No man is in a position to affirm that there is an antagonism between science and religion until he has first clearly determined the sphere and function of each, and can say distinctly _what science is and what religion is_. He may have utterly misconceived the nature of religion, or he may have misapprehended the function of science, and therefore the supposed antagonism may be purely imaginary. For example, Herbert Spencer says, "Every religion may be defined as an _à priori_ theory of the universe."[409] If this definition were correct, we could easily conceive how religion and modern science might come into collision, because the tendency of science at the present time is to occupy itself with "questions of origin"--that is, with "theories of the origin of things," instead of being, as Spencer defines it, "a systematic collection of facts, ascertained with precision, and so classified and generalized as to reveal the uniform relations of co-existence and succession among phenomena, and thus give _prevision_." This is the legitimate sphere of all that science which can lay any claim to be regarded as "exact science." When it transcends this limit it ceases to be science and becomes philosophy--a philosophy which will be more or less valid and legitimate as it recognizes the authority and submits to the guidance of _à priori_ ideas of the reason. But is Mr. Spencer's definition of religion correct? We think not. Indeed, it would be difficult to give a definition of religion wider from the mark. He might with just as much propriety have said that religion is an _à priori_ theory of the origin of language, of government, of trade, or of music. Either Mr. Spencer must have made this definition for an unworthy purpose, or he must be in utter darkness as to the nature of religion. One needs only to cast a hasty glance over the history of ancient religions, or to consider with an unprejudiced mind any of the contemporaneous forms of religion, to be convinced that religion is, and always has been, a mode of life determined by the sense of dependence upon a Supreme Power.[410] Religion has always been a matter of practical interest and personal concernment, and has no more to do with "theories of the universe" than with theories of light, or theories of electricity, or theories of political economy.