Sunday, December 30, 2007

Acts and Rules

My way of viewing the world has been greatly influenced by computers. I see computing machines as close analogues of how human minds perform. There is a CPU, various subprocessors, and I/O. The mysterious and fuzzy-bounded consciousness is the ghost in the CPU. Complicated tasks requiring a lot of processing power that need to be performed in real time, such as walking, assembling the input signals into meaningful patterns, using language, or playing an instrument, and basically most habits of thought and behavior, are all offloaded onto dedicated subprocessors, though most reside in the CPU while they are first being learned. Of course, the whole thing is analog, and made out of gelatin rather than printed on silicon chips, and swimming in chemicals that control the relative strength and stability of different algorithmic figures, but in sum the mechanical mind is much like the real thing.

The central question for ethics is always the same for the CPU: what do I do now? If we are good machines, we make good choices. Ethical questions are judgments, but the only one that matters at any given moment is the judgment among alternative actions that one might do. Lots of insights might go into that, but it all comes down to the final choice.

Most of the time the choice is to follow along some default habit, so making the right choice is usually a function of forming the right defaults. Set up the subprocessors so the CPU can run efficiently. Kant called this character. Thomas wrote about it. We can't function without it. There's too much information to sort through, so you have to identify salient patterns, weed out irrelevant distinctions, and generalize like crazy. Deep Blue does it when planning its next move. We need to do it even more.

So from time to time you hear the question put to freshman philosophy students whether you should follow the rules, or decide on an act by act basis what is right. There are tensions here. We generally want people to follow legitimate rules rather than risk their acting according to idiosyncratic criteria that are much less reliable, going so far as to deride carefully customized decisionmaking as "situational ethics." On the other hand, there are bad regimes of rules too. And there are other tensions.

Ultimately, all decisions will be made at some level based on both. An individual's set of internalized general rules will determine whether the ethical choice presents salient features that make it suitable for conscious, individuated decisionmaking. If societal rules are legitimate, it is presumed likely that they will seldom be in tension with the conscience of a good individual, and that when they are, this will trigger attentive choice whether to violate the rule. At least for questions of first impression -- repeated instances will start to follow well-worn mental pathways and fall back into the mechanisms of habit. Of course there are many regimes of rules, but most moral systems likewise have a high degree of overlap.

When a decision must be made where utilitarian, teleological analysis of the particular situation appears to favor a different choice than what "the rules" would prescribe, should the rules be followed? The answer is no, but it is no only provided that the consequences of the rule violation are fully factored into the decision. Violate the rule only if it is worth it. This does not mean only if it can be rationalized, but only if it is really worth it. This would be a non-answer except that there are reasonable criteria on which both sides of the balance can be analyzed.

On one hand, the benefit must be calculated according to principles of good. The following of the rule must cause harm or deny the benefit of the violation, as would be perceived from a neutral perspective without favoritism to the interests of the decider (and compensating for the decider's bias and ignorance of other interests and inability to foresee all effects.)

Likewise, the effect of the violation can be analyzed: what is the likelihood that the violation will become known and be emulated so that the rule is undermined? How is the violation likely to be explained? Is the rule undermined a legitimate one, time tested, and generally sound? A sincere one (as opposed to one honored mostly in the breach)? Is it complicated? Is it important? Is it unassailably secure, or not even viable? Does it approach being universal and absolute or is it widely controverted and full of exceptions? Is the case for the exception generalizable on broad terms, or is it highly anomalous? Is it likely to be seen as an anomaly? Is the violative act prone to being preceived as part of an exception formed on other potential patterns of generalization that run beyond or counter to its actual basis?

Friday, December 21, 2007

The Genesis of the Soul through Reification

I was just thinking the other day:

The Greeks didn't come up with the soul, but from their thinking, one can easily see where the idea comes from.

The ancient philosophers thought that objects acted the way they did in part because it was essential to their nature; events had a material cause in the subjects of the action performing in accordance with a predisposition. Complex materials were explained by Anaxagoras (or maybe it was Anaximander; I get them confused) as mixtures of elements that provided them with their essential properties. Wet matter contained water; the release of the water rendered the matter dry. Earth provided hardness and weight.

This idea continues to pervade lay impressions of the world even though science has largely rejected it. For example, people think it surprising that sodium and chlorine, neither of them a very savory material (a solid that explodes on contact with water, and a corrosive poision gas) unite to form table salt. The fact is that the properties of a substance depend on much more than the properties of components. Look at diamonds versus graphite. Both contain nothing but carbon. It is the arrangement of the atoms into larger structures that gives them their particular properties, not some missing or differing material component from one to another.

If properties and tendencies really belonged to object in material form, then you could take any positive or negative characteristic and, as they say, "bottle it." Sex appeal, money smarts, strength, slipperiness, heat, cold, invisibility: I'll take a liter of each.

