Monday, November 14, 2011

Why Consequentialism?

People sometimes ask me why I'm a consequentialist. This is a difficult question to answer productively, since direct introspection merely reveals my deep-rooted sense that non-consequentialist views just don't make sense. There's probably no single argument that's responsible for this intuitive response. But it might at least be fun to brainstorm a few considerations that could plausibly lead one to favour consequentialism...

(1) The 'fundamentality of value' intuition: It seems very plausible that morality (insofar as it's worth caring about) is fundamentally concerned with making the world a better place. This seems a more attractive conception of action-guiding normativity than the old-fashioned conception of morality as a list of "do"s and "do not"s.

Put another way: Suppose you have a choice between two actions, one of which makes things better, and the other makes things (comparatively) worse. Doesn't that seem to settle the question of which action is most worth choosing? We may wonder: How could it be wrong to choose the action that (predictably) makes things turn out best?

(2) Skepticism about doing/allowing and related distinctions. Whether it's doing vs allowing, or intended vs merely foreseen, the kinds of distinctions that deontologists rely upon just don't seem significant enough to be able to pull such normative weight. As such, non-consequentialist views end up looking like mere flimsy rationalizations for status quo bias.

(3) The Paradox of Deontology: It seems somehow incoherent to hold that one shouldn't perform certain kinds of actions even to prevent the occurrence of more such bad actions. (As G.A. Cohen put it, "if such sacrifice and violation are so horrendous, why should we not be concerned to minimize their occurrence?") See also Parfit's argument that common-sense morality is collectively self-defeating.

(4) "God's eye view" arguments: It seems that an ideal (benevolent, omniscient) observer would want us to perform the actions that make things turn out best. And it's plausible that the prescriptions of such an ideal observer would coincide with those of morality (why would they differ?). Similar remarks apply to the Veil of Ignorance.

(5) Equal Concern: In light of points 2 and 4 above, consequentialist impartiality seems like the most principled way to treat everyone equally (thereby satisfying an appealing and plausible candidate, rivaling point #1, for 'what morality is fundamentally about'). Again, the rivals seem to rest on either some kind of status quo bias, or else confused notions about what constitutes treating people with equal concern.

Have I missed anything? What do you consider the strongest reasons in favour of consequentialism?


  1. Hmm.

    Point 1 seems very dubious. I would say morality is fundamentally about making/doing the better/good, choice/action. The non-consequentialist intuition here would be that it is the action itself that is better or good not how the world may turn out. In fact a static description of the world is neither better or worse, merely the relations between people.

    As for point 3, the non-consequentialist would say that not commiting a bad action is the responsibilty of person A towards person B. Even though the bad action would prevent person B from doing the same bad action to someone else, it is the responsibility of Person B to do or not do that action. Unless individuals are morally responsible for their actions moral responsibility would stretch back to the very first person who enabled a person to do a bad thing. This seems utterly implausible.

  2. "direct introspection merely reveals my deep-rooted sense that non-consequentialist views just don't make sense."

    Same. What do you make of Joshua Greene's suggestion that the neuroscience can show (more or less) that consequentialism is based on reasoning and deontology on emotional + post hoc rationalisation?

  3. This comment has been removed by the author.

  4. I would add that once you don't have a doing/allowing distinction or an intending/foreseeing distinction (or some other such justification for constraints), it seems hard to accept that it could be permissible to do (at least significantly) less than the best.

  5. Saranga - Admittedly, not everyone will be swayed by these considerations. But for those attracted to consequentialism, these are likely some of the major reasons why.

    re: (1), I find it incredible to deny that a "static" world of suffering is worse (less desirable) than a world of flourishing individuals. We should prefer the latter to the former, even if we never have the opportunity to act on this preference.

    A slightly more technical way of making point #1 is the teleological conception of action: it seems (to consequentialists) that action is goal-directed. The morally fitting agent first has the right goals, and subsequently acts so as to bring about those goals.

    re: (3), I'm not sure we can parcel out responsibility so neatly. See the murderer at the door. Obviously there's some sense in which the murderer himself is primarily responsible, but that's no excuse for others to aid him when they could have (foreseeably) prevented him from finding and killing his victim. (Note: 'expected value' consequentialists will think it matters whether downstream bad results are foreseeable consequences of your current actions. This generally constrains how far back you can trace responsibility, and in a much more plausible fashion than just letting off the hook everyone besides the immediate actor.)

  6. David - I'm generally suspicious of the immediate relevance of attempted geneological "debunkings" of philosophical beliefs, as I don't think such origin stories can do much to show that the view in question is actually false. (In this case, for example, why think that 'reasoning' is necessarily a better guide to truth than 'emotion'? Even consequentialists may expect that a lack of adequate sympathy could often lead people morally astray.)

    Such debunking stories prove useful further "downstream", so to speak... Once we've established (via argument) some philosophical question, our conclusions may be reinforced by a story of how the opposing view is typically caused by factors that (on our view) aren't truth-conducive. I just think it's typically question-begging to bring up such issues at the start of inquiry, since which factors are truth-conducive obviously depends on the unsettled question of which view is true!

    Robin - what do you mean by "Morality is not binary"? Some actions are both right and wrong?

    There are many competing (and incompatible) accounts of what considerations are morally relevant, and how they should be weighed against each other. If we are to avoid making moral decisions arbitrarily, we must seek some principled way of adjudicating between these competing considerations. And that is just what a moral theory aims to do. It represents the theorist's best considered judgment of how to adjudicate between rival moral considerations.

