A question for those better tutored in (or perhaps more sympathetic to) consequentialism than I: How are we to understand the claim that consequentialists count the "value of the action itself" as part of the value of an outcome? I’ve encountered this claim numerous times, but I confess I don’t understand how to square an action’s having value in itself with the consequentialist thesis that only the results or consequences of an action determine its deontic status.
I quote William Shaw as a typical expositor of this claim:
When consequentialists refer to the results or consequences of an action, they have in mind the entire upshot of an action, that is, its overall outcome. They are concerned with whether, and to what extent, the world is better or worse because the agent has elected a given course of conduct. Thus, consequentialists take into account whatever value, if any, the action has in itself, not merely the value of its subsequent effects. ("The Consequentialist Persepective", in Dreier (ed.), Contemporary Debates in Moral Theory (Blackwell, 2006), p. 6)
Shaw then proceeds to point out that:
(a) consequences of an action include not only its causal effects, but also the results of refraining from acting, so that, e.g., my not giving a panhandler money does not cause him to sleep in the street that night, but is nevertheless a consequence of my not giving him money, and
(b) the boundary between an action and its consequences can be rendered fuzzy by how we describe the action.
Granting these two points, I still don’t understand Shaw’s conclusion that consequentialism recommends that we "assess and compare the overall outcomes of the various actions we could perform, and these outcomes include the positive or negative value, if any, of each action viewed by itself as well as the positive or negative value of its subsequent effects." (p. 7, emphasis added) The trouble I’m having is seeing how an action can have value that contributes to the value of its outcome without that value being either a causal consequence of the action itself or some counterfactual consequence of the action (i.e., referring back to (a) above, the consequences of my not doing what I could have done but did not in fact do). Suppose I break a promise: What value does the action have that contributes to the value of its outcome? I might end up diminishing whatever trust existed between me and the promisee, but that’s a causal outcome. I might express disrespect for the promisee or her interests or autonomy, and that could be seen as a property of the action rather than a subsequent effect of the action. But surely it’s easier for consequentialists to say that what has value here is the promisee understanding or experiencing that disrespect (a straightforward consequence), not my expressing it through my act of promise breaking. Talking about actions having value in themselves sounds suspiciously deontological to my ears.
Perhaps I’m just being dense here, but I’d find it useful to have concrete examples of ‘the action itself’ making a contribution to the value of its outcome. Thanks!
How about the event/action described correctly as ‘the killing of Smith’? The killing of Smith seems itself disvaluable apart from its further causal consequences. Or we could let the killing of Smith be an (improper) consequence of itself. Suppose I kill Smith at instant t. Now take the temporal interval (0, t]. It does seem true that the total value in (0, t] is lower than it would have been had I not killed Smith at t. If that’s true, as it seems to be, then the event at t is disvaluable apart from its (proper) consequences.
Michael, suppose you play a set of tennis this afternoon, and you play very well.
One consequence of this action of yours is that you feel good about yourself. Another is that you impress your opponent. These may both be good outcomes, from your point of view. But if we leave it at that, we are missing something pretty important: the playing itself. You had fun, and you achieved excellence; arguably (plausibly, to my mind) the actions themselves contributed constitutively to your welfare.
Is that a good enough example?
Jamie —
I’m not sure that your example doesn’t rather illustrate Michael’s concern about consequentialism; if I understand him correctly, “having fun”, “achieving excellence”, and the like all seem to be translatable by someone with that kind of concern into consequences of the act, rather than “properties of the act in itself.”
Can’t a consequentalist say that your breaking the promise has negative value because promising breaking is itself intrinsically bad? In other words, the claim being made here is a modifcation of the theory of the good: instances of promise breaking are intrinsically bad, while instances of promise keeping are intrinsically good. This, I take it, is what Mike A. was getting at when he wrote:
The killing of Smith seems itself disvaluable apart from its further causal consequences.
Anyway, that is how I have always understood claims like Shaw’s. Of course, no reason has been offered to accept this theory of the good, but consequentialism certainly seems compatible with it.
I’ve always understood it exactly as Scott notes. Given that acts have properties, and given that there’s no reason in principle why you can’t adopt any theory of the good that you like, properties of acts can themselves be counted as either valuable or disvaluable.
I’m not sure these comments are fully addressing Michael C’s concern.
Michael C wants to know how ‘the action itself’ can make a contribution to _the value of its outcome_. He is not just wondering how it can be that ‘the action itself’ can have value. After all, it might be that ‘the action itself’ has value without making a contribution to the value of its outcome. This would be the case if ‘the action itself’ were something wholly separate from its outcome. And since it does seem natural to separate actions from their outcomes, Michael C may be pointing to a real problem for the consequentialist even if it is made perfectly clear that ‘the action itself’ has value.
I think there are two main ways to solve Michael C’s problem. The first is simply to include the action itself among its consequences. To take this route, you just say that one of the consequences of doing X is that X has been performed. Then, as long as ‘the action itself’ has value, it contributes to the value of its consequences by contributing to the value of the fact that it has been performed. (A possible problem with this solution is that it might be that ‘the action itself’ is something relevantly different from the fact that the action has been performed, such that ‘the action itself’ might fail to contribute to the value of its consequences even if the fact that the action has been performed does contribute to the value of its consequences. In that case, this move does not solve Michael C’s problem.) The second way to solve Michael C’s problem is to sever the connection between consequentialism and consequences. If you take this route, you can say that consequentialists want to maximize (or satisfy a minimum level of, etc.) _value_, whether that value technically attaches to consequences or not. Then consequentialists will be free to include the value of ‘the action itself’ in their calculations even if ‘the action itself’ is, strictly speaking, a thing apart from its consequences. (A possible problem with this route is simply that it might deprive consequentialists of the right to the name ‘consequentialist.’)
