hem_,--a method by which, according
to his publishers notice, "The ordinary unphilosophical explanation of this
class of words is discarded, and a simple, intelligible, common-sense view
of the matter now _for the first time_ substituted,"--I know not what
novelty there is in it, that is not also just so much _error_. "Compare,"
says he, "these two sentences: 'I saw _whom_ I wanted to see;' 'I saw what
I wanted to see. If _what_ in the latter is equivalent to _that which_ or
_the thing which, whom_, in the former is equivalent to _him whom_, or _the
person whom_."--_Butler's Practical Gram._, p. 51. The former example being
simply elliptical of the antecedent, he judges the latter to be so too; and
infers, "that _what_ is nothing more than a relative pronoun, and includes
nothing else."--_Ib._ This conclusion is not well drawn, because the two
examples are _not analogous_; and whoever thus finds "that _what_ is
nothing more than a relative," ought also to find it is something less,--a
mere adjective. "I saw _the person whom_ I wanted to see," is a sentence
that _can scarcely spare_ the antecedent and retain the sense; "I saw
_what_ I wanted to see," is one which _cannot receive_ an antecedent,
without changing both the sense and the construction. One may say, "I saw
what _things_ I wanted to see;" but this, in stead of giving _what_ an
_antecedent_, makes it an _adjective_, while it _retains the force of a
relative_. Or he _may insert_ a noun before _what_, agreeably to the
solution of Butler; as, "I saw _the things_, what I wanted to see:" or, if
he please, both before and after; as, "I saw _the things_, what _things_ I
wanted to see." But still, in either case, _what_ is no "simple relative;"
for it here seems equivalent to the phrase, _so many as_. Or, again, he may
omit the comma, and say, "I saw _the thing_ what I wanted to see;" but
this, if it be not a vulgarism, will only mean, "I saw _the thing to be_
what I wanted to see." So that this method of parsing the pronoun what, is
manifestly no improvement, but rather a perversion and misinterpretation.

But, for further proof of his position, Butler adduces instances of what he
calls "_the relative_ THAT _with the antecedent omitted_. A few examples of
this," he says, "will help us to ascertain the nature of _what_. 'We speak
_that_ we do know,' _Bible_. [_John_, iii, 11.] 'I am _that_ I am.'
_Bible_. [_Exod._, iii, 14.] 'Eschewe _that_ wicked is.' _Gower_. 'Is it
pos