« ElőzőTovább »
call their relative or interrogative sense. What is this common something? What is it that distinguishes the relative who from the personal pronoun he? A little consideration will be sufficient to convince us that there is nothing, strictly speaking, more essentially relative in the former than in the latter of these words: both of them equally refer to some person previously mentioned or understood ; both of them are equally incomplete when standing by themselves as one of the terms of a proposition. It is as necessary, in order to complete the sense, to supply the name of some person if we say 'He was a great man,' as if it had been 'Who was a great man V It is obvious, then, that the distinctive name of relative is an ill-chosen one, nor would it ever have been given to this pronoun if grammarians had understood its true nature. We shall, in the first place, examine into the different modes of relation effected by these two pronouns, and we shall then endeavour to arrive at an exact notion of their meaning, not, as is usually done, by merely explaining the occasions on which it will be proper to use each, but by laying down expressions which may actually be substituted for the pronouns in discourse without altering the sense, at the same time taking care that neither of the words sought to be explained shall occur in its own explanation. It is in this precaution that the great difficulty consists; it is, however, absolutely necessary, in order to avoid being entangled in a vicious circle.
The connexion in meaning between he and who is so close, that it can hardly escape the attention of any inquirer; but we are not aware that the exact difference between the two has been hitherto pointed out. Harris, and other writers on grammar after him, have asserted that the relative may always be resolved into the personal pronoun and a conjunction. Thus, in the example given in the Hermes,
'Light is a body, which moves rapidly,' which is resolved into and it, and the passage becomes
'Light is a body, and it moves rapidly.' It is obvious that the conjunction and is here entirely redundant: it adds nothing to the meaning; the sense of the original sentence would"be as exactly rendered by simply substituting it for which —
Light is a body, it moves rapidly. In fact, the last version approaches rather nearer to the original than that in which the and is inserted; the mind treats the two propositions of which it consists more as a whole when not connected by a conjunction; the moment an and is introduced the sense is interrupted, and the propositions become entirely distinct. The assertion of Harris then leads to the unavoidable conclusion
that that we may always substitute the personal for the relative, a conclusion so obviously false, that it should be enough to convince us that he has not hit on the true solution of the question. We grant that either of the latter versions conveys, upon the whole, the same information with the original passage, but it is a great oversight to suppose that every corresponding term employed must necessarily be identical in meaning. The office of the relative in the original, though of similar nature, is perfectly distinct from that of the personal pronoun by which it is replaced in the other versions: which refers to body; it to light—
Light is a body, which body moves rapidly.
Light is a body, and it (light) moves rapidly. We should also observe that the personal pronoun it actually represents its antecedent light, it stands in its place; and the sense would be the same whether we use the word it or light—' it moves rapidly,' or 'light moves rapidly.' The relative which, on the contrary, has not simply the same meaning with its antecedent; it cannot be replaced by the word body, to which it refers: we cannot say
'Light is a body—a body moves rapidly:' for the second proposition is not true of all bodies, and we only intend to assert that the particular body called light moves rapidly —i. e. the body mentioned in the antecedent proposition, or that portion of the predicated genus body which agrees with the subject of the proposition, light.
Hence we may form this distinction between the personal and the relative pronoun, that the former actually represents or stands in the place of one term of an antecedent proposition—the latter represents such portion of it only, as agrees with the remaining term of the same proposition.
We now come to the consideration of the actual meaning of the personal and relative pronouns; and as it is obvious that the personal is the simpler of the two, we shall, in the first place, examine the pronoun he, she, it. It is frequently said that these words have no meaning of their own, but that they take in each particular case the meaning of the nouns to which they refer, or in whose place they stand. But this is an error; whetheg, with some grammarians, we consider the personal pronoun as a word used to stand in the place of the noun, the subject of discourse, in order to prevent the too frequent repetition of it in the same sentence ; or, with others, as used to distinguish the person spoken of from the person speaking, and the person spoken to: the meaning of the pronoun he, she, it, will in all cases be the same— the person or thing spoken of.
Thus the sentence ' Isaac bade Esau procure him venison'is
equivalent to' Isaac bade Esau procure the man mentioned venison.' If we prolong the sentence a little, and add, 'and promised that when he had brought it he would give him his blessing,' the pas- > sage becomes, 'Isaac bade Esau procure the man mentioned venison, and promised that when the man mentioned had brought the thing mentioned, the man mentioned would give the man mentioned the blessing of the man mentioned.' Here we are involved in hopeless confusion from there being two men mentioned, and nothing to point out which of them is intended every time 'the man mentioned' occurs.
