting the reader to participate directly in > constructing the meaning of the text? Imperatives, because they suggest > the writer's right to directly command the reader, also imply a closer > relationship between writer and audience. Sentence fragments increase > the informality of a text, suggesting a closer social relationship > between writer and audience than strict adherence to complete sentences. > This is why we see lots of fragments in ads, but few in the text of a > magazine article. It seems to me that Johanna is describing the grammatical choices which accomplished writers make. On Friday, my department spent all day looking at first year students essays. We do this to make sure that we are all evaluating student writing the same way. In the future, this may become a barrier exam. Almost all of the least successful essays had sentence fragments, but those sentence fragments had absolutely nothing to do with the student trying to be informal. It appeared to me they occurred in certain parts of the text because the student either did not control more appropriate grammar to use, or, because of time pressure and the cognitive demands of writing down her ideas, was unable to realize she had just used a sentence fragment. In the essays I read on Friday, students were not using 1st and 2nd person pronouns to be more subjective and questions were not used to have a communication with the reader. I agree with the following (although I would substitute cohesion for coherence). > I remain convinced that the most interesting thing we can teach about > sentence-level grammar is the way it contributes to the coherence and > style of a text. This is something writing teachers will clamor for. We > are certainly beginning to approach grammar teaching in this way (as > with Kolln's book on rhetorical effects of grammatical choices, and > Dykstra's work on speech structure vs. writing structure), but we also > have a long way to go. For anyone interested in understanding the acquisition of writing as the use of grammatical structures not found in the spoken language, I recommend the work of Katherine Perera (1984) Children's Writing and Reading or her chapter in Fletcher and Garman's (1986) Language Acquisition 2nd Edition on The acquisition of writing. Bob Yates, Central Missouri State University 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/