Wednesday, May 28, 2008

An Excellent Example of Professional Tunnelvision, Segue Into Healthy Debates

I've often noticed that outside their area of expertise, scientists and professors are often as incompetent as layman. This is actually true in general for professionals. They do not apply the principles they learn in their profession to life outside the profession.

Unfortunately, this is even true of logicians. I have an excellent example. Specifically, the phrase 'counter-example.' (Hello, irony. Fine day, isn't it?)

You would expect, from the normal definition of 'counter' for a counter-example to be in some sense opposite to an example. If you're a logician, however, this apparently escapes you.

It took me a good half-hour to come up with a coherent theory of what they meant by counter-example. In this case, 'counter' means 'exactly the same as the regular example, but more obviously false.' The argument structure is exactly the same, and this is how I can tell they are good logicians; they rarely screw up when identifying the argument type.

Unfortunately, their skills apparently end there. For example,
"Every event must have a cause, that is, every event bears the effect relation to some other event, which is its cause.
Therefore, there must be a cause of every event, that is, some event bears the inverse of the effect relation—which is the cause relation—to every other event. This first cause is what we call "God"."
They fucked up the argument. What they're actually arguing is not that God is a first cause, but that God is every cause, which indeed seems like a fallacy. The actual argument goes like this;
"Every event must have a cause, that is, every event bears the effect relation to some other event, which is its cause.
Therefore, there must be a cause of every event, that is, for each event, there exists a second (in logic, not in time) event that bears the inverse of the effect relation—which is the cause relation—to the first event. Events are ordered in time and time is finite. Therefore, there must have been a first event, which means there must have been a first cause. This first cause is what we call "God"."
(Notably, I also fucked up the argument - twice, in separate ways - but I corrected myself before publishing it.) As we can see, this makes a great deal more sense, and must be refuted by actual sophisticated argument, something the Dawkins-esque atheist apparently cannot believe is required when arguing with theists. For instance, you have to note that for this to work, God cannot himself be an event, or he would require a cause, thus proving that his cause is God, not God himself, and so on forever. But this means that not every cause is an event, contradicting a premise. There are probably similarly sophisticated rebuttals to my rebuttals. (Reasoning out consequences is a technique that can replace knowledge of literally all specific logical fallacies, though in both cases your opponents must consent to lay out their premises clearly.)

Similarly, when attempting to form my theory of what they meant, I did, sadly, check their glossary, but of course I still expected a logician to actually mean something related to 'counter' when they say 'counter.' How foolish of me. I thought the counter for an example of a fallacy would in some way be a non-fallacy example, or perhaps vice-versa. Or perhaps the counter-example was an example of where the fallacy existed but the conclusion was still true, showing how people might come to believe in fallacies, while at the same time demonstrating, through the regular example, how the reasoning is false. But no, my hopes were, as usual, far more sophisticated, more insightful, more consistent, and more useful than the reality. (At least I come up with good ideas.)

If you (also) check the glossary, the flaw is even more egregious, because they also list the standard definition of counter-example; an example that counters an argument.
  1. Lying is an effective tactic for lawyers/politicians
  2. A lying lawyer/politician will defeat an honest one in an otherwise fair contest.
  3. All successful lawyers/politicians are liars.
You know, the type of counter-example that would disprove categorical arguments like this; a single successful, honest, lawyer. (Note that I don't have to find out if the argument is invalid or if one of the premises is just untrue; the counter-example does all the work for me.)

Another way you can test yourself (the previous note being there to ensure you know how to use counter-examples, a service I wish more people would provide - again with my ideas vs reality.) is that I had an implicit premise.

Implicit premise: most contests are fair, or the unfairness is evenly distributed between most contestants. It seems unlikely that scrupulous lawyers would somehow have an unfair advantage, though yet again, my ideas are far more sophisticated, consistent, and useful than the reality - we should totally implement unfair advantages for honesty, perhaps with something akin to bounties for scrupulous behavior. Obviously, if everyone is honest the advantage disappears, while if most people are dishonest, we end up designing a system where the resources attending scruples can defeat the advantages attending dishonesty.


I have learned something by studying this data in perspective with other recent data, specifically by comparing the God-debate with the debates I'm currently engaging in. If indeed your opponents in a debate with submit their premises for review, it becomes easy to point out logical fallacies - which supplies the motive for not doing so. Having pointed them out, a debate can continue by having the rebuttee repair their argument and submit it again, either until the argument is sound, they cannot repair it, or the debaters disagree on whether a particular premise is true or false. This last would mean that the data sets of the debaters is incongruent, and must be fixed, usually by science. Naturally, even debaters with the same axioms but different data sets (both being imperfect, of course) can validly come to different conclusions.

This is what a healthy debate is like. Does your debate look like this? If not, how do you repair it?

Conversely, one of the debaters may eventually come to an argument that is sound, yet which their opponent believes is untrue. That is, the opponent cannot find a flaw in the premises or logic, but still objects to the conclusion.

Alternatively, they may object to the logic itself - there may be a step that one debator finds inevitable while the other finds objectionable. This is actually the basis of real irrationality

This is why I attempt to lay out my arguments, especially my argument about consciousness, as formally as possible, and welcome all attempts to correct the logical presentation.

If indeed there is irrationality in the debate, it's the only way to definitively find out.

No comments: