Empiricism and Ethics

One thing that enables us to make progress in a science is whether statements made within the discipline lend themselves to empirical testing. I think it might be useful to think about what implications this has for ethics. It seems to me that the statements that are made by prominent ethicists, such as Immanuel Kant, John Stuart Mill, Aristotle and John Rawls, fail to produce consensus. In any case there is no widespread belief that our ideas about right and wrong are improving in the same way that our ideas about Physics, Astronomy, Biology and the other empirical sciences are.

In many ways the world is a better place than it was in the past. Contrary to popular opinion, the world is likely more peaceful than it was during most of the past. Slavery is no longer legal. The role and status of women has improved. However, I suspect that these changes were not caused by any improvement in our thinking about right and wrong. They may simply be the result of an improvement in our material wellbeing that is the result of advances in other sciences combined with the effect of the accumulation of capital, or the material things used for production. Prosperity has made us less likely to risk our lives through violence. Replacing human muscle power with machines has increased the wages of women relative to those of men, thus leading to their economic independence.

What sort of statements about ethics lend themselves to empirical testing? I think of a system of moral and ethical values as a way of assigning blame and credit to actions. There is no empirical observation that can contradict a statement of the form "Action X is worthy of blame." Being worthy of blame is not a directly observable property. From this many conclude that there can be no empirical basis for preferring one system of moral and ethical values over another.

However, there are empirical observations that would cause an individual to prefer one system of moral and ethical values over another. Every individual has values. Empirical observations can tell us how we can act so that our values will be more likely to be realized. This enables us to use empirical evidence to support interesting statements about ethics. We can make it more likely that individuals will change their systems of moral and ethical values, if we convince them that it will be in their best interests to do so.

The question that will come up is whether we should do so. Perhaps we will make the world as a whole worse off through this approach. Are we not encouraging selfishness? People might choose moral and ethical values that enable them to profit at other people's expense. But this is not the case. In considering a system of moral and ethical values we are confining ourselves to assignments of blame and credit. In order to make such an assignment stick, individuals will have to get other individuals to agree with their assessments. Generally they will be unable to do so unless these other people are persuaded that doing so will be in their own best interests. This will usually be the case because moral and ethical values tend to be abstract. I will support the protection of your rights because they are principles that protect everyone including me. If there were a moral principle that would only protect you, there would be no reason for other people to support it.

One error that many ethicists make is that they attempt to explicitly design systems of moral and ethical values to serve the greater good. The problem with this is that statements about such systems are not subject to any kind of empirical verification. There is no way to test them. Suppose that I want to claim that my system of moral and ethical values would benefit the world in some way. In order to get people to implement this new and improved system, I would have to convince everyone to adopt it. They would have to accept my claim without evidence, since there can be no evidence until my system is implemented. Furthermore, once my system had been implemented there would be no way to tell what would have happened if my system had not been implemented. In order to tell what would happen if it had not been implemented, we would have to decide not to implement it. Hence there is no way to compare the results of implementing my system of moral and ethical values with leaving things as they are.

But if we are working with the individual, we need only convince the individual. Since it is likely that some individuals will be convinced while others will not, we will be able to compare the results of adopting one system of moral and ethical values to those obtained under another set of such values. We might even be able to do an experiment. Suppose we find a sample of people who are sitting on the fence on some issue in ethics. We could recruit them for an experiment in ethics. We could get half of them to adopt one position and the other half to take the other. Then we could compare the results.