citizen, and, THE man who always, &c., A and THE in connection with the conjunctive wно perform functions of a similar nature. But this function seems to be merely to indicate a lesser (in the case of a) or greater (in the case of the) degree of determinateness, or, in the language of the common grammars, to point out the word to which the conjunctive refers and to which it joints the accessory proposition. But THIS and THAT, ANY, SUCH, ALL, in fact most determinatives perform the very same function.

There seems not a shadow of apology for considering these two little words as forming a distinct part of speech, or division of the signs of our language, and then entering on the vain search for some function pertaining exclusively to them, by which they may be discriminated from all other classes of words. Such a course presents strong temptation to the exercise of ingenuity in finding grounds of distinction where none actually exist, and, thus, to pervert as well as perplex the grammar of the language. Injudicious classification-the application, especially, of distinctive names to things which are not in fact different (in the aspect in which they are regarded in classifying them), naturally leads to false speculation. It is not a sufficient apology in such cases to allege that the classification can do no harm, because the things classified remain unchanged by the classification. The establishment of a class is the implied assertion of a distinctive difference-a difference worthy too of the attention of an inquirer. Where such difference does not exist, the implication that it exists is an error, and may, like other errors, which in themselves appear trivial, lead in the end to more important errors.

If we wish to account for the fact that the articles have been placed foremost in the list of the parts of speech, we must look for the explanation, as before, to the Greek grammarians. These grammarians very properly placed the nouns at the head of their classification of words, and finding it convenient to indicate the genders of these nouns by prefixing the article which accorded with them in gender, they were led to give the declension of the article ó, i, ro, the first place in their treatises on language. The article is still used in our Greek lexicons to indicate the genders of nouns, instead of using, as we do in Latin dictionaries and those of the modern languages, abbreviations for masculine, feminine and neuter.

The most intelligent of our modern grammarians are unanimous in the decision, that it is improper to consider the articles a distinct class of words, or part of speech. See Dr. Robinson's Translation of Buttman's Greek Grammar, pp. 120, 121, note, &c., &c.

THE DETERMINATIVE AN OR A.-Contrary to what is asserted, or

implied in most of our grammars, the original form of this word is an, and a is a contraction. It is not strictly correct to say that "a becomes an before a vowel and a silent h." We should rather say that an becomes a before a syllable beginning with a consonantal sound.

A is employed before words commencing with a consonantal sound; that is, 1st, before all words commencing with the sounds represented by the letters called consonants, in the written language; 2d, before words commencing with an aspirated h, as, a hand, a hammer, &c.; 3d, before all words commencing with what is called the long sound of u-equivalent to the sound given to the combination of the semivowel y with the vowels ou in the word you, or in the word youth, as, a union, a university, &c. On the contrary, before u short, or purely vocal, as well as before the other vowels, and before silent h, an is employed; as, an uncle, an animal, an hour, &c.

An is perhaps the weakest of the determinative adjectives of all the determinatives the least determinative.*

When used with a noun it indicates that a single individual of the species, of which the noun is a general sign, is intended to be designated. Thus the word man used alone means the whole species, or race of men, as in the words of the poet, "The proper study of mankind is MAN." But a man indicates a single individual of the race. The a prefixed shows, 1st, that we do not intend to include a whole class; nor 2d, a number of the class, but a single individual of a class, and "farther it saith not." What particular individual is meant is left wholly undetermined.

The word an with its contract form a descends to us from the AngloSaxon an, or ane, the word expressive of unity in that language, and from which we have, with a little variation of sound and orthography, our modern sign of, unity, the numeral one. Ane or ae to express unity is a form still extant in the Scottish dialect, little differing in orthography and pronunciation from the Anglo-Saxon sign of unity, on the one hand, and our modern English determinative, on the other. Indeed, an more nearly resembles the Anglo-Saxon numeral in form than one, which corresponds exactly with the parent word in sense.

In the present usage of our language, AN and ONE agree in this, that both imply individuality. But they differ in this, that an implies, as we have already said, that an individual of a class is indicated, as dis

Hence many of our grammarians have been led to commit the solecism of classing an among the definitives, and then calling it the indefinite article-the indefinite definitive!

tinguished from the whole class of which the noun is the common appellation; whereas one implies that a single individual is meant, as distinguished from a number of individuals of the same class. The force of the two words and their distinct functions can be most clearly exhibited by the aid of examples. For this purpose we present the following questions, with suitable answers: Can a man perform that piece of work? No; but A HORSE can perform it. Can ONE man per-form that piece of work? No; but TWO MEN can perform it? In the first question the emphasis is on the word man, in the second on the determinative word one. The inquiry in the first case is, whether man (a human being) can do the work; in the latter, whether one man or more than one are necessary to perform it.

It will be seen from this that the employment of the determinative an (differing in form from the numeral one) introduces a convenient distinction in our language. Yet, in languages which employ the numeral for both purposes, no peculiar inconvenience is experienced, since a variation of emphasis is sufficient (in spoken language) to mark the discrimination of meaning. For example, the words un homme (in French), by laying a slight stress of voice on un, imply one man; without this stress, and giving preponderant force to homme, a man.

