The difficulty that presents itself lies mainly in the confusion between facts and our interpretation of facts. If A. declares that he saw a ghost, he does not relate a fact, but his interpretation of a fact. Let us suppose that he tells his story again to a third person D., who is a psychologist. D. most likely will not think him a liar. D. will accept the statement _bona fide_ as a mere interpretation of a fact and will inquire after the causes that produced the hallucination. He may be able, possibly, to lay bare the facts disfigured by the wrong interpretation of A. And having clearly stated the objective state of things he may with the assistance of his experience explain the origin of the whole process, partly from the mental condition and the physiological constitution of A., partly from individual circumstances that gave rise to the hallucination. He will not doubt that something extraordinary has happened to Mr. A. The latter's mind has been, and perhaps still is in an abnormal state. And as to B.'s believing the ghost story, Mr. D. will not think that he is insane; though we may presume that he will regard B.'s views of the world as resting upon unfirm grounds; and he will not believe him to be a man of critical ability. The notion is very common among idealists that we can never go beyond our subjective states of consciousness. This would be tantamount to saying that there is no difference between dreams and real life, except that a dream is cut off by awaking while life lasts comparatively much longer and ceases with death, which may also be an awakening from a dream. In that case hallucinations would be of the same value as sensations. Both would be interpretations of facts for which we do not have an objective criterion of truth. Interpretations of facts would be the sole facts, and it would be quite indifferent whether they were misinterpretations or correct interpretations. Take a simple instance. We see a tree. The perception of a tree is an interpretation of a set of facts. Interpretations of facts, whether correct or not, are of course also facts. Thus the perception of a tree is a fact which, if all matter were transparent, would, physiologically considered, appear to the eye of an observer as special vibrations in the brain. But the peculiarity of this fact is that it represents other facts. The question is no longer whether there is a perception of a tree taking place in a brain, but whether this perception is true, i. e., whether it agrees with the facts represented. Every perception has a meaning beyond itself; every perception is a fact representing other facts, and the question of truth or untruth has reference to the agreement between representations and facts represented. Professor Mach says in his essay "The Analysis of Sensations" (_The Monist_, Vol. I. No. I, p. 65): "Bodies do not produce sensations, but complexes of sensations (complexes of elements) form bodies."[53]