Chapter I: Of Concord (3)

_Note_ 10.--Adverbs, as Lowth observes, are generally placed
before the adjective to which they refer. This rule, however,
admits a few exceptions. The adverb _enough_ is always placed
after its adjective, as, “the reward was small enough.” The
proper position of this adverb, indeed, seems to be immediately
after the adjective; it is frequently, however, placed at some
distance from it, as, “a large house enough.” Usage is, indeed,
somewhat divided on this point, Mr. Baker, and a few others,
pleading for the following arrangement, “a large enough house.” The
former collocation, however, seems far the more general; and is
recommended by that rule, by which the substantive and adjective
should be placed in juxta-position, or as near as possible to each
other. The latter is defended by the principle, that the qualifying
adverb should be placed close to the adjective, whose signification
it modifies. This collocation is generally, however, pronounced a
Scotticism; but it is not peculiar to Scotch writers.

RULE XXI.--Conjunctions have no government.

_Note_ 1.--In giving this rule, I differ from all other
grammarians, who have erroneously, as I conceive, assigned them a
regimen. Some conjunctions, says Lowth, govern the indicative, and
some the subjunctive mood. This I affirm without hesitation to be
a great mistake; for not a single example, I venture to assert,
can be produced, in which the verb is divested of its indicative
form, in consequence of its being subjoined to any conjunction.
The Latins had a form of the verb, which they properly enough
denominated the subjunctive mood; because, where the meaning
was unconditionally assertive, they employed this form, if the
clause was preceded by some particular conjunctive or adverbial
term. Thus, when they said, _adeo benevolus erat, ut omnes eum
amarent_, “he was so benevolent, that all men loved him,” though
the assertion in the latter clause, be evidently unconditional, as
the English shows, they changed the indicative into another form,
because the verb is preceded by the conjunction _ut_. No similar
example can be produced in English.

Lowth informs us, that, when hypothesis, conditionality, or
contingency is implied, the mood should be subjunctive; if
certainty, or something determinate and absolute be signified,
the verb should be indicative. Now surely, if the sense require a
form different from the indicative, the verb cannot be said to be
under the government of the conjunction; for the verb assumes that
form, not because preceded by the conjunctive term, or because it
is under its government, but because the sentiment to be expressed
requires that phraseology. Whether the conditional, or what Lowth
terms the subjunctive, be a distinct form of the verb, or only an
elliptical mode of expression, we have already inquired. See p. 126.

_Note_ 2.--Mr. Harris says, that the chief difference between
prepositions and conjunctions is, that the former couple words, and
the latter sentences. This opinion is erroneous; for conjunctions
frequently couple words, as in the following example: “A man of
wisdom and virtue is a perfect character.” Here it is not implied,
that “a man of wisdom is a perfect character;” but “a man of
wisdom combined with virtue, or a man of wisdom and virtue.” That
conjunctions, indeed, do not couple at all, in that sense, at
least, in which grammarians have understood the term, Mr. Tooke
seems to have incontestably proved. That they sometimes couple
sentences, or that instances may be produced, in which Harris’s
definition will appear correct, the following example will serve
as an evidence: “You, and I, and John rode to town;” _i.e._ “you
rode,” “and I rode,” “and John rode.” But to assert, that this is
their distinctive property, is to affirm what may be disproved by
numberless examples. If we say, “two and two are four.” Are two
four, and two four? “A B, B C, and C A, form a triangle.” Is A B a
triangle? or B C? or C A? “John and Mary are a handsome couple.” Is
John a couple? and Mary a couple? The common theory, therefore, is
false; nor is it to be doubted, that conjunctions are, in respect
to signification, and were originally in regard to their regimen,
verbs, or words compounded of nouns and attributives. In explaining
them, however, I divided them, as the reader may remember, into
the several classes of adversative, concessive, conditional, &c.
This I did, not only in conformity to general usage, and that he
might not be a stranger to the names assigned to them; but likewise
for this reason, that, though they originally formed no distinct
species of words, but were either verbs, or compounds of nouns and
verbs, they have now assumed another character, and are construed
in a different manner. It is necessary, however, that he should be
acquainted not only with their present use, but also with their
primitive import, and classification.

