very names they used. But as to the tracing of those figures to the springs in human nature from which they flow, extremely little hath as yet been attempted. Nay, the names that have been given are but few, and by consequence very generical.-Each class, the metaphor and the metonymy in particular, is capable of being divided into several tribes, to which no names have yet been assigned. It was affirmed that the tropes and figures of eloquence are found to be the same upon the main in all ages and nations. The words upon the main were added, because though the most and the principal of them are entirely the same, there are a few which presuppose a certain refinement of thought not natural to a rude and illiterate people. Such in particular is that species of the metonymy, the concrete for the abstract, and possibly some others. We shall afterwards perhaps have occasion to remark, that the modern improvements in ridicule have given rise to some which cannot properly be ranged under any of the classes above mentioned; to which, therefore, no name hath as yet been appropriated, and of which I am not sure whether antiquity can furnish us with an example. SECTION III.—Words considered as Sounds. When I entered on the consideration of vivacity as depending on the choice of words, I observed that the words may be either proper terms or rhetorical tropes; and whether the one or the other, they may be regarded not only as signs but as sounds, and consequently as capable in certain cases of bearing, in some degree, a natural resemblance or affinity to the things signified. The two first articles, proper terms and rhetorical tropes, I have discussed already, regarding only the sense and application of the words, whether used literally or figuratively. It remains now to consider them in regard to the sound, and the affinity to the subject of which the sound is susceptible. When, as Pope expresseth it," the sound is made an echo to the sense?," there is added in a certain degree, to the association arising from custom, the influence of resemblance between the signs and the things signified; and this doubtless tends to strengthen the impression made by the discourse. This subject, I acknowledge, hath been very much canvassed by critics; I shall therefore be the briefer in my remarks, confining myself chiefly to the two following points. First, I shall inquire what kinds of things language is capable of imitating by its sound, and in what degree it is capable; 9 Essay on Criticism. secondly, what rank ought to be assigned to this species of excellence, and in what cases it ought to be attempted. PART I.-What are articulate Sounds capable of imitating, and in what Degree? First, I shall inquire what kinds of things language is capable of imitating by its sound, and in what degree it is capable. And here it is natural to think, that the imitative power of language must be greatest when the subject itself is things audible. One sound may surely have a greater resemblance to another sound, than it can have to any thing of a different nature. In the description therefore of the terrible thunder, whirlwind and tempest, or of the cooling zephyr and the gentle gale, or of any other thing that is sonorous, the imitation that may be made by the sound of the description will certainly be more perfect than can well be expected in what concerns things purely intelligible, or visible, or tangible. Yet even here the resemblance, if we consider it abstractly, is very faint. The human voice is, doubtless, capable of imitating, to a considerable degree of exactness, almost any sound whatever. But our present inquiry is solely about what may be imitated by articulate sounds, for articulation greatly confines the natural powers of the voice; neither do we inquire what an extraordinary pronunciation may effectuate, but what power in this respect the letters of the alphabet have, when combined into syllables, and these into words, and these again into sentences, uttered audibly indeed and distinctly, but without any uncommon effort. Nay, the orator, in this species of imitation, is still more limited. He is not at liberty to select whatever articulate sounds he can find to be fittest for imitating those concerning which he is discoursing. That he may be understood, he is under a necessity of confining himself to such sounds as are rendered by use the signs of the things he would suggest by them. If there be a variety of these signs, which commonly cannot be great, he hath some scope for selection, but not otherwise. Yet so remote is the resemblance here at best, that in no language, ancient or modern, are the meanings of any words, except perhaps those expressing the cries of some animals, discoverable, on the bare hearing, to one who doth not understand the language. Indeed, when the subject is articulate sound, the speaker or the writer may do more than produce a resemblance, he may even render the expression an example of that which he affirms. Of this kind precisely are the three last lines of the following quotation from Pope: These equal syllables alone require, But this manner, which, it must be owned, hath a very good. effect in enlivening the expression, is not imitation, though it hath sometimes been mistaken for it, or rather confounded with it. As to sounds inarticulate, a proper imitation of them hath been attempted in