Rules Against vs. Aversions To Harming the Innocent

304 days until classes start. 7 days until I introduce myself to the department.

Yes, I get nervous.

I had a realization this morning. I have an association with a large number of lawyers. A number of those lawyers will probably have an interest in a philosophical argument concerning the justification of punishment, and even have some familiarity with the literature. What I realized is that I can present the arguments I am writing up in response to Boonin to them for comments and criticisms.

Anyway, to continue with my response to Boonin's book.

In Part I I, presented a moral aversion theory of punishment.

In Part II I, looked at Boonin's definition of punishment and showed how moral aversion theory was a theory of punishment on his definition.

In Part III, I presented the consequentialist theory of punishment and some of the major problems with it - particularly the problem that it sometimes justifies punishing the innocent when it can do the most good.

In Part IV, I discussed the details of a moral aversion to harming the innocent.

Now here, in Part V, I defend this moral aversion theory from objections that Boonin raises against rule-utilitarian theories, which he (incorrectly) asserts are just as applicable to motive-utilitarian theories.

Rules Against vs. Aversions To Harming the Innocent

Boonin brings up several objections to a rule-utilitarian defense of punishment that he claims would be just as applicable to motive-utilitarian theories. Moral aversion theory, where aversions, rather than rules, are justified by their utility would be an example of a motive-utilitarian theory.

One of these objections says that utilitarianism sometimes justifies punishing the innocent. He illustrates this point by asking us to imagine setting up a secret committee that would be charged with determining when it would be useful to frame and punish an innocent person. Upon deciding that, perhaps due to a series of crimes, it would be useful to frame a person for those crimes and have a very public trial and punishment, it would then set events in motion to bring about that end. It would find somebody to frame, frame that person is a very public trial, punish that person, and thereby create sufficient deterrence to reduce the overall number of crimes. The social benefit will outweigh the social cost when following those rules.

This suggests that the rule utilitarian best rule is not, “Do not punish the innocent.” It is, “Do not punish the innocent unless it is deemed useful to punish them by a secret committee that determines beyond a reasonable doubt that it will improve the overall public good."

We may have to flesh out these rules with a few more details to make then work, but it seems likely that there is some set of rules that would sometimes justify punishing the innocent for utilitarian reasons. Whatever those rules happen to be, rule-utilitarianism would then be charged with justifying something (framing an innocent person for a crime and punishing him) that is, itself, objectionable.

This is one of the objections to rule utilitarianism that Boonin asserts is just as applicable to any type of motive-utilitarian theory. Insofar as punishing the innocent is sometimes useful, there must be some set of motives that would have us punish the innocent person in those circumstances where it is useful. A flat prohibition on punishing the innocent would not likely be the best option.

However, motives and rules have some significant differences – particularly when the motives we are talking about are moral aversions.

First, motives are persistent over time. You can easily create a rule to not eat fish except on the day of the full moon. However, it is much more difficult (if it is even possible at all) to not LIKE fish except on the day of the full moon. You can have a rule that expires at midnight on December 31, but it would be nearly impossible to have an aversion that expires at midnight on December 31. Comparably, our aversion to harming the innocent will not easily allow for exceptions where we can simply "shut it off because, in this instance, punishing the innocent will produce good consequences". The aversion will persist through those instances, providing a reason not to harm the innocent - and to not harm them BECAUSE they are innocent.

Second, we can easily create complex rules where we cannot easily create complex motives. While it may be possible that a person could be afraid of spiders except in her Aunt Jane's house or except for Stanley's pet spider Jake, it is very difficult to intentionally engineer aversions with these exceptions. Where the ease of making a rule may make it worthwhile to do so, creating an aversion to match the rule may require far more work than it is worth. Furthermore, it seems safe to say that our most complex rules can be far more complex than our most complex socially engineered aversions.

Third, moral aversions are taught in a society to all individuals. One of the defining characteristics about moral principles is that they are universal – they apply to all people. That same defining characteristic applies to moral aversions. Insofar as they are moral, they are aversions that everybody should be caused to have. One can have a secret rule creating a secret committee that determines when innocent people may be framed for a crime and punished for the public good, but this would require that everybody lack the aversion to punishing the innocent when they are on such a committee.

