Means/ends free-for-all

Yesterday I made an assertion that I took to be fairly uncontroversial. I said that most ethical systems acknowledge that ends can sometimes justify means. By "ethical systems" I mean both philosophical schools of ethics and cultural norms.

What is it for an end to justify a means to that end? That's a complicated question that could be interpreted in many different ways. However, this discussion is about whether it is ever acceptable to do something morally wrong for the sake of a significantly greater good. Note about terminology: I'm being deliberately vague in my descriptions of the relevant moral concepts I'm invoking. Terms like "acceptable"/"justified"/"right"/"permissible" have precise technical meanings within philosophical discourses. However, for the sake of this discussion, I want to be as inclusive as possible so as to highlight certain fundamental similarities between various schools of thought.

Note about consensus: Obviously, ethics isn't a democracy and consensus isn't justification. However, analysis has to start somewhere in ethics. One way is to systematically provoke your moral intuitions and attempt to create a rational framework that fits them best. Another approach is to start with your preexisting ethical concepts and analyze them to discover the full implications of those ideas (i.e. concepts like good, right, justification etc.), still another method is to observe the real-life practices that we recognize as ethical and attempt to discern what these examples have in common. Whatever methods you choose, consensus is often a helpful starting point. If you want to understand the general principles of right and wrong, it helps to start with what your community takes to be the clearest examplars of these ideas. Maybe you'll have to revisit your initial assumptions because they don't stand up under scrutiny, but you've got to start someone. It's easier to inquire about the nature of wrong if you start with something that strikes you and everyone else as unequivocally wrong, say, murder.

Getting back to the theme of today's discussion, here are some relatively uncontroversial examples of ends justifying means:

i) Stealing food in order to survive
ii) Lying to a would-be assassin about where your friend is hiding in order to prevent his murder

Philosophers disagree about how to describe tradeoffs like these, but their pragmatic advice is always the same: take it and live; lie and save a life. Offhand, the only major exceptions within analytic philosophy are hardcore Kantians who argue that it is unacceptable to lie even to save a life. This is a position that most contemporary Kantians find uncomfortable. Some of Kant's critics think this result is sufficiently absurd to discredit Kantian ethics.

Utilitarian/consequentialist philosophers have no objection to morally questionable acts that clearly generate a better overall result. Generally speaking, they hold that any means are acceptable if they contribute to the overall end (be it maximizing utility, satisfying preferences, or whatever). As we will discuss below, this flexible moral accounting can have implications so stark that they lead many philosophers to question the soundness of consequentialist ethics.

Virtue ethics is compatible with a range of attitudes towards ends justifying means. Virtue ethicists are often concerned with identifying the sets of virtues that we observe in everyday life. They urge us to identify admirable people and try to understand what combination of characteristics contribute to their flourishing. It's worth noting that our role models are seldom so rigid as to starve themselves to death rather than eat a purloined chocolate bar.

(Questions about the morality of stealing to survive wouldn't arise within most ethics of care. But ethics of care generate their own means/ends conflicts. Readers are encouraged to think of their own examples. Maybe I'll get a chance to write more about ethics of care later on.)

Almost everyone would tell a starving person to go ahead and eat the damned chocolate bar. The question is how to describe the moral situation that arose.

Some people will say that it just isn't wrong to take somebody else's food in order to survive. However, if they also maintain that it's usually wrong to take other people's property without permission, they are arguing that the end (survival) justifies the means (unauthorized taking).

Others will say that there are some extreme situations in which morality has no authority. They'll say that moral principles are supposed to provide reasons to act, but we shouldn't assume that those reasons always override other reasons a person might have.

Just as most ethicists agree that there are some acceptable means/ends tradeoffs, there's an equally strong consensus that the ends don't always justify the means. Consequentialists are locked in an unending struggle to explain away the seemingly horrific implications of their theories. Most of us are uncomfortable with an ethical system that could sanction murder for the redistribution of transplantable organs. But as we discussed earlier, strict Kantians and other strict duty-based ethicists often find themselves short on principled explanations for why we shouldn't stick to the rules in the face of unthinkable consequences. Does morality require us to allow the destruction of the entire world rather than harm one hair on an innocent child's head? If that's what our behavioral code tells us, are we still persuing anything we'd recognize as ethics?