We may here notice a use of the word ONE resembling that of the indefinite (more properly, less definite) article, in such expressions as "One Simon, a tanner," "One Mnason of Cyprus," &c. This use, certainly, more nearly resembles that of the article an, than the common use of the numeral. It may be doubted whether this word ONE is not the indeterminate pronoun described

in § 155: 25. It strongly indicates indeterminateness, and hence is often employed contemptuously to insinuate a reproach of obscurity, want of notoriety and social importance. The word is, we believe, in this use employed only with the names of persons. Thus used it is equivalent to the Greek indefinite pronoun Tis, and the Latin quidam.

The word any is of kindred meaning as well as of kindred origin with an. It is formed from the Anglo-Saxon numeral ane or aen, with the addition of the affix ig, and was originally written aenig, meaning one like. This word may be regarded perhaps as more loosely determinative than an. Though originally implying unity, it is, unlike an, often employed before plural nouns. We can say any men, as well as any man. We may observe the distinction between an or a, any and some in the following examples: Can a boy do that? No; but A man can. Can ANY boy do that? No; but SOME boys can. Some is commonly in present usage employed with plurals, anciently it was used with nouns of the singular form with the force of the Latin aliquis; as

"Some man will say," &c. We have examples of this usage in the compound words somebody, something.

All three words an, any, some, agree in this that they indicate the partition of the class of objects represented by the nouns to which they are applied, and that only a part (in the case of an only an individual) is embraced under the expression. If the noun is accompanied by a limiting or descriptive complement, they indicate that only a part (an individual in the case of an) of the objects represented by the noun so limited is embraced: for example, a wise man, means an individual of the class indicated by the words wise man. It is always implied that there are more individuals of the same kind not embraced by the expression.

In this respect an and the other determinatives now mentioned are markedly different from those which we are about to consider.

Before passing from an or a, we may remark an apparent exception to the assertion that it indicates individuality in the fact that it is placed before plural nouns modified by numerals. Thus we say, a dozen men, or a dozen; a thousand men, a hundred men ; &c. In all such cases the word dozen, or hundred, or thousand is regarded as expressing a collectire unit. The proof is at hand. We can equally say, one dozen, one hundred, one thousand, &c.

THE DETERMINATIVE THE.-This word is used before both singular and plural nouns to indicate that they are to be taken in a strictly determinate sense. It indicates that the object or objects represented by the noun, as limited either by an expressed or implied modification, are embraced in their complete totality. Thus, the man of integrity, indicates the class man, as limited by the words of integrity, in its complete totality. If we make any assertion about the man of integrity, it ought to apply to every individual of this whole class. The phrase, a man of integrity, agreeably to what we have said in considering an, implies the partition of the class, and that only an individual is embraced in the expression. There are numerous instances in which these two forms of expression may be employed indifferently in expressing the same truth. Thus, A man of integrity would not do such an action; and, THE man of integrity would not do such an action. These two assertions are nearly equivalent, for, if an individual of integrity taken at random would not do a certain action, it is obvious to infer that the whole class (included under the expression, the man of integrity) would not do it. But this is not explicitly declared in the assertion. Something is left for the exercise of the hearer's judgment-something to be inferred.

That the two expressions are not in themselves (without the assistance of inference) exactly equivalent may be readily discovered by attending to similar forms of expression in which the subject matter is different. For example, A man whom we saw yesterday came to my house this morning, and THE man whom we saw yesterday came, &c. In the first of these forms it is indicated, that the man whom we saw yesterday is only one of a class-an indefinite individual of a number whom we saw; but, in the latter form, a definite man, about whom we have an understanding-whom we both know, as the man whom we saw yesterday, is indicated.

We must, if we wish to understand this subject, guard carefully against confounding the function performed by the determinatives with the function performed by the more intimate limiting words. We believe that the determinative the is never employed in our language except with a noun that is otherwise limited either expressly or by implication. The determinative THE indicates the fact of such limitation, the mutual recognition of such limitation by the speaker and the party addressed, and further that the object or class of objects represented by the noun so limited is embraced in its totality by the expression. When some grammarians say that the article the "limits a noun and shows how far its signification extends," they seem to confound together the function performed by the determinative and the functions performed by the more intimate modifications, expressed or implied, which together with the noun taken as a whole compound conception come within the determinative influence of the article. It is, perhaps, more correct to say that the article is placed before a noun because it is already limited, than to say that it is placed before the noun for the purpose of limiting it. It rather refers to a limitation expressed or implied than expresses one itself, and it thus exercises a force similar to that of the relative pronouns. In fact the closest relation in origin and function exists in most languages between determinatives and relatives. In some cases the same word, like our that, performs both functions.

We reiterate, at the hazard of being charged with repetition, that the noun, which is the name of a class, is first limited by what we have called its more intimate modifications (descriptive adjectives, genitives, &c.), expressed or understood and coming, as is taken for granted, within the cognizance of the party addressed; thus generally forming a subclass by means of the limitations, which subclass, it is further indicated by the article, is to be taken in its totality. A subclass, we say, is generally formed, but sometimes the determinate expression embraces only an individual object limited as above de