Consequentialism came into the conversation at my debate with David Smalley (Sunday, August 9). Only a short critique of this theory was possible at that time. I want to express more clearly what this theory is and why it’s a terrible option.
Consequentialism is a broad term for any ethical system that judges the morality of actions solely by their consequences. The term was first coined by Elizabeth Anscombe in “Modern Moral Philosophy” (1958) as a catch-all for the different brands of utilitarianism such as Jeremy Bentham’s quantitative model, J.S. Mill’s qualitative model, and G.E. Moore’s rule utilitarianism. Anscombe, at the time, was criticizing consequentialism and Kantian alternatives for being too action-based and impersonal–as opposed to character based, and personal. She is considered a herald of the oncoming Virtue Ethics revival signaled in large part by Alisdair Macintyre’s After Virtue (1981). All of these broad categories for ethics, still apply today: consequentialism, Kantianism (Kantian Ethics), and Virtue Ethics–the latter of these being, in many ways, a critique of consequentialism.
But why would Kantian ethics and Virtue ethics arise as critiques of consequentialism? What’s wrong with consequentialism?
With some nods to Kantian ethics, and some nods to Virtue ethics, the following critique of consequentialism is still, largely, an independent attack on this deeply flawed system. One need not ascribe to Kantian ethics, or Virtue ethics, or any other particular system to be able to accept the critique that follows. But first, some positives should be noted, lest we err in special pleading.
- Consequentialism allows for a relatively simple, largely practical model of ethics–rather than ascribing, for example, to a religious ethical system which has different laws for different eras (Old Testament and New Testament), different and conflicting accounts for ethical grounding (Divine Command Theory vs. Ethical Essentialism), different epistemologies (Natural law vs. Special Revelation).
- Consequentialism does not specify or explicitly require any God or supernature.
- Consequentialism, in the form of utilitarianism, treats people as broadly equal
- The ends do seem to possess some moral weight.
- Consequentialism, in the form of utilitarianism, identifies pleasure within goodness so that even animals can be ethically valuable since they can feel pleasure and pain.
- In some cases, the ends do seem to justify the means–such as torturing prisoners to avert a terrorist strike.
Now, there may be other benefits not named here that should be included in this list. But this sample should suffice to show how consequentialism can be attractive to novice and advanced ethicists. However, when the drawbacks are also weighed, it should be obvious whether consequentialism is found wanting. Mene mene tekel upharsin.
- Consequentialism allows people to be treated as “means to an end.”
This was a key critique from Immanuel Kant. Kant’s deontological ethic proposed that people are ethical subjects–ends in themselves–not to be treated as objects which some social engineer can rightly manipulate for the “greater good.” Consequentialism, at least in it’s figurehead system of utilitarianism, locates “the good” not in people, or life, or or human nature, but in pleasure. Since people are distinct from and separable from pleasure, people are not intrinsically valuable. People can be means or ends. When people are treated as means some horrible stuff can happen. Consequentialism was a key component in justifying the U.S. eugenics legislation at the turn of the 20th century, and in the 3rd Reich in Germany in WWII, and again in Apartheid South Africa.
- The ends do not justify the means.
As David Smalley admitted in the course of the debate last Sunday, the ends do not justify the means. There is a qualification to be made here. Consequentialism does not require dilemma contexts before using something (including people) as a “means” to an end. If one man is having some low-level of pleasure but is otherwise pretty sad and boring but dozens of people can experience great pleasure over that person’s death, then consequentialism can be used to justify killing him. The ethically exalted “ends,” in this case, greater pleasure for the most people, can lord over anyone who is standing in the way of that greater good. Remember that consequentialism does not explicitly require dilemma contexts, where the only options available are, for example murder vs. assault-and-battery. One need not be pitted between two apparent evils before having ethical permission to do what would otherwise seem to be “evil” (like assault-and-battery). For example, a man is attempting murder but will cease and desist if a passerby were to punch him a few times. Both the consequentialist, and non-consequentialist can agree that this thug should get punched. But the consequentialist can go much further and affirm that another assault when there’s no pressing evil to prevent but just some wildly entertaing fun for the masses. Greater good was had by many, so the means was justified–right?!
- Too few “good” elements
Consequentialism suffers from too scarce a diet of goods. Besides the “ends” (consequences, results, outcomes, etc.) there are also motives, agents, character, virtue, actions, means, manner, and circumstances that can all lend goodness to a moral event.