Let us now substitute he and it for the man and the thing mentioned, and the sense becomes clear at once. We should observe, however, that the advantage gained by the substitution is purely mechanical: in point of grammatical construction, the ambiguity still remains, and the pronoun, equally with the expression of the man mentioned, may apply indifferently either to Isaac or Esau; but the sentence is so much shortened by the substitution that the mind is now enabled to retain the whole of it with all the circumstances of the case at once within its view, and thus no difficulty occurs in referring each he and him to its proper antecedent. The pronoun being thus an abbreviated form of a longer sentence, we may conclude that there was a period in the history of mankind before that abbreviation took place, though so convenient a contrivance would in all probability have been very early introduced as one of the first steps in what we may call the civilization of language. This being the case, it is very unlikely that men should have invented arbitrary sounds to express these relations, which were already capable of being sufficiently represented by the periphrasis given above. It is more probable that, like the organic remains of the material world, these particles were formed of the most striking portions of the sentences which they represent, whilst the more perishable parts have mouldered away. In some respects the fossil remains have met with a more fortunate destiny than these relics of the immaterial world, for, whilst the former have for the most part been preserved by the protecting soil in which they were embedded, so that a skilful anatomist has little difficulty in deciding to what portion of the skeleton of living animals they correspond, the latter, from their everyday and universal use, have been worn, until, like pebbles on the beach, they have lost every corner and distinctive mark, and hardly a vestige remains to indicate their original form. Yet even here we are not left entirely without traces which may enable us to form some conjecture of the origin of one or two of these pronouns. We have seen that he and it represent sentences meaning the man, and the thing mentioned. Now the most striking parts—the bones, as it were, of these sentences, and therefore the parts most likely to remain as symbols of the whole, are the nouns man and thing. Accordingly, we find that, he is in German er, and this is not unlikely to have arisen from the primitive word, from which comes vir (Latin) and wer (Anglo Saxon). If this be true, the Latin is, he, would stand for ir, like honos for honor, and other instances of the same change of letter. This conjecture would, at first sight, seem to be much strengthened by the word for he in the northern Gothic languages, namely hann—a word which would appear naturally to point to the word man as its origin, but we are led by the analogy of the Swedish annan where the Germans have ander, as well as by other circumstances, to doubt whether hann and er may not be radically the same. In the Bas Breton the pronoun is ken, whilst wan is den. Again: the word it might easily arise out of the Gothic vailds—Anglo Saxon wiht—a thing; from whence our word nought, no-it, or nothing. Horne Tooke sought for the origin of it in the other member of the same sentence with ourselves, deriving it from the participle of haitan, to name, in old English hight, in the sense of ' the said.' But this does not appear tenable ground for several reasons, amongst others because hight means named, called, and not mentioned or spoken of. Besides, this derivation would make it apply equally well to the masculine as to the neuter gender: it might represent ' the said man' as well as the 'said thing.' We are not aware that any derivation has been suggested for she, but there is little doubt that it must be from some word signifying a woman; and no German scholar will be a loss for a plausible guess.
We shall derive some assistance in the analysis of the relative from the following table of the personal, relative, and demonstrative pronouns in a few of the principal languages of the IndoEuropean stock—
Personal. Interrogative Demonstrative.
It is impossible to look at this table without perceiving that there is a certain relation between the three classes of pronouns; and a little attention will suffice to convince us, notwithstanding slight variations from the rule, that in general the relative and demonstrative are derived from the personal pronoun by the addition of certain prefixes, which vary in the different tongues according to the peculiar tendency of each to prefer certain sounds. Thus in Sanscrit, Sclavonic, Irish, and Ionic Greek, the relative prefix is a k, in ordinary Greek, Welch, and Breton, a p,—in Scotch and Latin, qu,—in Gothic and Anglo Saxon, hv,—in German, w,—and English, wh. The identity of the plan on which the relative is formed in almost all these languages, as well as the nature of the letters employed as prefix, shows that this prefix is of common origin. It is probable that the original sound was nearly that of the Scotch qu, viz. a strongly aspirated sound between a w and a k, from which, by gradually softening down the aspirate, the transition was easy through the sound of the English wh in what, to the a in the German wus on the one side, and through the Latin qu to the simple k of the Sanscrit, Sclavonic, and Gaelic, on the other side. In other words the simple aspirate has remained, while the w has been lost, as in who, where the w is not pronounced, and where unquestionably, if this change had taken place at an earlier period m the history of the language—. when spelling was more unsettled—the w would have been omitted in writing as it is now in speaking. The greatest change appears to be that from k to the p in Greek and Welch, and to our ears there seems certainly to be little resemblance between these two letters. When, however, we see that in dialects so nearly allied, as Welch and Gaelic—and Ionic, and common Greek—they correspond to each other, we cannot doubt that there is an organic connexion between the two sounds. If we take the following series, of which p and k form the two extremities, p, b, v, w, gw, g, k, we shall see that there are no two of the consecutive sounds that are not constantly interchanged, and hence it is natural that the