only a kind of opinion or consensus of men about a conventional distinction between what they choose to call truth and what they choose to call error." This view, as the author has confessed, he once tried to hold. Still this meaningless doctrine of relativity is not the same as the view that contents itself with the postulates before discussed. That view might take, and for the author at one time did take, the possible and intelligible form thus expressible : “Truth and error, though really distinguishable, are for us distinguished only through our postulates, in so far as relates to past and future time" Such views, while not denying that there is real truth, despair of the attainability for us of more than momentary truth. But the doctrine of Total Relativity, this view above expressed, differs from genuine skepticism. It tries to put even skepticism to rest, by declaring the opinion, that there is error, to be itself an error. This is not merely a moderate expression of human limitations, but jargon, and therefore formidable, because jargon is always unanswerable. When the famous Cretan declared all statements made by Cretans to be in all cases lies, his declaration was hard to refute, because it was such honest-seeming nonsense. Even so with the statement that declares the very existence of error to be an erroneously believed fancy. No consensus of men can make an error erroneous. We can only find or commit an error, not create it. When we commit an error, we say what was an error already. If our skeptical view in previous chapters seemed to regard truth and error as mere objects of our postulates, that was only because, to our skepticism, the real truth, the real error, about any real past and future, seemed beyond our reach, so that we had to content ourselves with postulates. But that real error exists is absolutely indubitable.

This being the case, it is evident that even the most thorough-going skepticism is full of assumptions. If I say, "There may be no money in that purse yonder," I assume the existence of the purse yonder in order to make just that particular doubt possible. Of course, however, just that doubt may be rendered meaningless by the discovery of the actual non-existence of that particular purse. If there is no purse yonder, then it is nonsensical either to affirm or to deny that it contains money. And so if the purse of which I speak is an hallucination of mine, then the doubt about whether, as an actually existent purse, it has money in it, is deprived of sense. My real error in that case would lie in supposing the purse itself to exist. If, however, I abandon the first doubt, and go on to doubt the real existence of the purse, I equally assume a room, or some other environment, or at all events the universe, as existent, in order to give sense to my question whether the purse has any being in this environment or in this universe. But if I go yet further, and doubt whether there is any universe at all outside of my thought, what does my doubt yet mean? If it is to be a doubt with any real sense, it must be a doubt still with an object before it. It seems then to imply an assumed order of being, in which there are at least two elements, my lonely thought about an universe, and an empty environment of this thought, in which there is, in fact, no universe. But this empty environment, whose nature is such that my thought does wrong to suppose it to be an universe, what is that? Surely if the doubt is to have meaning, this idea needs further examination. The absolute skepticism is thus full of assumptions.

The first European thinker who seems to have discussed our present problem was Plato, in a toomuch-neglected passage of the "Theætetus," 1 where Socrates, replying to the second definition of knowledge given by Theætetus, namely, knowledge is True Opinion, answers that his great difficulty has often been to see how any opinion can possibly be false. The conclusion reached by Plato is no very definite one, but the discussion is deeply suggestive. And we cannot do better here than to pray that the shade of the mighty Greek may deign to save us now in our distress, and to show us the true nature of error.

III.

Logicians are agreed that single ideas, thoughts viewed apart from judgments, are neither true nor false. Only a judgment can be false. And if a reasoning process is said to be false, the real error lies still in an actual or suppressed assertion. A fallacy is a false assertion that a certain conclusion follows from certain premises. Error is therefore generally defined as a judgment that does not agree with its object. In the erroneous judgment, sub1 Plato, Th., p. 187 sqq.

ject and predicate are so combined as, in the object, the corresponding elements are not combined. And thus the judgment comes to be false. Now, in this definition, nothing is doubtful or obscure save the one thing, namely, the assumed relation between the judgment and its object. The definition assumes as quite clear that a judgment has an object, wherewith it can agree or not agree. And what is meant by the agreement would not be obscure, if we could see what is meant by the object, and by the possession of this object implied in the pronoun its. What then is meant by its object? The difficulties involved in this phrase begin to appear as soon as you look closer. First then the object of the assertion is as such supposed to be neither the subject nor the predicate thereof. It is external to the judgment. It has a nature of its own. Furthermore, not all judgments have the same object, so that objects are very numerous. But from the infinity of real or of possible objects the judgment somehow picks out its own. Thus then for a judgment to have an object, there must be something about the judgment that shows what one of the external objects that are beyond itself this judgment does pick out as its own. But this something that gives the judgment its object can only be the intention wherewith the judgment is accompanied. A judgment has as object only what it intends to have as object. It has to conform only to that to which it wants to conform. But the essence of an intention is the knowledge of what one intends. One can, for instance, intend a deed or any of its consequences only in so far as he

:

foresees them. I cannot be said to intend the accidental or the remote or even the immediate consequences of anything that I do, unless I foresaw that they would follow; and this is true however much the lawyers and judges may find it practically necessary to hold me responsible for these consequences. Even so we all find it practically useful to regard one of our fellows as in error in case his assertions, as we understand them, seem to us to lead to consequences that we do not approve. But our criticisms of his opinions, just like legal judgments of his acts, are not intended to be exact. Common sense will admit that, unless a man is thinking of the object of which I suppose him to be thinking, he makes no real error by merely failing to agree with the object that I have in mind. If the knights in the fable judge each other to be wrong, that is because each knight takes the other's shield to be identical with the shield as he himself has it in mind. In fact neither of them is in error, unless his assertion is false for the shield as he intended to make it his object.

So then judgments err only by disagreeing with their intended objects, and they can intend an object only in so far forth as this object is known to the thought that makes the judgment. Such, it would seem, is the consequence of the common-sense view. But in this case a judgment can be in error only if it is knowingly in error. That also, as it seems, follows from the common-sense suppositions. Or, if we will have it in syllogistic form:

Everything intended is something known. The