I think it is worth attending to the citation from Shaw. On the one hand, the citation implies that instrumentalism is primary for consequentialism; the instrumental relation between acts and consequences is what makes an act valuable. On the other hand, to be ‘concerned with whether the ‘world is better or worse because the agent elected’ some course of conduct implies that the level of value is primary for consequentialism. If the instrumental relation is primary for consequentialism, then we have a ‘productive’ theory. The value of the outcomes produced determines moral status of the act but the act, and presumably properties of the agent (since an act entails an agent), are not valuable intrinsically valuable.
If value, is primary, then it is open to the consequentialist to say that acts, properties of the act (presumably properties of the agent also) determine moral status of the act in addition to its outcomes. I am sympathetic to the view that this may no longer be ‘consequentialism’ or that it may be a case of non-standard of ‘consequences.’
David and Robert,
Consider what Mike C. calls Shaw’s point (a): “consequences of an action include not only its causal effects, but also the results of refraining from acting, so that, e.g., my not giving a panhandler money does not cause him to sleep in the street that night, but is nevertheless a consequence of my not giving him money.” Do you deny that philosophers have always used the term ‘consequentialism’ to apply to theories that included such non-causal effects as part of an act’s consequences? And if ‘consequentialism’ has always been used to describe a theory that assesses more than the causal consequences of an act, why would including such non-causal effects deprive consequentialists of the right to the name ‘consequentialist’?
Doug, I guess I’m less sure than you are about the causal consequences of omissions. On some analyses of causation it looks like I might have caused the panhandler to sleep in the street. Maybe the events are linked in the right counterfactual way; I can’t tell. But then maybe my omission is a probabilistic cause, if you want causation to be compatible with liberatarian freedom. On the main point, I’m pretty sure that consequentialists vary in what they consider to be the consequences of their actions. Some do restrict consequences to causal consequences, others have a much broader view.
Mike A.,
Point taken on the causal consequences of omissions.
Regarding the main point, however, you write, “I’m pretty sure that consequentialists vary in what they consider to be the consequences of their actions. Some do restrict consequences to causal consequences, others have a much broader view.”
I’m curious. Can you name names: philosophers (consequentialists or non-consequentialists) who have explicitly said that ‘consequentialism’ is to refer only to theories that assess the value of solely an act’s causal consequences? Just off the top of my head, I can name a number of philosophers who have explicitly said consequentialists do include non-causal consequences in their assessments of outcomes: Bernard Williams, Samuel Scheffler, David Sosa, Amartya Sen, Jamie Dreier, Doug Portmore, Campbell Brown, William Shaw, David Brink, etc. I’m sure I could find dozens more if I looked into it. But I can’t think of anyone who has explicitly said that consequentialists can’t include non-causal consequences in their assessments of outcomes. But maybe I just have a selective memory. So I would be genuinely curious to know who explicitly takes the opposite view.
JH Sobel’s ‘Utilitarianisms: Simple and General’ _Inquiry_ restricts the relevant consequences of actions A to those that are strictly depend on the performance of A. There are numerous other restrictions on consequences that arise in the context of utilitarianism and past and future mistakes. Holly Goldman in ‘Dated Rightness and Moral Imperfection’ PR, and her ‘Doing the Best We Can’ (in Values and Morals, (ed.) J. Kim), M. McKinsey ‘Levels of Obligation’ PS, Sobel, ‘Utilitarianism and Past and Future Mistakes’ Nous. Fred Feldman, ‘World Utilitarianism’ (Analysis and Metaphysics, (ed.) K. Lehrer) P.S. Greenspan, ‘Conditional Oughts and Hypothetical Imperatives’, JP., Lars Bergstrom’s, _Alternatives and Consequences of Actions_ (Stockholm: Almqvist & Wiksell). There is also relevant work by Richmond Thomason and Lennart Aqvist (I’ll get the references if you like). In any event here’s a progression of utilitarian principles you’ll find in this context:
P. An action x ought to take place iff. (i) x is contained in a life that is optimum among lives open to the agent at this agent’s first moment of choice (ii) no other agent-identical action incompatible with x satisfies (i).
Dag Prawitz (‘The Alternatives to an Action’, Theoria, 75(?)) has defended something like (P) that is insensitive to past and future mistakes of utilitarian agents. What matters are the consequences of one’s actions in the best utilitarian history rather than the consequences of one’s actions in the actual sub-optimal history.
B. An action x ought to take place iff. x is contained in a life that is open to the agent of x at the action’s first moment (ii) same as in (P).
The consequences that matter to (B) are those that are best as measured from the first moment of the action. (B) is sensitive to past mistakes but not future mistakes. (B) would have me ignore the fact that utilitarian agents will in fact make mistakes and fail to bring about the best future possible at the first moment of my action. Those facts about the future do not count as part of the consequences of what I do now. Lar’s Bergstrom has defended something like (B) and so has Fred Feldman (Doing the Best We Can)
S. An action x ought to take place iff. x is contained in a life optimum among lives securable by the agent at the actions first moment (ii) same as above.
I think you have in mind something like (S) that makes relevant to the assessment of an action everything that would follow the action (including all the mistakes that would be made and have been made). Goldman defends something like (S) and so does Sobel in some places. In any case, these are some of the references that you asked for.
Mike A.,
Thanks for these references. But do any of these authors explicitly say that the term ‘consequentialism’ is to be reserved for only those theories that assess the value of only the causal consequences of acts? I certainly don’t doubt that some philosophers think that when it comes to assessing the deontic statuses of actions the only relevant consequences are those that are the causal effects of the acts themselves. I thought, though, that the issue was whether someone who denied this might still legitimately be called a consequentialist in the sense that the term ‘consequentialism’ is ordinarily used.