Here are some much, much more controversial examples of means/ends tradeoffs:

iii) Just wars (Non-pacifists, define "just" as the list of criteria that a war would have to meet in order for you to support it.)
iv) "Ticking time bomb" torture scenarios
v) Hostage-taking cases in which the kidnapper threatens to kill 20 people unless you shoot 1 innocent person

Yesterday's discussion was really fun. So, I'll turn the floor over to you guys. Enjoy!

Comments

[togolosh]

There was a particularly awful example of (v) reported on NPR in the last week. An Iraqi father whose son was collaborating with US troops was given the choice (by insurgents) of killing his own son or having his whole family killed. He killed his son.

[Njorl]

I guess the easiest war to justify is one of self-defense, that you can win, against an immoral aggressor.

Questions arise when removing these criteria one by one.

Is it right to be a moral aggressor? If there is a greater good that can only be achieved by war, is it justified. If the Nazis did not start a war, but did attempt the extermination of the Jews, would war be justified? I would say yes, if other means were tried - diplomatic pressure, subversion of the citizenry etc. - and failed.

This brings up the converse. Is it right to fight back against a justified aggressor? Are the citizens of the above alternate-world Nazi Germany justified in defending their nation against aggressors? This is a bit harder. I can't accept in this situation that both sides could be justified in fighting. If the invaders are genuinely morally justified, and you are brave ennough to fight, then you are brave enough to fight against your immoral government. Of course, the justice of the situation is not likely to be clear to those who must make the decision to fight or not.

The last criterion is probably the most difficult. Is it right to fight a war that in all other ways is justified if you can not possibly win? I would say it depends on the consequences. When Ghengis Khan set out to conquer a city, he offered them a choice - submit or die. Usually, there was no chance whatsoever of successful resistance. In this case, resistance was unjust. Submission to the immoral demand resulted in heavy taxation, but continued life. When Tamerlane was in control of the Mongols, he eliminated the "submit or" part of the equation. It was just "die". The chance of victory was often every bit as nonexistant, but resistance in this case was justified.

[dan]

going back to yesterday's discussion, which I had to abandon once my beloved Braves came on (and needless to say, after the game I was too devastated for ethics), ADR sez

The idea that moral systems are in any more or less trouble than ethical, metaphysical, scientific, etc. systems when it comes to the "well, most people would agree that..." is odd, if not blindly faithful (vestigial scientism, perhaps). If anything, ethics is admirable for its willingness to admit that, sooner or later, every system has to rest on reasonable human judgement.

well, fine, but at that point ethics is purely an aesthetic field. Which is actually a call that I'm pretty comfortable with.

Not to be too picky, here, but Lindsay's comment that set me off yesterday was "Sometimes ends do justify morally questionable means." Today she characterizes her claim as "most ethical systems acknowledge that ends can sometimes justify means." To me, there's a huge difference between those claims. My position, for what it's worth, is that it's perfectly reasonable and straightforward to have a debate about whether, within the context of ethical system X, practice Y is acceptable; and of course Lindsay is right that most ethical systems etc. I just don't see how it's possible to make the jump from "most ethical systems acknowledge that it's ok" to "ethically it is ok."

This entire discussion--this entire field of philosophy--is predicated on the idea that human activity should be guided by some sort of moral principles, and implicit in the discussion--certainly implicit in Lindsay's proposed methodology--is that there are some quasi-Platonic ethical principles that should guide us, which we can ferret out by, e.g., drawing conclusions from our existing feelings about admirable acts, or studying admirable people, or systematically provoking our own moral intuitions. I can imagine absolutely no reason to believe that there are metaphysically significant ethical principles out there, so I think that if you start looking for what the consensus view of ethical behavior is, you'll wind up finding the consensus view of what ethical behavior is. No more. Which, to wrap this up, is the difference between saying "ends can justify means" and "most ethical systems acknowledge that ends can justify means."

To put it slightly differently, I have no problem agreeing that most people believe that ends can justify means. My problem is acknowledging that anything turns on that consensus, aside from giving us a basis to predict how society will react to various acts, and no doubt to informing our aesthetic judgments about actions (which is to say, yeah, I believe that morality is a con, but I still wouldn't torture a puppy, and am repulsed by the idea of killing a person so that 20 people may receive organ transplants).

