we have been aiming at a parsing of sentences, and have been mainly concerned with the kinds of rules which enable us to identity the constituents of sentences, and their structures.
these rules can be used un analysis, but if they were formulated precisely enough,