There were different words for soul. Anima was the theoretical construct that distinguished living (animate) from unliving (inanimate) matter. You could have another for sentience, and another for conscience, or motion or repiration, or reproduction. If properties stemmed from material components alone, any of these properties would be things that were in living people (many of them in animals as well) and absent from dead people. If they were material, and ceased to be in a body, then they had to either have been destroyed, or suppressed into a potential but inactive or invisible state, or have gone somewhere else, existing as pure essence without the body.

It sometimes makes sense to think of some properties this way, because they truly are conserved. If one billiard ball hits a larger, heavier billiard ball dead on, it will impart its momentum to the second ball. With some loss for friction, momentum is conserved. Speed will not be conserved, because the larger mass will move slower. You can speak of the momentum being transferred. You can think of the speed being potential, with the possibility of coming out if the heavy ball hits a lighter one.

But other properties are pretty much gone. They change form so totally, it is not really useful to talk about them moving or hiding. When the ball falls off the table and hits the ground, it transfers its momentum to the Earth: it has just hit a super-massive billiard ball. Its speed? I'm sorry, but the speed of the billiard ball is dead. You're not going to re-extract that from the ground. Vaporize a painting of the last supper with an A-bomb and see if you can find that last-supperiness somewhere in the radioactive mist. John McCain's chances of being elected president in 2000: gone.

Mass is conserved pretty easily, although little bits of mass are coming and going all the time, changing back and forth into energy. Patterns and arrangements, shapes, structures, information generally, are much more fragile. Processes, once interrupted, may not easily be restarted. Of course the information isn't really gone. But its recoverability into a recognizable form is likely to be so attenuated as to render it the practical equivalent.

In any event, there is no intact, all-together chunk of concentrated image of Jesus and the apostles in the ionized smoke of the atom blast. The information is scattered, diffused, unrecognizable, and no human means can reassemble it. Likewise the motion of the past billiard ball sitting on the floor, or the shape of the ice sculpture that melted, or the softness and slipperiness of the graphite once it has become a diamond. You can't bottle these things. Nor the animate property of living matter. It's gone.

Sunday, December 02, 2007

Jim Kimball Gets the Treatment

I heard a story today on the radio concerning a guy in Idaho. Here is a link with broadcast details and how to order. Here is a full article from another source on the guy. I wanted to keep a link to this because it is exactly like what we get here all the time.

This guy, Jim Kimball, is on the sex offender registry because he was charged with a sex offense, which was in fact dismissed. Apparently, in Idaho you can be a registered sex offender without having been convicted of a sex offense, and the burden is on you if you want the record expunged. The offense was an isolated incident of consensual sexual contact about 15 years ago, involving a teenage girl that he flirted with in his early 20s and never actually had intercourse with.

A local news station hyped his story. Some high points:
  • They initially failed to report that he was not convicted, or that the allegation was 15 years old, even though these were matters of public record.
  • They failed to ever seek his side of the story.
  • They pushed to air in spite of concerns raised by the reporter that this would ruin someone's life and that the details had not yet emerged to justify the story.
  • They misleadingly illustrated the story by showing children ten years younger than the alleged victim in his case.
  • They used special effects like washing out the color from his photograph and zooming in slowly on the eyes, to manufacture a sense of menace.
  • When other news agencies universally decided not to cover the story, because it was not real news in any way, the station used the fact that it had exercised unsound judgment as a selling-point, repeatedly emphasizing that it had an exclusive.
  • They managed to get him fired; expose him to his nine-year old daughter whom he and his wife had planned to tell when she was older; traumatized is wife; forced him to miss school events for his children; drove away his friends because people were misled into thinking he was a child sexual predator.
I have some doubts about the use of actuarial tests to lock people up, but I do think they have some scientifically proven predictive value. If you were to apply a test like the STATIC-99 to Kimball, I have no doubt he would fall into the lowest risk category. Guys like this tend to be off-the-scale low risk. There is no evidence I know of that they are at any greater risk of offense than the general population.

I do want to correct one assertion from the KCBI report linked to above. It is stated in the report that an offender's recidivism rate rises over time. This is just bullshit. The cumulative likelihood of course rises, by definition, but the annual rate of recidivism decreases steadily. It decreases even more at older ages, and (for serial offenders) after roughly the first six offenses. The only time risk tends to rise over time, is when there has been recent penalty or treatment whose effects are still fresh and are wearing off, which is a small factor in the long term. That's something that would be meaningless in this case. Otherwise, just think: would we feel that a person just released from prison for a sex offense would be the low risk guy, and the guy who's been good for 25 years is the one you have to watch? It runs completely counter to the principle of induction.