_But do any of these authors explicitly say that the term ‘consequentialism’ is to be reserved for only those theories that assess the value of only the causal consequences of acts?_
Oh, no. Definitely not. If that is the issue, then you have all of these on your side of it. I thought the question was whether any utilitarian/consequentialist took the position that the relevant consequences were just the causal ones. The answer to that I think is yes, and there are all sorts of views in between on which consequences matter. But no one I know has suggested that the term ‘consequentialism’ applies only to theories focus on causal consequences. That looks like a semantic claim to me, and a mistaken one.
Yes, right, it is that semantic claim, which we both argee is mistaken, that I was objecting to. I take it that David and Robert were both making this semantic claim, each in the last sentence of their comments above.
I’m surprised no one’s yet mentioned Moore (the ‘original’ consequentialist) in connection to Michael’s question. His own view seems pretty attractive: “In asserting that the action is *the* best thing to do [hence, ‘right’], we assert that it together with its consequences presents a greater sum of intrinsic value han any possible alternative…[it is to assert] more good or less evil will exist in the world if [the line of conduct] be adopted…” (PE p. 25) He then goes on to spell it out. Maybe he’s wrong that this is what we’re ‘asserting’. But the basic idea seems to get at the core idea that actions can have intrinsic as well as instrumental value, and both have to go into our computation of what to do.
It seems to me that some of the anxiety stemming from the semantic issue derives from the worry that, if we allow a theory that recognizes the values of non-causal consequences (e.g. the logical fact that a promise has been kept, or broken) to count as consequentialist, then we will have no way of drawing a bright line between consequentialist and non-consequentialist theories. As Michael C. put it in his original post, “Talking about actions having value in themselves sounds suspiciously deontological to my ears.” The worry, then, is that (for example) someone who thought that agents must always keep their promises would simply count as a consequentialist who put a very high positive value on the keeping of promises.
But as Scheffler pointed out (I believe) in The Rejection of Consequentialism, there is another and better way of drawing the bright line. What unites genuinely consequentialist theories is the restriction that the value of an outcome must be impersonal, in the sense that it does not vary depending on whose point of view we are considering things from. Thus, a consequentialist of the sort above would think that, while the positive value of keeping one’s promise would generally make promise-keepings highly valuable (and thus morally obligatory), an agent who faced the choice between breaking one of her own promises, or bringing about a state of affairs in which two other agents broke their promises ought to prefer the former alternative. A deontologist, on the other hand, would hold that one ought to keep one’s own promises, even where the outcomes are not optimal.
This can get tricky, since many deontologists will want to place a limit on how bad consequences can get (and thus, might want to insert the phrase “except in extreme circumstances” before “one ought to…” above), and allowing that exceptionally bad outcomes can outweigh the obligation to keep one’s promises should not disqualify one as a deontologist. (Also, some theorists – Richard Fumerton, for instance – insist on muddying the waters by referring to egoism, for example, as a consequentialist theory.)
But I think the general approach is the right one: the deontologist will place a special significance on the positive and negative values attached to one’s own acts, while the consequentialist will not.
I certainly did not intend to endorse what Doug P is calling ‘the semantic claim,’ and am having trouble finding a sentence in my comment that could be reasonably construed as an endorsement of it. In any case, here is a slightly clearer version of (part of) what I was trying to say.
It is possible for the consequentialist to say that
(a) ‘The act itself’ has or brings about some value; but
(b) The value brought about by the act itself fails to contribute to the value of the consequences of the act; nevertheless
(c) We still ought to take the value of the act itself into account when making the relevant calculations (e.g. calculations that determine what one ought to do).
My claim was that, if consequentialists say these three things, then they enable themselves to take the value of ‘the act itself’ into account when determining things like whether an act ought to be performed. A possible problem with taking this route, I said, is that (b) and (c) might be thought to rob consequentialists of the right to the name ‘consequentialist’ (since if they take this route, something other than a consequence would matter to the rightness of an act). I did not say that consequentialists must or ought to take this route — it was just one of two possible routes I described. Nor would ‘the semantic claim’ follow if consequentialists did take this route.
Troy,
You note that Mike C. says, “Talking about actions having value in themselves sounds suspiciously deontological to my ears.” That doesn’t sound deontological to my ear. After all, the claim that certain actions have intrinsic value is a claim about what’s good, not a claim about what’s right. And, as I see it, deontology is a certain class of theories about what’s right. Deontological theories are, as I see it, neutral respect to what is good.
Also, I think there are other ways of drawing a bright line between consequentialist and non-consequentialist theories that don’t involve, as you suggest, restricting consequentialism to impersonal (agent-neutral) value. For instance, we might hold that a theory is consequentialist if and only if it holds that the rightness of an action is solely determined by how its outcome (broadly construed) ranks relative to those of the available alternatives on some (agent-relative or agent-neutral) evaluative ranking of outcomes. There is, of course, a “danger” that Kant and myself will both agree that it is wrong to break a promise even to prevent two others from committing comparable promise-breakings. But there is no danger that Kant will “simply count as a consequentialist who put a very high positive value on the keeping of promises,” for Kant holds that such promise-breakings are wrong not because their outcomes don’t rank as high as the alternatives, but because they violate the categorical imperative. Consequentialists and non-consequentialists can agree on which acts are right and which acts are wrong; they just can’t agree on what makes acts right and wrong.
David,
Thanks for the clarification. Let me try to clarify my point. If the term ‘consequentialism’ has ordinarily been used to denote all and only theories that hold that the deontic statuses of actions depend only on the value of their total outcomes (where the total outcome associated with an act includes the act itself), then anyone who holds that the deontic statuses of actions depend only on the value of their total outcomes has the right to be called ‘consequentialist’ whether or not they accept your (b) and (c). You can complain that they shouldn’t be called “consequentialist” if they don’t hold that the deontic statuses of actions depends only on the value of their consequences. But isn’t that like complaining that clay pigeons shouldn’t be called ‘clay pigeons’ because they are not really pigeons. It seems to me that they should be called whatever they are called. And ditto for consequentialists.
