Whatever were the difficulties, therefore, which embarrassed the first invention of nouns adjective, the same, and many more, must have embarrassed that of prepositions. If mankind, therefore, in the first formation of languages, seem to have, for some time, evaded the necessity of nouns adjective, by varying the termination of the names of substances, according as these varied in some of their most important qualities, they would much more find themselves under the necessity of evading, by some similar contrivance, the yet more difficult invention of prepositions. The different cases in the ancient languages is a contrivance of precisely the same kind. The genitive and dative cases, in Greek and Latin, evidently supply the place of the prepositions; and by a variation in the noun substantive, which stands for the co-relative term, express the relation which subsists between what is denoted by that noun substantive, and what is expressed by some other word in the sentence. In these expressions, for example, fructus arboris, the fruit of the tree; sacer Herculi, sacred to Hercules; the variations made in the co-relative words, arbor and Hercules, express the same relations which are expressed in English by the prepositions of and to.
To express a relation in this manner, did not require any effort of abstraction. It was not here expressed by a peculiar word denoting relation and nothing but relation, but by a variation upon the co-relative term. It was expressed here, as it appears in nature, not as something separated and detached, but as thoroughly mixed and blended with the co-relative object.
To express relation in this manner, did not require any effort of generalization. The words arboris and Herculi, while they involve in their signification the same relation expressed by the English prepositions of and to, are not, like those prepositions, general words, which can be applied to express the same relation between whatever other objects it might be observed to subsist.
To express relation in this manner did not require any effort of comparison. The words arboris and Herculi are not general words intended to denote a particular species of relations which the inventors of those expressions meant, in consequence of some sort of comparison, to separate and distinguish from every other sort of relation. The example, indeed, of this contrivance would soon probably be followed, and whoever had occasion to express a similar relation between any other objects would be very apt to do it by making a similar variation on the name of the co-relative object. This, I say, would probably, or rather certainly happen; but it would happen without any intention or foresight in those who first set the example, and who never meant to establish any general rule. The general rule would establish itself insensibly, and by slow degrees, in consequence of that love of analogy and similarity of sound, which is the foundation of by far the greater part of the rules of grammar.
To express relation therefore, by a variation in the name of the co-relative object, requiring neither abstraction, nor generalization, nor comparison of any kind, would, at first, be much more natural and easy, than to express it by those general words called prepositions, of which the first invention must have demanded some degree of all those operations.
The number of cases is different in different languages. There are five in the Greek, six in the Latin, and there are said to be ten in the Armenian language. It must have naturally happened that there should be a greater or a smaller number of cases, according as in the terminations of nouns substantive the first formers of any language happened to have established a greater or a smaller number of variations, in order to express the different relations they had occasion to take notice of, before the invention of those more general and abstract prepositions which could supply their place.
It is, perhaps, worth while to observe that those prepositions, which in modern languages hold the place of the ancient cases, are, of all others, the most general, and abstract, and metaphysical; and of consequence, would probably be the last invented. Ask any man of common acuteness, What relation is expressed by the preposition above? He will readily answer, that of superiority. By the preposition below? He will as quickly reply, that of inferiority. But ask him, what relation is expressed by the preposition of, and, if he has not beforehand employed his thoughts a good deal upon these subjects, you may safely allow him a week to consider of his answer. The prepositions above and below do not denote any of the relations expressed by the cases in the ancient languages. But the preposition of, denotes the same relation, which is in them expressed by the genitive case; and which, it is easy to observe, is of a very metaphysical nature. The preposition of, denotes relation in general, considered in concrete with the co-relative object. It marks that the noun substantive which goes before it, is somehow or other related to that which comes after it, but without in any respect ascertaining, as is done by the preposition above, what is the peculiar nature of that relation. We often apply it, therefore, to express the most opposite relations; because, the most opposite relations agree so far that each of them comprehends in it the general idea or nature of a relation. We say, the father of the son, and the son of the father; the fir-trees of the forest, and the forest of the fir-trees. The relation in which the father stands to the son, is, it is evident, a quite opposite relation to that in which the son stands to the father; that in which the parts stand to the whole, is quite opposite to that in which the whole stands to the parts. The word of, however, serves very well to denote all those relations, because in itself it denotes no particular relation, but only relation in general; and so far as any particular relation is collected from such expressions, it is inferred by the mind, not from the preposition itself, but from the nature and arrangement of the substantives, between which the preposition is placed.
What I have said concerning the preposition of, may in some measure be applied to the prepositions, to, for, with, by, and to whatever other prepositions are made use of in modern languages, to supply the place of the ancient cases. They all of them express very abstract and metaphysical relations, which any man, who takes the trouble to try it, will find it extremely difficult to express by nouns substantive, in the same manner as we may express the relation denoted by the preposition above, by the noun substantive superiority. They all of them, however, express some specific relation, and are, consequently, none of them so abstract as the preposition of, which may be regarded as by far the most metaphysical of all prepositions. The prepositions therefore, which are capable of supplying the place of the ancient cases, being more abstract than the other prepositions, would naturally be of more difficult invention. The relations at the same time which those prepositions express, are, of all others, those which we have most frequent occasion to mention. The prepositions above, below, near, within, without, against, &c. are much more rarely made use of, in modern languages, than the prepositions of, to, for, with, from, by. A preposition of the former kind will not occur twice in a page; we can scarce compose a single sentence without the assistance of one or two of the latter. If these latter prepositions, therefore, which supply the place of the cases, would be of such difficult invention on account of their abstractedness, some expedient, to supply their place, must have been of indispensable necessity, on account of the frequent occasion which men have to take notice of the relations which they denote. But there is no expedient so obvious, as that of varying the termination of one of the principal words.
It is, perhaps, unnecessary to observe, that there are some of the cases in the ancient languages, which, for particular reasons, cannot be represented by any prepositions. These are the nominative, accusative, and vocative cases. In those modern languages, which do not admit of any such variety in the terminations of their nouns substantive, the correspondent relations are expressed by the place of the words, and by the order and construction of the sentence.
As men have frequently occasion to make mention of multitudes as well as of single objects, it became necessary that they should have some method of expressing number. Number may be expressed either by a particular word, expressing number in general, such as the words many, more, &c. or by some variation upon the words which express the things numbered. It is this last expedient which mankind would probably have recourse to, in the infancy of language. Number, considered in general, without relation to any particular set of objects numbered, is one of the most abstract and metaphysical ideas, which the mind of man is capable of forming; and, consequently, is not an idea, which would readily occur to rude mortals, who were just beginning to form a language. They would naturally, therefore, distinguish when they talked of a single, and when they talked of a multitude of objects, not by any metaphysical adjectives, such as the English, a, an, many, but by a variation upon the termination of the word which signified the objects numbered. Hence the origin of the singular and plural numbers, in all the ancient languages; and the same distinction has likewise been retained in all the modern languages, at least, in the greater part of words.