The only Platonic truth is that the Braves cannot hit in the postseason. Everything else is blather.

[Lindsay Beyerstein]

Dan, you're right that I asserted that ends sometimes justify means without proper qualification. Nothing follows from the mere fact of consensus. However, broadly shared intuitions are often a useful starting point.

It's pretty uncontroversial that a person should take bread to feed his starving family. Any ethical theory has to address that intuition somehow. Either by attempting to show that the intuition is wrong, or by adjusting the overall theory of ethics to accomadate it.

[Dan]

"However, broadly shared intuitions are often a useful starting point."

What basis do you have for making that claim? What makes you think that examining broadly shared intuitions does anything other than reveal broadly shared intuitions? What are they "useful" for? What turns on their use? If 9 out of 10 dentists prefer not to torture people, what have we learned?

[Gary Sugar]

Sorry, I again come close to agreeing with dan; so I again can't be helpful or interesting with this question. I don't believe in an absolute morality; so for me, the question only asks whether I'd do something I dislike to get something I really want. Well I do that every day.

[dan]

If we posit that there are moral principles which should guide action, then (hopefully) examining broadly held assumptions would be a useful starting point for determining what those principles are.

However,
a) I see no reason to believe there are such principles,
b) If such principles exist, I cannot imagine how we could establish their ontological status,
c) if such principles exist and can be verified, I don't understand where they could derive their authority (unless we posit a god, which I won't, for now)
[caveat: I enjoy Kant enough to think that _maybe_ if we could discover the princples through pure reason, they would have some claim to moral authority on that basis. But of course, were that the case, we wouldn't be examining broadly held social consensus to begin with. And I hope it goes without saying I don't hold out any hope for that project to begin with]

[C.J.Colucci]

I think it was Lenin who said "If ends don't justify means, what does?"

[Lindsay Beyerstein]

Dan, I know you're skeptical about the methodological underpinnings of analytic ethics. I'm sympathetic to those critiques. That's probably why I'm not an ethicist and prefer instead to play one on the internet.

But how else are you going to start inquiring about ethics, if not by consulting your preexisting ethical intuitions. The alternative might be to audit your pre-theoretical ethical concepts, but then you have to deal with the fact that you learned to apply those terms by watching putative examples of ethical and unethical behavior as defined by the culture in which you learned to speak.

Also, as I think you pointed out in the previous thread, it's impossible to have a conversation with someone unless you identify certain shared assumptions. If I want to talk ethics with a broad audience, it helps to pick broadly shared intuitions.

Suppose I chose a more controversial example. "Is it wrong to take a job in the capitalist system in order to finance the Socialist Revolution?" We'd never get around to a means/ends discussion because there's no consensus about whether the capitalist system is good, or whether a Socialist revolution is desirable, or whether it's possible to advance socialism by participating in capitalism.

My original statement might be better phrased as a conditional: "If you think that it is permissible for a starving person to steal food in order to survive, and you accept a general moral prohibition against theft, then you're agreeing that at least one end can justify at least one otherwise impermissible means to that end."

[Gary Sugar]

I again won't go quite as far as Dan but still reach pretty much the same conclusion. I think I can deduce some practical advice that's good for me, just not anything that's absolutely good. Anything that helps me continue in existence is good; anything that hinders my continuance is bad. If I establish what I mean by happiness, I can say anything that increases my happiness is good. And with a little epistemology, I can say anything that increases my understanding is good.

I better admit that nothing I'm saying is new. These are the fundamental principles in Spinoza's ethics.

[John Emerson]

The "stealing in order to survive" case is an example of why I think that ethics is subsidiary to politics and power. It assumes a system of property relations and prohibitions deriving from them, while also apparently assuming the property-owner's right to deny food to a starving person. These principles are historically specific rather than universal.

Going beyond the initial limited case (a starving person non-violently taking an unguarded unit of food), you can extend the question to anyone whose survival depends on violating any social convention, or harming someone else in any way.

