I have a question about one of the inconsistencies of the Reed- Kellogg system of diagramming sentences. I am wondering if anyone can offer a plausible explanation for this particular flaw in the system (and I am aware that there are many other flaws): (a) Passive sentences are diagrammed according to their surface form, even though a text like Kolln's, for example, classifies the sentences according to the pattern of their 'underlying active versions.' (b) Sentences having undergone 'question' or 'there' transformations, however, are diagrammed according to their 'underlying versions,' and not their surface forms. I suppose that part of the problem may be that the Reed-Kellogg system was developed before the idea of transformations overtook linguistics and we began looking at active & passive sentences as related at a deeper syntactic level. But then that raises the question of why the Reed-Kellogg system shows a relation between the the question and statement forms of a sentence, implying a syntactic relation at a level beneath the surface forms. Having done a little sketching of an alternative diagram for passives, I find it difficult to find a satisfactory alternative using a Reed-Kellogg format. Is such an major inconsistency good reason for abandoning the system altogether? Are there any good reasons for using the system at all? ********************************************************************** R. Michael Medley VPH 211 Ph: (712) 737-7047 Assistant Professor Northwestern College Department of English Orange City, IA 51041 **********************************************************************