David,
(a)-(c) makes the dispute verbal. Effectively you’re saying that we should consider the value of the action (itself) in deciding what to do, but we should not call the value of the action itself a consequence of the action. Fine.
I have no idea why an action cannot be a consequence of itself, but it doesn’t matter. On your view we are still considering the value of the action itself, so your view has no implications for consequentialist deliberation.
Doug,
You write that “we might hold that a theory is consequentialist if and only if it holds that the rightness of an action is solely determined by how its outcome (broadly construed) ranks relative to those of the available alternatives on some (agent-relative or agent-neutral) evaluative ranking of outcomes” and that “Kant holds that such promise-breakings are wrong not because their outcomes don’t rank as high as the alternatives, but because they violate the categorical imperative.” But if the ranking in question is simply the ranking of which actions are morally preferable to which other actions, then the breaking of a promise will rank lower than the keeping of a promise on Kant’s ranking, precisely because the former violates the categorical imperative. Of course, you might respond ‘Yes, but that’s not an evaluative ranking.’ But I’m not sure that it isn’t—at least if we construe the notion of value somewhat broadly (and this whole thread is about the possibility of construing things broadly!). After all, there is surely some sense of ‘bad’ in which it is true to say that Kant thinks that violations of the categorical imperative are bad. So this sort of approach seems to me to lose its grip on what is supposed to be a deep difference between consequentialist and deontological theories—unless, of course, you can define the precise sense of ‘bad’ in which a promise-breaking can be held to be ‘wrong,’ but not ‘bad,’ from a Kantian perspective. (Or, to be fair, in which it is allowed to be bad but its badness is taken to be irrelevant to its wrongness.)
Troy,
I don’t want to get into Kant scholarship and what Kant can and cannot say. But here’s the distinction that I’m trying to make: a non-consequentialist can say that breaking a promise is wrong even if its outcome ranks lower than that of its alternative on the relevant evaluative ranking, whereas a consequentialist cannot. For the consequentialist, but not for the non-consequentialist, deontic status has to track value. Isn’t this a clear distinction that doesn’t require restricting consequentialist to an agent-neutral evaluative ranking of outcomes?
Thanks everyone for the timely and insightful replies. I won’t be able to address each one specifically. To my surprise, my question was less innocuous than I thought.
In general, I think it is a legitimate worry (echoed by Troy, David, and Andrew) that incorporating the intrinsic value of actions into the value of outcomes ends up eliding the differences between consequentialists and their opponents. “Do that which has (creates, produces, realizes, etc.) the most value” is a claim that, taken in broadest terms, most every moral theorist, consequentialist or otherwise, would affirm. What makes this still more acute is (as Troy observes) that deontologists, particularly of the prima facie rather than asbolutist stripe, can say that consequences have value that contributes to an action’s moral status (without perhaps exhausting its status). So how are deontologists and consequentialists distinct? The going wisdom (I thought) is that they differ in where value (or at least the value relevant to the determination of an action’s deontic status) is located. But that gets harder to see once consequentialists add actions’ intrinsic value to the value of outcomes.
Also, Jamie’s examples are the kinds that occurred to me as well, but (following Andrew and Troy a bit here) doesn’t a more elegant theory result if consequentialists treat these values as consequences in one of the two senses I mentioned initially rather than as values intrinsic to the action? It does seem to me that such values can be re-translated into either effects of actions or non-causal consequences. This would allow consequentialists to hold the simpler view that only states of affairs, not actions, have intrinsic value.
And there is something strange about talking about actions having “intrinsic” value; I’m not sure that deontologists uniformly accept it. Doug points out that deontology is a theory about the right, not the good. And there are some deontologists (of the perhaps suspect, ‘intuitionist’, a priori self-evident moral principles kind) that might say that an action’s moral status is determined by its intrinsic value, where the intrinsic value is due to the action’s being an act of a certain type (a killing of an innocent, a promise breaking, etc.) But some deontologists (or non-consequentialists, let’s say) will hold that an action’s moral status is determined by other properties (a broken promise harms another, violates the categorical imperative, etc.) that are not obviously ‘intrinsic’ to the action. Perhaps this is a long way round to saying that I’m not sure we have a clear grip on how (or whether) an action has intrinsic value.
Of course, there’s no museum we go to in order to find who’s a consequentialist and who isn’t. But expositions of consequentialisms like Shaw’s muddy things up in my mind.
Doug, you write, “here’s the distinction that I’m trying to make: a non-consequentialist can say that breaking a promise is wrong even if its outcome ranks lower than that of its alternative on the relevant evaluative ranking, whereas a consequentialist cannot.” I (as you know) agree with you that we don’t want to characterize consequentialism as always involving agent-neutrality. But I also sympathize with Troy’s point about your characterization. As I understand it, the worry is that depending on what we’re including in our rankings, some non-consequentialists might not be able to say that “breaking a promise is wrong even if its outcome [i.e., the outcome of keeping the promise, right?] ranks lower than that of its alternative on the relevant evaluative ranking.” For the non-consequentialist might have a value system that ranks actions (wholly or partly) in terms of their deontic statuses, in which case if breaking a promise is wrong then it will rank lower than keeping the promise. This view then would be non-consequentialist (by hypothesis), but would not be able to say that breaking a promise is wrong even if the oucome of keeping the promise ranks lower than its alternatives.
For this reason, I think we should follow those who restrict the values included in the evaluative ranking to non-deontic (or as it’s often put, non-moral) value, when drawing the consequentialist/non-consequentialist distinction. As I think you’ve pointed out to me in the past, this might make some views, like Sen’s, not-necessarily-consequentialist. But it seems to require the fewest revisions to our overall classificatory scheme.
