The effect is twofold; an appearance of security and stability is given to the attitude, which thus seems to be justified; and at the same time it is no longer so necessary to sustain this attitude by the more difficult means peculiar to the arts, or to pay full attention to form. The reader can be relied upon to do more than his share. That neither effect is desirable is easily seen. The attitude for the sake of which the belief is introduced is thereby made not more but less stable. Remove the belief, once it has affected the attitude; the attitude collapses. It may later be restored by more appropriate means, but that is another matter. And all such beliefs are very likely to be removed; their logical connections with other beliefs scientifically entertained are, to say the least, shaky. In the second place these attitudes, produced not by the appropriate means but, as it were by a short cut, through beliefs, are rarely so healthy, so vigorous and full of life as the others. Unlike attitudes normally produced they usually require an increased stimulus every time that they are reinstated. The belief has to grow more and more fervent, more and more convinced, in order to produce the same attitude. The believer has to pass from one paroxysm of conviction to another, enduring each time a greater strain. This substitution of an intellectual formula for the poem or work of art is of course most easily observed in the case of religion, where the temptation is greatest. In place of an experience, which is a direct response to a certain selection of the possibilities of stimulation, we have a highly indirect response, made, not to the actual influences of the world upon us, but to a special kind of belief as to some particular state of affairs.[101] There is a suppressed conditional clause implicit in all poetry. If things were such and such then . . . and so the response develops. The amplitude and fineness of the response, its sanction and authority, in other words, depend upon this freedom from actual assertion in all cases in which the belief is questionable on any ground whatsoever. For any such assertion involves suppressions, of indefinite extent, which may be fatal to the wholeness, the integrity of the experience. And the assertion is almost always unnecessary; if we look closely we find that the greatest poets, as poets, though frequently not as critics, refrain from assertion. But it is easy, by what seems only a slight change of approach, to make the initial step an act of faith, and to make the whole response dependent upon a belief as to a matter of fact. Even when the belief is true, the damage done to the whole experience may be great, in the case of a person whose reasons for this belief are inadequate, for example, and the increased temporary vivacity which is the cause of perversion is no sufficient compensation. As a convenient example it may be permissible to refer to the