The task of reporting on the full range of current views on either morphology or syntax, let alone both, is a completely daunting one, and could not possibly be accomplished in anything like a serious way within the scope of a single talk. Instead of attempting it, therefore, I will present some of my own views on some areas where the two can usefully be compared, and suggest that the result of that comparison does not come out quite the way it is often assumed to. A long tradition, reaching back to the beginnings of scientific linguistics in the early twentieth century, has seen morphology and syntax as quite similar areas of structure, at least in terms of the appropriate theoretical principles which underlie them. In structuralist theory, syntax was often seen as an essentially morphological study, extended to structures above the word but other-wise exhibiting the same general organization and accessible to the same methods. Somewhat ironically, perhaps, generative theory long maintained the converse: that morphology was just the syntax of word-internal domains, with no real properties of its own. The last three decades or so have seen a revival of interest in morphology for its own sake, and with that, a recognition that word structure really does have some properties that distinguish it from sentence structure. A close look, indeed, suggests that there are virtually no principles of any substantive sort that the two share, apart from a vocabulary of features that serve at the interface between syntax and the lexicon (Anderson 1992: Ch. 3). Despite this, the assumption persists in the field that morphology is 'really' like syntax, in that the structure of words consists of a hierarchical arrangement of fundamentally concatenative units ('morphemes') quite parallel to the arrangement of words into phrases. On that picture, the fundamental operations of both morphology and syntax consist in (a) selecting a set of content units - morphemes or words, as the case might be; (b) grouping these into larger structures - words or phrases; and then (c) computing the appropriate phonetic forms which should be assigned to the result. I will suggest below that this conception of our knowledge of words is significantly flawed, and that morphology has a somewhat different character from that implicit in the traditional division of the field into the study of 'morphotactics' and 'allomorphy' So much, I think, is really just a summary of what we have come to know by taking morphology seriously as an area of inquiry in its own right. Somewhat less of a consensus exists, however, with respect to the other part of what I have to say. I will go on to suggest that in some areas outside the domain of word structure - specifically, the analysis of incorporation constructions, clitics, and 'Verb-second' structures, the insights we get from the study of words are also applicable, and that these traditionally 'syntactic' areas are thus more like morphology than is usually assumed.
