Do our moral beliefs need to be consistent?

by Emrys Westacott

We generally think it desirable for our moral and political opinions to be logically consistent. We view inconsistency as a failing. Why?

I'm not talking here about consistency between a person's beliefs and their actions. Failing to practice what we preach is the sort of inconsistency we call hypocrisy, and it's easy to see why we disapprove of that. Hypocrites are less trustworthy and predictable than people whose actions accord with their stated opinions. Nor am I talking about remaining consistent over time, never altering or abandoning one's earlier convictions. That's the sort of “foolish consistency” that Emerson ridiculed as “the hobgoblin of little minds.”

I'm talking about logical consistency between beliefs. Why do we care about this? Exposing inconsistency is a standard move in many an ethical argument. Take the debate about abortion, for instance. A standard argument for viewing abortion as immoral is that it is essentially no different from infanticide, which, as it is the premeditated killing of an innocent human being, meets the definition of murder. Note the form of the argument: if you think murder is wrong, then, to be consistent, you should think infanticide is wrong, in which case, to be consistent, you should think that abortion is wrong. On the other side, a common justification for permitting abortion rests on the idea that a woman has property rights over her own body. Essentially, the argument runs: if you agree that a woman's body is her own property, then consistency requires you to accept that she can do with it as she pleases, and if you agree that the fetus is a part of her body, then consistency requires you to accept that she can do as she pleases with the fetus.

Or take Peter Singer's well-known argument for why all of us who can afford to should give more to help the needy. We all agree it would be wrong to not save someone from drowning just because we didn't want to ruin our shoes. Well, Singer argues, if we think that, then we should also accept that we have a duty to save human lives if we can do so by making similar minor sacrifices–and many of us can do this by donating our disposable income to charity. Whether these lives are close by or far away is irrelevant. Again, the underlying strategy here is an appeal to consistency. If you think x, then you ought, for the sake of consistency, to think y. Many other arguments about moral matters take this form.

But why do we value consistency? In science and in our everyday beliefs about the way things are, there is a straightforward answer. Inconsistent beliefs, taken together, form a contradiction: a proposition that has the form “p and not p.” We assume that reality does not contain contradictions (an assumption first articulated by Parmenides). So we infer that an inconsistent set of beliefs cannot possibly be an accurate description of the way things are.

It may be a useful working map or model; we may not know at present how to improve on it; but so long as there are inconsistencies, we assume it cannot be the final account, the definitive truth. (Note: I am not assuming here that a realist view of truth is philosophically satisfactory, only that it is the one most of us work with most of the time. I am also aware that in logic contradictory statements are false by definition, but that isn't why most people think they are false; the logician's definition reflects ordinary thinking, not vice-versa.)

When it comes to moral beliefs, however, this reason for valuing consistency doesn't apply. Well, to be fair, it does apply if you believe in an objective moral world order that makes our moral beliefs true or false. But in an increasingly secular age that position is unfashionable, to put it mildly, and hard to support without some dubious religious or metaphysical props.

So, if we don't think that our moral judgements describe an objective moral reality, are there other reasons for wanting them to be consistent? I can think of two. One is that we want to be rational, and consistency is the hallmark of rationality. This is a weak argument, though. Really, it just tries to bolster the idea that we should care about consistency by associating it with another term we automatically approve of—rationality. Besides, a concern for strict logical consistency is not the only way of being rational. There is also a pragmatic kind of rationality, where we are concerned with finding the best means to secure a desired end. And sometimes, often perhaps, this sort of rationality should take precedence. For instance, it may be hard to justify theoretically why we allow athletes to buy some sorts of advantages, like enhanced vision through eye surgery, but not others such as higher stamina levels through blood transfusions. But instead of looking for some subtle difference that supposedly justifies differential treatment, we should perhaps simply acknowledge that the reasons are essentially pragmatic, having to do with the likely consequences for certain sports if these procedures are allowed or banned.

The other reason for caring about consistency is more significant. Being consistent in the way we treat people is at the heart of our notions of impartiality and fairness. Thus, a leading argument for legalizing same-sex marriage is that if heterosexuals can marry the person they love, then gays should be free to as well. This is clearly an appeal to consistency. What is at stake here, though, is not primarily the theoretical coherence of our beliefs but the practical discrimination experienced by certain members of society.

