|
|
| date | 2000-11-21:15:23 |
| Physics |
That'll teach me to go writing stuff based on half-remembered papers from the late '80's. The problem with the solar corona (the outer layer of the sun's atmosphere) is still not solved, it turns out. There are a few references on the Web to the problem: 1) Solar overview 2) Solar corona 3) Shuttle experiments It's clear there's something weird going on, but it isn't clear if the claim made in the paper I was thinking of--that the phenomenon is a gravitationally-driven Maxwell's Demon--is supportable. It appears that there is more electromagetic involvement than expected, which kind of makes sense given that the temperature of the corona is something like a million degrees, and therefore pretty heavily ionized. One thing I didn't mention explicitly yesterday was that the presence of long-range forces like gravity in the thermodynamics of the solar corona give an "out" to the 2nd Law of Thermodynamics, which is why it's possible to even contemplate the existence of a demon in the solar atmosphere. The presence of these forces effectively takes the system out of thermodynamic equilibrium, and the validity of the 2nd Law depends on things being in thermodynamic equilibrium. There is therefore an interesting border-land between systems that are essentially ballistic in their behavior and systems that are essentially thermal in their behavior where it isn't clear (at least to me) that demon's can't exist. Maxwell's Demon is an instance of a "perpetual motion machine of the second kind", which are generally held to be impossible because they violate the 2nd Law. Perpetual motion machines of the first kind are far less interesting, because they violate the law of conservation of energy, which has but a single loophole and that one is thoroughly plugged. Energy is conserved so long as the laws of nature do not vary with time, and it is known empirically that they don't. The 2nd Law of Thermodynamics, however, applies in a more limited context, and I think it's fair to wonder: if we step outside that context might we find something interesting? The field of non-equilibrium thermodynamics is ferociously complex, but one of the deep truths it reminds us of is that the 2nd Law is at its root statisitcal: it says that heat does not usually flow from a cold body to a hot one spontaneously. "Usually" has a lot of force: you'd have to wait many, many times the lifetime of the universe for your cold cup of coffee to spontaneously absorb heat from the room, but the fact remains that it isn't impossible. Indeed, the Poincare' recurrence theorem, which states that a closed mechanical system will always return to a configuration arbitrarily close to its starting configuration, suggests that it must happen, eventually. One might say then that building a perpetual motion machine of the second kind is certainly possible, but so far as we know building a reliable perpetual motion machine of the second kind is impossible. On the other hand, a world that tolerates the level of reliability supplied by Microsoft might just be ready for a source of miniscule amounts of energy that only works every few hundred million years. |
| SelfConciousness |
The reason for all this muttering about thermodynamics is twofold: one is that there seems to be a deep connection between the kind of unknowability we get in the quantum world with thermodynamic phenomena, which are largely based on unknowably. The second is that I don't know much about statistical mechanics (I recall a lecture where the prof derived the basic equations and then commented, "After that it's all just fiddling around with the partition function") and it seems like a good time to learn. My thoughts on probability and thermodynamics are pretty free-form just now, more like notes to myself than anything else. In the next little while I will revisit them in a more systematic way, pulling out small peices that will form the backbone of whatever argument I wind up making. My approach has always been to do this kind of survey of the landscape first; the difference now is that Caro has shown me how to order the results of the survey a lot more systematically, which is the next step. |
| Metaphysics |
Richard Taylor in Metaphysics seems to want to argue strongly in favor of fate. In doing so, he reveals one of the oddities of classical logical that I'm still wrapping my head around, which is the idea that truths are timeless. I don't agree with this, and I don't see what force or meaning it can have. It assumes without justification a particular model of time and change that I don't think stands up to very much scrutiny at all. In particular, I think that such a model is incapable of dealing with probability. In the essay, "How the Brain Does Probable Reasoning" Edwin Jaynes points out the inadequacy of a logic that cannot deal with probability, and lays out a basis for probability theory along Bayesian lines that is quite remarkable. He then shows how this generalizes classical logic to deal with cases where the truth is not known; and it may even apply to cases where the truth is unknowable. Jayne's approach takes plausibility as the unanalyzed, atomic quality that forms the empirical foundatoin for the concept PROBABILITY. Probability, in his view, is the formal quantification of plausibility. On this basis, it seems reasonable to consider a logic where truth and falsity are limiting cases of plausible and implausible, so the idea of eternal truths--things that are true now and forever--is a limiting case of a limiting case. So far from being unavoidable, it would seem that Taylor's view of truth is quite narrow. ======================== One funny thing about Taylor's argument is that he claims that belief in fatalism has some necessary consequences for people who adopt it, including freedom from temptation. As he himself seems to want to believe in fatalism--which he defines as the belief that the future is as fixed and unchangeable as the past--it is not clear to what he thinks "temptation" refers, as clearly a creature that has no choices cannot be tempted, for what is temptation if not temptation to choose? Fatalism, far from being the moral pancea Taylor presents it as, can't lead to anything but amoralism, for only creatures that have choice have any need of morality. Note that having choice does not imply a need of morality: cats can clearly make choices, but are equally clearly devoid of morals, and seem none-the-worse for it. But not having choice certainly implies not needing (or being capable of) morality, as the only role of morality is to guide us in making moral choices. |
| Reading |
More thoughts on David Lodge's Out of the Shelter. The "shelter" of the title is a bomb shelter, where Timothy spends many nights as a child in London during the Blitz. People who go out of the shelter are risking death, and some do die. For a book written in the 1960's, Lodge is surprisingly aware of the failure of area bombing, which is still a matter of controversy today. Huge areas of Germany were leveled by Allied bombing raids, yet war production in Germany kept increasing until sometime in 1944. I've never seen a clear explanation of how this could be, and people who say the bombing was wrong-headed on the one hand seem to want to make a case that it was morally wrong because it produced enormous devestation and killed all sorts of people, and on the other hand want to make a case that is was morally wrong because it produced no devestation and killed hardly anyone. Both of these can't be true, though I've stated the case facetiously. The question is: how could the bombing be both massively destructive and completely or largely ineffective? ================== Why is Diomedes almost never talked about? He dominates the action in the first half of the Iliad, although he never does much but kill people, although his encounter with the son of a friend of his father's, in the midst of battle, is one of the most memorable scenes in the book. The son is fighting with the Trojans, but when they discover the the connection between their families they decide to go off and kill others instead. Ah, war was friendlier in those days. |