avouches in the same breath in which he gives that "nature" to a
participle and its adverb! For, by a false comma after _much_, he cuts his
first "_substantive phrase_" absurdly in two; and doubtless supposes a
false ellipsis of _by_ before the participle _performing_. Of his method of
resolving the second example, some notice has already been taken, in
Observations 4th and 5th on Rule 5th. Though he pretends that the whole
phrase is in the objective case, "the truth is, the assertion grammatically
affects the first word only;" which in one aspect he regards as a noun, and
in an other as a participle: whereas he himself, on the preceding page, had
adopted from Lowth a different doctrine, and cautioned the learner against
treating words in _ing_, "as if they were of an _amphibious_ species,
partly nouns and partly _verbs_;" that is, "partly nouns and partly
_participles_;" for, according to Murray, Lowth, and many others,
participles are verbs. The term, "_substantive phrase_," itself a solecism,
was invented merely to cloak this otherwise bald inconsistency. Copying
Lowth again, the great Compiler defines a phrase to be "two or more words
rightly put together;" and, surely, if we have a well-digested system of
grammar, whatsoever words are rightly put together, may be regularly parsed
by it. But how can one indivisible word be consistently made two different
parts of speech at once? And is not this the situation of every transitive
participle that is made either the _subject_ or the _object_ of a verb?
Adjuncts never alter either the nature or the construction of the words on
which they depend; and participial nouns differ from participles in both.
The former express actions _as things_; the latter generally attribute them
to their agents or recipients.

OBS. 21.--The Latin gerund is "a kind of verbal noun, partaking of the
nature of a participle."--_Webster's Dict._ "A gerund is a participial
noun, of the neuter gender, and singular-number, declinable like a
substantive, having no vocative, construed like a substantive, and
governing the case of its verb."--_Grant's Lat. Gram._, p. 70. In the Latin
gerund thus defined, there is an appearance of ancient classical authority
for that "amphibious species" of words of which so much notice has already
been taken. Our participle in _ing_, when governed by a preposition,
undoubtedly corresponds very nearly, both in sense and construction, to
this Latin gerund; the principal diff