Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Posting-Version: version B 2.10.1 6/24/83; site umcp-cs.UUCP Path: utzoo!linus!philabs!cmcl2!seismo!umcp-cs!mangoe From: mangoe@umcp-cs.UUCP (Charley Wingate) Newsgroups: net.philosophy Subject: Re: Mechanism and Determinism Message-ID: <1185@umcp-cs.UUCP> Date: Sat, 10-Aug-85 20:53:22 EDT Article-I.D.: umcp-cs.1185 Posted: Sat Aug 10 20:53:22 1985 Date-Received: Mon, 12-Aug-85 08:33:11 EDT References: <573@mmintl.UUCP> Organization: U of Maryland, Computer Science Dept., College Park, MD Lines: 82 In article <573@mmintl.UUCP> franka@mmintl.UUCP (Frank Adams) writes: >There has been an underlying in a number of recent discussions recently >which I think is invalid. This is, to put it baldly, that randomness >equals freedom. For example, Charley Wingate states that the hypothesis >that some behavior reflects quantum fluctuations has no evidence against >it, and seems to feel that he has proved something thereby. > >Unless quantum mechanics is very wrong, random events are part of the way >the universe works. Determinism is a dead issue. But this does not imply >that the behavior of complex systems is not reducible to the interactions >of their components. > >Let me define what I will mean by a mechanistic system. A system is >mechanistic if its components, at a sufficiently fine level of detail, >can have their behavior in any situation described completely by a >probability distribution, and the behavior of the system is *in principle* >describable by this behavior, given the (initial) relationships of the >system. > >The modern equivalent of determinism is to assert that the entire universe >is mechanistic. I see no reason to regard a resident of a mechanistic >universe as any more "free" than the resident of a strictly deterministic >universe. Both run equally counter to my sujective sense of free will. As I understand it, free will is essentially the assertion that the process of consciousness can in fact decide; it is the antithesis of the assertion that all human behavior is caused. It seems to me that acausality can occur on two different levels: the processing mechanism itself can be acausal, and their can be a lack of causation at the information level. The second clearly produces precisely what we all seem to agree to call free will. Frank questions whether or not the first does. The first problem comes when you approach the question of probability. The question of the level of detail is indeed quite important; clearly one cannot deal with anything larger than the behavior of an individual. Even if group behavior is constrained by a probability distribution, there is nothing preventing each individual from exerting freedom in choosing where he lies in the distribution; in addition, there is clearly a feedback loop from the distrbution back to the behavior which severely limits the form that the distribution can be expressed in, and which adds a strong uncertainty element. One can therefore ask whether this same objection could be applied to individual behavior. I think it does apply. There is still the same feedback loop, and again, at any one time the individual only has to show up at one point on the curve. The distribution may in fact only describe the tendencies of the deciding apparatus, in which case it's clearly improper to say that the distribution constrains the choice. To deny free will, therefore, I think you must show two things: (1) That random fluctuations in processing are unimportant. (2) That all behavior can be traced inevitably to external causes. This leads us to to a long list of positions: (A) Soulism: A supernatural entity is the ultimate cause of behavior. (B) Physical Soulism: A physical entity unlike matter or energy as we know it is the ultimate cause of behavior. (C) Quantum indeterminism: Quantum events are so important that the sources of behavior are truly random. (D) Natal Independence: The initial presence of randomness in the brain precludes the development of causality in the mind. (E) Combinations of C and D. (F) Various combinations of the denial of the above. I think we can ignore the first two. The third essentially denies mechanism, since it asserts that gross behavior is truly not predictable, even probabilistically. One can argue at length whether or not this constitutes free will. Position D, however, is the most potent. It essentially asserts that, however the brain works, there is an initial random component at birth which does not dissipate. In this case, processing can be mechanistic, yet there is still free will in the strongest sense. C WIngate