is the art of speaking and writing with propriety” who that really values clearness and accuracy of expression, can think the want of them excusable in models prescribed for the exercise of parsing?  And is it not better to maintain the distinction above named, than to interlace our syntactical parsing with broken allusions to the definitions which pertain to etymology?  If it is, this new mode of parsing, which Kirkham claims to have invented, and Smith pretends to have got from Germany, whatever boast may be made of it, is essentially defective and very immethodical.[219] This remark applies not merely to the forms above cited, respecting the pronoun what, but to the whole method of parsing adopted by the author of “English Grammar in Familiar Lectures.”

OBS. 35.—­The forms of etymological parsing which I have adopted, being designed to train the pupil, in the first place, by a succession of easy steps, to a rapid and accurate description of the several species of words, and a ready habit of fully defining the technical terms employed in such descriptions, will be found to differ more from the forms of syntactical parsing, than do those of perhaps any other grammarian.  The definitions, which constitute so large a portion of the former, being omitted as soon as they are thoroughly learned, give place in the latter, to the facts and principles of syntax.  Thus have we fullness in the one part, conciseness in the other, order and distinctness in both.  The separation of etymology from syntax, however, though judiciously adopted by almost all grammarians, is in itself a mere matter of convenience.  No one will pretend that these two parts of grammar are in their nature totally distinct and independent.  Hence, though a due regard to method demands the maintenance of this ancient and still usual division of the subject, we not unfrequently, in treating of the classes and modifications of words, exhibit contingently some of the principles of their construction.  This, however, is very different from a purposed blending of the two parts, than which nothing can be more unwise.

OBS. 36.—­The great peculiarity of the pronoun what, or of its compound whatever or whatsoever, is a peculiarity of construction, rather than of etymology.  Hence, in etymological parsing, it may be sufficient to notice it only as a relative, though the construction be double.  It is in fact a relative; but it is one that reverses the order of the antecedent, whenever the noun is inserted with it.  But as the noun is usually suppressed, and as the supplying of it is attended with an obvious difficulty, arising from the transposition, we cut the matter short, by declaring the word to have, as it appears to have, a double syntactical relation.  Of the foregoing example, therefore—­viz., “From what is recorded,” &c.,—­a pupil of mine, in parsing etymologically, would say thus:  “What is a relative pronoun,