To assume that one can make the patient forget such worries as these without first discovering what they are, is obviously fatuity at its grossest. Moreover, as we have seen, it is quite insufficient merely to discover that the patient is “suffering from hallucinations” or delusions and then to tell him to dismiss them from his mind. To suppose that, without understanding the nature of and the specific reasons for the development of a particular hallucination, one can “make the patient forget” his interpretation of a real experience which has appealed to him night and day for weeks, or banish a delusion which is gradually becoming systematised and rationalised—_i.e._, intimately interwoven into the tissues of the whole of his experience—is an assumption which has no foundation in fact. The point cannot be too much emphasised that many of these patients are quite sane, if conduct be regarded as the criterion of sanity; but they are growing afraid of the appearance of these abnormal phenomena, and take them for signs of incipient—or, more usually perhaps, of established—insanity. Hence follows the important corollary that while treatment by isolation has obvious advantages in certain cases, in the particular group of patients which we are now discussing it is often dangerous, for the reasons already emphasised in the last chapter. The presence of such mental phenomena is usually confided to the physician only after great hesitation, and such worrying experiences are common in cases of insomnia and other disorders, which, though troublesome, do not appear to be grave. It is therefore possible that isolation may have serious effects in many cases in which its net result seems merely to be that the patient is no better. It is granted then that in some instances (by no means all), the patient may be temporarily pained by the dragging into daylight of the causes of his worry, but it is usually a case of _reculer pour mieux sauter_. This procedure is often inevitable in the medical treatment of many disorders which have become complicated to any considerable extent.