How these words were degraded from their original rank, and deemed
insignificant, while some, perhaps, lost their syntactical power,
is a matter, I conceive, of no difficult inquiry. For, when the
verbs, to which any of these words belonged, became obsolete,
the words themselves, thus separated from their parent stock,
and stripped of that consequence and authority which they thence
derived, their extraction becoming daily more dubious, and their
original value more obscure, sunk by degrees into inferior note,
and at last dwindled into comparative insignificance. Besides, many
of them, doubtless, were transplanted into our language without the
_radices_; their etymology, therefore, being little known, their
primitive character, and real import, would soon be involved in
increasing darkness.

It is to be considered, also, that those who have dispensed
the laws of grammar in our language, or assumed the office of
critics, have been generally such as, though perhaps sufficiently
conversant in Greek and Latin, were entirely unacquainted with
the Northern languages. Accustomed, therefore, to render the
conjunctions and prepositions in Greek or Latin, by synonymous
English words, and unacquainted with the true character of these
vernacular terms, their _etymons_ being obsolete, or having never
been used in our language, it is easy to conceive how they would
naturally assign to the English words the same character and the
same name which were affixed to the synonymous Latin terms. Nay,
this has been so much the case, that we have ascribed an ambiguous
character to several English words, referring them now to one
class, then to another, merely because they agree in signification
with certain Greek and Latin terms, which have been severally
referred by classical grammarians to different orders. That the
word _whether_ has uniformly, in our language, the same import and
the same character, denoting “which of the two,” there can be no
doubt; yet, because this word answers sometimes to _an_, _anne_,
_num_, and sometimes to _uter_, grammarians and lexicographers have
accounted it both a conjunction and a pronoun. _Utrum_ in Latin has
shared the same fate. So far, indeed, has this spirit been carried,
that we will not admit _except_, _according_, _concerning_,
_respecting_, with many similar terms, to be verbs or participles,
because _præter_, _secundum_, _de_, are prepositions. It is from
this propensity to assimilate ours with the Latin language, that
all these errors have arisen.

That the words now termed prepositions and conjunctions were
originally verbs, or nouns, or compounds of these, Tooke has, in
my judgment, incontrovertibly proved. This being admitted, it
appears to me highly probable, that they were primitively construed
as such, joined either with the nominative or the objective case,
as the verbs had either a transitive or intransitive meaning;
and that they were followed by either single words or clauses.
This, however, is merely conjecture, founded indeed in the nature
of the words, but not supported by any evidence. In process of
time, in consequence of that assimilation which naturally takes
place between a living language and a dead one, much read, much
written, and much admired, these words, when their origin became
obscure, would, as I have remarked, be divested of their primitive
character, and be considered as belonging to those classes, to
which the synonymous Latin words were referred. Hence their
regimen would likewise undergo a change. It would appear awkward
and vicious to say now, “I saw nobody but he;” is not improbable,
however, that the mode of expression was originally, “I saw
nobody, be out he,” _i.e._ “he be out.” But I am now indulging
in conjecture, the very error which chiefly has misled us in our
grammatical researches. One thing, however, is certain, that
several words, which were originally employed as prepositions
or conjunctions indifferently, have now acquired a more fixed
character, and are used but seldom in a double capacity. Of this
the word _without_ is an example. Thus, it was not unusual to
say, “without you go, I will not,” where the term of exclusion,
though in truth a preposition prefixed to a clause, was considered
as a conjunction synonymous with _nisi_. This usage, unless in
conversation, is now almost entirely relinquished; and the term
_without_ is now generally employed as a preposition, being
prefixed to single words. It is likewise certain that in respect
to signification there is no difference between conjunctions and
prepositions: _vidi neminem nisi eum_, is equivalent to _vidi
neminem præter eum_. In like manner, “I saw nobody but him,” is
synonymous with “I saw nobody besides him;” in which examples
the conjunctions _nisi_ and _but_ are perfectly synonymous with
_præter_ and _besides_, which are termed prepositions.