- Its general norms admit too many exceptions and conflicts to be effective as norms.
Consequentialism is what some have called a “generalist” theory, identifying ethical goods by a “general rule/norm.” Bentham called this rule the “rule of utility,” i.e., “act so as to achieve the greatest pleasure for the most people.” Now one does not have to run this rule across too many scenarios before finding some hard cases which reveal that “pleasure” is only sometimes closely identified with “good,” and people can enjoy things which intuitively or by common consensus are deemed evil. The many exceptions to this general rule compromise its force as a rule.
- Utilitarian acts have no intrinsic value
Another strange consequence of consequentialism is that actions have no intrinsic value so failed attempts and inconsequential acts are not good, hollow. Yet we know that a failed attempt to rescue someone from drowning is still a good thing.
- It’s too Impersonal
Utilitarianism, the sole example of consequentialism to achieve notoriety, was first formulated for political applications by Jeremy Bentham in the 18th century. He seemed to have in mind a strong sense of justice, and that tends to mean equal treatment before the law. Stated another way, the effort is to be impersonal and unbiased. In some ways this effort to treat everyone equally makes great sense, especially when the context is jury trials for example. However, there is an odd drawback to this model which has proven to be a painful failing for utilitarianism. Ethics demands some level of personal preference, or subjective and unequal treatment. In other words, a parent should give priority to his own child. You or I would never be able to go out to a movie, or go to a concert, since that money or time could have been used serving in a soup kitchen somewhere or donating to starving children in a 3rd world country. I should not feed my kids 3 square meals, but instead my kids and I should all split 1 meal a day so that the money saved can be used for children who wouldn’t get any meals today. Now, I’m all for charity and self-sacrifice. But consequentialism has no built-in rubric for identifying rightful levels of duty to different kinds of people–my family first, friends second, strangers third, etc.–since all people are to be treated equal. That equality makes great sense politically, but would be horribly ill-suited on smaller scale. Yet even at the political level, there can be rightful demarcations between which people a government should care for most–namely, it’s own citizens. If the U.S. government protects it’s own citizens by military and police forces, but don’t owe the same manner or degree of protection to non-citizens. It’s not that the Government should treat non-citizens as if they have no value, but rather that there are practical limits to what time, energy, and resources can be spent and a government’s duty (i.e., akin to Kantian deontology) is to it’s own citizens first.
- Consequentialism Fails to Avoid Absolutism
David Smalley makes clear in his book, Baptized Atheist, that he (at the time of writing) rejects absolutist ethical systems. And it seems that consequentialism offers an escape from “absolutism.” Unfortunately, consequentialism still ends up introducing an ethical foundation, an “absolute” rule that ends up determining all the ethical choices. Consequentialism avoids some kinds of absolutism, for example, Kantian ethics or Divine Command Theory, but it fails to distinguish itself from absolutism generally since it effectively replaces one set of rules for another. Absolutes abound in consequentialism such as the rule of utility, the Bentham’s felicity calculus, hedonism (pleasure-focus), or the “ends justify the means.” And when we consider G.E. Moore’s rule utilitarianism, a bevy of duties and absolutes are introduced.
- Discriminates Against Minorities
Since the rule for consequentialism (at least in it’s utilitarian forms) is to generate the “greatest good for the most people,” it inherently discriminates against minority groups. For example, one can justify oppressing or killing minority groups, or dissenters so long there are more people experiencing pleasure than are not.
- Absolutes are needed to discern “better” and “worse” ethical options.
We’ve already seen how utilitarians fail to escape absolutist rules, but they also fail to escape the need for absolute reference points besides their bevy or rules and principles. There’s no way to tell what is the “greater good” without knowing what is the greatest good, but a perfect standard of good would itself be an absolute.
