When people criticize the idea that ‘the ends justify the means’, it seems to me that the real objection usually has to do with the distribution of outcomes between different agents, rather than the standard of assessing the morality of something according to the consequences it produces. Moral codes based on the outcomes of decisions are called ‘teleological’ whereas those based on rules about behaviour are called ‘deontological.’ Some have argued that a view based on consequences is likely to produce injustices, so it is more appropriate to base morality on set rules, such as the defence of individual rights. I don’t think it is necessary to make that jump. Indeed, I think the transition from an outcomes based view to a rights based view is likely to lead to less effective moral deliberation.
Consider some examples of ‘the ends justifying the means:’
- A person infected with a contagious, lethal illness is killed to stop them from accidentally infecting others.
- Reduced unemployment benefits drive 20% of those previously receiving benefits to get jobs, while making 5% poorer than before.
The relevant moral factors all seem to be based around consequences. How urgent was it to kill the infected person? How soon would they have died of the disease? What kind of jobs did the 20% get? Did they end up better off, all told? How much worse off did the 5% end up? In general, it seems that our objections to the ends justifying the means boil down to two kinds of objections: that the decision made undermined an important procedure or institution, or that those who were made to sacrifice welfare either gave up an excessive amount or were already badly off.
The procedural exception is certainly very important. Say a police officer is also a member of a drug gang and witnesses a fellow gang member killing someone. He is unwilling to testify to his involvement, so he plants false evidence leading to the man’s conviction. Here, we would legitimately object to the corruption of the police force and the impartial treatment of evidence. Even if doing so produced the ‘right’ outcome in this case – the conviction of a guilty person – the degree to which the means undermined the system made it unacceptable. Note that this is still an argument about consequences. If the police force operated in this way, we would expect future societal welfare to be lower than if the police behaved with integrity.
The exception based on the division of costs is similarly convincing, and similarly teleological in nature. Just think about the kind of justifications that could be employed in the case of the diseased person. Say they were about to board an airplane for a long flight, and the only way to prevent them from doing so would be lethal. Any argument would be about whether a more ‘cautious’ approach detrimental to the individual produces a better overall outcome than an approach that more aggressively asserts the larger interests of the group. The argument against excessive burdens on one individual or entity is similar. We recognize that forcing someone to sacrifice one of their last units of wealth is a greater imposition than making them give up one of their first units, since people pushed below a bare minimum level of subsistence suffer more than those pushed from greater to lesser affluence. We also recognize that the minimum moral action is of a higher magnitude when someone is in a truly desperate situation: we may not be morally obliged to provide aid to someone with the sniffles, but we may well be for someone who has just had a stroke or heart attack. Ultimately, those moral imperatives derive from the set of all outcomes associated with each choice.
In short, moving to a system based on rights forces us to adjudicate between them, and doing that necessarily brings us back into the realm of consequences. Say that I have the right to free speech and you have the right to privacy. How do we adjudicate between them? Using rules based on the consequences arising from different arrangements. It may be important to protect my right to criticize you if you are a public figure, for instance. It may also be especially important to protect the privacy of children and minors. Ultimately, rights just enumerate the moral issues that need to be considered. Consequentialist or teleological analyses give us our best insight into what is, or is not, moral.
Consequentialism and ‘public service’ ethics
The fourth ‘family’ is especially odd. It basically centres around the rejection of the phrase ‘the end justifies the means.’ What they mean by this, essentially, is not to circumvent procedures that exist for good reasons to achieve some narrow objective. What seems foolish about it is the fact that the ethical yardstick remains the ends. It is inappropriate to fast-track an excellent seeming job candidate past normal checks because of the risk that your intuition is wrong, and the possibility doing so will undermine the system. Both objections are ultimately based on a comparison between two sets of means (sloppy and rigorous) and two sets of outcomes.
It is indeed very easy to argue that all moral decisions are about consequences when you collapse any betterness claim into a claim about consequences. Let’s say we face the choice of moving from the distribution 10,1 to either the distribution 100,1 or 15,15. Although it partly depends what those numbers actually stand for – that is, what the lives of the two people are actually like – it seems obvious to me that a) 15,15 is better than 100,1 and b) 100,1 is better than 15,15 in terms of the consequences of the choice. Also, teleological is not a synonym of consequentialist: teleological means aiming at a goal, consequentialist means concerned with consequences.
All these arguments for consequentialism beg the question. But this isn’t by accident – consequentialism has to presuppose in advance a notion of the Good which it’s basically impossible to argue for. As soon as you suppose such a notion, everything appears to be done “for the sake of it”, and thus appears, insofar as the good is quantifiable, as consequentalism. Insofar as it is only qualifiable, it would appear as deontology, and insofar as it is uncodifiable, as virtue.
For example, in your police officer argument, you offer no argument as to why the fact that “justice” is not produced in this case is actually an argument about consequences. You just hope we were basing our reason on future welfare, and not on the upholding of justice.
Similarly, even the thesis statement “Indeed, I think the transition from an outcomes based view to a rights based view is likely to lead to less effective moral deliberation.”, is basically an example of circular reasoning, because what counts as moral deliberation will give the answer to your question.
