254 INTRODUCTION TO PHILOSOPHY OF SCIENCE 



order: empirical foundation 



It seems obvious at the outset that nature does exhibit 

 situations which can be spoken of as orders, or arrangements, 

 and the initial problem is to select events of this kind and 

 examine their properties. Common sense is assured that 

 complexes such as the following should be spoken of as 

 series : the days of the week, the months of the year, points 

 on a line, the floors of a building, the colors in the spectrum 

 arranged in the usual way, individuals arranged according 

 to height, automobiles arranged according to cost, and so 

 on. Spatial and temporal relations are predominant in 

 serial situations, though they are not usually supposed to 

 define the orders in question. For example, the individuals 

 in a group may be said to have serial properties by virtue 

 of their differences in weight, yet individuals need not be 

 arranged in an actual order with the heaviest at one end, 

 the lightest at the other, and all intervening individuals 

 placed properly on the line joining the two. In the same 

 way, one may think of his pleasures as having an ordering 

 of intensity, though no physical arrangement of them would 

 be possible. Order, therefore, seems to be applicable to com- 

 plexes whose elements exhibit certain characteristic relation- 

 ships, and the problem is to determine the nature of these 

 relationships. 



Agreement can be reached, in the first place, on the fact 

 that order is applicable only to events exhibiting a certain 

 complexity. A single, unitary event may not be ordered; 

 such an event may itself have position in a series, but it is 

 then considered in its relation to other events, and the total 

 group of events exhibits order. Serial arrangement is pos- 

 sible only with reference to the elements of a complex event. 

 But how great must the complexity be; e.g., would a two- 

 term complex suffice? The answer to this question depends 

 upon the level at which the empirical analysis has been 

 made. There is some justification for the contention that 

 order exists wherever there is an asymmetrical relation; l 



1 For the definition of this notion see p. 123. 