- “End” is ambiguous
Consequentialism treats ends with special interest, since those outcomes/results/effects are supposedly the good-making property for all of ethics. Yet this category of “ends” is woefully ambiguous. Suppose, for example, you were to help an old lady across the street. Normally we would consider that action “good.” It was pleasing to you for helping, to her for being helped, and to the drivers for having a capable assistant to keep feeble old ladies from causing car accidents. But you may have also served the “end” of achieving a merit badge in boy scouts–the “community service” merit badge. It may have also served a different end; she waits for young men to help her walk around because she likes to caress their arms as they walk; she gets a devilish little thrill out of it and most of her helpers don’t even realize it. Still another end was achieved; she survived another day and so her drug dealing nephew didn’t get discovered living in her basement where she has been aiding and abetting him so she can pay her rent. Now, when you help the lady cross the street, one may think that act was “good,” but it’s difficult to render a judgment since it’s results vary widely from a pleasant gesture, to a perverted pathology, and even contributing to criminal activity. Surely this odd example reveals how we need a wide range of discrminating criteria besides merely “outcomes,” since character, virtue, intention, foresight, knowledge, etc. are evidently good-making properties or good-making circumstances. Consequentialism has a difficult time expanding to include these other things (besides outcomes) since each concession beyond mere “outcomes” makes the system less and less consequentialist.
- How long is the long run?
Further illustrating the ambiguity of “ends” is a temporal question: How long is the long run? Ethical events, for the consequentialist, are to be judged by their outcomes, but how far must one forecast into the future in judging the morality of an act? Helping that old lady across the street is “good” insofar as you have helped her avert a car-accident. But if she, as a results, survives for several more years.
- Suspended “good”
Consequentialism also has the odd result of suspending ethical evaluations till the “ends” arrive. If you donate money to a reputable charity by sending the check in the mail, you may intend to achieve good, act to achieve good, and there are many people and circumstances participating incidentally to help accomplish that potentially good act–but that moral event is neither good nor evil unless and until the “results are in” (whatever that stopping point may be; see #11 above).
- Character Is Good
One of the strengths of Virtue ethics is that it rightly challenges other systems which locate the good entirely in behaviors. Virtue ethics allows for some measure of goodness in behaviors but it goes far deeper to identify goodness, also, in the virtues and accumulated character which direct one’s actions. Consequentialism has no robust or thoroughgoing categories to account for the goodness of character. Consequentialism treats ethics in terms of public actions. Consequentialists cannot, within their system, treat the intentions, virtues and character that motivate an action as “good” since intentions and virtues and character are causes but consequences are effects.
- Consequentialism puts the locus of ethics beyond our reach
By locating goodness in the outcomes, consequentialism has set our ethical task in the future. Yet no matter how well we may plan, or how reasonable our estimations may be, there is always an element of mystery to the future. Unforeseeable events control the real ethics of any given situation, since we can plan, and work, and act with the best knowledge available, yet a hurricane or a traffic accident, or a death in the family, ruin it all for us. We do not hold the future, but we can live and act in the present and we can control our motivations, our thoughts, and our actions now. If we are to be ethically responsible for our actions it would only make sense for that ethical component to those actions to be at least somehow within our control. Every action eventuates a train of consequences, and we cannot fully predict where that train may lead, or what harm it might cause on the way. It would only make sense then to identify ethical duties to the extent that we have control over that train of actions.
- Pleasure can be bad and pain can be good.
Utilitarianism ascribes goodness to pleasure, specifically, “the greatest pleasure for the most people.” Consequentialism could, potentially, expand or replace that hedonistic framing with other things like “justice,” or “love.” But so far, the only consequentialist schemas I’ve yet encountered are hedonistic, framing goodness strictly in terms of pleasure.
- The Utilitarian fallacy
Perhaps the most glaring fault of utilitarianism is what’s called the “Utilitarian Fallacy.” Put simply, utilitarianism can (and has) been used to justify all sorts of known evils. Rather than using that calculus to “show” that those ends are good; those reprehensible ends can be used to show that the calculus is bad. Many comic book villains, or action hero villains, demonstrate the cold calculated logic of this fallacy–once people are means, instead of ends in themselves, a person can justify most any slaughter so long as some projected future goal is thought to be good enough to justify it.
