Friday, May 23, 2008

Ethics and free will

It is often argued that the notion of ethics fails without "free will". This view is nonsense. It is true, however, that our notion of ethics has to change substantially without free will, but change it must. If we abandon the notion of "free will" then we must change our ethical thinking to be about causal efficacy rather than an incoherent and irrational notion of deserts.

"Free will" is a fundamentally vacuous, unfalsifiable notion. In all of my philosophical investigations, I've never seen a definition of free will that would permit me to determine if I myself — much less anyone else — actually had free will. Just like "God", free will seems to mean something, but when we investigate the notion carefully, it evaporates into vacuous platitudes.

Any specific behavior is the result of some causal process, physical or non-physical, or it is not the result of some causal process. If it is the result of a causal process, our traditional notion of free will is facially contradicted. But what does it mean that a behavior is not the result of a causal process? The opposite of a causal process is a random event, an event that cannot be correlated to anything previous in time. If we cannot hold someone accountable for a caused behavior, by what virtue do we hold someone accountable for a random behavior? The view that the antinomy between causality and randomness makes free will fundamentally ineffable or mysterious just digs the philosophical hole deeper: By what virtue do we hold someone accountable based on a phenomenon we not only do not understand but cannot understand?

The notion of consciousness as a non-physical phenomenon fails to rescue free will per se. Non-physical consciousness would just establish non-physical causality. It is the causality per se that contradicts free will, not the physical causality.

(There's an additional confusion about causality that's peripherally relevant: determinism and predictability: A deterministic system is often held to be ipso facto a mechanistic system, i.e. a system whose behavior is completely stereotyped, rigid and predictable. This notion is nonsense. First, there's good evidence from Quantum Mechanics that all systems are in some sense mixed, the result of an interaction between random quantum events and deterministic physical laws. Secondly, chaos theory has conclusively shown that even systems with very simple, rigidly deterministic behavior are radically, perhaps even fundamentally unpredictable. The notion that determinism leads to "sphexish" behavior is definitely contradicted by both facts and logic.)

So we have to adjust our notion of ethics: We must hold people accountable, at least in some sense, even though their behavior is causal or a mixture of causality and randomness. How can we do so?

We can see an amusing example of the issue in the scene from Fawlty Towers (I'm firmly convinced that comedians are, in general, better philosophers than most philosophers) where Basil beats his car with a stick because it won't start. Basil's action is obviously absurd and funny. But it's funny not because the car doesn't have free will and is therefore not responsible for its failure, it's funny because hitting a car with a stick is a spectacularly ineffective way of making it start.

When my car fails, I do indeed hold the car itself "accountable". If it doesn't perform as I require, then I fix the car. I don't merely conclude that since it doesn't have free will, there's nothing I can do about it. Human beings are, of course, vastly more complicated than cars, but the principle applies.

The way to think about ethics in the absence of free will is just in this causal sense. Use scientific reasoning to figure out the causal mechanisms that underlie human behavior, and then use that understanding to intervene and bring about the results we desire. In just the same sense we use science to determine the laws of physics, and then engineers manipulate the laws of physics to build desirable technology.

Taking this causal view, it becomes incoherent to imprison or punish murderers because in some mystical, ineffable and unfalsifiable sense they "deserve" punishment. We should imprison or punish murderers because that's an effective way of persuading people not to engage in the undesirable activity of killing their fellow human beings. And if imprisonment and punishment were not effective, or some other mutually exclusive activity were more effective, then it would positively irrational to continue imprison and punish murderers. Not because we condone murder; precisely the opposite. If we really do condemn murder, then the rational choice is do that which is most effective at reducing it.


  1. Suppose O.J. Simpson didn't really kill his wife, but it would be really effective to tell the public that he did and then imprison him? Would that be acceptable?

    On the other hand, what if imprisoning Josef Fritzl wouldn't have any significant effect on people, because (say) no one else wants to lock their daughter in the cellar and rape her for 24 years. Should we let him go unpunished?

  2. Tea: Suppose O.J. Simpson didn't really kill his wife, but it would be really effective to tell the public that he did and then imprison him?

    Effective at what? In what sense are you asserting that imprisoning an innocent person might be "effective"?

    Did you miss the part about scientifically understanding the causal mechanisms underlying human behavior? Oh, right, you're a philosophy student.

    Saying that something might be effective is nonsense: show me how it actually is effective, and what it's effective at, and we can talk.

    It's "effective" in a very limited, narrow sense to have everyone use pogo-sticks to commute to work, but in a larger sense, apprehensible to those of us with triple digit IQs, pogo-sticks are not at all an effective system of transportation.

    What is it about the study of philosophy that turns (presumably) normally intelligent people into complete fucking morons?

  3. It's also completely retarded to use O.J. Simpson as an example of imprisoning an innocent man because a) he actually did kill his wife and b) he was actually acquitted.

    O.J. Simpson was acquitted because the justice system is (somewhat) effective at limiting the power of the government and the police to imprison people without a good evidentiary case regardless of their guilt or innocence in actual fact.

  4. Hey thanks, that was fun. You interpreted my questions perfectly by referring to them as "assertions", your dealing with thought experiments is brilliant, and anyway, you're great at debating philosophical problems.

  5. If we don't have free will, then isn't it the case that we can't choose whether or not to punish someone, or on what basis to punish him or her? To say that we can make such choices implies that we have the power to choose other than we currently do, which seems to imply free will, as it's traditionally understood.

