Shall we, without such adherence to a particular pattern, assume it to be either indisputable or, at least, most probable that an extensive corruption of the text can hardly have been avoided[43]; and shall we, in consequence, hold the received text provisionally, and subject to excision or to amendment according to any particular theory concerning Homer, his age, its manners and institutions, which we may ourselves have thought fit to follow or construct? Shall we admit as authoritative, the excisions of Aristarchus or the Alexandrian critics, and the _obeli_ which he has placed against verses which he suspected? Or shall we proceed, as a general rule, upon the belief, that the received text of Homer is in general sound and trustworthy, so far, at least, as to be very greatly preferable to any reconstructed or altered form whatever, in which it has hitherto been produced or proposed for our acceptance? My decided preference is for the fourth and last of these alternatives: with the observation, however, in passing, that the third does not essentially differ from it with respect to the great body of the Poems, so far as we know what the Alexandrian text really was. I prefer this course as by far the safest: as the only one which can be entered upon with such an amount of preliminary assent, as to secure a free and unbiassed consideration of Homeric questions upon a ground held in common: and as, therefore, the only one, by means of which it can be hoped to attain to solid and material results as the reward of inquiry. In order fairly to raise the issue, the two following propositions may be stated as fitting canons of Homeric study:-- 1. That we should adopt the text itself as the basis of all Homeric inquiry, and not any preconceived theory, nor any arbitrary standard of criticism, referable to particular periods, schools, or persons. 2. That as we proceed in any work of construction by evidence drawn from the text, we should avoid the temptation to solve difficulties found to lie in our way, by denouncing particular portions of it as corrupt or interpolated: should never set it aside except upon the closest examination of the particular passage questioned; should use sparingly the liberty even of arraying presumptions against it; and should always let the reader understand both when and why it is questioned. Now, let us consider these rules, and the method which it is proposed by means of them to apply, _a._ With reference to the failure of other methods. _b._ With reference to the antecedent probabilities for or against the general soundness of the text. _c._ With reference to the internal evidence of soundness or unsoundness afforded by the text itself.