by Tim Sommers
Even as we want to do the right thing, we may wonder if there is “really” a right thing to do. Through most of the twentieth-century most Anglo-American philosophers were some sort of subjectivist or other. Since they focused on language, the way that they tended to put it was something like this. Ethical statements look like straight-forward propositions that might be true or false, but in fact they are simply expressions or descriptions of our emotions or preferences. J.L. Mackie’s “error-theory” version, for example, implied that when I say ‘Donald Trump is a horrible person’ what I really mean is ‘I don’t like Donald Trump’. If we really believed that claims about what is right or wrong, good or bad, or just or unjust, were just subjective expressions of our own idiosyncratic emotions and desires, then our shared public discourse, and our shared public life, obviously, would look very different. One of Nietzsche’s “terrible truths” is that most of our thinking about right and wrong is just a hangover from Christianity that will eventually dissipate. We are like the cartoon character who has gone over a cliff but is not yet falling only because we haven’t looked down. Yet.
On the other hand, there is Sam Harris’s widely-read book from a few years ago: The Moral Landscape: How Science Can Determine Human Values. Now, if “science” could determine human values, depending on exactly how it did so, we might well have an answer to Nietzsche and subjectivism. Unfortunately, if I had been asked to review that book, I might have followed Wittgenstein who once said of a book that he would agree with it if you put the entire text in brackets and wrote in front “It is not the case that…” Just one example: Most philosophers would tell you that the project of offering a rigorous methodological distinction between a science and a non-science or even a pseudoscience, the so-called “demarcation problem”, is hopeless. So, Harris’s central claim that “science” will save ethics is either tautological – because whatever objective methods we develop to answer ethical questions will be, by an expansive enough definition, some kind of science – or false – since none of the existing sciences – physics, chemistry, or even biology – are likely ever to answer ethical questions.
I am being unfair to Harris, though. He is, in his own way, onto one source of objectivity in ethics when he talks about the centrality of “human flourishing”. One of the arguments that David Hume, the great godfather of subjectivist ethics, made in support of subjectivism about ethics is that there is nothing in the world, no objective features, that support, or are picked-out by, the label “good” or “bad”. If I witness an act, a robbery or an assault, say, I can describe all of its real, objective features without any evaluative labels at all. Ethics just isn’t out there in the world. It’s all in your head.
But leave aside for a moment whether anything is good or bad simpliciter, surely some things are good or bad for you. The emotional distress of the robbery victim, or the injuries of the assaulted, are real enough features of our shared, objective world. What’s good or bad for people is not wholly subjective.
Here’s a way to deny that. What makes something good or bad is simply whether or not I prefer it. But this is not really plausible – even though it’s a useful over-simplification in welfare economics. Surely, it’s true that part of what makes some things good or bad for me is whether or not I prefer them. But this is not the whole story. Sometimes I prefer things that are bad for me. But there seems to be an objective component, at least, to what is good or bad for me, an objective component to human welfare. A good way to think about “welfare” is to compare it to “health”. Health is studied quite rigorously by doctors and scientists. Even though we choose sometimes, with full knowledge of the bad consequences, the unhealthy option, we are not tempted, therefore, to describe “health” as a subjective notion. Human beings are animals. Some things are healthy or unhealthy for them. Judgements about health do have an evaluative dimension. Mountain climbing and smoking are unhealthy because, among other things, they increase your risk of death. Your love of smoking or mountain climbing may be great enough to lead you to risk death to pursue them, but they are still, objectively, unhealthy. We can think of welfare in much the same way. We sometimes do things that are not to our benefit, but this doesn’t show that there is no objective content to claims about what, in fact, benefits us. Quite the contrary. It suggests that we already intuitively grasp that what is good or bad for us is (at least partly) objective. Hence, one source of objectivity in ethics, we might say, is that welfare, like health, is at least partly objective.Even if that is true, one might respond, it only proves that what is good for me is objective, not that what is good is. Neither Plato or Aristotle would see it this way. For them, what is good for me and what is good in general are the same. But in modern thought these two ideas have come apart. We tend to think almost the opposite, in fact. We think that the right thing to do is typically something we must sacrifice (to some degree) our own interest to do. Why this is the case, is too much to get to here. But how should we think about the difference between reasoning about what to do and reasoning about what is good?
