former, history will be found to be equally inadmissible in the constitution of the latter.

The assertion now hazarded, on the incompatibility of a purely historical subject with a poetical romance, will not require to be discussed with much intricacy of argument before it is established. And though the assertion may appear at first rather paradoxical, it is, however, a fact, that to embody productions of this kind with history, and thus to give them an absolute foundation in reality, would tend only to diminish those qualities of truth, probability, individuality, and embellishment, which make up the notion of that ideal beauty, by which the poet may be supposed bound to regulate his fictions.

The essence of the poetical romance consists in a wildness of fiction, which derives its appearance of truth, not from our knowledge, but credulity: the fictitious parts of such compositions can of course derive little improvement from a forced alliance with that science which possessing no varieties of change, is confined to the straight line of real occurrence. Over facts which have once occurred we have no power of alteration; we may

misrepresent, but we cannot virtually change them: it must of course pervert and destroy the nature of such materials, in any production whatever, to blend them with fictitious circumstances. When we join those discordant ingredients, not by incorporation, but in succession, such an union must be equally unpromising of a successful issue; as it must tend rather to bring discredit on that part of the composition which we must believe as being true, than give probability to that part which we must doubt as being preternatural. In this mixture, we can be as little said to improve the general effect which arises from the verisimilitude of the entire subject, as the verisimilitude produced in any of its parts; for what is partially fictitious, cannot be collectively true.

This reasoning is equally conclusive when, in point of extent and magnitude, the historick part of the subject bears no proportion to the fictitious. Of a very different nature is the power which the mind possesses over real and over imaginary occurrences: those, we have observed, it cannot alter; over these it exercises a power of varying them even to an unlimited degree. By whatever modes of combination these heterogeneous materials may

be connected, it must be therefore pretty evident, that the part which is fictitious must bend and accommodate itself, as being more ductile, to that which is real and unalterable; and that this circumstance will give the work that marked turn of feature, which is to determine its character as historick or marvellous. Fiction is, under its most fascinating appearance, of a rare and subtle nature; it may be rendered at once beautiful and considerable, from the extent to which it may be drawn out, and the exility with which it may be superinduced on the exteriour of any subject; but reality takes a more forcible hold of the observation, from the prominence and solidity with which it stands above the level of the surface. A romance, therefore, constructed on a historick subject, becomes a regular claimant, from its nature, to the title of a production founded on fact; and regarded in this light, what is historical in its composition must at least fix the æra, and determine the bounds of its subject. The case of the poetical romance becomes, in this view, analogous to that of the historick poem; both must be considered the expansion, in poetical language, of a certain number of facts, and of facts whereof the reader is supposed to possess a steady

view, and a perfect knowledge. If, therefore, the mind rejects, as an imperfection, the licence of alteration in the one species of production, how much more will it revolt against that unbounded fiction, which the other does not take as an appendage, but claims as a principal component in its productions?

It is almost superfluous to remark, that from the present consideration of the poetical romance, that case is excluded where the real occurrences, on which the work is founded, have undergone such alterations as prevents them from being known: for such a plan, though said, and with much propriety, to be founded on fact, must be considered purely fictitious, as facts appear no longer in its composition.

But in assuming the case of a romance being founded on history, if the subject of the poem appears to receive no improvement in its truth or probability, it is impossible it can be benefited by the alliance in any other particular. To give it a dash of individual nature," which gives a strength of colouring

w See Sir J. Reynolds' Notes on Du Fresnoy's Art of Painting as quoted by Mr. Twining, in his Notes to Aristotle's Poeticks, p. 509.

to all the compositions of art, may be conceived within the power of history, which is occasionally dedicated to the particular occurrences incident to extraordinary persons. But any accessions which it could derive in this respect, it may acquire from other sources, or may appropriate from the science in consideration, without being bound to adopt such attendant circumstances, as will make facts the basis of its subject, or give the production the title of being historical. Nor can the romantick poem derive any improvement from history in point of embellishment. For a liberty of giving such a direction to the facts and circumstances of the work as may suit the poet's caprice, being implied in the very nature of this species of composition as being fictitious, it follows, from the power which the mind possesses of improving in conception on almost every object submitted to the imagination, that the romantick poet inherits a greater power of contributing to our delight by realizing an imaginary creation, than he could possibly attain by following the direct track of real events, however splendid and dignified.

But though the poet is thus debarred from giving his composition a general alliance with