generalized, and thus become applicable to less analytical. Having watched the prothe expression of various concepts. If, in- pensity of pigs to scratch the soil with their stead of calling all the remaining metaphors noses, some of the Aryans called the plough verbal, I preferred to call them poetical, it the pig, the ploughshare, the pig's snout. was partly because verbal is now generally Thus Panini tells us that potram in Sanskrit supposed to exclude nominal, partly because meant both a pig and a plough: Halâyudha I wanted to imply that these metaphors con- states that protham is the name of the snouts stituted preeminently the innate poetry of both of plough and pig. Plutarch goes a step language. These metaphors, the unconscious further, and asserts that the first idea of a poetry of language, were originally as much plough came from watching the pig burrowan act of poetical genius performed by a for- ing, and that hence the ploughshare was called gotten poet as was any metaphorical expres-vis. It is curious that the Latin porca, a sion of Shakspeare or Goethe. But from our ridge between two furrows, is derived from point of view there is a difference, and a very important difference, between a metaphor that hasbeen so completely absorbed into the blood of a language as no longer to be felt as a metaphor, and others which we use with a conscious feeling that they are our own work or the work of some one else, and that they require a kind of excuse, or even an interpre tation. Aristotle (Poet. c. 21) calls such metaphors artificial (Teñоiημéva), as when some poets call the horns "small branches" (epvvyes,) or a priest "one who prays" (apητîР.)

porcus, and that the German Furche (furicha), furrow, is connected with farah, boar. In Sanskrit we find vrika, the name for wolf, used in the sense of plough; but this may be due to a radical metaphor, vrika being derived from vrask, to tear. In many languages the living principle within us is called spirit (breath); to die is expressed by to wither, to scheme by to spin, a doubt by a knot, kind by warm, unkind by cold, etc.

All this I call poetical metaphor, and it interested me as being a most important element I confined my observations chiefly to a con- in the growth of language. What we genersideration of metaphors which have become ally call metaphors, and what Dr. Brinkmann part and parcel of a language, what Dr. is chiefly concerned with, are no doubt poetiBrinkmann would call incarnate metaphors, cal too, and perhaps, if poetical means what is such as when the central spot of the eye is done by professed poets, even more truly called the pupil, the little girl-in Spanish, la poetical than what I call so. But they belong niña de los ojos; or when a machine for bat- to a later stratum of language and thought. tering is called a battering-ram (aries); or an- | If I call a man a lion, in the sense of dandy; other for lifting is called a crane. Such or a dog, in the sense of a wretch, these are metaphors are very numerous. Thus the incarnate metaphors, and their study belongs name of donkey, in German, Esel, is used in to the science of language. But if I say "he English as the name of a support for pictures was like a lion in fight," or "he was a lion in (easel). In Spanish la borrica del hato, "the fight," if I call him "Cœur de lion," these she donkey of a bundle of clothes," is used are individual metaphors, and their study be to signify a shepherd's wallet. In Greek don longs to rhetoric. It may sometimes be diffikey (ovos) is used for windlass, the upper mill-cult to draw a sharp line between the two, stone, and a distaff. When the Aryans had but that is due to the very nature of metadiscovered that the soil, after having been raked up, proved more fertile, and when they had contrived some crude kind of plough, the essential part of which consisted in a piece of wood, stone, or metal that tore open the soil, how were they to call it? Such words as the Sanskrit go-darana, earth-cleaver, are late. Ancient languages were shorter and

phors. Though all originally the work of individuals, their acceptance and popularity depend on the taste of others; and it is often, therefore, a mere question of time whether they become incorporated in the spoken lan guage or remain outside. Frequently a modern poet does but revive the latent metaphors of language, or furbish them up till they

show once more their original intentions. If | latio, explains it by brevior similitudo, an we say "to plough the sea," in French, silloner | abridged comparison; and this has remained la mer, in Italian, solcare il mare, in Spanish, for centuries the recognized definition of the arar la mar, in Latin, perarare aquas, sulcare term. By similitudo Quintilian means such vada carina, we only repeat the old radical expressions as when we say that a man acted metaphor which gave to the root ar the like a lion, by metaphora when we say more meanings of stirring, ploughing, and rowing. briefly the man is a lion. In addition to these Frequently a modern metaphor fades and he admits two other kinds of trope, viz., the hardens so quickly that we forget that it ever | synecdoche and metonymy. When we are meant was a metaphor. Who thinks of a steel-pen as a feather, or of shares, when they rise and fall, as portions of capital? Yet these are metaphors of very modern date.

But though for the purposes which I had chiefly in view when treating of the origin of mythology, the division of metaphors into radical and poetical, as explained by myself, seemed most convenient, a more detailed classification of metaphors may be useful for studying some deeper and wider strata in the growth of human thought and language.

