Objects of Representations and Kant's Second Analogy

Objects of Representations and Kant's Second Analogy STEVEN M. BAYNE IT IS UNIVERSALLYacknowledged that in the Second Analogy Kant attempted to prove a law of causality. In addition to disagreement over whether or not his argument actually is a proof of a law of causality, however, there are a number of other disagreements concerning the Second Analogy. The first of these concerns the exact nature of the law of causality which Kant attempts to prove. Another disagreement concerns the exact nature of the strategy which Kant utilizes in his attempt to prove the law of causality. In what follows I begin by very briefly examining the positions which have been adopted concerning these two disagreements. I then argue that Kant attempts to prove that the truth of the causal principle is something which is required by an adequate understanding of the nature of events. I argue, that in addition to ordinary physical objects such as a house, kant also holds that events such as a ship floating down the stream or a sample of water freezing are objects of representations . Furthermore, Kant argues that in order for an event tobe an object of representations, it must be subject to the law of causality. Finally I conclude by considering some objections to this interpretation. 1. What exactly is the causal principle which Kant hoped to prove in the Second Analogy? There are two main possibilities. The first possibility is the EveryEvent -Some-Cause, thesis. This first possibility is simply the principle that all events have some cause. EESC = For every event e there is some event c and c causes e.2 The names I have assigned to these three possibilitiesare basicallythe same as those used by L. W. Beck.See"A Prussian Hume and a ScottishKant,"inEssaysonKant and Hume (NewHaven: Yale University Press, 1978), 1~o. ' Commentators who have adopted the EESC formulation include the following: Henry E. Allison(Kant'sTranscendentalIdealism),LewisWhite Beck("APrussian Hume and a ScottishKant" [38q 382 JOURNAL OF THE HISTORY OF PHILOSOPHY 3e:3 JULY ~994 The second possibility is the Same-Cause-Same-Effect thesis. This is the principle we get if we add to EESC the clause that if one individual event causes a second individual event, then whenever any event of the first type occurs, an event of the second type will follow.s SCSE = 1) For every event e there is some event c and c causes e, and ~) Ifc causes e, then there is some event type C of which c is an instance and some event type E of which e is an instance such that whenever an event of type C occurs an event of type E will follow.4 Although neither of these formulations of the causal principle requires a great deal of explanation, it will be helpful to take note of some of the similarities and differences between them. The first point of importance is that of these formulations EESC is the only one which deals exclusively with particular events. That is, EESC makes no mention of event types. This is important because any claim which asserts that some causal generalization does or does not hold must be a claim which mentions not only particular events, but also event types.5 So, according to EESC, a successful proof of the causal principle will leave all options concerning causal generalizations open. That is, proving the causal principle, on this reading, will not guarantee that there are any true causal generalizations, nor will it rule them out. Proving the causal principle, then, will only guarantee that for any particular event there is some particular event which causes it. It will also be noted, from what has just been said, that the EESC formulation will, since it makes no claim one way or the other concerning causal in Essays on Kant and Hume), Norman Kemp Smith (A Commentary to Kant's "Critique of Pure Reason"). sThis statement should be taken to have the same meaning as the second clause in the formalized version. What the formalized version makes clear is that this formulation should not be taken to require that there is a true generalizationwhich connects everyevent type of...