the same piece, in the subsequent lines, and with tolerable success, at least in the concluding couplet: Soft is the strain when Zephyr gently blows, And the smooth stream in smoother numbers flows; An attempt of the same kind of conformity of the sound to the sense, is perhaps but too discernible in the following quotation from Dryden : O'er all the dreary coasts, Dreadful gleams, Sullen moans, Hollow groans, And cries of injured ghosts". Milton's description of the opening of hell-gates ought not here to be overlooked. The same author has, in another performance, given an excellent specimen in this way, Grate on their scrannel pipes of wretched straw5. He succeeds the better here, that what he says is evidently accompanied with a design of exciting contempt. This induceth us to make allowance for his leaving the beaten road in search of epithets. In this passage of the Odyssey 1 Essay on Criticism. 2 Ibid. 4 Paradise Lost, B. ii. Ode on St. Cecilia's Day. His bloody hand Snatch'd two unhappy of my martial band; the sound, but not the abruptness of the crash, is, I imagine, better imitated than in the original, which on account of both, especially the last, was much admired by the critic of Halicarnassus. An excellent attempt in this way we have in a poem of Dyer; -The pilgrim oft At dead of night mid his oraison hears Rattling around, loud thundering to the moon". But the best example to be found in our language is, in my opinion, the following lines of Mr, Pope, What! like Sir Richard, rumbling, rough, and fierce, With arms, and George, and Brunswick crowd the verse, Let Carolina smoothe the tuneful lay, Lull with Amelia's liquid name the nine, The success here is the greater that the author appears through the whole to deride the immoderate affectation of this over-rated beauty, with which some modern poetasters are so completely dazzled. On the whole, the specimens produced, though perhaps as good as any of the kind extant in our language, serve to evince rather how little than how much can be done in this way, and how great scope there is here for the fancy to influence the judgment. But there are other subjects beside sound, to which language is capable of bearing some resemblance. Time and motion, for example, or whatever can admit the epithets of quick and slow, is capable in some degree of being imitated by speech. In language there are long and short syllables, one of the former being equal or nearly equal to two of the latter. As these may be variously combined in a sentence, and syllables of either kind may be made more or less to predominate, the sentence may be rendered by the sound more or less expressive of celerity or tardiness. And though even here the power of speech seems to be much limited, there being but two degrees 6 Pope's Od. In Homer thus, Συν δε δυω μαρψας, ὥστε σκυλακας ποτι γαιη Κοπτ * Ruins of Rome, Dodsley's Collection, vol. i. • Sat. 1. in syllables, whereas the natural degrees of quickness or slowness in motion or action may be infinitely varied, yet on this subject the imitative power of articulate sounds seems to be greater and more distinctive than on any other. This appears to particular advantage in verse, when, without violating the rules of prosody, a greater or a less number of syllables is made to suit the time. Take the following example from Milton, When the merry bells ring round, In this passage the third line, though consisting of ten syllables, is, by means of two anapasts, pronounced, without hurting the measure, in the same time with an iambic line of eight syllables, and therefore well adapted in sound to the airy diversion he is describing. At the same time it must be owned, that some languages have in this particular a remarkable superiority over others. In English the iambic verse, which is the commonest, admits here and there the insertion of a spondee, for protracting, or of an anapæst, as in the example quoted, for quickening the expression1. But, in my opinion, Greek and Latin have here an advantage, at least in their heroic measure, over all modern tongues. Accordingly Homer and Virgil furnish us with some excellent specimens in this way. But that we may know what our own tongue and metre is capable of effecting let us recur to our own poets, and first of all to the celebrated translator of the Grecian bard. I have made choice of him the rather as he was perfectly sensible of this beauty in the original, which he copied, and endeavoured, as much as the materials he had to work upon would permit him, to exhibit it in his version. Let us take for an example the punishment of Sisyphus in the other world, a passage which had on this very account been much admired in Homer by all the critics both ancient and modern. 9 L'Allegro. Up the high hill he heaves a huge round stone; The huge round stone resulting with a bound, Thunders impetuous down, and smokes along the ground2. 1 Perhaps the feet employed in ancient poetry are not in strict propriety applicable to the measures adopted by the English prosody. It is not my business at present to enter into this curious question. It suffices that I think there is a rhythmus in our verse plainly discernible by the ear, and which, as it at least bears some analogy to the Greek and Latin feet, makes this application of their names sufficiently intelligible. 2 In Greek thus, Λααν ανα ωθεσκε ποτι γεφον |