ceivable." The reader will remember that we had to distinguish between real thinking and picture-thinking, thought and imagination. Sometimes more especially in the chapter on "Ultimate Religious Ideas "Spencer uses "conceivable" in one sense, sometimes in the other. It is evident that the conceivable and the imaginable are different mental products: thus even a pleasure or pain, considered as mere feeling, is not a thing of which we can form any kind of mental picture; but we can conceive it, for we know what the term means. In his Principles of Psychology Spencer lays down a test of truth, which he calls "the Universal Postulate," whose application-if it were reliable-would be frustrated by the confusion between thinking and imagining,—a confusion which it is very easy to fall into, in spite of great care taken to avoid it. He says that an ultimate truth is that whose

opposite" is "inconceivable." Now all arguments depending on the mere psychological experiment of trying whether something is "conceivable" or not, are most untrustworthy, mainly through the confusion just

mentioned; yet Spencer makes these arguments the bases of his system.

Apart from this difficulty, the "Universal Postulate" is no reliable test of truth. Among the various statements of it which Spencer has given, the following is one of the most precise :

An abortive effort to conceive the negation of a proposition shows that the cognition expressed is one of which the predicate invariably exists along with its subject; and the discovery that the predicate always exists along with its subject, is the discovery that this cognition is one which we are compelled to accept.

One of our ablest living psychologists, commenting on this passage, has well said that this is no test of truth; it is only a test of belief :

It may always happen that what is found to be unthinkable at one time may become thinkable at another, and vice versa. New data and new points of view are always capable of working a change; to fall back on the unthinkableness of the opposite as a test of truth is simply to shut our eyes to further evidence; logically, Mr Spencer's position is identical with that attributed to ladies-it simply amounts to saying that "a thing is because it is." It makes the mere existence of a belief, or at any rate of a full assurance, the evidence of its truth.1

1 G. F. Stout, Analytic Psychology, vol. ii. p. 241.

causes.

It is almost startling to find that Spencer's position here is the same as that of Cardinal Newman, who makes "indefectibility of certitude" the test of truth. The "indefectibility" is equivalent to "inconceivability of the opposite." And as we saw, it amounts to no more than a very strong mental impulse to hold certain beliefs: to find in this impulse, as a mere fact, a reason for the belief, is to say that there are no reasons -only feelings, impulses, and psychological In any case Spencer seems to make a rather arbitrary selection among the “firm assurances" of consciousness. One of the firmest is the conviction of every one that "himself" is more than the mere succession of his thoughts and feelings: this Spencer rejects as illusion. Another is the conviction that in experience we have a direct hold upon reality in many of its details, and not merely upon a substitute coming between us and things as they are; but Spencer affirms that we know only appearance, not reality. The view which we have advanced is that the reality is known through the phenomenon; the knowledge

may be very partial and very fragmentary, but it is none the less true of reality, until it is corrected by fuller truth.

The second great difference between Spencer and Comte is in their respective views of the surviving element in religion. Spencer thinks it is only the feeling of awe in the presence of an Infinite Mystery; this, he says, is "the truly religious element in religion," which "has always been good." "The consciousness of an inscrutable power manifested to us through all phenomena," a power "whose nature transcends intuition and is beyond imagination," "gives the religious sentiment the widest possible sphere of action." The chapter in which Spencer expounds this conclusion is one which can hardly fail to make a deep impression on the reader. I must quote in full its most important passage:

Those who espouse the alternative position [to agnosticism, i.e., Theism] make the erroneous assumption that the choice is between personality and something lower than personality; whereas the choice is rather between personality and something higher. Is it not possible that there is a mode of Being as much transcending

intelligence and will as these transcend mechanical motion? It is true that we are totally unable to conceive any such higher mode of being. But this is not a reason for questioning its existence; it is rather the reverse. Have we not seen how utterly incompetent our minds are to form even an approach to a conception of that which underlies phenomena? Is it not proved that this incompetency is the incompetency of the Conditioned to grasp the Unconditioned ? Does it not follow that the Ultimate Cause cannot in any respect be conceived by us because it is in every respect greater than can be conceived? And may we not, therefore, rightly refrain from assigning to it any attributes whatever, on the ground that such attributes, derived as they must be from our own natures, are not elevations but degradations?

Indeed it seems somewhat strange that men should suppose the highest worship to lie in assimilating the object of their worship to themselves. Not in asserting a transcendent difference, but in asserting a transcendent likeness, consists the element of their creed which they think essential. It is true that from the time when the rudest savages imagined the causes of all things to be creatures of flesh and blood like themselves, down to our own time, the degree of assumed likeness has been diminishing. But though a bodily form and substance, similar to that of man, has long since ceased among cultivated races to be a literally conceived attribute of the Ultimate Cause-though the grosser human desires have also been rejected as unfit elements of the conception-though there is some hesitation in ascribing even the higher human feelings, save in greatly idealised shapes, yet it is still thought not only proper, but imperative, to ascribe the most abstract qualities of our nature. To think of the Creative Power as in all