asked silently to countenance as useful substitutes for right reason.  ‘Erroneous motives,’ as Condorcet has expressed this matter, ’have an additional drawback attached to them, the habit which they strengthen of reasoning ill.  The more important the subject on which you reason ill, and the more you busy yourself about it, by so much the more dangerous do the influences of such a habit become.  It is especially on subjects analogous to that on which you reason wrongly, or which you connect with it by habit, that such a defect extends most powerfully and most rapidly.  Hence it is extremely hard for the man who believes himself obliged to conform in his conduct to what he considers truths useful to men, but who attributes the obligation to erroneous motives, to reason very correctly on the truths themselves; the more attention he pays to such motives, and the more importance he comes to attach to them, the more likely he will be to go wrong.’[9] So, in short, superstition does an immense harm by enfeebling rational ways of thinking; it does a little good by accidentally endorsing rational conclusions in one or two matters.  And yet, though the evil which it is said to repair is a trifle beside the evil which it is admitted to inflict, the balance of expediencies is after all declared to be such as to warrant us in calling errors useful!

III.  A third objection now presents itself to me, which I wish to state as strongly as possible.  ’Even if a false opinion cannot in itself be more useful than a true one, whatever good habits may seem to be connected with it, yet,’ it may be contended, ’relatively to the general mental attitude of a set of men, to their other notions and maxims, the false opinion may entail less harm than would be wrought by its mere demolition.  There are false opinions so intimately bound up with the whole way of thinking and feeling, that to introduce one or two detached true opinions in their stead, would, even if it were possible, only serve to break up that coherency of character and conduct which it is one of the chief objects of moralists and the great art of living to produce.  For a true opinion does not necessarily bring in its train all the other true opinions that are logically connected with it.  On the contrary, it is only too notorious a fact in the history of belief, that not merely individuals but whole societies are capable of holding at one and the same time contradictory opinions and mutually destructive principles.  On the other hand, neither does a false opinion involve practically all the evil consequences deducible from it.  For the results of human inconsistency are not all unhappy, and if we do not always act up to virtuous principle, no more do we always work out to its remotest inference every vicious principle.  Not insincerity, but inconsistency, has constantly turned the adherents of persecuting precepts into friends of tolerant practice.’

’It is a comparatively small thing to persuade a superstitious person to abandon this or that article of his superstition.  You have no security that the rejection of the one article which you have displaced will lead to the rejection of any other, and it is quite possible that it may lead to all the more fervid an adhesion to what remains behind.  Error, therefore, in view of such considerations may surely be allowed to have at least a provisional utility.’