Pragmatism and the whole story

In attempting to criticize a pragmatic approach to ethics, Tea offers the following arguments:

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?

This counterargument is hardly original; Tea should not bear all the blame for its stupidity.

The argument fails. It's a narrow view of pragmatism: If an action has some particular desirable effect, then the action is itself desirable. But an action can have other undesirable effects in addition to the desirable effect. In just the same sense some action might lack some particular desirable effect, but it might have other effects that are indeed desirable. That we cannot judge any action or decision on the basis of a single effect is not a valid argument that we cannot judge an action on the totality of its effects.

When we consider anything on pragmatic, efficacious grounds, we want to consider all the effects. It is "effective", for example, to hammer a nail into a board using my computer: The nail will indeed be hammered in. On the other hand, if I hammer too many nails in with my computer, I'll eventually find it very difficult to blog (some readers might find this result an added benefit). On the other hand, if the zombies were attacking my house and I didn't have a hammer, the value of getting the nails in might well justify losing the ordinary use of my computer.

In Tea's first instance (ignoring the witless irony of employing O.J. Simpson as an example of someone unjustly punished for an act he himself did not commit), she ignores the fact that there are a number of competing pragmatic demands we place on our justice system. We want to reactively keep actual perpetrators of undesirable activities from repeating those activities and we want to proactively discourage others from doing similar actions. We also want to prevent the government from using its coercive powers in undesirable ways. Individuals want to have assurance that if they act lawfully, they will not be punished. And it is the case that some perpetrators of undesirable actions themselves — many of those who commit "crimes of passion" — require severe penance to expiate their own guilt (which is why many people who commit crimes actually confess).

To evaluate Tea's argument carefully we have to ask: in precisely what sense would it be "effective" to tell the public that O.J. did indeed kill his wife and her friend? If the government is in the habit of just telling people in general that people did or did not commit crimes without justifying that conclusion, they would create great distrust in their judgment. Even supposedly absolute monarchs have found it desirable to create in the population some degree of rational trust in their judgment. Contrawise, if the government were in the habit of making an evidentiary case, failing to do so in a specific instance would immediately engender distrust.

In case our incredulity over using O.J. as an example of unjust punishment clouds our judgment, Tea presents an alternative case:

Rulers have known for a long time that by publicly cutting off a poor [and presumably innocent] 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 would be nice to know precisely which rulers Tea is referring to, but again, the argument fails. In this case, both the punished farmer and the actual thief know that the farmer has been arbitrarily punished. Surely the thief himself has not been deterred from stealing; he knows that someone else will suffer for his actions. And the farmers as a whole, knowing punishment lacks specific justification, are deterred from stealing only by fear, not by the apprehension that punishing actual theft accrues to their mutual benefit. Clearly the basis in mutual benefit is superior, and arbitrary punishment is mutually exclusive with punishment with a basis in fact. So, while arbitrary punishment might be locally effective, it is ineffective in a larger sense.

Furthermore, we can with relative ease make the case for permitting at least some punishment of the innocent. If we are going to reactively punish people for performing certain actions (and we do), and even if we are committed to punishing people only after a sincere and honest evidentiary justification to the best of our abilities (and we are, more or less), we know that because we are finite, imperfect human beings, and we will devote only finite time to making our evidentiary justification, it is certain* that we will make mistakes (and we do), and we cannot always err on the side of caution.

*Well, almost certain; it's possible but extremely unlikely that by pure chance we might avoid punishing any innocent person.

If it were truly inherently unacceptable that any innocent person be punished, then our present day justice system would itself be inherently unacceptable. But it is not. We have decided that there is an acceptable level of risk, that the positive value of reactively punishing people outweighs at least some number of innocent people punished.

Tea offers an additional case that simply reverses the question:

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?

First, Tea's hypothetical seems to miss the point entirely. We do not punish some behavior to deter others from performing the exact same behavior under exactly the same circumstance. We punish people (in part) to deter others from performing similar actions, and the essential features of what constitutes "similarity" are publicly described and consistently determinable.

Regardless of the novelty of Fritzl's crime, it is still the case that he has shown profound indifference to or enjoyment of the suffering of others. We cannot allow people with characteristic to operate autonomously unless they choose intellectually to fake or mimic concern for the well-being of others to avoid punishment, which Fritzl has obviously failed to do. Whether we "punish" him, we certainly cannot simply let him go free. If he is rational — i.e. he knew and apprehended at least to some degree the risks of capture and punishment — then by performing the actions he consented to the punishment if caught. If he is irrational, if he was utterly unable to apprehend the risks of capture and punishment, then he is insane, probably incurably, and we must therefore confine him.

Whether we are talking about engineering, law, ethics or meta-ethics, it is unacceptably eliminative to consider only a single possible consequence or effect of any action, decision, or behavior to construct its justification. We must always consider the totality of effects, at least those effects we can — to the best of our abilities — reasonably foresee.