Word for Wednesday: Consequentialism definition

Consequentialism is the claim that we should judge the rightness of an action by the desirability of its consequences, as opposed, say, to its conformity with some concept of virtue or honour. As I mentioned last week, it is very closely tied to utilitarianism and is most appealing as a criterion for public policy rather than individual ethics or jurisprudence.

The big question in relation to consequentialism is how we assess the consequences of an action. And as usual there seems a huge divide between economists (particularly decision theorists) and philosophers on this point.

Speaking for the economists, I would say that the only operational form of consequentialism is one that judges actions based on the best possible ex ante evaluation of their uncertain consequences. This in turn, naturally ties the dominant theory of choice under uncertainty (expected utility theory) into what I’ve claimed (in the face of strong criticism from Lawrence Solum, Gary Sauer-Thompson and Eric Tam) to be the dominant public philosophy, namely utilitarianism. Most of my professional work is concerned with generalisations of and alternatives to expected utility theory which give rise to alternative versions of consequentialism, including that of Rawls.

Judging by this entry in the Stanford Encyclopedia of Philosophy, kindly sent to me by Brian Weatherson, the dominant view in philosophical circles seems to be that we should focus on actual consequences.

This view has the obvious problem that it doesn’t give us any actual guidance, since we’re uncertain about the future, and therefore can’t tell which choice will have the best consequence. Less obviously, it can’t even be used in retrospect to assess the rightness of an action, since we can’t know what would have happened if we had made some alternative decision. This is claimed to be the traditional utilitarian view, but I don’t believe that hardheads like Bentham and James Mill would have advocated a non-operational policy criterion.

The discussion of this issue in the Stanford Encyclopedia entry seems to me to be exceptionally naive, especially since the main examples selected are in areas that have been the subject of intense study by economists. First, there is a toy example about someone feeding ‘rotten meat’ to their children, written without any reference to the vast literature on food safety, risk perceptions and public policy.

Then there’s the statement, made by way of a concession, that

The criterion of a good stock investment is its total return, but the best decision procedure still might be to reduce risk by buying an index fund or blue-chip stocks

I would say, in the damning phrase attributed to Wolfgang Pauli, that this “isn’t even wrong”. In a standard finance model, the question “given the realisation of some given state of nature in the future, what investment would yield the maximum return” doesn’t have a well-defined answer. If I knew which security was going to yield the maximum return in the given state, and I faced no liquidity constraint, I could make an unbounded profit by going short in all other assets and using the proceeds to buy the preferred security. That’s why questions like “what stock investment yields maximum return” aren’t even discussed in finance theory.

As I say, I find the discussion here exceptionally naive. But I have no doubt that, to any philosophers in the audience, I’ve just demonstrated the naivety of economic thinking, and I eagerly await their subtle refutations.

8 thoughts on “Word for Wednesday: Consequentialism definition

  1. According to recent reports, British intelligence in Northern Ireland knowingly allowed their spy in the IRA, code named Stakeknife, to murder informants and others, because to stop him would have given away the fact that he was an informant.

    They made the ex ante evaluation that allowing him to murder innocent people, and murder people who themselves had risked their lives and their families’ lives in the fight against terrorism, was all worth it, given the consequences — progress in the fight against terrorism.

    A pretty tough call, no?

  2. A tough call indeed, and one that raises a lot of issues. I’d argue that intelligencies agencies making such decisions don’t take enough account of the damage they do to the rule of law, and should therefore be subject to legal/political controls that would preclude such decisions in all but the most extreme cases.

    But as some of the previous examples have shown, public policy decisionmakers can’t get away from making decisions that cause the deaths of some innocent people in order to save others. A lot of anti-consequentialists try to avoid facing this fact.

  3. Ah, but if they are constrained by legal/political controls, then they can’t just simply make decisions that have the best consequences, unless you include respect for the constraints of law as part of the good consequences, in which you are reduced to saying whatever you do is, by definition, the action with the best consequences. Which makes it all rather meaningless and not a useful guide for decision making.

  4. As I’ll explain in a later post, I think your argument would be valid in the context of individual ethics, but it is not valid in relation to public policy, where there is a principal-agent problem.

    The fact that public policy should be aimed at the best consequences does not imply that each individual who makes public policy decisions should have discretion to do whatever they think will have the best outcome.

    But this is probably something that needs a whole post, rather than a comment.

  5. Also something that needs a whole post, but:

    I’ve often thought that public policy in the aggregate should be aimed at the best consequences (as any policy, actually), but that the best consequences are usually the result of people acting as though they were governed by some sort of deontological theory.

    In other words, act like Kantians to reap the rewards of utilitarianism.

    Make sense?

  6. Decision Procedures vs. Criteria
    John Quiggin criticizes the Stanford Encyclopedia of Philosophy’s entry on consequentialism. Specifically, he’s not happy with their emphasis on actual consequences as opposed to expected consequences. Then he says:As I say, I find the discussion here …

  7. Are the universities an instrument?
    It is now taken for granted in the policy making culture in Australia that our higher education institutions are, and should be, an instrument for wealth creation. The purpose of deploying the instrument, say the utilitarians, is to ensure comparative …

  8. Risk and Reagan
    Since the obituaries and eulogies for Ronald Reagan have now been read, I think it’s reasonable to take a critical look at his historical contribution. It’s often argued that Reagan accelerated the end of the Cold War by raising US…

Comments are closed.