Just because consequences are important reasons for acting doesn’t mean that moral oughts derive from consequences. Its perfectly plausible that only some consequences are relavent for consideration, and in that case, we need a set of rules to tell us which consequences to consider or not. In that case, it looks like rules are prior to consequences. So, I could use all these examples to try to prove that rules are prior to ends.
Its perfectly plausible that only some consequences are relavent for consideration, and in that case, we need a set of rules to tell us which consequences to consider or not.
I agree that some consequences should not be taken into account – for instance, the displeasure of bigots who were displeased when the vote was extended to women. That being said, the reasons those consequences are invalid are themselves consequential: if society did value them, overall outcomes would be worse.
Also, teleological is not a synonym of consequentialist: teleological means aiming at a goal, consequentialist means concerned with consequences.
True, it is simply unwieldy to use a word like ‘teleological’ too frequently in a piece of writing.
Although it partly depends what those numbers actually stand for – that is, what the lives of the two people are actually like – it seems obvious to me that a) 15,15 is better than 100,1 and b) 100,1 is better than 15,15 in terms of the consequences of the choice.
What non-consequentialist framework would provide clearer guidance?
“if society did value them, overall outcomes would be worse.”
Then your analysis reduces “the good” to what “society values”.
The problem with this is a) it’s full bore cultural relativism. You can’t say a value is wrong. You can only criticize the Nazi’s for poorly fullfilling their own values.
b) what proof do you have that the thing society values is quantifiable? Are social values always quantifiable, or sometimes qualifiable, and sometimes neither? If that was the case, you couldn’t conclude consequentialism had any more inherent priority than virtue ethics or deontology anyway.
“What non-consequentialist framework would provide clearer guidance?”
Again, only if the good is quantifiable (“15”, “100”), will consequentialism be a candidate for providing guidance.
Tristan,
I certainly didn’t mean to imply support for cultural relativism. There are definitely kinds of societies that are morally superior to others, partly because they are founded on values that produce a good level of human welfare along with protection against the things (tyranny, oppression, discrimination, genocide) that are most strongly associated with bad human lives.
I agree that quantification is a big problem, and one that can never really be solved. In the end, we need to make guesses about the magnitude of consequences based on what we can know.
It still seems to me that nobody has presented an alternative deontological framework that provides decent guidance on what it means to be moral.
If you can’t “solve” the problem of quantification, then you at least need to demonstrate that moral goodness is quantifiable at all, and that we should believe that the problem is “solvable” in principle, even if not in practice. Otherwise there is no immediate reason to think consequences provide us better guidance than a deontological framework.
There is a lot of confusion as to what counts as deontology. Sure, it means rules, but what sort of rules? We call Kant a “deontologist” but he has only one rule. So, if you say consequentialists have a “rule” in the form of the greatest happiness principle, then there are no more “rules” in Kantian morality than in Mill’s.
We can say it means “the good is prior to rules, rules are made for the sake of the good”. But, to be serious, this is true for deontology aswell. It follows from the nature of the good, for deontologists, that it can be determined by rule(s). Any deontologist would admit that if the good is a quantity, then deontological frameworks need a consequentialist justification (i.e. you need to prove that your set of rules will result in the greatest good compared to other sets of rules).
I think we’d have a lot more luck at figuring out what it means to act morally by using the principle of demonstration – and Hume agrees with me on this. If you want to know what it is to be moral, find some moral people, and see what they do. You could demand I give a justification for calling them “moral” before deciding which ones are moral and which ones arn’t, but that would be the fallacy of equivocating circular reasoning with vicious circular reasoning – if the goal is clarification of concepts, then a circular argument is not invalid – it is invalid only for the determination of concepts.
If you can’t “solve” the problem of quantification, then you at least need to demonstrate that moral goodness is quantifiable at all, and that we should believe that the problem is “solvable” in principle, even if not in practice. Otherwise there is no immediate reason to think consequences provide us better guidance than a deontological framework.
I don’t think the process can ever really be quantifiable in a mathematical sense. It is more like the weighing of issues a judge must conduct when deciding a civil case. They include the interests of those directly and indirectly involved, the incentives being created, the strength of evidence, etc.
There is a lot of confusion as to what counts as deontology. Sure, it means rules, but what sort of rules? We call Kant a “deontologist” but he has only one rule. So, if you say consequentialists have a “rule” in the form of the greatest happiness principle, then there are no more “rules” in Kantian morality than in Mill’s.
I agree that they are not as distinct as sometimes portrayed. For instance, there are lots of cases in which it makes sense to adopt a rule as shorthand for what would arise from a consideration of consequences.
If you want to know what it is to be moral, find some moral people, and see what they do.
This is not too far off the judging model I mention above. Looking at examples certainly helps us clarify our moral reasoning. For instance, we can evaluate our domestic laws against those that exist in other states – both the sort that we consider admirable and those we consider deplorable.
A question related to the ones above is the importance of intention. How do we deal with people who are doing harm from our own perspective while they are making a genuine attempt to do good according to their own ethical understanding? Where does the moral authority that would let us aid or resist them arise from?