第16章 「非論証的推論」n.10 - 各出来事の集合はその「物」の瞬間的状態
ある事象の系のどの一つ事象からでもそれに近接して起る事象について何らかのことが推論できるという特性(property 属性)を持つ時、そのような事象の系を私は「因果の線」 (causal line)と名づける。「物(thing)」の概念を常識にとって有用なものにし、「物質matter)」の概念を科学にとって有用なものにしたのは、上のような因果の線が存在するという事実である。また、現代物理学をし 「物質」の概念は不十分なものであると考えさせるにいたったものは、この因果の線は近似的なものであり、不変でなく、普遍的なものではない、という事実である。 |
Chapter 16: Non-Demonstrative Inference , n.10It used to be the custom among philosophers to think that causal laws can be stated in the form 'A causes B', interpreted as meaning that whenever an event of a certain kind A occurs, it is followed by an event of another specified kind, B. It was held by many that a causal sequence involves something more than invariability and must have some character that can be called ‘necessity'. Many empiricists, however, denied this and thought that nothing was involved except invariable sequence. This whole point of view, however, could never have persisted among philosophers if they had had any acquaintance with science. Causal laws must be either not invariable or such as state only tendencies. In classical dynamics they take the form of differential equations, stating acceleration, not actual occurrences. In modern physics the laws have become statistical: they do not state what will happen in any particular case, but only different things, each of which will happen in an assigned proportion of cases. For such reasons, causation is no longer what it used to be in the books of old-fashioned philosophers. Nevertheless, it still retains an essential place. Take, for example, what we mean by a single 'thing' which is more or less persistent. This ‘thing’ must really consist of a series of sets of occurrences, each set characterizing what we may call a momentary state of the ‘thing'. The states of the ‘thing' at different times are, often, though not always, connected by means of laws which can be stated without mentioning other ‘things’. If this were not the case, scientific knowledge could never get a start. Unless we can know something without knowing everything, it is obvious that we can never know something. And this applies, not only to particular events, but also to the laws connecting events. In physics, atoms and molecules persist for a time, and, if they did not, the conception of motion would become meaningless. A human body persists for a time, although the atoms and molecules of which it is composed are not always the same. A photon which travels from a star to a human eye persists throughout its journey, and, if it did not, we should not be able to state what we mean by seeing a star. But all these kinds of persistence are only usual, not invariable, and the causal laws with which science begins must state only an approximation to what usually happens. Whether, in the end, something more exact is attainable, we do not know. What I think we can say is something like this: given any event, there is usually, at any neighbouring time and in some neighbouring place, an event very like the given event; and, as a rule, it is possible to discover some law approximately determining its small difference from the given event. Some such principle is necessary to explain the approximate persistence of many 'things', and also to explain the difference between perceiving A and perceiving B - for example, if A and B are stars, both of which we are seeing.I give the name of 'causal line' to a series of events having the property that from any one of them something can be inferred as to neighbouring events in the series. It is the fact that such causal lines exist which has made the conception of 'things' useful to common sense, and the conception of 'matter' useful to physics. It is the fact that such causal lines are approximate, impermanent, and not universal which has caused modem physics to regard the conception of 'matter' as unsatisfactory. |