I’d like to point out the first dfn of ‘consequence’ in the OED is: ‘A thing or circumstance which follows as an effect or result from something preceding.” If we are going to take the ‘action’ which precedes as a consequence, then we are not using standard English. I don’t know of consequentialists who have restricted consequences as the OED does – but it is worth noting that if consequentialist ethical theories are to include the ‘action,’ then we re-define a concept theoretical purpose. Nothing wrong with that but we should be alert to what we are doing. However, the semantic point is not the main issue here. It is how to conceive of what has value for purposes of moral evaluation.
Michael,
The Big Mistake, in my opinion, is to suppose that there is one difference that divides the domain of ethical theories into ‘consequentialists’ and ‘non-consequentialists’. There are several. One is the distinction between views that count only causal consequences and those that allow non-causal consequences. Another is the distinction between views that are strictly agent-neutral and those that have some agent-centered elements. Then there is the similar distinction between time-neutral and time-centered. And so on. Not one of these distinctions justifies the label ‘consequentialist’, but it’s too late to fix that.
I honestly see no meaningful difference. Suppose one theory says that there is this to say in favor of Serena’s tennis-playing-act: that inasmuch as it is excellent activity in accordance with virtue, it is itself valuable. A second theory says that there is this to say in favor of Serena’s tennis-playing-act: that it has as a valuable consequence the state of affairs that an excellent activity in accordance with virtue has taken place. It is hard for me to believe that one of these theories is more elegant or simpler than the other. (The wording of the second is more awkward, but I take it that’s not the sort of elegance or simplicity you had in mind.)
“Do that which has (creates, produces, realizes, etc.) the most value” is a claim that, taken in broadest terms, most every moral theorist, consequentialist or otherwise, would affirm.”
Untrue. I can’t name a deontological theory which makes a claim even close to this. Indeed the distinctive feature of deontological theories is that the value produced/associated with actions does not *alone* determine the rightness/wrongness of actions. On deontological theories, the value produced/associated with actions is in general relevant to determining the rightness (or, moral status, generally) of actions, but (unlike teleological theories) it is not the only thing relevant. We are required to do what does not maximize value on social contract theories (Rawls, Gauthier, Nozick etc.), on divine command theories (Alston, R.M. Adams) Kantian and neo-Kantian theories (Tom Hill), and so on and on. Attempts to reduce deontological theories to teleological theories inevitably conflate value and rightness. Peter Vallentyne shows this convincingly (to my mind) in his dissertation and subsequent work in the area.
Jamie,
I like your point about the Big Mistake. But we could look at Michael’s question (or, at any rate, a similar question) from an evaluative rather than descriptive angle: what is the best way to make sense of the consequentialism/non-consequentialism distinction? If this is the task, then we would want to arbitrate between different uses, because, depending on our desiderata, some clearly are better than others. For example, some ways of drawing the distinction put views in their traditionally ascribed camps more often than other ways.
(Then, in an ideal world, it wouldn’t be too late, and we could have a big vote establishing that we’ll only use the best distinction, and live clearly-ever-after.)
A question for Doug.
You say,
Can you tell me, what is the ‘relevant’ ranking? One might have thought it was the ranking R which is such that an act is wrong iff R ranks the act lower than some alternative. But then the thing that, according to you, non-consequentialists can say, but consequentialists cannot, would be just incoherent.
Josh and Campbell,
The evaluative ranking is a ranking in terms of non-deontic value just in the sense that the ranking determines the deontic statuses of actions and not vice versa. This is not to say that the sort of evaluative ranking that consequentialists employ can’t appeal to moral concepts like that of virtue and hold, for instance, that an outcome ranks higher, other things being equal, if the virtuous get a larger share of the aggregate utility than the vicious do. For this reason, I would hesitate to say that consequentialists must appeal to only non-moral value.
On my view what makes a ranking evaluative is that it is a ranking in terms of what people have reason to desire and prefer. So, on my view, one outcome O1 outranks another outcome O2 on a given person’s ranking if and only if that person has better object-given reasons to intrinsically prefer O1 to O2 than to intrinsically prefer O2 to O1. Consequentialists can and do, though, disagree about what the relevant substantive ranking is. I take it that utilitarians hold that people always have better object-given reasons to prefer an outcome with more aggregate utility to one with less aggregate utility. But we might disagree about whether this ranking in terms of aggregate utility is the relevant one, the one that accurately reflects what people have best reason to prefer. The point I was trying to make was only that, on my version of the distinction, the non-consequentialist, but not the consequentialist, can hold both that it would be wrong for me to break a promise even to prevent two others from breaking their promises and that I ought always prefer, other things being equal, a world with fewer promise breakings. Unlike consequentialists, non-consequentialists can hold that what I have reason to do can come apart from what I have reason to want. This is a clear distinction that doesn’t restrict consequentialists to agent-neutral value.
I agree with Jamie’s point about the Big Mistake, and I don’t claim to be giving a definition of ‘consequentialism’ as it is ordinarily used.
Doug,
That’s interesting. I think I understand your view better now. Let me check.
It seems that what you mean by ‘evaluative ranking’ is not what I would think of as a ranking, or at least not merely a ranking. I think of a ranking simply as a relation, a set of ordered pairs, where the members of the pairs are outcomes, or consequences, or whatever is being ranked. A ranking tells us, for each pair of things in its domain, whether one is ranked above the other (and if so, which is the higher ranked), or whether they are ranked together, or whether neither is ranked vis-a-vis the other. But that’s all a ranking, by itself, can tell us.
If that’s what a ranking is, then it seems your ‘evaluative ranking’ is not a ranking. You say, for example, that an evaluative ranking can ‘appeal to moral concepts’. But a ranking cannot do that. A set of ordered pairs does not appeal to any concepts, let alone moral ones.
