do not like to leave any thing to chance or
to discretion, when we have a _clear principle_ for our guide."--_Ib._,
243. This author's "clear principle" is merely his own confident
assumption, that every form of figurative or implied agreement, every thing
which the old grammarians denominated _zeugma_, is at once to be condemned
as a solecism. He is however supported by an other late writer of much
greater merit. See _Churchill's New Gram._, pp. 142 and 312.

OBS. 7.--If, in lieu of their fictitious examples, our grammarians would
give us actual quotations from reputable authors, their instructions would
doubtless gain something in accuracy, and still more in authority. "_I or
they were offended by it_," and, "_I, or thou, or he, is the author of
it_," are expressions that I shall not defend. They imply an _egotistical_
speaker, who either does not know, or will not tell, whether he is
_offended_ or not,--whether he _is the author_ or not! Again, there are
expressions that are unobjectionable, and yet not conformable to any of the
rules just quoted. That nominatives may be correctly connected by _or_ or
_nor_ without an express agreement of the verb with each of them, is a
point which can be proved to as full certainty as almost any other in
grammar; Churchill, Cobbett, and Peirce to the contrary notwithstanding.
But with which of the nominatives the verb shall expressly agree, or to
which of them it may most properly be understood, is a matter not easy to
be settled by any _sure_ general rule. Nor is the lack of such a rule a
very important defect, though the inculcation of a false or imperfect one
may be. So judged at least the ancient grammarians, who noticed and named
almost every possible form of the zeugma, without censuring any as being
ungrammatical. In the Institutes of English Grammar, I noted first the
usual form of this concord, and then the allowable exceptions; but a few
late writers, we see, denounce every form of it, exceptions and all: and,
standing alone in their notions of the figure, value their own authority
more than that of all other critics together.

OBS. 8.--In English, as in other languages, when a verb has discordant
nominatives connected disjunctively, it most commonly agrees expressly with
that which is nearest, and only by implication, with the more remote; as,
"When some word or words _are_ dependent on the attribute."--_Webster's
Philos. Gram._, p. 153. "To the first of these qualities