It may be asked, if then prepositions and conjunctions be alike
verbs, or nouns, or compounds of these, and if many prepositions
and conjunctions be in point of meaning identical, what forms the
ground of distinction between them? It is simply this, that the
former are prefixed to single words only, as nouns and pronouns,
or to clauses involving an infinitive mood[133], the infinitive
being strictly the name of the verb; and that they have a regimen;
while the latter are prefixed to clauses, and have no regimen.
This is the only distinction between prepositions and conjunctions
as discriminated in modern use. Their original character is
sufficiently established by Mr. Tooke.

I have said that some of these words have, in our language, an
ambiguous character, being employed both as prepositions and
conjunctions. Of this the word _than_ is an example. Priestley
seems to consider it as a preposition, and pleads in favour of the
following expression, “you are taller than him,” not “taller than
he.” “Since it is allowed,” says the Doctor, “that the oblique case
should follow prepositions, and since the comparative degree of an
adjective, and the particle _than_, have certainly between them
the force of a preposition, expressing the relation of one word
to another, they ought to require the oblique case of the pronoun
following, so that, _greater than me_ will be more grammatical than
_greater than I_.” Here I cannot concur with the learned author.
The same argument would prove that _major quam me_, would be more
grammatical than _major quam ego_; a conclusion which is opposed
by universal authority. The truth is, _than_ must be either a
conjunction or a preposition, or both. If a conjunction, it can
have no government, any more than the Latin _quam_; unless we
confound the distinction which has been just now explained, and is
universally admitted, namely, that conjunctions are distinguished
from prepositions, by their having no government. If it be a
preposition, no argument is necessary to prove that it may be
joined with an objective case; for such is the distinguishing
character of prepositions. If it be either a preposition or a
conjunction, it follows, that it may be construed either with or
without a regimen. Lowth, with greater propriety, considers it as a
conjunction; and Campbell, in his “Rhetoric,” recommends this usage
as the only means of preventing that ambiguity, which necessarily
arises from the employment of this word as a preposition only. For,
if we use it as a preposition, we should say, “I love you better
than him,” whether it be meant “I love you better than I love him,”
or “I love you better than he does.” By using it as a conjunction,
the ambiguity is prevented. For, if the former sentiment be
implied, we say, “I love you better than him,” _i.e._ “than I love
him;” if the latter, we say, “I love you better than he,” _i.e._
“than he loves you.” Whatever may have been the original character
or syntax of this word, since usage is now divided, some writers
employing it as a conjunction, and others as a preposition, the
grammarian may, consistently with his duty, plead for that usage
only, which prevents ambiguity.

The rule here recommended is generally violated, when _than_
is joined with the relative pronoun, as, “Alfred, than whom
a greater king never reigned.” “Beelzebub, than whom, Satan
excepted, none higher sat.” Salmon has attempted to account for
this almost universal phraseology, by saying, that the expression
is elliptical, being the same as, “than compared with whom.” This
explanation is forced and unnatural. It is likewise unnecessary.
The simple fact is, that the word _than_ was formerly used as a
preposition, and, I believe, more frequently than it is now. Hence,
doubtless, arose this phraseology.

RULE XXII.--Derivatives are generally construed like their primitives; as, “it was a happy thing _for_ this country, that the Pretender was defeated;” or “happily _for_ this country the Pretender was defeated.” Thus also, “to compare _with_,” and “in comparison _with_ riches;”--“to depend _on_,” and his “dependence on the court.”

RULE XXIII.--One negative destroys another; or two negatives are equivalent to an affirmative; as, “nor have I no money, which I can spare;” that is, “I have money, which I can spare.”--“Nor was the king unacquainted with his designs;” that is, “he was acquainted.”

_Note_ 1.--Here our language accords with the Latin. In Greek and
French, two negatives render the negation stronger.

_Note_ 2.--This rule is violated in such examples as this, “Nor
is danger ever apprehended in such a government, no more than we
commonly apprehend danger from thunder or earthquakes.” It should
be _any more_.

