Meta-meta-ethics

(NOTE: for those who don't wish to read my summary of moral intuitionism, skip till after the second set of asterisks.)

I wish to propose a meta-meta-ethical view that will give reason for existence of meta-ethical theories.

What does that mean? Well, ethics asks evaluative questions. Such as: "What actions or circumstances are good or bad?" and "What actions or circumstances are right or wrong?"

Meta-ethics asks questions about the nature of evaluative truths. What is good and bad? What is their nature? What do we mean when we say those words (or should mean if we are to make any sense)? How do we know what actions are good or bad? How can we find out?

A meta-meta-ethical analysis asks what the role of making meta-ethical theories is. And how we know that those are true.

One simple answer is: by using reason. We can just see if a meta-ethical theory makes any sense internally. (Just like we would with, say, a geometric theorem.)

For instance, Kant says that a categorical imperative is a kind of imperative that would be binding on you if you knew it to be binding on everyone. Which you can verify by imagining a world in which such an imperative is not binding on anyone. Is this view coherent? Does it make sense? One can ask questions like: Well, yes, it makes sense that were absolute binding rules to exist, they would take form of categorical imperatives. But who says they exist? Also: how are we to judge whether a world in which everyone is free to lie or murder is a good world? Should we use our emotions? Why is that a good method for evaluating whether a world is good or bad?

In other words, we can test the internal coherence of a specific meta-ethical theory.

But people oftentimes don't do just that. They do that and they also test the conclusions of the theory. For instance: is torturing puppies for fun moral? Some might argue that it passes every one of Kant's moral maxims. Is lying to a Nazi asking you where the Jews are hiding immoral? Some argue that according to Kant's maxims it is immoral, and one shouldn't lie even in such a situation.

When people hear such conclusions, they infer that there is something wrong with Kant's meta-ethical theory. Or at least that it is incomplete. Because it leads to absurd conclusions.

But why should that be a test of a meta-ethical theory?

* * *

It seems that people already know what the correct moral answers are — at least in some cases. Even without having a coherent meta-ethical theory on their hands. How do they know? And if they do, what's the use of having meta-ethical theories?

Well, this is where the theory of moral intuitionism comes in. I've discussed it before, and I was introduced to it through writings of a moral philosopher Michael Huemer (whose book, Moral Intuitionism, I am currently reading and recommend to everyone interested in the subject), but, briefly, the idea is that there are moral truths about the world and our actions which we perceive intuitively.

What are these truths metaphysically? I haven't read that far in Huemer's book yet, but from his essays, it seems that he believes that the answer doesn't make sense. It assumes that moral truths must be defined in terms of some other set of truths, like natural truths. But, first, this would defy Humean is–ought gap, and, second, there is no reason why we cannot think of moral truths to be a separate set of truths, besides the natural ones.*

(I expressed my frustration and skepticism with this analysis before and offered an alternative explanation as to the metaphysics of moral truths. But I will leave that issue alone for now. And I am not completely sure I was right.)

How do we know about moral truths? Well, we perceive them. Intuitively. We have a perceptual modality called moral intuition, and through it we perceive certain actions as "choice-worthy" or not. And the idea called phenomenal conservatism says that unless we have a good reason to do otherwise, we should take our perceptions seriously. Because: well, they are there. We know about them. That is some sort of knowledge. Unless you have another sort of knowledge (usually from other perceptions) that denies this knowledge, it seems unreasonable not take the latter seriously.

* * *

So, how can moral intuitionism be a meta-meta-ethical theory? Doesn't it sound like a meta-ethical theory in itself?

Well, to answer this question, we have to humble ourselves a little regarding our physical and mathematical theories. We perceive physical and mathematical "truths" to be exactly such: true (we hope). Of course, with physics, we often re-evaluate our models later. And then again. With math we don't re-evaluate them, but we just make new ones if we need to. (So, in physics, "F=ma" is corrected by Einstein. In Math, Eucledean geometry exists in parallel with Riemannian geometry.)