- Consequentialism Struggles with Intrinsic goods
I asked David at one point what he thought was “good.” This is both a simple and a hard question, since it can be answered either with examples of good things (as David did), and there are plenty of easily found examples of “good” that he and I agree on. Or it can be answered with a precising definition, seeking to get to the heart of this difficult topic. One of the more compelling answers to this sublime question is from Aristotle who says that “the good” is intrinsic value, that is, something to be sought for it’s own sake. We seek food for nourishment and flavor, we seek marriage for companionship and love, we seek a television for viewing pleasure–each of these things are sought for the sake of some other use they serve. But we seek goodness for no other reason than that it is good. Goodness is it’s own value. Value is not extrinsically imposed on goodness, goodness is intrinsically valuable in it’s own right. It does not drawit’s worth from how useful it is for some other purpose. No, goodness is itself the end to which all other things are “aimed.” Consequentialism struggles in this regard because it’s operating categories of “utility,” “usefulness,” and “consequences,” expose it’s extrinsic approach to goodness. Consequentialism is metaphysically weak, lacking any robust built-in justification for why any particular thing is ultimately good. It’s goods are only extrinsic, where the “goodness” is it’s usefulness toward some other goal. Sure, it has an operating rule–the rule of utility–and that lends some sense of grounding and reliability, but that rule deals only in instrumental/extrinsic values; “X is good if it achieves pleasurable outcomes”–but what then makes pleasure good? As we said above, some pleasures can be bad (such as “enjoying the torture of babies”) and some pains can be good (such as painful alerts warning of danger). Here, Utilitarianism dwindles down to old-fashioned hedonism, and hedonism has been soundly rebuked as early as Aristotle’s Nichomachean Ethics. In summary, utilitarianism is too extrinsic, treating all goods like they are only extrinsically traced to the various ways people experience pleasure, when instead there are many “goods” abiding at the level of character and beneath any public actions.
- Consequentialism Suffers Logical contradictions
The same event can be “good” and “bad” in the same sense depending on whether the immediate results or the distant results are considered. It does not make sense for the same historical event to be equally good and bad, overall, and change back and forth between “good” and “bad” as different and distal effects occur.
- Maximizing the pleasure for the most people is overly demanding.
Besides the “impersonal” problem above (#6), the rubric of consequentialism loses simplicity when faced with real world decisions. One could never go to a movie or eat out because that money could always have gone to meet someone else’s more basic needs. Applied consistently, one can hardly enjoy a single day since someone else is hurting worse than you, and he or she needs your resources more than you do. One cannot rightly prize family above strangers, since feeding 12 children in New Guinea might be cheaper than feeding two of your own children here. Unless some qualifications are added, utilitarianism demands we feed the twelve instead.
- Consequentialism Suffers the Naturalistic Fallacy
Consequentialism, to the extent that it is a naturalistic approach to ethics, suffers the yet unresolved “is-ought” problem, AKA: the naturalistic fallacy. In short, we cannot assume that natural facts give rise to moral facts, “Is” can generate “oughts.” Now both categories do seem to exist, natural and moral facts. But naturalists have a bridging problem since nature is ultimately non-mental and even human minds would, at best, give rise to culture-relative ethics (i.e., conventionalism). Meanwhile, we take it that it is a fact (not just a preference, or convention) that “rape is evil.” There does not yet seem to be a sound solution to how one bridges from non-mental, undesigned, naturalistically-defined “nature” into the transcendent realm of “morality.”
- Consequentialism has no real metaethic
Consequentialism is, in some ways, only the veneer of an ethical system. This is not an island, cropping up from miles of grounding underneath. Consequentialism is a dingy on the ethical ocean. In many cases, if not most cases, consequentialists cannot espouse consequentialism as the deep or complete summary of their ethics. To do that one must presume a definition of “good” from the outset, usually hedonism (“good is pleasure”) yet that very admission reveals that there’s another whole ethical system underneath consequentialism. And to the extent that that system is flawed or inadequate, consequentialism may likewise suffer. Hedonism, for example, suffers the glaring problem of how some pleasure are manifestly evil, and some pains do not seem closely tied to suffering or evil. This problem can, potentially, be averted if one supplements consequentialism with a more robust metaethic from some other system, but so far, whenever that supplement is naturalistic it suffers from the naturalistic fallacy and other problems mentioned above.
- In consequentialism one can’t do “goodness for goodness’ sake” since good’s are accidental not intrinsic.
Because of the extrinsic-intrinsic problem above, utilitarianism, if true, would make it impossible to do “good for goodness’ sake.” All goods are found in the outcomes not in anything intrinsically. For consequentialists (and against, utilitarianism is the only example yet presented of consequentialism), goodness is always only an accident of outcome never an intrinsic property of any character trait, or action. One cannot do justice because justice is good, one can only do justice–which is, at the time, neutral–because it might bring about some foreseen pleasure in the future.