The oldest division of metaphors dates from the time of Aristotle.

to understand the many from the one, the whole from the part, the genus from the species, the result from the antecedents, and vice versá, that with him is synecdoche; when we put one name for another, such as Homer for Homer's poems, that is metonymy.

This classification has answered its object very well, particularly as it was intended chiefly for rhetorical purposes. But as we acquire a fuller understanding of certain processes of the mind and language, it often happens that the old classification and the old technical terms prove inadequate and that we have nevertheless to retain them, though in a modified sense. Thus the name of metaphor is certainly objectionable, except when we restrict it to individual poetical metaphors, because it seems to imply a conscious transference of a name from one object to another, both previously known, both previously named. Such transference takes place both in modern and ancient writers, as when, for instance, Gibbon says, "Some seeds of knowledge might be cast upon a fruitful soil!" Such a metaphor is poetical and intentional. This is already less so in a passage quoted by Aristotle in his Poetica, when the sun is spoken of as oneipwv beokтioтav pλóya, "sowing the divine light!" For, as Aristotle hints himself, the metaphor here is not quite involuntary, because the Greek language had no separate verb to express the act of strewing or scattering the light, and nothing remained but to use oneípe, to sow.

He takes μcrabopá in a very wide sense, calling by that name every transference of a word, (1) from the genus to the species, as if we say, "to stand" of a ship, instead of "being at anchor;" (2) from the species to the genus, if we say a "thousand," instead of "many;" (3) from one species to another species, if we say xaλký ảñò þʊxǹv ȧpúσas, "with the weapon lifting the soul as water with a pitcher from the well," or reμv ȧтeiρéi xadký, "cutting with the unyielding weapon," for in both cases the special ȧpúer and Téμvé are used in the sense of taking away; and (4), according to analogy. Aristotle gives here as an instance "the goblet of Ares:" and he adds, "as the goblet stands to Dionysos in the same relation as the shield to Ares, the former is used for the latter." Another instance is, if we call the evening the old age of the day, or old age the evening of life. It was this last transference, however, that "according to This is a very important remark, and a analogy," which in later times monopolized closer examination of ancient metaphors the name of metaphora-Berkeley uses analogy teaches us that poverty of language was a very as synonymous with metaphor-while tropus important, nay the most important element was used in the more general sense which in the formation. Language had need of Aristotle had assigned to metaphora. Thus metaphors, had in fact to borrow, because it Quintilian, rendering metaphora by trans- was too poor, or, as Cicero says, hae transla

tiones quasi mutationes sunt, cum quod non But though it would be more correct to habeas, aliunde sumas. He distinguishes these call ancient metaphors transformations or metaphors from others, which he calls paulo transitions rather than transferences, it will audaciores, quæ non inopiam indicant, sed be necessary to retain the old technical term, orationi splendoris aliquid arcessunt.” only guarding against its etymological meanWhen there was no word to express a nas-ing being taken for its real definition. After cent idea, what could be done but to take the next best? Man was driven to speak metaphorically, whether he liked it or not. It was not because he could not restrain his poetical imagination, but rather because he had to strain it to the very utmost, in order to find expres ion for the ever-increasing wants of his mind. Suppose man had advanced as far as platting or weaving; it would be very natural that, after setting lines to catch birds he should when he had to describe his day's work, be reminded of the words for platting or weaving. Weaving would thus take the sense of putting snares, and when a new word was wanted for setting snares-that is, for tricking, cheating, luring, inveigling a person by false words-nothing, again, was more natural than to take a word of a similar import, and to use, for instance, vaive, to weave, in the sense of plotting. Thus Homer says, πυκινὸν δόλον ὑφαίνειν, μῆτιν ὑφαίνειν, etc., i. e. to weave a plot. This metaphor spread very widely, and we may discover it even in our own word subtle, Lat. subtilis, which comes from subtexere, to weave beneath, like téla for texla.

Metaphor, therefore, ought no longer to be understood as simply the premeditated act of a poet, as a conscious transference of a word from one object to another. This is modern, fanciful, individual metaphor, while the old metaphor was much more frequently a matter of necessity, and in most cases not so much the transference of a word from one concept to another, as the creation or determination of a new concept by means of an old name. A poet who transfers the name of tear to the dew has already clear names and concepts both for tear and dew. But the old framers of language who for the first time used "to weave" in the sense of plotting had before this neither concept nor name for plotting; they created or fixed the new concept and widened the old name at one and the same time.