RULE XXIV.--Interjections are joined with the objective case of the pronoun of the first person, and with the nominative of the pronoun of the second, as, “ah me,” “oh me,” “ah thou wretch,” “O thou who dwellest.”

_Syntax_ being that part of grammar, which teaches rules not only
for the concord and government, but also for the order of words
in clauses and sentences, I shall subjoin the few following brief
directions for the guidance of the scholar, respecting arrangement.

1st. The collocation should never invert the natural order of
events, or violate the principles of reason and metaphysical
propriety. It is obvious, for example, that no person can write,
who cannot read. The ability to do the former necessarily implies
a capacity to do the latter. It is preposterous, therefore, to
say with Addison, “There will be few in the next generation, who
will not at least be able to write and read.” He should have
said, “to read and write.” “He was the son of a mother, who had
nursed him with maternal tenderness, and had borne him in an hour
of the deepest affliction.” The natural order of events should
have dictated the reverse arrangement. There would be a manifest
impropriety in saying “Our father is well, and alive;” the former
state necessarily implying the latter. In the following passage,
however, it is perhaps excusable, the answers particularly
corresponding to the questions: Joseph says to his brothers, “Is
your father well? The old man, of whom ye spake, is he yet alive?”
They answer, “Thy servant, our father, is in good health; he is yet
alive.” This error was termed by the ancient grammarians _hysteron
proteron_; and, though not so palpably as in the preceding
examples, it occurs much more frequently than an inattentive reader
is apt to imagine.

2nd. The English language admits but few inflexions, and therefore
little or no room for variety of arrangement. The connection of
one word with another is not to be perceived, as in Greek and
Latin, by correspondence of termination, but by relative position.
This renders it indispensably necessary, that those words which
are intimately related by sense one to another, should be closely
connected by collocation. “The cunning of Hannibal was too powerful
for the Pergamenians, who by the same kind of stratagem had
frequently obtained great victories at land.” The relative here, by
its position, must be understood as referring to the Pergamenians;
whereas it is intended to refer to Hannibal. The relative clause,
therefore, should have followed the name of the Carthaginian. “His
picture, in distemper, of calumny, borrowed from the description
of one painted by Apelles, was supposed to be a satire on that
cardinal.”--_Walpole._ The error here is obvious. He should have
said, “His picture of calumny.” “It is folly to pretend to arm
ourselves against the accidents of life, by heaping up treasures,
which nothing can protect us against, but the good providence of
our heavenly Father.”--_Sherlock._ Here the grammatical antecedent
is _treasures_; but it is intended to be _accidents_. The relative
is removed from its proper subject.

3rd. As the converse of the preceding rule, it may be observed,
that those words should be separated, which in juxta-position
may, at first sight, or first hearing, possibly convey a meaning
which the speaker or writer does not intend. “I, like a well-bred
man, who is never disposed to mortify or to offend, praised both
sorts of food.” As the two introductory words are capable of two
meanings, would it not be better to say, “Like a well-bred man
... I praised both sorts of food.” I am aware, that the other
collocation is preferable, where a particular stress is to be laid
on the principal subject; but ambiguity is an error, which should
be studiously avoided, and the meaning should not be left to the
determination of a comma.

4th. From the preceding rules, it follows as a corollary, that
no clause should be so placed in a sentence, as to be referable
either to what precedes, or what follows. “The knight, seeing his
habitation reduced to so small a compass, and himself in a manner
shut out of his own house, _on the death of his mother_, ordered
all the apartments to be flung open.” The clause in italics is
ambiguously placed.

5th. When each of two arrangements is equally favourable to
perspicuity, and equally consistent with metaphysical propriety,
that should be preferred which is the more agreeable to the ear.

6th. Harsh and abrupt cadences should be avoided; and in elevated
style, the clauses should swell towards the close of the sentence.
This latter rule, however, which requires some limitations, belongs
to the province of the rhetorician, rather than to that of the
grammarian.