- Utilitarianism is Relativistic
Because of the extrinsic manner, and the subjectivity of pleasure, utilitarianism comingles with relativism. In at least some cases, utilitarianism is nothing but a form of relativism. But there’s a reason that the majority of atheists in philosophical ethics are not relativists. It’s because relativism is pretty abhorrent when pressed to its logical extremes. cultural relativism discriminates against minority ethical systems and against moral revolutionaries like Martin Luther King or Ghandi. Relativism trivializes all morality, taking it out of the realm of objective knowledge and factual stuff, and placing it instead into the category of opinions, instincts, and sentiments. Individualistic forms of relativism have no ultimate arbiters between people who morally disagree. And cultural relativism can’t help arbitrate between feuding cultures either. Relativism also attempts to prove a negative, claiming to know (within a reasoanble estimation) that there are no absolutes even though no relativist alive has studied every single claimed “moral absolute.” cultural relativism ends up making “the good” into a “might makes rightt” or “group vote,” despite the fact that we know, from history and experience, that the majority can be wrong, and the stronger party can be evil. Relativism can also be used to justify most anything. And lastly, we know better than relativism. We know that some things are just plain wrong, no matter what the consequentialist or relativistic rubric may have us think.
Phew! It feels good to get all that off my chest. Ethics is messy stuff and there is no ethical system out there that is free of problems. But hopefully you can see from this list that consequentialism is a deeply problematic system that looks like it would need too much “patching up” to be worth our purchase.
5 thoughts on “The Evil Outcomes of Consequentialism”
Interesting read! I don’t agree with much of it, but I thought it was well-discussed. There are a few of your points I would like to specifically hit:
1. I don’t see a problem with this, so long as people end up getting treated better than they otherwise would have. Imagine you are going to be placed into one of two societies. The first one will treat you as an end rather than a means to an end, but you’ll be treated badly. The second will treat you as a means to an end, but you’ll have a greater likelihood of being treated well. Which would you pick? I think most would pick the second. In fact, in these scenarios, I think “Which society will give me the greatest chance of being treated well?” is the only question we would care about.
2. Sometimes the ends do justify the means. If you shoot an armed nighttime intruder to protect your family, what will be your justification for your actions? It will probably be something to the effect of “If I hadn’t shot him, he could have killed us all.” That is pointing to the ends to justify the means. It seems very odd to me to say that the outcome of our actions has no bearing on the justification of our actions.
3. One can be a consequentialist about any of these things. One could be a virtue consequentialist, holding the view that whatever actions maximize virtue are good actions. Consequentialism is simply a view about how we should evaluate actions. It doesn’t specify any specific ingredient.
5. This is an oversimplification of utilitarianism. It is perfectly consistent with many forms of utilitarianism to say that we should take actions that produce the greatest expected value of X, where X is the good we are pursuing. Trying to save a drowning child has a high expected value of good, even if we only have a 50% chance of being successful. A 50% chance of acheiving a great good is a worthy pursuit.
6. Our aversion to this sort of impersonal view of morality is likely a product of our own evolutionary history, nothing more. It doesn’t seem obvious to me that governments should value their own citizens over other citizens. Rather, this simply sounds like the sort of “in group/out group” choices our evolutionary ancestors made for protection.
8. Consequentialism doesn’t simply say that any action that is better than any other alternative is justified. Killing all of the minorities isn’t justified simply because it would be better than killing all of the majorities. Rather, the act of killing all minorities must be evaluated in both a counterfactual sense (What if the minorities hadn’t been killed?) and a best substitute sense (What other actions could have been taken rather than killing the minorities?). Killing all minorities would, in the vast majority of scenarios, fail these tests.
11. We face a great deal of uncertainty when projecting outcomes into the far future. This uncertainty probably lets us off the hook for “down the chain” reactions that happen many years into the future. If I help an old lady across the street, the foreseeable aspects of my actions contain more good than bad. It’s possible that she will go on to save a child from drowning, and that child will grow up to become a horrible terrorist who kills thousands of people. But that outcome isn’t foreseeable from my epistemic vantage point, and the weighted average value of all possible worlds where I help the lady is almost certainly greater than the weighted average value of possible worlds where I let her get hit by a car.
12. See my critique from #11. Many forms of consequentialism allow us to evaluate behaviors based on the expected value of the outcomes.