But perhaps another equally correct way of thinking about both physical and mathematical models is not in terms of their absolute truth (or a degree of approximation thereof), but in terms of their usefulness. It is useful to imagine the world being countable in discrete numbers. It is useful to imagine shapes of all objects as being drawn on a sheet of paper. It is useful to think of the world as full of moving particles.

Until it's not useful anymore. And then, on that level, we apply a different model. What do I mean by "useful"? Useful in what sense? Only in the sense of explanatory power. I can explain how heat flows from one room to another if I model it in terms of many particles' kinetic energy. Such a model is more useful than one having some mysterious substance phlogiston flowing from one place to another, because the first model fits and predicts more observable phenomena.

The same goes for physical concepts like field, mass, energy, force, spin, voltage, and so on. They are just lego pieces that we invented; a perceptual glue to hold our observations together and make them fit into one shape. It's not really clear to me, for example, whether field really exists out there, as a property of space, or whether it's just something we made up to explain charged particles' behavior when they are a certain distance from other charged particles.

Why do we need the models? Because we can't observe all aspects of nature directly. So, we must deduce existence of something from those observations that we can make directly. And those deductions must be judged in terms of their usefulness in terms of their predictive and explanatory power.


What does this have to do with ethics? Well: I know it's wrong to kill humans out of pure convenience (or to prevent inconvenience). I know it's wrong to kill newborn babies. I know it's not wrong to kill one's cancer tumor (lo aleinu) or remove a mole. All these things I know from intuition.

I don't know whether it's wrong to kill fetuses. That's a gray area for my intuition.

For that reason, I need to construct a meta-ethical theory that will somehow incorporate my intuitive knowledge about ethics of "everyday" killing and will give me an answer about the morality of abortion. The data for this theory (that will either verify or break it) will be the moral facts I already know intuitively.

For instance, if we say that it's wrong to kill someone because you're extinguishing consciousness, then it seems OK to kill a fetus before his nervous system is developed enough to be conscious. Well... is it moral to put someone under anesthesia? Is it moral to kill someone, once he is under anesthesia? The answers to these questions seem intuitively obvious, and these answers break the "extinguishing consciousness" theory.

Note that "extinguishing consciousness" theory is not itself internally complete. It doesn't explain why extinguishing consciousness would be wrong. It only attempts to systematize our existing beliefs in some cohesive whole and make predictions about moral truths not immediately accessible to our intuition. Unfortunately, as I said, it is not a good meta-ethical theory, because it is broken by the data: our intuitive knowledge.

This way, moral intuitionism serves not only as a meta-ethical theory (it explains about the nature of moral truths and how we are to know them), but also as a meta-meta-ethical theory: it explains how meta-ethical theories can be useful even if we are not to think of them as "real truths" as opposed to mere models.

_____________
* Also, as an aside, if someone asks you: "What is water?", the most straightforward answer is: "What I perceive to be water: a clear, odorless, tasteless, drinkable liquid, with certain observable physical and chemical properties". Based on those characteristics, I know I am dealing with water.

If you answer: "Well, water is H2O", that invites the question: "What is hydrogen? What is oxygen? What are atoms?" You can explain in terms of electrons and protons, and so on... but eventually you will have to stop. You will have to say that there is a set of phenomena in the universe that behave a certain way, but why they behave that way we don't know for sure yet.

So, does this mean that you fundamentally don't know what water is? That seems strange. Also, imagine that — as unlikely as it sounds — it was proven that actually water is not H2O and H2O is not water. What would that mean? Would we continue calling the H2O water? Seemingly not. We would continue calling water the phenomenon of clear, odorless, tasteless, drinkable liquid that freezes at 0 degrees Celcius.

Basically, this analysis shows us that our perceptions are not chopped liver. Now apply the same analysis to "what is good" and "what is evil", and the answer "that which I perceive as something that ought not be done in principle" doesn't seem so silly anymore.