Perhaps, then, you have the following in mind. An evaluative ranking is not a ranking alone, but rather a ranking plus something else: a principle that characterises the ranking. For example, the principle might be, ‘X is ranked above Y iff the total utility in X is greater than that in Y, and X and Y are ranked together iff the total utility in X equals that in Y.’ A principle tells us more than a ranking by itself. It tells us that the ranking is coinstantiated with some other relation: in our example, the relation of ‘having greater total utility’. It reveals some kind of order or pattern in the ranking. One might say it explains the ranking. (A ranking is purely extensional, a principle intensional.)
Is that right? And if so, can you say a little more about the role played by the principle component of the evaluative ranking in your definition of consequentialism?
Hi Campbell,
Yes, that’s right, and thanks keeping me honest. Consequentialism, as I see it, conjoins a ranking principle (which generates a ranking) with a principle that specifies a function that determines the deontic statuses of actions in terms of that ranking (e.g., x is morally permissible just when and because its outcome is not outranked by that of any of its alternatives). If you want a more careful (although probably not careful enough) explication of my view, it can be found in my papers “Consequentializing Moral Theories” and “Dual-Ranking Act-Consequentialism,” at least one of which you’ve already read.
In any case, I’m not sure what more you want me to say. I’ve already told you that the ranking is coinstantiated with some other relation: namely, the relation of ‘having better object-given reason to intrinsically prefer’. Given your example, I would have thought that that was enough.
Jamie has already mentioned the Big Mistake. But there’s also so the Nearly-as-big Mistake: thinking that the content of some philosophical doctrine must be given by the meaning of the words used to name it. The word ‘consequentialism’ may be linguistically derived from ‘consequence’, but it doesn’t follow that consequentialism is a doctrine especially about consequences. (One might say, I suppose, that the mistake is failing to see that ‘consequentialism’ is a name rather than a description.) Perhaps this means that consequentialism is inaptly named, but, as Jamie says, sometimes it’s an inapt name that sticks. Recently I was reading the preface of David Lewis’s On the Plurality of Worlds, where he says that ‘modal realism’ is not a very apt name for the doctrine he defends in the book, but that he’ll stick with it anyway, because that’s the name people have come to use. It would be very odd to claim that modal realism is not the view famously defended by Lewis, but rather some other view for which ‘modal realism’ would have been a more apt name.
Doug,
Sorry, my last question was a little vague. I was thinking that your reason for defining consequentialism in terms of a ranking principle, rather than merely a ranking, was to build some substantive content into it. But now I see that’s not right. The substantive content, on your definition, comes from the requirement that the ranking coincide with what the agent has most object-given reason to want.
But now I worry that this makes your definition really quite revisionary. On your definition, it seems, one could be a utilitarian without being a consequentialist. (Either that or one could believe that we ought to maximise total utility without being a utilitarian.) That seems to be a conceptual possibility at least. But isn’t it just a philosophical platitude that utilitarianism is a form of consequentialism?
Campbell,
Yes, my definition is quite revisionary. But you haven’t convinced me that someone can, on my view, be a utilitarian but not a consequentialist. I certainly think that someone can hold that we ought to maximize total utility without being a utilitarian. Of course, this goes back to old disagreements that we’ve had: see the first few posts of this blog.
It’s interesting that you ask, “But isn’t it just a philosophical platitude that utilitarianism is a form of consequentialism?” Dan Jacobson has an excellent paper arguing that this is false. Indeed, he argues that Mill was a utilitarian but not a consequentialist. Unfortunately, it’s not available on the web as far as I can tell, but you might write him and ask to see it if you’re interested.
I do know of the Jacobson paper (though, I confess, I haven’t read it). It came up during my job interview for BGSU. I had just said, ‘Well, of course, everybody thinks that utilitarianism is a form of consequentialism’, when Dan interrupted and said, ‘Actually, I don’t think that, and I have a paper arguing that no one should think that’ (or words to that effect). That kind of stopped me in my tracks!
But I still think it’s a platitude. (One dissenter does not a platitude destroy.) I might be displaying my historical ignorance here, but I have a vague recollection that the person who coined the term ‘consequentialism’ explicitly intended it as name for some larger category of moral view of which utilitarianism was a member. And I defy anyone who has taught intro to ethics to truthfully say that they did not once tell their students that utilitarianism is a form of consequentialism.
I also think it’s a platitude that a utilitarian is someone who believes we ought to maximise utility. I think it would needlessly clutter the ‘conceptual landscape’ if we had to leave space for non-utilitarian views according to which we ought to maximise utility.
But I’d like to hear what others think.
In any case, am I right in thinking that, on your way of defining consequentialism, at least one of these two alleged platitudes must be false?
Campbell,
The term ‘consequentialism’ is due to Anscombe, “Modern moral philosophy.” A quote:
“The denial of any distinction between foreseen and intended consequences, as far as responsibility is concerned, was not made by Sidgwick in developing any one “method of ethics”; he made this important move on behalf of everybody and just on its own account; and I think it plausible to suggest that this move on the part of Sidgwick explains the difference between old‑fashioned Utilitarianism and that consequentialism, as I name it, which marks him and every English academic moral philosopher since him. By it, the kind of consideration which would formerly have been regarded as a temptation, the kind of consideration urged upon men by wives and flattering friends, was given a status by moral philosophers in their theories.”
On the face of it, this doesn’t capture how we now understand the difference(s) between U and C, since both views are skeptical of any distinction between intended and foreseen consequences. And Anscombe is not clear in this quote whether (a) she thinks it’s utilitarians who recognize this distinction, but consequentialists who do not, or vice versa, and (b) if she is conceiving U as a species of the consequentialist genus or not.