Zack, thank you for your feedback. I’ll try to address your thoughtful responses. I appreciate the effort put into a thorough assessment.
“1. I don’t see a problem with this [treating people as means instead of ends], so long as people end up getting treated better than they otherwise would have.”
It sounds like you just affirmed slavery, so long as people were rescued from an overall worse situation, say between warring tribes in Africa, into a society with more upward mobility and a higher standard of living than before. If they are “treated better than they otherwise would have been” then that still doesn’t justify doing unnecessary evil to them. You’d do better to appeal to a dilemma context, where a person has ONLY BAD OPTIONS to choose from (ex., prison camp versus indentured servitude). I’d suggest dilemma contexts can make many other ethical systems appear to be similar to consequentialism when they, in fact, are principally different.
“2. Sometimes the ends do justify the means. If you shoot an armed nighttime intruder to protect your family,…”
You gave an example of a dilemma scenario, i.e., kill or be killed. Just because many ethical systems can agree on a choice in a given scenario doesn’t mean they agree on the REASONS or the METAETHICS behind that choice. To put it another way, many ethical systems can agree that the “ends” have some moral weight, but they do not agree with consequentialism that the “ends” have ALL the moral weight. That’s what consequentialists need to prove, and so far, have not been able to prove that very well.
“3. One can be a consequentialist about any of these things.”
If you agree with me that non-consequences can serve the same role as consequences, then I don’t know what you mean by consequentialism. Your response suggests to me you don’t understand this critique, or else you are working on a different definition of “consequentialism” than I am. Perhaps I can clarify. I’m suggesting there are things, besides the “ends/consequences” which can and should impact our moral calculations. This is fundamentally antiethical to consequentialism. If one allows that non-consequences can be good-making properties then it’s no longer consequentialism. Perhaps we are equivocating on the word “consequentialism.”
“5. [the notion that ‘Utilitarian acts have no intrinsic value’] is an oversimplification of utilitarianism”
As I use the term “utiltiarian acts” i’m referring to some action performed within a moral event, where the “good-making property” for that act is located entirely in the outcomes (‘the ends”). If you disagree with this claim, it sounds like you are saying that something other than the outcomes can serve as a good-making property. If so, I think you might not be talking about utilitarianism. You might be talking about an accidental virtue ethicist, or some other school of thought where the ends, alone, do not justify the means. Remember, most every ethical system admits some moral weight to the outcomes, we just don’t grant that that outcomes are the only good-making property. In this way, I tend to suspect that many of the nuanced and qualified versions of consequentialism, aren’t really consequentialism at heart.
“6. Our aversion to this sort of impersonal view of morality is likely a product of our own evolutionary history, nothing more.”
This sounds like a reductive fallacy (a.k.a., “nothing buttery” fallacy). to say it’s “nothing but” some evolutionary accident is virtually impossible to prove, and amounts to a universal negative. You’d be safer to hedge your statement a little bit instead of putting confidence in, what appears to be, evolutionary ethics/social evolution. I’d suggest evolution has nothing whatsoever to tell us about morality, even if I thought some Darwinian or neo-darwinian evolutionary model were true. meanwhile, in philosophical ethics, it’s a stiff and powerful critique to identify where ethics is cold, robotic, and impersonal. That’s the implicit critique of consequentialist thinking the doomsday/dystopian scenarios we see in, for example, Age of Ultron, Terminator, Matrix, etc. When people are treated as objects and not subjects, we see some of the worst atrocities in human history. If you’ve read much about the Nazi rationalizations for the “final solution,” or the “tuskegee experiments,” or the eugenics projects in the early 20th century U.S. then you’ve seen how it’s a bad precedent to treat people as objects instead of subjects. this is also why Kant distilled this principle as the core of ethics, and it’s widely accepted and admitted across most ethical systems – people should not be objectified.
“8. Consequentialism doesn’t simply say that any action that is better than any other alternative is justified.”
Again, it sounds like you are not thinking consistently with consequentialism. with utilitarianism, the central guiding point is “greatest pleasure for the most people.” I don’t see how your explanation shows that your view of consequentialism fits with utilitarianism’s core principle. I don’t think you’ve effectively rebutted this point. And I’m not sure you understand it. You seem to suggest that alternative possibilities need to be considered. But, let’s say there is one scenario where 90% of the population can have 100 more units of pleasure, per person, by reducing the well-being of the remaining 10% by 100 units each. We’re talking a 80,000 unit net gain in pleasure. Now, I’m not saying utilitarianism requires that we take that action, but it at least permits it. And in democratically run and populists systems, we know that the majority often rules in these sorts of things whether that be slavery, sexism, trans/homophobia, etc. I don’t see why we’d need to appeal to to a dilemma context to identify if a given action is morally permitted within utilitarianism.