Fourth, creating a universal moral aversion requires that it be a type of aversion that it is possible to teach universally (or very nearly so). It would have to be something that could be taught to a variety of people with a variety of experiences and a variety of intellectual capabilities. It will be far easier to teach an aversion to punishing the innocent than it is to teach an aversion to punish the innocent except when one thinks it will promote the public good.

Fifth, rules can be broken, while aversions can be overridden only when a stronger motive or a combination of motives override it. Boonin talks about a rule in baseball that a runner must stay in the baselines when going from one base to another in order to score a run. He provides an example where the runner breaks the rule when he sees a child choking on a hot dog and he alone seems to know the Heimlich maneuver. He thereby breaks the rule. However, he can only do so intentionally if he has a motive for doing so. We then need to look at that motive - at its expected consequences, and whether or not people generally have reason to promote or inhibit it. The player can step outside of a set of rules, but he can never step outside of a set of motives.

We need to examine the aversion to punishing the innocent in light of these differences.

It is a simple aversion to teach, and it can be taught publicly and universally.

Once taught, it will persist in the motivational even as other concerns weigh against it. It will continue to motivate the judge who has an opportunity to punish an innocent person for the public good, and a legislator considering establishing a committee to identify opportunities to punish the innocent for the public good. They will each have an aversion to punishing the innocent precisely because it is punishing the innocent such that the public good has to be very great indeed for the agent to even consider such an act.

Furthermore, the judge and the legislator will know that people generally have many and strong reasons to condemn, and even to punish those who harm the innocent, as a way of promoting this aversion. Indeed, the judge and the legislator are among those who have many and strong reasons to promote that aversion – to condemn and even call for the punishment of those who would punish the innocent. This would mean self-condemnation if they were to engage in such an act.

This aversion to punishing the innocent may allow anybody to punish the innocent if the punishment is small - and we do, indeed, allow it when it involves displays of anger and condemnation in words and tone alone. However, it would motivate agents to take steps - and to require that others take steps - to make sure that the individual is guilty if the harm is great. It would motivate agents to require the use of a set of practices that would include presenting evidence for and against to an impartial jury. It would motivate an aversion to allowing each individual to decide on and carry out punishment because we know from experience that a lot of people make mistakes in this area. Furthermore, in the same way that we have no reason to trust their judgment regarding the execution of private justice, they have no reason to trust ours.

An aversion to harming the innocent even provides motivation to punishing the guilty – since that is the method by which we activate the reward centers of the brains of the person being punished specifically, and people generally, to create the aversions that protect the innocent from harm.

Can we not do better by promoting an aversion to doing harm, rather than an aversion to harming the innocent?

The question arises here as to how we would do that. If we are denied the use of the tool of either creating a state of affairs to which agents are averse, or denying what they desire, as a way to activate the reward centers of the brain and create the aversion, then how is the aversion to be created?

I have mentioned that we can create these aversions with stories and parables. However, it may well be the case that at least the risk of real harm is necessary to fully trigger the desired effect. One has to think that people can actually be harmed for performing such an act. Furthermore, they would have be stories in which people are intentionally harmed because they intentionally harmed others, and it would be very difficult to create a coherent story that met that criterion.

In short, motives are quite different from rules. Those differences protect a motive-utilitarian defense of punishment from arguments that would be effective against a rule-utilitarian theory. It explains and justifies a simple, public, universal aversion to harming the innocent. It explains and justifies institutions such as the jury trial and practices such as the presumption of innocence. It is consistent with the fact that, if there is enough at stake, it may be legitimate to punish the innocent (e.g., to save humanity from extinction). At the same time, it still asserts that agents always should be averse to harming the innocent because they are innocent, and even when harming them can promote some other good.

Boonin provides some other arguments against rule and motive utilitarian theories that are applicable to this moral aversion theory. In my next post, I wish to examine some of those other objections.