these preliminary remarks, a classification of ancient metaphors will become less difficult. FUNDAMENTAL METAPHOR.-There is, first of all, a whole class of metaphors which arise from a deep necessity of thought. Of these I have often spoken before, and need not dwell on them now, particularly as they have lately been discussed with great philosophical in. sight by Professor Noiré in his Logos. There was no way of conceiving or naming anything objective except after the similitude of the subjective, or of ourselves. Not only animals must be conceived as acting like ourselves, as pointing, retrieving, rejoicing, grieving, will. ing, or resisting, but all inanimate objects had to be interpreted in the same way. The sun rises and sets, the moon grows and wanes, the clouds fly, the river runs, the mountains stand, the trees die, the sea smiles. Homer calls even a lance furious (μatμwwoa), and a stone shameless (ἀναιδής.) This fundamental metaphor, however, dates back so far in the growth of our thoughts and words that it is hardly ever felt as a metaphor. It is at the root of all mythology, and had been perceived as such long ago, before the science of comparative mythology was even dreamt of. Thus Reid wrote: "Our first thoughts seem to be that the objects in which we perceive motion have understanding and power as we have. 'Savages,' says the Abbé Raynal, 'wherever they see motion which they cannot account for, there they suppose a soul.' All men may be considered as a savages in this respect, until they are capable of instruction, and using their faculties in a more perfect manner than savages do. The Abbé Raynal's observation is sufficiently confirmed both from fact and from the structure of a languages. Ruder nations do really believe sun, moon, and stars, earth, sea, and air, fountains and lakes, to have understanding and active power. To pay homage to them, and implore their favor, is a kind of idolatry natural to savages. All

languages carry in their structure the marks | for feeling. Sometimes the name of the inof their being formed when this belief pre- strument is made to convey the effect provailed.' With certain limitations this is quite duced by it, as when the Greek word xapakтýρ true, but mythology is but one out of many an instrument for graving, is used for the manifestations in which fundamental meta- mark produced by it, then for any mark, and phor shows itself. lastly for the peculiar nature or character of

called sensuous knowledge the foot of a table, or the foot of a mountain before we gave it a name. The carpenter who made the foot knew it as a piece of wood, as a stick, as properly shaped, whether square or round. But until he conceived it as something sup

the body, he did not know it as a foot, and it is impossible to say which came first, concept or name, in what must have been an almost instantaneous process.

gubernator in the language of sailors and others had dwindled down to a mere director, no actual transference was necessary. Guber nator had been so far emptied of its original contents, its intension had shriveled up so much that it was naturally applicable to ever so many persons, provided they acted a lead-porting the top of a table, as a foot supports ing part in the management of any affairs. There is, for instance, a great difference between calling a ruler a steersman, a gubernator, and calling the same man a column of the state. First of all, the latter simile belongs probably to a much later time, when columns had become not only useful, but also ornamental. Secondly, column would have to dwindle down very much before it could fall into the same wide genus as minister of state. Here, therefore, a real poetical ransference seems to have taken place, and when Pope, in his translation of the Odyssey, introduces this simile

"Now from my fond embrace by tempest torn Our other column of the state is borne,"

we feel at once a change of atmosphere, for Homer would certainly not have spoken of a column of the state, nor would he have represented such a column as torn from his mother's fond embrace by tempest.

A poet, no doubt, might dispense with this slow process of aphaeresis or apoptosis; he might not wait for the gradual dropping off of claws and wings and feathers before he called the sun a golden bird. But with the majority of mankind metaphor is mostly produced by the gradual fading of the colors of our percepts, and even by the vanishing of the outlines of their shadows, i. e., of our concepts. This gives us abstract, hence general names, and these general names, without any metaphorical effort, become applicable to a large number of new objects, and are afterward called metaphors.

How quickly language, even in modern times, can generalize, we see in a number of idiomatic and proverbial expressions in which one single case is used to convey wide inferIf we speak of the moons of Jupiter, moon ences and very general lessons. The Spanish is no longer our measurer of time, but it has language is particularly rich in such proverbs faded into a mere satellite, a companion of a and metaphors, and they have been carefully planet. It has become a very general name, collected by Spanish scholars. The Dictionand, as such, it proved applicable to the satel-ary of the Spanish Academy is well known lites of Jupiter or of any other planet. A for its wealth of metaphorical expressions, foot had originally a very full intension. It meant the member of a living body, made of flesh and bone and muscles, with five toes, and used for locomotion. It was meant for a human foot, and implied very soon a certain length. But many of its attributes not being attended to, foot became applicable to the locomotive organs of other animals, of quadrupeds, insects, birds, till at last it lost even the attribute of locomotion, and, as the foot of a table, or the foot of a mountain, signified what is most lifeless and motionless.

And here again we see very clearly how language and thought march hand in hand. It was not that we did not know by what is

most of which are carefully and successfully explained. The number of Spanish proverbs is said to amount to no less than twenty-four thousand. Instead of saying, "What service have you rendered me?" the Spaniard says, Qué hijo me has sacado de pila? "Which son have you taken for me from the font?" Instead of saying Why? he may say, Por qué carga de agua? "For what load of water?" When we say, "Tell this story to another person," he says, "A otro perro con eso hueso," "Go to another dog with that bone." The Spanish language abounds in similar expressions which in one sense may all be called metaphorical, because they are all based on