“11. We face a great deal of uncertainty when projecting outcomes into the far future.”
That’s my point. And that’s good reason to keep utilitarians on the hook. Since outcomes are future-oriented, utilitarianism is flawed for basing it’s ethical evaluation in something where, at best, we are only making educated guesses. Ethics involves too much guess-work already, it seems even more precarious when we realize that immediate outcomes aren’t certain, but utiltiarians still identify the core good-making property in something that’s subject to arbitrary and fickle fortune (i.e., the future). Furthermore, you didn’t really address the heart of my point. A given moral action can be “good” in the short run, “bad” in the mid-term,” “really bad” in the long run, but “really really good” in the extra-long run. Supposing that we are able to foresee (reasonably so) each of these stages, it seems a bit arbitrary to drop anchor at any one point and say “This stage of outcomes is what defines whether that action was good.” That’s hard to swallow because these different outcomes contradict each other ethically.
12. . . .
In light of my last response, and your reversion to comments on 11., I’d suggest you haven’t really addressed the core contention here. Expected outcomes are not outcomes. This may be a subtle deviation away from consequentialism. meanwhile, I reiterate my point from 11. above.
1. The slavery scenario IS a dilemma context. And you’re right: I do think slavery is the preferable option, at least in the hypothetical scenario presented here (I have no idea if it is an historic fact that slaves would have been worse off in Africa. I doubt it.). I think you are undervaluing the weight of what it means to say that a slave would have actually been better off as a slave than a free person in Africa. That assessment must include the subjective experience of slavery, not just superficial, objective assessments of the situation. If it is actually true that the slave, by his own subjective assessment, was better off as a slave than as a free African, why are you convinced he is wrong? Or, perhaps more accurately, why do you not think it is justifiable to give such a person what he or she prefers? Imagine there is a magic burglar who knows all future-facing propositions. In fact, he knows that if he were to rob my house today, I would go down to the police station to report the crime. He also knows that, on the way to the police station, I would meet the love of my life. Meeting her would improve my life immensely. What you are saying is that the all-knowing burglar should not rob my house, even though doing so would bring the greatest improvement to my happiness.
2. Consequentialism doesn’t say that the ends have all the moral weight. That’s my point. “The ends justify the means” is a misunderstanding. People should instead understand consequentialism as saying “The ends, when weighed against the means, determine the moral status of an action.” To turn the question around, non-consequentialists must answer why we should sometimes take actions that produce a world with less of what we desire and more of what we don’t desire.
3. I should have been clearer. Consequentialism is simply a view about how we should do the “moral math.” It says nothing about what things we should seek to maximize. Specific views within consequentialsim (such as utilitarianism) do, but consequentialism in general does not. We could devise a consequentialist system that seeks to maximize specific actions, certain values, etc. We could be a consequentialist about art, whereby we seek to maximize the amount of art produced.
5. Our disagreement is about why non-good producing acts are still good in certain scenarios (such as a failed rescue attempt). No, consequentialism would not say that such acts are INTRINSICALLY good. But I don’t see why that is bad. Consequentialism would still say failed rescue attempts are generally good because they represent a possibility of producing a major good, such as saving a life. Intuitively, I don’t have a problem with that.
6. You are suggesting that a moral theory that is impersonal is probably incorrect, and I understand you to be pointing to our intuitions as evidence. At least, that’s how I understood your argument. If it is indeed an argument about our intuitions, then evolutionary psychology certainly cannot be ignored. Keep in mind, I am referring to what I see as likely, not what is necessary. I am not saying it is necessarily the case that there are no other explanations for such an intuition. Rather, I am saying that, based on what we know about evolutionary psychology and the importance of pre-human primate in-group/out-group distinctions, it seems likely that we hold these intuitions due to our evolutionary past. That at least should reduce the importance we give to our disdain for impersonal moral explanations.