To say that people should be willing to die rather than do wrong does assume the rightness of the social order within this "wrong" is defined. (Or perhaps it assumes the duty of obedience to instiututed rules as such, even wrong rules). But when you start asking whether the very social order is right, within which right and wrong are defined (e.g., stealing property or not), then you open a whole can of worms. "Was X justified in killing Y" can be a routinely answerable question within given structures, but questions like "Is society A justified in killing persons of the type Y, e.g. murderers" do not have a given framework. (In other words, a system of ethics makes it possible to judge act; but what is it that makes it possible to judge ethical systems?)

Ethics may or may not be competent to judge whole societies, but regardless, ethical judgements about whole societies are usually ineffectual, and always are slow to be put into effect. When they are put in effect, furthermore, it is usually the result of power politics and simply ends up instituting a new power regime, with new ethical rules and new ethical blemishes.

P.S. What my post is "about" is not whether it is possible to judge societies. It's about the limitations of academic ethics as a tool for judging societies, and the consequent limitation of the scope of academic ethics, which assumes instituted systems.

[dan]

LB says

"But how else are you going to start inquiring about ethics, if not by consulting your preexisting ethical intuitions."

My answer is "why bother inquiring about ethics, if you're not convinced it can possibly lead anywhere?"

To which you presumably respond

"If I want to talk ethics with a broad audience, it helps to pick broadly shared intuitions."

Which is a pretty good answer. For my money, there are few endeavors less rewarding than talking ethics with a broad audience. I guess that's why I'm not a blogger, I just pester them in comments.

[Trystero]

Could someone here explain the difference between moral and ethical? I was trying to sort it out for someone who speaks English as a second language, and I think I got about halfway there: murder is immoral; nepotism is unethical. So, morals are some kind of absolute principles, while ethics are a code of behaviour preventing people from taking unfair advantage?

It also seems like we use these words a little differently in general conversation than y'all are using them here, so are the academic meanings different from general usage?

[dan]

Trystero

good question. I, for one, have been incredibly sloppy about that, and I apologize. It's been a long while since I was as careful as I should be with my philosophical language.

ethics can have at least two meanings. First, it applies to the rules of conduct within particular endeavors, and especially within particular professions. Thus lawyers, doctors, etc., have codes of ethics, as opposed to codes of morality. Using this sense of "ethics," nepotism could be described as immoral or unethical; if you wanted to stress the harm that incompetent employees do to the public, you'd probably describe it as immoral; if you want to stress that it is unseemly or somehow beneath the standards of conduct we should expect from politicians, you'd probably describe it as unethical, although that's more a question of shading than a requirement of the language. However, more broadly, in philosophy, "ethics" is the study of morality, or perhaps more precisely the study of moral systems and rules. So, weirdly, one meaning of ethics is narrower than morality and the other is arguably broader. Speaking only for myself, I've been talking about the broader meaning.

[Lindsay Beyerstein]

I use the terms "moral" and "ethical" interchangeably. If you're reading philosophy, assume that the two terms are synonymous unless the author stipulates otherwise.

Sometimes "ethics" is used to talk about role-specific codes of conduct for particular types of people, e.g., journalistic ethics, medical ethics, business ethics. Some of the rules of professional ethics more conventional than normative. (For example, someone could argue that assisted suicide is permissible, but that it is unethical for a physician to assist a patient in this regard because she is bound by professional oaths not to do so. One could go on to argue physician assisted suicide could become an ethical act for a doctor if the AMA voted to redefine the doctor/patient relationship, or whatever.)

In informal written English, "immoral" tends to be applied to more serious transgressions than "unethical". "Moral" sometimes religiously-based and/or socially conservative rules, especially rules, often rules about sexuality.

In American politics "moral values" is code for conservative stances on social issues like abortion, homosexuality, pornography, etc.

[Gary Sugar]

I agree that in philosophy there's no consistent distinction between "ethics" and "morality" beyond definitions given by this or that philosopher. The greek "ethos" emphasized personality or character slightly more; the latin "mores" emphasized feelings or conscience slightly more. But I'd never assume a writer intended those meanings; and half of what we read is translated anyway. Common or modern usage is all over the place and no help in my opinion.

[FoolishOwl]

C.J., I don't know if Lenin said, "If ends don't justify means, what does?" However, I was just about to say that, before going on to talk about Trotsky's "Their Morals and Ours."

