Vivek's blog: The Fallacy of Cause & Effect, Part 2

Monday, January 2, 2012

The Fallacy of Cause & Effect, Part 2

As we learn more and more about what causes actions, we face a paradox: we find ourselves less and less able to deal with the consequences of those actions.

Why? Because we try not to punish people for actions they had no control over.

This seems like a good thing. If you're holding a knife and I fall into you, injuring myself, you're not charged with assault. Well and good.

But what if you're holding a knife while sleepwalking, and you injure me with the knife. Is that your fault? The courts are divided on this one - especially in situations where you might have a motive to injure me - but there is certainly precedent for acquittal based on sleepwalking. The New Scientist claims that advances in medicine might hold the answer for sleepwalking and other "unconscious" crimes.

OK, so if you're not awake and aware, you might not be responsible for your actions. But what if you were fully conscious, but you weren't in a fit mental state? You could be experiencing a psychotic episode, in which you were completely unable to distinguish right from wrong. But you don't even need this criterion.

Some defendants have successfully argued that they were unable to control their actions owing to irresistible impulses (a form of temporary insanity). When subjected to an irresistible impulse, the argument goes, the person who committed a criminal act should avoid conviction because they were unable to control their actions. Crimes of vengeance (eg an enraged spouse attacking their unfaithful partner) sometimes see an "irresistible impulse" defense.

On occasion, a number of factors come together to exculpate an individual. Such was the case of the infamous "Twinkie defense," wherein the lawyers of Dan White - who shot and killed San Francisco mayor George Moscone and supervisor Harvey Milk - argued that White suffered from depression, the symptoms of which may have been exacerbated by consumption of junk food. White was acquitted (and later committed suicide).

These instances point to an underlying principle: as we more profoundly understand the proximate causes of a physical action (such as a chemical imbalance in someone's mind), we are less able to ascribe "intent" (of which free will is a necessary component).

In short, we can provide simple biological, chemical, or physical explanations for these actions. So we can't say they were the product of a "free" mind. And, therefore, we can't punish them.

Pretty soon, this is going to become problematic - because we are going to be explaining a LOT more behavior via medicine. We've already identified certain genotypes congruent with increased aggression. People can't choose their genes. (No one is saying that behavior is determined by genes alone; they are, however, an important component in causing action.)

What if we could explain literally every transgression by identifying underlying biology and chemistry that caused those transgressions? We'll never get to the point where we can predict individual action, but if we can retroactively identify the proximate causes behind action, how can we apportion punishment?

Science has gotten us into this problem. Can it help us out of it? Maybe. Perhaps we can use research to evaluate the effects of punishment itself. If we view punishment as retribution for crime ("balancing the scales of justice"), then the more we take "intent" out of the equation, the more we have to reƫvaluate our justification for punishment.

However, if punishment is primarily a way to forestall undesirable behavior (through a deterring effect) or to change behavior (by preventing recidivism), then science can help - a lot. Because in this world, we'd have clear goals for punishment: reducing incidence of a particular behavior in a population, and reducing recurrence of behavior in an individual. Well-designed experiments could test the effects of given punishments (or rehabilitation techniques).

Policy could be grounded in the effectiveness of meeting these goals, rather than in political platitudes ("Tough on crime!" or "Zero tolerance!" or "Keep us safe!"). Theoretically, our society could be safer, more dignified, and more efficient. And we wouldn't have to fret over the chemical state of an individual's mind at a given point in time.

Or we could keep doing it the hard way.

(Thanks to Joel Kraut on Flickr for the Twinkie image. This post has only a bit to do with my earlier post, The Fallacy of Cause & Effect, Part 1.)


  1. Immediately following your tweet on this, the Dalai Lama tweeted "While revenge weakens society, forgiveness gives it strength".

    We are going to have to be very clear about the role of the legal system re: publishment/retribution vs prevention/deterrence. And given that the first serious attempt to replace eye-for-an-eye with turn-the-other-cheek was 2000-odd years ago and we still seem fuzzy on why/whether we punish miscreants, I don't hold out much hope for a sudden outbreak of rational thinking on this.

    Try to check out the documentary series All Watched Over by Machines of Loving Grace ( The thesis of the third episode is pertinent: that our collective embrace of a new Determinism based on the Selfish Gene hypothesis ("I'm not to blame, honest guv - it was my genes what did it") is a result of our despair of ever being able to change the world for the better, or even escape the unintended catastrophes the flapping of our little butterfly wings visit on the world...


  2. Hi Dave! All true - and all good points. I do want to note that I don't want to say that we need to abandon the idea of punishment for a "let's all just forgive each other" philosophy. That viewpoint would make for bad policy. But I do think that the reasoning applies in a utilitarian and a rights-based framework. If someone committed a crime on a sinking ship, the punishment calculation should be the same in either famework.


Please keep comments as concise and relevant as possible. Limit profanity, except if it's really funny.