This essay challenges several patterns of thinking common in twentieth-century linguistics. The most pervasive of these is our habit of looking at language from the point of view of the speaker. When we take, instead, that of the hearer, matters fall into place in a new way. In syntax, we are led to examine the evidence available to hearers for interpreting what they hear, and this reveals both the true nature and the locus existendi of “deep structure”. Chomsky's 1957 diagnosis of the then prevalent syntactic theory is upheld, though his proposed remedy is not. The principle of Gestalt perception yields a characterization of the word quite different from Bloomfield's classic definition, lending support of new kind to Pike's mid-century views of the relation between phonemics and grammar. In morphology, assuming the hearer's standpoint forces the abondonment of the “atomic morpheme” that has prevailed in America since the post-Bloomfieldians, together with much of classical morphophonemics, and by a domino effect this in turn undermines much of generative phonology.