The late, great Derek Parfit used to present a chart in class that distinguished between theoretical reason and three distinct kinds of practical reason. Theoretical reasoning is reasoning about what to believe. Practical reasoning is reasoning about what to do. Instrumental practical reasoning is reasoning about what means to take to our ends. If we take some goal as a given, then reason about how to achieve it, we are doing instrumental reasoning. (In the way that one might think that the skeleton of theoretical reasoning is logic, one might say the skeleton of instrumental reasoning is the axioms of rational decision theory.) But there is also prudential practical reasoning, reasoning about what ends I should have and pursue. At a minimum, I can weigh ends against each other in terms of which will fulfill more of my preferences, but we can also (Parfit argued) reason about which ends are better or more important. Finally, there is ethical practical reasoning: reasoning about how to take other’s ends into account. Ethics, on this view, requires, or even consists of, a certain kind of impartiality. I don’t count my interests as less important than the interests of others, but I also don’t count them as more important.
What kind of reasoning is this? It’s most basic forms are familiar enough. As Kant said, “Originality in ethics is no virtue.” We might think that we should do whatever leads to the greatest benefit for the greatest number. Or we might think that we should always act according to principles that we expect everyone to follow. Many complications then ensue. But whatever mix of these, or other, principles we end up endorsing, do we have any reason to think of any of this as a kind of reasoning, really?
Hume rejected practical reason altogether. He said, “It is not irrational to prefer the destruction of the world to the pricking of my finger.” But surely it is. Instrumental reasoning all by itself will tell you that means of preventing the pricking of your finger that result in the destruction of the whole world are irrational since the destruction of the whole world will be worse for you and your ends than the finger pricking. Hume’s point is that it is not irrational because it is not a kind of reasoning. But surely reasoning about how to use means to achieve your ends is reasoning.
That’s the first step and the biggest leap. Parfit believed that this was the thin edge of the wedge that gets us to ethics as kind of reasoning. If we can get you to take that step, the rest is easier. If there is practical reasoning of any sort, why deny that ethics is a kind of reasoning? Here’s the next step. If we accept fitting means to our ends as kind of reasoning, isn’t adjusting our ends, and even comparing and weighting them, also, a kind of reasoning? And if we can reason about which ends are most valuable and how they are to be weighed, can’t we reason about how our ends are weighed against the ends to others?
Most of ethics is about fleshing out how to take others ends into accounts, as I’ve said. Ethics is about what impartiality between the ends of persons really means. However, while I’ve argued that welfare is really out there in the world, these principles really are all in our head. But they are not subjective. They are the products of reason.
If that sounds unlikely, even mystical, compare mathematics. Mathematics is also all in our head. The principles lead to axioms which lead to proofs. But they begin as intuitions to be balanced and fit together. All the rigor comes later. Consider the claim I made earlier about instrumental practical reasoning. There are whole disciplines – rational decision theory and microeconomics, at least – devoted to systematizing how we think about reasoning about means to ends. There is controversy there and competing axioms are proposed. But almost no one thinks that developing an axiomatic approach to instrumental reasoning is subjective.
Here’s a weird but revealing question, from George Stock’s The Book of Questions, that I have often posed to my classes. Given the choice between saving the lives of 1000 strangers in a foreign country, 500 Americans on an airplane, or your best friend, whom would you save? In my experience, most people think you should save the 1000. A small but significant group of people say “your best friend”. Almost no one ever picks the 500. Why? Maybe, it’s just the wrong answer. There are consistent, opposed lines of reasoning about the principles that ought to guide are actions that take you to either a 1000 people or your best friend – and which is right is a hard and interesting question. But it’s hard to reason your way to a consistent principle supporting the other choice. This doesn’t show that ethics is easy or even “solvable”. To me at least, it does suggest that we can reason about principles in way that is not merely or wholly subjective.
So here are at least two potential sources of objectivity that can inform our ethics. On the one hand, welfare, what is good and bad for persons, is (at least partly) objective – it’s really out there, part of “the furniture of the world” (Quine). On the other hand, reasoning about what to do is a kind of reasoning. Like all reasoning, it goes on inside our heads, but, like all reasoning, it (potentially) has a kind of objective rigor to it.
One of many questions that we haven’t answered is this. Why should I care about other people’s ends at all? Why not just do what is good for me? Egoism, however, is not a challenge to the objectivity of ethics, but to it’s content.