  6. Just in case you're interested in debating the topic anyway… Since you’re clearly too bright to be able to deal with counterfactuals, how about this: those of us with the IQ of Forrest Gump are capable of understanding that punishing innocent people at least sometimes has the desired effect on the masses. Rulers have known for a long time that by publicly cutting off a poor farmer's hand as a punishment for stealing was going to affect other poor farmers and make it less likely that they would steal in the future. It really didn't matter to the king in question whether the guy whose hand he just cut off really did steal - it was all about the message.

    My question to you was whether you’re ok with such implications. Your response seems to imply “no”, but then again, I’m not that sharp…

  7. Lau: If we don't have free will, then isn't it the case that we can't choose whether or not to punish someone, or on what basis to punish him or her?

    What do you mean by "choose"? Why should we associate "choice" with "free will"? A choice must itself be a causal process.

    Tea: You interpreted my questions perfectly by referring to them as "assertions"...

    I am, as are my readers, able to see an assertion disguised as a question.

    Since you’re clearly too bright to be able to deal with counterfactuals...

    Counterfactual universals, such as you suggest, are not very interesting: If pigs had wings, they'd be chickens.

    [P]unishing innocent people at least sometimes has the desired effect on the masses.

    Punishing innocent people has one desired effect. It also has other undesired effects. We must consider, to the best of our ability, the totality of the effects. Simply saying that some action has some particular desirable effect is not a counterargument if the action has other undesirable effects.

    I’m not that sharp…

    No, you're not. Keep studying, though, there's hope for you yet.

  8. what a pathetic little prick you are... i wouldn't mind your arrogance if it were at least justified, but you're as thick as a rock.

  9. Tea: Awwwww... po' little baby. Don't let the door hit you in the ass on your way out.

  10. As I think you hinted, even if one introduces the randomness that quantum mechanics suggest, we still have no more control over randomness then we do a strictly deterministic environment.

    As far as punishment is concerned, just because current criminal laws are in place as if people do have free will doesn't prove or mean that people do. How I see it is, laws and penalties are in place to condition us to act a certain way, if one doesn't act that certain way anyways because other conditions have determined them not to, then they are determined to face the penalties. Them having to face the penalties is designed to reinforce and strengthen the stimuli to increase the conditioning effectiveness of these laws for others.

    A murderers thoughts are determined by the conditions of his environment, and the determined thoughts determined his actions. "He" is just the product of his biology's reaction to his accumulated experiences with his environment. We can't blame him for being determined to that course. What we are in effect doing with laws and consequences is conditioning people to behave in a way we have deemed appropriate. If they act against the conditions we've set forth anyways, we know that there were conditions greater then the ones we've put in place that caused that individual to behave the way they did. By making them faces the consequences we strengthen the effect of the conditions we've put in place as it serves as a model for others and that individual for future incidents. But was he really to blame, no.

    My concern is with determining what conditions are stronger or work better to condition people to behave how we thing they "ought" to, and how to deal with them when they don't act accordingly. The laws and consequences like we have now. Or rehabilitation, which is really just reconditioning the individual. Or perhaps both. When we punish people for their actions which they have no control over, they end up being "collateral damage" in a cause that has a greater agenda then just them.

  11. I know this is an old post, but I some how just came across it and I had what I think might be an interesting insight.

    Even though thought is a mixture of random events and events with causes, the brain is sufficiently complex and poorly understood to produce results that approximate the traditional idea of free will quite well. Shouldn't this allow free will to be used as a useful approximation for practical purposes, in much the same way that equations relating to heat transfer are useful, even though they simply closely approximate the average results of small scale interactions in most real world cases, rather than accurately representing what is really going on?

    If this is irrelevant, flawed, completely wrong, or just plain stupid, please explain how and why so that I can "debug" my thought processes.

  12. Even though thought is a mixture of random events and events with causes, the brain is sufficiently complex and poorly understood to produce results that approximate the traditional idea of free will quite well.

    The problem is that I have no idea what the "traditional idea of free will" really is. It seems to be contradictory: an event is freely willed if it is neither caused nor not caused.

  13. Free and will seem contradictory to me as well. What about the idea of conscious vs unconscious will . As in our unconscious mind has causal efficacy and our conscious (what we call I) mind has far less. As such are decisions are made far in advance. Therefore the "I" doesn't really "do" anything that is all that significant in the world. I know that at least some neurologist take this idea seriously. I don't believe there is a consensus. Call me Moniker.

    1. our no "are" . By "far" I mean minutes to days. And by significant I mean relative to the unconscious as in the "I" tags along for the ride. By the way I have no formal education in medical, or psychological sciences.


Please pick a handle or moniker for your comment. It's much easier to address someone by a name or pseudonym than simply "hey you". I have the option of requiring a "hard" identity, but I don't want to turn that on... yet.

With few exceptions, I will not respond or reply to anonymous comments, and I may delete them. I keep a copy of all comments; if you want the text of your comment to repost with something vaguely resembling an identity, email me.

No spam, pr0n, commercial advertising, insanity, lies, repetition or off-topic comments. Creationists, Global Warming deniers, anti-vaxers, Randians, and Libertarians are automatically presumed to be idiots; Christians and Muslims might get the benefit of the doubt, if I'm in a good mood.

See the Debate Flowchart for some basic rules.

Sourced factual corrections are always published and acknowledged.

I will respond or not respond to comments as the mood takes me. See my latest comment policy for details. I am not a pseudonomous-American: my real name is Larry.

Comments may be moderated from time to time. When I do moderate comments, anonymous comments are far more likely to be rejected.

I've already answered some typical comments.

I have jqMath enabled for the blog. If you have a dollar sign (\$) in your comment, put a \\ in front of it: \\\$, unless you want to include a formula in your comment.