On rereading that essay, it's clear that Trotsky poses that very question (though not in those words) and says, if it's not the ends that justify the means, then it must be God. (And I thought I was so clever in thinking up something that it turns out I read years ago.) Anyway, if you reject metaphysical claims, then it must be the ends that justify the means.

Furthermore, ends and means are dialectically related. You can't just choose any means to achieve an end -- you have to choose the means that will actually achieve that end. Furthermore, you've got to see means and ends as a totality -- what are the full effects of the means chosen, what is the full nature of the end achieved?

[Njorl]

"Furthermore, you've got to see means and ends as a totality -- what are the full effects of the means chosen, what is the full nature of the end achieved?"

I strongly agree. The fact that "the means" have occurred must be considered one of "the ends". Provided this is done, if the ends are justified, then the means are justified.

I see no need for a metaphysical basis for morality. A moral code is the nebulously realized set of rules for continued survival of the culture that embraces it. It is not necessary that a culture understand this about their moral code. Its authority stems from its success. The culture exists therefore the morality is sound. This is purely empirical, so there is always the danger of counter-examples existing which may destroy the culture. More likely, the moral code will be alterred. The culture will become agitated and adjust.

When viewed in this vein, the ends/means discussion is a bit easier. It is not quite so simple as choosing whatever action leaves the most people living but it does save you the trouble of worrying about damaging immortal souls. Using dubious means to achieve apparently favorable outcomes can have unforseen, unpleasant complications. The idea that the ends do not justify the means is shorthand. It means that any individual does not have the right to set aside the culture's moral code to bring about a situation that they see as preferable.

Does that mean that you can't steal food to survive? No. Any individual has only limited understanding of the true moral code. Because circumstances are infinitely variable, the moral code is infinitely detailed. No one can know all of the details. What we know are rules. Rules are imperfect approximations to the moral code. Rules may be violated because of their imprecision. One may use ends to justify breaking rules, but it is not possible to use ends to justify violations of moral code.

[Njorl]

"Furthermore, you've got to see means and ends as a totality -- what are the full effects of the means chosen, what is the full nature of the end achieved?"

I strongly agree. The fact that "the means" have occurred must be considered one of "the ends". Provided this is done, if the ends are justified, then the means are justified.

I see no need for a metaphysical basis for morality. A moral code is the nebulously realized set of rules for continued survival of the culture that embraces it. It is not necessary that a culture understand this about their moral code. Its authority stems from its success. The culture exists therefore the morality is sound. This is purely empirical, so there is always the danger of counter-examples existing which may destroy the culture. More likely, the moral code will be alterred. The culture will become agitated and adjust.

When viewed in this vein, the ends/means discussion is a bit easier. It is not quite so simple as choosing whatever action leaves the most people living but it does save you the trouble of worrying about damaging immortal souls. Using dubious means to achieve apparently favorable outcomes can have unforseen, unpleasant complications. The idea that the ends do not justify the means is shorthand. It means that any individual does not have the right to set aside the culture's moral code to bring about a situation that they see as preferable.

Does that mean that you can't steal food to survive? No. Any individual has only limited understanding of the true moral code. Because circumstances are infinitely variable, the moral code is infinitely detailed. No one can know all of the details. What we know are rules. Rules are imperfect approximations to the moral code. Rules may be violated because of their imprecision. One may use ends to justify breaking rules, but it is not possible to use ends to justify violations of moral code.

[pdf23ds]

dan said:
"if such principles exist and can be verified, I don't understand where they could derive their authority (unless we posit a god, which I won't, for now)"

Now, now. One of the more compelling, to my eyes, critiques of Christian ethics is that there really *is* no way to derive a real moral imperative from the desires of even a Christian-type god. What do you think of that?

[FoolishOwl]

I think I don't know what you mean, pdf. Are you saying that God's will is unknowable? Because the point of Christian ethics is that obedience to God's will is the moral law.

[pdf23ds]

Oh, no. Assuming that we know God's will. The argument is that if I say that God's will isn't binding on me, that I don't care what he wants me to do, there's no argument that could convince me (as an impartial and rational hypothetical individual) that it is.