More definitely, abstraction is indispensable if one experience is to be applicable in other experiences. Every concrete experience in its totality is unique; it is itself, non-reduplicable. Taken in its full concreteness, it yields no instruction, it throws no light. What is called abstraction means that some phase of it is selected for the sake of the aid it gives in grasping something else. Taken by itself, it is a mangled fragment, a poor substitute for the living whole from which it is extracted. But viewed teleologically or practically, it represents the only way in which one experience can be made of any value for another--the only way in which something enlightening can be secured. What is called false or vicious abstractionism signifies that the _function_ of the detached fragment is forgotten and neglected, so that it is esteemed barely in itself as something of a higher order than the muddy and irregular concrete from which it was wrenched. Looked at functionally, not structurally and statically, abstraction means that something has been released from one experience for transfer to another. Abstraction is liberation. The more theoretical, the more abstract, an abstraction, or the farther away it is from anything experienced in its concreteness, the better fitted it is to deal with any one of the indefinite variety of things that may later present themselves. Ancient mathematics and physics were much nearer the gross concrete experience than are modern. For that very reason they were more impotent in affording any insight into and control over such concretes as present themselves in new and unexpected forms. Abstraction and generalization have always been recognized as close kin. It may be said that they are the negative and positive sides of the same function. Abstraction sets free some factor so that it may be used. Generalization is the use. It carries over and extends. It is always in some sense a leap in the dark. It is an adventure. There can be no assurance in advance that what is extracted from one concrete can be fruitfully extended to another individual case. Since these other cases are individual and concrete they _must_ be dissimilar. The trait of flying is detached from the concrete bird. This abstraction is then carried over to the bat, and it is expected in view of the application of the quality to have some of the other traits of the bird. This trivial instance indicates the essence of generalization, and also illustrates the riskiness of the proceeding. It transfers, extends, applies, a result of some former experience to the reception and interpretation of a new one. Deductive processes define, delimit, purify and set in order the conceptions through which this enriching and directive operation is carried on, but they cannot, however perfect, guarantee the outcome.