Thanks, Michael. I think the quotation confirms what I said: I was indeed only showing my ignorance!
But I have more evidence in support of my claim that it’s a platitude that utilitarianism is a form of consequentialism. With the aid of Google, I had a look at what has been said on the web about the relation between the two views. Below is a list of quotations taken from various webpages, all supporting my claim.
Campbell,
What are the two “platitudes” that you have in mind? Here are three possibilities given what you’ve said above.
(1) “one could [not] be a utilitarian without being a consequentialist”
(2) “one could [not] believe that we ought to maximise total utility without being a utilitarian”
(3) “a utilitarian is someone who believes we ought to maximise utility”
On my way of defining consequentialism, at least one of (1) and (2) is false. But I see no reason for thinking that (2) is a platitude. It seems to me that what makes one a utilitarian is not that one believes that we ought to maximise total utility, but that one takes the principle “Always maximize utility” to be a moral principle that isn’t simply derivative of some more fundamental moral principle. So (3) is true insofar as all utilitarians believe that we ought to maximize utility, but (2) is false insofar as utilitarians are not the only ones who believe that we ought to maximize utility.
Campbell,
You should read Dan’s paper. He would deny that your google search is evidence in support of your claim. He would argue that although ‘consequentialism’ just means whatever philosophers have meant by it, “utilitarianism was a movement in the history of ideas. Hence that appellation must be understood broadly enough to include the views of classical utilitarians” (i.e., Mill). He then argues that Mill was not a consequentialist as philosophers use the term ‘consequentialism’. So it could be that all those people you quote above are just mistaken about what Mill’s view was and, consequently, what utilitarianism is.
It is worth remembering that on Anscombe’s usage Ross counts as a consequentialist; so it is doubtful that her usage is consistent with most contemporary philosophers.
In case anyone cares about this, Cora Diamond has an interesting paper aimed at clarifying Anscombe’s usage. In it she argues that Mill is not an Anscombe-style Consequentialist – her argument hinges on an interpretation of the concept of tendency that shows up in Mill’s discussion of an action’s tendency to produce consequences.
Doug,
If lots of philosphers say that P, and they say it in such a way as to suggest that it’s perfectly obvious and uncontroversial that P, then I’d say that’s pretty good evidence that P is a platitude among philosophers.
Campbell,
It’s evidence that P is a platitude in the sense of being a trite and banal remark, but not in the sense of being a truism (an obvious and undoubted truth).
Sorry, but I can’t help quoting Ambrose Bierce’s Devil’s Dictonary. Wonder if there is an entry in “Quiddities” – my copy is at the office. Anyway, here is Bierce:
PLATITUDE, n.
The fundamental element and special glory of popular literature. A thought that snores in words that smoke. The wisdom of a million fools in the diction of a dullard. A fossil sentiment in artificial rock. A moral without the fable. All that is mortal of a departed truth. A demi-tasse of milk-and-mortality. The Pope’s-nose of a featherless peacock. A jelly-fish withering on the shore of the sea of thought. The cackle surviving the egg. A desiccated epigram.
Doug,
Yes, that’s the sense of ‘platitude’ I had in mind. If a statement is trite and banal, then that’s pretty good evidence that it’s true, especially if it’s just a conceptual claim about the relation of two technical terms in philosophy. But it’s not conclusive evidence. Perhaps Jacobson has revealed an incompatibility between two platitudes: (1) that utilitarianism is a form of consequentialism, and (2) that Mill was a utilitarian. These are incompatible, Jacobson claims, because Mill was not a consequentialist. In that case, some revisionism might be called for. But then, I think, I’d be more inclined to give up (2).
I’m curious now. Why, according to Jacobson, was Mill not a consequentialist?
Doug, another thought:
Earlier you wrote,
This might be where I differ with Jacobson. I think ‘utilitarianism’ also means just whatever philosophers have meant by it. Though what they have meant by it may have changed with time. Perhaps, then, we should distinguish ‘classical utilitarianism’, which means whatever Mill et al meant by it, and ‘contemporary utilitarianism’, which means whatever philosophers nowadays mean by it. My claim is only that contemporary utilitarianism is a form of consequentialism. I say this is a platitude in the philosophical sense; that is, it is one of those claims consistency with which is a central desiderata in any analysis of the terms involved.
According to Dan, Mill wasn’t a utilitarian because he accepted what is nowadays called the self-other asymmetry.
Wouldn’t you say that it is a platitude that Mill accepted utilitarianism, as philosophers nowadays use the term ‘utilitarianism’? Couldn’t I do a google search and get the same sort of evidence for this claim that you got for yours? And isn’t this claim just as much a platitude as the claim that utilitarianism is a form of consequentialism? So if Dan is right in his interpretation of Mill, then you have two platitudes that are equally platitudinous but are logically inconsistent. If that’s the case, I don’t think the fact that your claim is trite and banal is sufficient evidence for thinking that it’s true, especially when there is a contrary claim that is just as trite and banal.
He wasn’t a utilitarian? I assume you meant to say ‘consequentialist’. If the claim is that he wasn’t a utilitarian, then that’s no counterexample to my claim that all utilitarians are consequentialists.
No, I think that’s a substantive claim, which could be false. For one thing, I’m pretty sure that nowadays philosophers use ‘utilitarianism’ to mean a view that rejects the ‘self-other asymmetry’ (though I guess I’m not entirely sure what you mean by this). So if Mill accepted that asymmetry, as Jacobson claims, then he didn’t accept the view that philosophers nowadays call ‘utilitarianism’. Moreover, when philosophers nowadays speak of ‘utilitarianism’ they often have in mind ideas developed in formal areas of study, like utility theory; they’re thinking of ‘utility functions’ and the like. But, presumably, Mill didn’t know about such things.
