ON FLOATING IDEAS AND THE IMAGINAET. 471 observed, no consistency is required even in his imaginary character. 1 The control in a word has become theoretical as well as moral. The playing dog knows, we may say, that beyond a certain point he must not. The playing boy knows this in all his playing, and in some cases he knows no more. But in other cases he knows also that beyond a certain point the thing is not. He has here a world of imagination, qualify- ing the real world but always subject to and restrained by that world. 2 These two controls, the moral and theoretical, are in much play so joined and blended that to separate their several effects would be hard or impossible. In 'playing a part,' on the stage or again in real life, this intimate mingling may be observed. We have first of all the letting- go of certain activities subject to a certain moral restraint. But we have in addition the entrance, for ourselves or for others or for both, into the sphere of pretence make-believe and illusion. This entrance is limited by our consciousness of the real fact, and again by conventional rules wherever and so far as these exist. And to what extent the control is before the mind, and how far illusion actually is present, depends in every case upon the conditions and the individual. 3 Thus pretence and make-believe do not belong to the general essence of play. They are obviously present often where there is no playing and where they are used consci- 1 So of course mutati-s mutandis with the girl and her doll. On the above point the reader may be referred to. Prof. Sully's Human Mind, i., 384. 2 1 do not mean that in playing the moral or theoretical control must be kept always before my mind. As we saw before, it is enough that this control should be ready at any moment to come in, and that any suggestion of excess should at once bring it before me, or at least bring it into action. 3 The amount of actual illusion is said, for instance, to differ widely with different actors. See Mr. Archer's well-known collection of facts in his Masks or Faces. Again flirting, the amatory game, is an instance where it is not easy to distinguish between the two kinds of control, theoretical and moral. The amount of illusion or pretence varies widely in various cases, and in many cases probably amounts to nothing. You may have simply the letting loose of certain sexual feelings and actions without pretence or illusion but within a certain moral limit The be- ginning of this is easy to observe, for example, among dogs. The main essence of the affair is in short not illusion but limit. That is why (as Prof. Groos rightly observes in The Play of Man, p. 253) we do not in the same way play at eating, for there short of the satisfaction of appetite the means are not by themselves sufficiently agreeable. But to a certain extent, I should say, we may play at eating, for instance at dessert. And children play thus habitually, I suppose because the real satisfaction is out of their power. But here of course the imaginary element comes in and is important.
