It will be agreed that a less cryptic account of good, if one can be given, which is in accordance with verifiable facts, would be preferable, even though no means were available for refuting the simpler theory. Upholders of this theory, however, have produced certain arguments to show that no other view of good is possible, and these must first be briefly examined. They provide, in addition, an excellent example of the misuse of psychological assumptions in research, for although a psychological approach is often of the utmost service, it can also be a source of obscurantism and over-confidence. The arguments against any naturalistic account depend upon the alleged results of directly inspecting what is before our minds when we judge that anything is good. If we substitute, it is maintained, any account of good whatever for ‘good’ in the assertion, ‘This is good’—for example, ‘This is desired’ or ‘This is approved’—we can detect that what is substituted is different from ‘good’, and that we are not then making the same judgment. This result, it is claimed, is confirmed by the fact that we can always ask, “Is what is desired, or what is approved, good?” however we may elaborate the account provided, and that this is always a genuine question which would be impossible were the substituted account actually the analysis of good. The persuasiveness of this refutation is found to vary enormously from individual to individual, for the results of the experiments upon which it relies differ. Those who have accustomed themselves to the belief that good is a supersensuous simple Idea readily discover the fraudulent character of any offered substitute, while those who hold some psychological theory of value, with equal ease identify their account with ‘good’. The further question, “When and under what conditions can judgments be distinguished?” arises, a question so difficult to answer that any argument becomes suspect which depends upon assuming that they can be infallibly recognised as different. If for any reason we wish to distinguish two judgments, we can persuade ourselves, in any case in which they are differently formulated, that they are different. Thus it has been thought that ‘_a_ exceeds _b_’ and ‘_a_ is greater than _b_’ are distinguishable, the first being supposed to state simply that _a_ has the relation ‘exceeds’ to _b_, while the second is supposed to state that _a_ has the relation ‘is’ to _greater_ which again has the relation ‘than’ to _b_.[18] The conclusion to be drawn from the application of such methods to the problem of the meaning of Good would seem to be that they are not competent to decide anything about it—by no means a valueless result.