help of fancy. Indeed without steadiness and direction to some end, a great fancy is one kind of madness." "Judgment without fancy," he concludes, "is wit" (this anticipates the extreme neo-classical use of the word wit), "but fancy without judgment, not."

Dryden betrays the influence of Hobbes when he says of the period of incubation of his "Rival Ladies": "Fancy was yet in its first work, moving the sleeping images of things towards the light, there to be distinguished and either chosen or rejected by judgment." Fancy or imagination (the words were still synonymous), as conceived by the English neo-classicists, often shows a strange vivacity for a faculty that is after all only "decaying sense." "Fancy without judgment," says Dryden, "is a hot-mouthed jade without a curb." "Fancy," writes Rymer in a similar vein, "leaps and frisks, and away she's gone; whilst reason rattles the chain and follows after." The following lines of Mulgrave are typical of the neo-classical notion of the relation between fancy and judgment:

As all is dullness when the Fancy's bad,
So without Judgment, Fancy is but mad.
Reason is that substantial, useful part

Which gains the Head, while t' other wins the Heart.1

The opposition established by the neo-classicist in passages of this kind is too mechanical. Fancy and judgment do not seem to coöperate but to war with one another. In case of doubt the neo-classicist is always ready to sacrifice fancy to the "substantial, useful part," and so he seems too negative and cool and prosaic in his reason, and this is because his reason is so largely a pro1 An Essay upon Poetry (1682).,

test against a previous romantic excess. What had been considered genius in the time of the "metaphysicals" had too often turned out to be only oddity. With this warning before them men kept their eyes fixed very closely on the model of normal human nature that had been set up, and imitated it very literally and timorously. A man was haunted by the fear that he might be "monstrous," and so, as Rymer put it, "satisfy nobody's maggot but his own." Correctness thus became a sort of tyranny. We suffer to the present day from this neoclassical failure to work out a sound conception of the imagination in its relation to good sense. Because the neo-classicist held the imagination lightly as compared with good sense the romantic rebels were led to hold good sense lightly as compared with imagination. The romantic view in short is too much the neo-classical view turned upside down; and, as Sainte-Beuve says, nothing resembles a hollow so much as a swelling.

III

Because the classicism against which romanticism rebelled was inadequate it does not follow that every type of classicism suffers from a similar inadequacy. The great movement away from imaginative unrestraint towards regularity and good sense took place in the main under French auspices. In general the French have been the chief exponents of the classic spirit in modern times. They themselves feel this so strongly that a certain group in France has of late years inclined to use interchangeably the words classicist and nationalist. But this is a grave confusion, for if the classic spirit is anything at all it is in its essence not local and national, but universal and hu

man. To be sure, any particular manifestation of classicism will of necessity contain elements that are less universal, elements that reflect merely a certain person or persons, or a certain age and country. This is a truth that we scarcely need to have preached to us; for with the growth of the historical method we have come to fix our attention almost exclusively on these local and relative elements. The complete critic will accept the historical method but be on his guard against its excess. He will see an element in man that is set above the local and the relative; he will learn to detect this abiding element through all the flux of circumstance; in Platonic language, he will perceive the One in the Many.

Formerly, it must be admitted, critics were not historical enough. They took to be of the essence of classicism what was merely its local coloring, especially the coloring it received from the French of the seventeenth century. If we wish to distinguish between essence and accident in the classic spirit we must get behind the French of the seventeenth century, behind the Italians of the sixteenth century who laid the foundations of neo-classical theory, behind the Romans who were the immediate models of most neo-classicists, to the source of classicism in Greece. Even in Greece the classic spirit is very much implicated in the local and the relative, yet in the life of no other people perhaps does what is universal in man shine forth more clearly from what is only local and relative. We still need, therefore, to return to Greece, not merely for the best practice, but for the best theory of classicism; for this is still found in spite of all its obscurities and incompleteness in the Poetics of Aristotle. If we have recourse to this treatise, however, it must be on condition that we

do not, like the critics of the Renaissance, deal with it in an abstract and dogmatic way (the form of the treatise it must be confessed gave them no slight encouragement), but in a spirit akin to Aristotle's own as revealed in the total body of his writings - a spirit that is at its best positive and experimental.

Aristotle not only deals positively and experimentally with the natural order and with man so far as he is a part of this order, but he deals in a similar fashion with a side of man that the modern positivist often overlooks. Like all the great Greeks Aristotle recognizes that man is the creature of two laws: he has an ordinary or natural self of impulse and desire and a human self that is known practically as a power of control over impulse and desire. If man is to become human he must not let impulse and desire run wild, but must oppose to everything excessive in his ordinary self, whether in thought or deed or emotion, the law of measure. This insistence on restraint and proportion is rightly taken to be of the essence not merely of the Greek spirit but of the classical spirit in general. The norm or standard that is to set bounds to the ordinary self is got at by different types of classicists in different ways and described variously: for example, as the human law, or the better self, or reason (a word to be discussed more fully later), or nature. Thus when Boileau says, "Let nature be your only study," he does not mean outer nature, nor again the nature of this or that individual, but representative human nature. Having decided what is normal either for man or some particular class of men the classicist takes this normal "nature" for his model and proceeds to imitate it. Whatever accords with the model he has thus set up he pronounces natural or

probable, whatever on the other hand departs too far from what he conceives to be the normal type or the normal sequence of cause and effect he holds to be "improbable" and unnatural or even, if it attains an extreme of abnormality, "monstrous." Whatever in conduct or character is duly restrained and proportionate with reference to the model is said to observe decorum. Probability and decorum are identical in some of their aspects and closely related in all.1 To recapitulate, a general nature, a core of normal experience, is affirmed by all classicists. From this central affirmation derives the doctrine of imitation, and from imitation in turn the doctrines of probability and decorum.

But though all classicists are alike in insisting on nature, imitation, probability and decorum, they differ widely, as I have already intimated, in what they understand by these terms. Let us consider first what Aristotle and the Greeks understand by them. The first point to observe is that according to Aristotle one is to get his general nature not on authority or second hand, but is to disengage it directly for himself from the jumble of particulars that he has before his eyes. He is not, says Aristotle, to imitate things as they are, but as they ought to be. Thus conceived imitation is a creative act. Through all the welter of the actual one penetrates to the real and so succeeds without ceasing to be individual in suggesting the universal. Poetry that is imitative in this sense is,

1 The French Academy discriminates in its Sentiments sur le Cid between two types of probability, "ordinary" and "extraordinary." Probability in general is more especially reserved for action. In the domain of action "ordinary" probability and decorum run very close together. It is, for example, both indecorous and improbable that Chimène in the Cid should marry her father's murderer.