The fact that people have committed unjustified actions due to a poor understanding of impersonal ethics is hardly a critique of impersonal ethical systems. Perhaps it gives us good reason to not try to convince our philosophically ignorant politicians to to be consequentialists, but it isn’t an argument against consequentialism.
8. I don’t think you understood my critique. You are using dilemma scenarios, and my point is that utilitarianism seeks to maximize the GREATEST good for the greatest number of people, not simply greater good than some other option. In a dilemma scenario, such as the one you presented where we can gain 80,000 units, sure, utilitarianism (in a crude, non-rule form) would have us prefer the action that increases happiness by 80,000 units IF those are the only two options. But my point is that this doesn’t then give us license to perform any action that, in isolation, increases happiness. There are likely many other alternatives that would increase happiness even more (such as increasing everyone’s happiness by 99 units. There are further ramifications of our actions that might outweigh short-term happiness increases (such as civil war or general social unrest). In short, utilitarianism (in a crude, non-rule form) can give us some unpleasant results in dilemma scenarios that don’t always persist in the real world, where we have more than two options.
One last thought on minorities and consequentialism: So far, you have been assessing only total utilitarianism. But there are many other forms, such as average utilitarianism. Maybe we should seek to make the average person as happy as possible rather than increase total happiness. One could also be a 10th percentile utilitarian or even a 1st percentile utilitarian, wherein we would seek to make the bottom 1% as happy as possible. It is even possible to imagine a system where one could be a Rawlsian and a utilitarian. The point here is that there are a lot of debates within utilitarianism that split along some of the issues you are presenting, but total utilitarianism is by no means a given.
11. In your scenario, we should pick the option that produces the greatest overall amount of good, ignoring total/avg/etc. utilitarian considerations. If the action is “really bad” in the furthest timeline, and that “really bad” is bad enough to outweigh the short-term good, then it’s easy: Don’t do the action. If the short-term produces 100 units, the mid-term produces -50 units and the long-term produces -200 units, the action would produce -150 units. There’s no contradiction. I guess I also don’t get why it is bad to say that we should still consider the future when making moral decisions. We do this all the time in other arenas of life. You don’t know what will happen when you get in your car and drive to work, for example. You might be hit and killed, or your car might be stolen and used in a terrorist attack. But based on your best knowledge of the likelihood of outcomes, the benefit of you going to work via car outweighs the risk of not doing so. If that is sufficient for practical rationality, why is it not sufficient in moral decision making? To be clear, I am not saying we shouldn’t adjust for such uncertainties. But the existence of uncertainty doesn’t seem to me to be reason to ignore the future effects of our actions.
12. I think the vast majority of consequentialists would endorse expected value decision making. Imagine I agree to haul a group of passengers down a mountainside road in a car that I know has faulty brakes. There is a 90% chance the brakes will fail and cause fatalities, but on this particular day, we make it down safely. Consequentialism doesn’t commit us to the idea that my act was a morally justifiable act simply because the outcome was positive. The expected value of my outcome was terrible, and I (along with my passengers) simply got lucky. A person who behaved like I behaved would not, over the long-run, maximize happiness and minimize suffering. Insofar as we know the likelihood of the various outcomes of our actions, we should evaluate the value of a given act based on the expected value of its outcome, not the actual outcome when all is said and done.
15. There are other forms of utilitarianism, such as preference utilitarianism. But I’m curious: Could you give me an example of a case where pleasure is bad? To clarify, I’m not asking for a scenario in which the way pleasure is being obtained is bad. I’m asking for a scenario in which the pleasure itself is bad (or conversely, the suffering itself is good)?
16. This is a misuse of consequentialism. It has as much teeth as “Religion can cause violence, so religion is bad.” Simply because people have been bad consequentialists doesn’t mean consequentialism as a whole is bad. Consequentialism still permits moral agents to consider epistemic limitations, and accepting current atrocities for far future goods is often a fool’s wager due to the inherent difficulty of understanding the far future ramificiations of our present actions.
18. This is a misrepresentation of consequentialism. All that matters is the net balance of all consequences produced by a given action. Of course, it is very difficult to foresee all consequences, but that doesn’t mean a consequentialist will believe a given action is both good and bad. Rather, there might be epistemic uncertainty regarding the status of a given action.
20 and 21. You are confusing meta-ethics with moral theories here. Consequentialism is a moral theory. It is possible for two consequentialists to have different views about meta-ethics.
This was a long response already, so I’ll leave it at that.