    It's a weird charge of "bias" that charges people with acting on their best considered judgment. (Charges of 'partisanship' seem especially misleading in this context, as I trust that theorists on all sides are genuinely truth-seeking, and not merely seeking rhetorical ammunition to make their "tribe" look better regardless of the facts.)

    You suggest that "a more compelling case would consider all arguments", but it's not like a proponent of a theory has never considered the kinds of arguments put forth by opposing views. Again, he has merely come to the considered conclusion that they are bad arguments. He may or may not be correct as a matter of substance. But I don't see anything procedurally illict about employing such philosophical discernment.

    Nick - indeed, though I'm tempted to go further and question whether the tradition categories ('permissible', etc.) are really carving the structure of "normative space" at its joints. I'm inclined towards a more "scalar" view on which we fundamentally just have more or less reason to perform various actions.

  7. Oops, that should read "Once we've settled (via argument) some philosophical question..."

  8. This comment has been removed by the author.

  9. Hi Richard,

    (1) and (4) have got me thinking.

    I don't have any settled meta-ethical views. I am pretty far from thinking of myself as endorsing consequentialism, but not for any worked-out reason.

    (1) and (4) seem sort of low on content. I would suggest that what makes a lot of people not accept consequentialism is *not* that claims/intuitions like (1) and (4) seem false, but that they don't seem substantial enough to be very informative or interesting. This is obviously connected to their reliance on a prior conception of what it is for consequences to be better than others.

    I realize that consequentialism understood schematically like this is still far from trivial. But I want to ask something like: is it really a view of metaethics, or just a (very open-ended) schema for one?

    That's a very unclear question, but perhaps it's what really separates consequentialists from a lot of their non-subscribing fellows. Thus the disagreement between *these* groups, insofar as there is one, might be regarded as meta-meta-ethical: is schematic consequentialism the *sort* of thing which should be occupying a space marked 'View of Meta-Ethics'? And perhaps this disagreement wouldn't seem very important once clarified. (A plea for meta-meta-ethics!)

    Let me make a confession in the interest of science: I have a mild aversion to consequentialism and statements of it. My not being a consquentialist, what ever that means, seems to have a lot to do with that, rather than anything to do with arguments, theoretical virtues, etc. It seems somehow shallow, or over-simple or something. (I think Nietzsche may have influenced me here.)

    I'm not trying to defend this reaction, nor am I entering it as a move in a debate. I just thought it might make thought-provoking data for consequentialists like yourself who find it hard to make sense of anti-consequentialism.

    (I have no idea what I'm talking about, but if Robin the A-Level Student can weigh in, so can I!)

  10. Tristan - right, there are many different possible theories that fall under the consequentialist umbrella. There are important further questions, e.g. regarding our theory of the good, but I'm bracketing those details for the moment. So really the issue here is whether we have good reasons for thinking that the correct theory must be some form of consequentialism (as opposed to some form of non-consequentialism). I think this marks a real (and important) disagreement within normative ethics. In particular, I take it that consequentialists and non-consequentialists will disagree over cases like the following: Should I kill one innocent person to prevent someone else from killing five innocent people?

    P.S. I understand consequentialism primarily as a first-order (or "normative ethical") theory, though admittedly the borderline between normative ethics and metaethics gets a little fuzzy in places.

  11. Richard -

    With regards to the first point. It seems the argument is that, if the action makes the world better then it is the morally right action. Given this we are compelled in to some analysis of "making the world better." This analysis can't find better to simply mean morally right because there would be a circularity. It appears entirely unclear to me what the right approach or playing field would look like to analyse this "making the world better."

    I think here the non-consequentialist thought is that, if an action is morally right then it makes the world better. Given this the analysis stays int he realm of morality and we simply int he project of finding out what is the morally right. The project could then define the right in terms of various other ideas such equality, liberty or whatever.

    With regards to point 3, while I agree that the enablers of morally wrong action are part of the chain of responsibility, the only obligation hey have if they feel like is to feel slightly guilty. Other than that the action of a murder is fundamentally and physically different to the inaction of a passer by who does not help or the. Action and inaction are just so metaphysically distinct that to me they in no way can share the same moral value. Also, the vagueness of who to include in the chain of responsibility itself seems enough to abandon this argument.

  12. Saranga - We may start with a welfarist understanding of "making the world better" as increasing the well-being of the world's inhabitants. It's an interesting open question whether there are non-welfarist values in addition (e.g. great cultural or aesthetic achievements).

  13. I agree with Robin. While I do understand the motivations for your position, I think that only a morbid rationalist could really accuse people with context-sensitive, nonideological responses to ethical situations as choosing "arbitrarily". I don't think that you could really mean *that*.

    By way of comparison, very few philosophers of science think that there is something like an algorithm for producing the most acceptable scientific theory. Multiple values (scope, parsimony, power, etc) determine which theory is best in any given context, and there seems no way to adjudicate in principle. Furthermore, as a matter of fact, the three biggest physical theories on offer stand in deep contradiction. Yet, we do not accuse scientists of acting arbitrarily when they choose, for personal or contextual reasons, to operate within one or another theory. It just "works" for them, and no further answer is required.

    Yet, for some reason, moral theorists want to hold ordinary people to a much harsher standard. Even a genuinely concerned, sensitive and thoughtful person is supposed submit his decision to Reason™, on pain of arbitrariness. Once we've reached this level of absurdity, something has gone wrong.


Visitors: check my comments policy first.
Non-Blogger users: If the comment form isn't working for you, email me your comment and I can post it on your behalf. (If your comment is too long, first try breaking it into two parts.)

Note: only a member of this blog may post a comment.