Now some people will claim that any inconsistency in the way we treat people is inherently unfair and hence wrong. To a large extent this is indeed our default way of thinking in everyday life. Nevertheless, there is a difference between thinking of consistency as intrinsically good and thinking of it as pragmatically good most of the time. And I would argue that the latter point of view is preferable. On this view, we should generally approve of treating people consistently because a world in which this notion of fairness prevails will be a world in which people live more happily; there is likely to be less conflict, greater social mobility, more efficient use of labor, less resentment, less selfishness, greater social cohesion, and so on. The alternative to thinking this way is to declare that consistency is simply good in itself and for its own sake. But then there is no satisfactory answer to the critic who asks why we would want to fetishize an abstract virtue–consistency–possibly even at the expense of concrete human well-being.

Of course, on the face of it we often seem to condone inconsistent treatment for pragmatic reasons. For instance, we discriminate against epileptics by not allowing them to operate locomotives. But one could argue that every supposedly pragmatic justification for such discrimination can be recast as an explanation as to why we are not really being inconsistent. Such explanations point to what we see as a relevant difference in the discriminatees compared to the rest of the population. We say, for instance, that the vulnerability of epileptics to seizures increases the risk of an accident to an unacceptable level. Thus, the policy does not treat people inconsistently at all; that would only be true if there were no relevant differences between the people receiving differential treatment.

That is a pretty good argument, and it covers many cases. But I don't think it covers all. There are times when our attempts to justify an apparent inconsistency in our judgements or practices look suspiciously like disingenuous rationalizations of ways of thinking or acting that we have become comfortable with, or that it would be problematic to abandon, or that we think carry significant practical benefits, or to which we can't think of a decent alternative. Consider, for instance, the laws regarding alcohol compared to other recreational drugs, or US foreign policy regarding various undemocratic countries, or court rulings that rest on dubious appeals to precedent or questionable interpretations of the constitution.

The considerations touched on above perhaps help to explain why we value logical consistency in our moral judgements . We value it in other areas as a necessary condition of truth; we see it as central to our notion of rationality; and in policy and practice it is closely linked to our conception of fairness. So maybe our commitment to consistency gets carried over from these spheres into the realm of moral theory and reflection. It then becomes something like a conversational convention, a rule that everyone recognizes and that helps give form to the discourse. Such conventions can certainly have instrumental value: dialogue must have forms just as games need rules. So a commitment to logical consistency may well be worth upholding much of the time; but it can still make sense sometimes to grant other considerations a higher value.

The analogy between conversational conventions and the rules of a game can illuminate this point. In show jumping, knocking over a fence detracts from the rider's overall score, but the fault needn't be decisive; it can be compensated for by other factors such as speed. Thinking along these lines, inconsistency might be viewed as a flaw in a position or a theory, but not necessarily as constituting a decisive objection. Logical consistency can be viewed as a desideratum in our moral beliefs, but it may sometimes be trumped by other values such as the practical benefits of enacting a certain policy.

An advantage of not insisting on logical consistency as a sine qua non of any acceptable moral position or ethical theory is that we will be more likely to give due weight to pragmatic considerations. Consider the abortion debate again. Much ink has been spilled constructing sophisticated arguments to show that allowing abortion is or is not consistent with certain other precepts we adhere to. But an alternative approach is to cut the Gordian knot by not worrying about that and simply asking instead: what are the likely consequences of allowing or prohibiting abortion? If prohibiting it is likely to produce more dangerous backstreet abortions, more unwanted children growing up in deprived circumstances, more single mothers mired in poverty, and so on, then these are reasons for ensuring that it be legal and available. If, on the other hand, its ready availability tends to put a heavy economic burden on the health care system, diminish our respect for human life, and foster less careful attitudes to sex which in turn increases the incidence of sexually transmitted diseases, then these are reasons for banning abortion.

To sum up: I'm not saying that we should stop caring at all about logical consistency in working out our positions on moral issues. But I think it is interesting and reasonable to ask why we do care. Moral philosophers, as theoreticians, naturally tend to focus on the theoretical coherence of statements and their implications. But morality isn't mathematics. It is perfectly rational, in one sense of the term, to prioritize practical consequences over logical consistency. Once we accept this, we will perhaps be more comfortable taking a pragmatic approach to moral problems, and feel free to do so without dissimulation or apology.

Like what you're reading? Don't keep it to yourself!
Share on Facebook
Facebook
Tweet about this on Twitter
Twitter
Share on Reddit
Reddit
Share on LinkedIn
Linkedin
Email this to someone
email