Where Philosophies Break Down

Do you find yourself possessing a philosophy that explains everything? That helps you out in every possible aspect of your life, that's just all around perfect? You'd recommend it to every single human, and fully expect it to enrich their lives after they've adopted it?

If so, I don't think you're looking at it hard enough.

Make a list of every occurrence your philosophy has helped you. Not where it theoretically would help you, where it has actually helped you in your past. Do these constitute a very broad range of possibilities? Are you still sure it's a perfect philosophy applicable to all of mankind?

There is no such thing as an Almighty Band-Aid. A good theory is not a monolithic one that covers every possible explanation; if you can explain everything, then you have no knowledge. The power of a theory is in what it doesn't explain, and in what it prohibits. Hence, in any decent philosophy, there should be a point where it breaks down and fails to provide help.

The Golden Rule, commonly stated as "do unto others as you would have done unto yourself", kind of falls apart when you're a masochist with sadist tendencies. It kind of falls apart when you rationalize that if you were a thief, you'd want society to draw and quarter you in public. It falls apart when you would give no honor to yourself, no lifting up your self-esteem, and thus fail to respect others.

Non-violence, at least to me, seems a bit more general and flexible, but it still has its breaking points. For example, assume a swarm of nanobots from outside the solar system is coming to ingest our sun for energy. Non-violence isn't going to work here. If an alien species wants to exterminate us as a species, it's a safe bet they won't be impressed by non-violence. If a sociopath is running rampant in society, killing because he thinks it's fun, not comprehending any reasons why he shouldn't (as he lacks the typical human cognitive architecture to process such feelings), he will merely laugh at non-violence.

The beauty of non-violence is that it generally works for functioning humans. Non-violence does not say bow down to the aggressor and give in to his demands in fear; non-violence demands you stand your ground, do not give way to fear, and if your opponent strikes you then you get up again and face him squarely. If he kills you, then he has your body, but he never got your obedience. It's the obedience they seek and crave; it's the bravery of refusing to give it that they admire and respect--for even criminals are still human, and lest they be plagued by brain dysfunctions, it's reasonable to assume non-violence will work better than violent behavior. Violence incites more violence, and few of us are Kung Fu Masters capable of ending a conflict soundly before it can really begin. (Besides, we have things called guns and vengeful children now. Even if you "win" the fight you might lose much more in the end.) I talk about the individual level of non-violence because I think Gandhi provided sufficient evidence for non-violent non-cooperation succeeding on a macro scale.

And still, non-violence is not invincible. I do not think it would work against genocides, non-fully functional human brains, and non-human brains. It becomes shaky when faced with an enemy commander skilled in rhetoric and capable of demonizing the enemy to such an extent the soldiers lose all sense of their intuitive balance of right and wrong. So the question here becomes: what do we do in these scenarios? My answer now is to fall back on the ever-so-popular method of "eye for an eye"--only strike in retaliation, don't initialize conflicts. There seems like there should be something better, though, as this has been around for so long and yet done so little to curb violence.

Anarchy with a capital-A has some failures as well. Namely, I'm uncertain how much of the population needs to be cooperative in the effort. Certainly the majority, certainly not everyone, but where is the line? Other failure modes can arise from overly optimistic views of humans. This is traditional of other anarchies and most "perfect" futures--it assumes everyone will do what the speaker thinks they should do. "Violence will simply not be present in the culture because no one will have reason to be violent." "Mob rule will not break out because no one will want to be part of a mob." "There will be no disputes over property when everything is owned by someone and everyone is rationally selfish according to this convoluted system (which contradicts some results of game theory)." I could go on.

Here are two thought experiments. The first one is to try and figure out where your theory breaks down. If you can't acknowledge it breaks down anywhere, even under ideal circumstances, I think you're doing something very wrong. If you think it might fail given such and such circumstances, think through it, as it probably does. Beware if you're creating a huge, elaborate justification for why it doesn't fail in those cases. If it doesn't fail, it should be a simple explanation why.

The second thought experiment is to see if you could apply your theory overnight to the world as it currently is and have everything turn out somewhat okay in the near future. Take all forms (that I'm aware of, including my own) of anarchy: if you removed the governments overnight, we can predict with high certainty that chaos will arise and mobs will form and we can only hope it will stabilize into something orderly that's not just another stupid government. If you find that your theory could be applied overnight, think through it again, and then ask yourself why it hasn't happened yet. If it can't be applied overnight, see if you can come up with a simple transition into it from the current state. My simple transition is "voting for Anarchy". The government doesn't vaporize, but it does take itself down, and must go down at a slow enough pace to prevent mob-rule and other failure modes of anarchy from breaking out. If you can come up with a simple transition, think about putting it into practice. Myself, I have more important things to do right now than run for office on the banner of Anarchy, but if someone else does, you know where my support is going.

Above all, be wary of becoming a Truth Guardian. Be willing to make changes, drastic changes, to your philosophy should the evidence suggest you need to. Strive for perfection, look for faults in your theory so that you may correct them and then see more faults. You may eventually see it's Swiss Cheese and requires scrapping, or you may find ever more subtler holes but the system as a whole remains consistent. If you want to have a chance of salvaging a Swiss Cheese system, then start modularizing the pieces instead of balancing everything on a black box or a shaky core. Use an open model, where should you discover the center is crap, you're free to swap it out for a better one and still keep the useful tools that were independent of the center.

Math is an excellent example of modularization: if I find tomorrow that imaginary numbers are just wrong for all of their uses (I can't imagine this being true, but play along), that doesn't destroy the whole arena of math. It would require some sciences to rederive things, but overall it wouldn't be much of a blow to humanity's continuous betterment of itself. To belabor the point, if your computer doesn't boot up, you can at least rule out the need to go and buy a brand new computer. You know the hardware inside is modularized so that if the hard drive dies, the motherboard, processor, RAM and graphics card will still be all right, and you just need to replace the hard drive. So make your philosophy like a computer, and then search for problems with its parts. Probability is not on your side if you expect the system to be without flaw.

Posted on 2009-10-13 by Jach

LaTeX allowed in comments, use $\\...\\$\$ to wrap inline and $$...$$ to wrap blocks.