ch
advance as I have made in technique--and I trust I have made some--I
have owed to the critical running analysis of the construction of
sentences, which has been my habit ever since I began to write. That
this is constant with me, subconsciously, is shown by the frequency
with which it passes into a conscious logical recasting of what I
read. To get antecedents and consequents as near one another as
possible; qualifying words or phrases as close as may be to that which
they qualify; an object near its verb; to avoid an adjective which
applies to one of two nouns being so placed as to seem to qualify
both; such minute details seem to me worthy of the utmost care, and I
think I can trace advance in these respects. My experiments tend to
show that the natural order of nominative, verb, object, is usually
preferable; and as a rule I find that adverbs and adverbial phrases
fall best between nominative and verb. Still, the desirability of
tying each period to its predecessor, as does the rhyme of the fourth
and fifth lines of a sonnet, will modify arrangement. In reading
another author, where such precaution as I name is neglected, a word
misplaced in its relation to the others of the sentence runs my mind
off the track, like an engine on a misplaced switch, and I dislike the
trouble of backing to get on the right rails. It is the same with my
own work, if time enough elapses between composition and subsequent
reading. Generally I make such time, either in manuscript or proofs;
but I am chagrined when I meet slips in the printed page, as I too
often do. There is no provision against such fault equal to laying the
text aside till it has become unfamiliar; but even this is not
certain, for construction, being consonant to your permanent mode of
thinking, may not when erroneous jar upon you as upon another.

In acquiring an automatic habit, which technique should become,
principles tend to crystallize into rules, and a few such I have;
counsels of perfection many of these, too often unrealized. I do not
like the same word repeated in the same paragraph, though this lays a
heavy tax on so-called synonymes. Assonances jar me, even two
terminations "tion" near together. I will not knowingly use "that" for
"which," except to avoid two "whiches" between the same two periods.
The split infinitive I abhor, more as a matter of taste than argument.
I recognize that it is at times very tempting to snuggle the adverb so
close to the ver