Dorothy Augustine's Geometries and Words: Linguistics and Philosophy: A Model of the Composing Process (College English, 43 [1981], 221-31) illustrates how radically our understanding of the composing process has changed from the linear schemes of the last generation.' However, this new understanding is not always applied to discussions of technical writing. In fact, technical writing is sometimes assumed to be a rhetorically simple process because the rhetorical context of the completed product, the document, is generally limited.2 This assumption is not borne out by practical experience. My own initiatory adventures as a technical writer have led me to the conviction that technical discourse of any seriousness is a structure necessarily created by the writer out of the elements of the writing situation. In other words, the writing situation cannot by itself determine for the writer or editor the meaning of the technical document to be produced; in a fundamental sense, technical discourse is a lamp upon rather than a mirror of the world it represents. Of course, not all technical writing is complex; the IRS form 1040A is simple, not only as product but also as process. Moreover, a given technical document