Campbell,
I’m sorry about the typo. I meant ‘consequentialist’.
On reflection, I agree with you that the claim that Mill accepted utilitarianism (as philosophers nowadays use the term) is a substantive claim and not a platitude in the philosophical sense. Thanks for setting me straight.
Here’s my comment:
I agree that actions have some value, but to someone else. Michael’s final statement, concerning the “intrinsic value” of actions negates his entire argument.
“Modus Ponens” shows that the consequent might justify a persons action. His foreseeing justifies what he did. In the example, someone didn’t give a bum money. What is the consequence, or consequences of the action? Are that someone’s actions justified?
If you are a Bible-believing person, probably not; then again, maybe so. Conceptually, and epistemically, maybe so; then again, maybe not. The argument concerning the consequence of not giving a bum money may depend on your religious, or secular set of maxims.
So I consider that Michael’s problematic might be better served with the solution that causality is a proposition to do the action, and the consequent is the hypothesis, i.e., the theorized result. By not giving a bum money, Michael theorized that something would happen, if that was the giver’s intention, or the giver didn’t have money to give, if that’s possible. Either way, there was a cause, and the cause provided the means to justify the action of not giving. What is the consequent? Who knows; he/she can only theorize what the consequent of not giving to the bum might be.
So I again ask his question and re-propose his proposition: do actions have intrinsic value?
My answer: no. All actions serve a consequence. It’s just the old question still unanswered; so take that!
Pardon the bad grammar.
Love them paranthetical statements.
I agree with Doug, too. I think that his “google search,” produced a virtuoso, epistemologic, proof that shows the consequence of narrow paths of logic. The result is a diametric dilemma: on the one hand you have a logical sequence of “modus ponens,” and on the other you have the restrictiveness of the logical progression. Quaint, but an effective, demonstration of regressionary logic, i.e., closed to the possiblities that are restricted by the psuche, and the limitations of psuche-oriented postulation.
“How now,” says the cow and you would try to convince me that the cow can speak. Back to Michael, and to afnother commentator: if the dilemma concerning an actions “inherent value,” is still a question in contemporary thought, or philosophy, why the confusion concerning the value of an action? Deontological thought belongs in a museum and the evidence shows this necessity.
All actions are consequent-oriented. Consequentialism is the only philosophical theory that works; No other theories do. A doer of an action knows that what he does will have a consequence, but she might not know what the consequence will be. Shelly Kagan wrote: “Intuitively, at least, it seems that many normative factors have a moral significance that is not exhausted by the good or bad results they involve” (NORMATIVE ETHICS 70), but is it harder to see the intuitive possibility, or to see the consequences of previous actions spur further action; thus consequent, then again action, ad redundatum? I’m sure you’ve all heard this latter point before.
Consequentialism is not narrow; it fits all the epistemic requirements for a proof-positive truth. Any order of proofs show a consequent; nobody can do an action and consider the action to have intrinsic value; because nobody, not God, nor an angel, or Satan, demon, or any man has ever done an action without a consequent result.
Michael asked about the intrinsic value inherent in “not giving a bum money,” and so who wouldn’t ask if there is any value in that action. There doesn’t seem to be any value to “not giving a bum money” from a teleological perspective; there could be some from a utilitarian perspective; not giving may be a result of a normative philosophy; it could be a hedonistic pleasure for some churlish person, but where’s the deontological value? Any one of several philosophical approaches could show value derived from the act of “not giving a bum money,” but I fail to see anything intrinsically valuable in this act. I don’t see any proof showing there is no consequent to the act, and I continue to suppose that all acts have only instrumental value. Take, again, the example of love: the concept of love would appear to have intrinsic value, but only when a being, God or man, works an action with love inherent in the action does the value connected to love issue out. Love, without God’s, or a person’s, application of the notion, is valueless in itself. Love applied is valuable.
Thus I render this example to show that “actions [can be] more important than belief” (Ode Lunardi – Language Major in a Mexico University) even though belief is a result of the application of faith and may be associated with an ontological train of inquiry, but who would say that the action of believing does not produce result(s). I pose one more question: what value does anyone associate with believing in love?
I finally theorize, and believe, that a person should disaccociate actions, and I’ve seen this from someone before, from entities, i.e., God, man, demons, any subject that is an entity, a noun so to speak. I’ve heard someone say that he can place instrinsic value on/in an entity, but he can’t place intrinsic value on/in an action.
An action, though I may refer to an action in the same way I would refer to a noun, is concerned with, to use a linguistic comparison, verbs. What is a verb? Any word that describes an action. What is an action? Any thing done by an entity. Where does the value go? To the entity or to the action? You answer this for yourselves.
Sorry, I have to correct something I wrote: I wrote, “The concept of love would appear to have intrinsic value, but only when a being, God or man, works an action with love inherent in the action does the value connected with love issue out.”
Actually I meant this: “The concept of love would appear to have intrinsic value, but only when a being, for example God or your mother, works out an action of love does the value connected with love issue out.”
Chai and “God speed.”
Chai, a tea in Africa, tastes good.
Modus ponens:
1. I think chai tastes good
2. Chai has to be a tea from Africa in order for it to taste good.
3. Chai tastes good.
Chai, a tea in Africa, tastes good.
Modus ponens:
1. I think chai tastes good
2. Chai has to be a tea from Africa in order for it to taste good.
3. Chai tastes good.
The dialectic of philosophy is that all philosophies have a polar appproach that the hegelian analysis would purport to be the only proof-positive circumstance by which any truth could be reached. Only by understanding bi-poloar opposites can one truly comprehend a synthesis to opposition. The gist is in the fact that there has to be opposition for there to be a synthesis. Because this reality exists, in the universe, one can conclude that there is an ultimate synthesis to every question that is asked in genuine search for the truth.
I think; therefore I record my own writing.