I changed my mind midway through this post. Hopefully it still makes sense... I started disagreeing with you based on the first two thoughts that come to mind, but I'm now beginning to think you may be right.
So it's hard to see how timeless cooperation could be morally significant, since morality usually deals with terminal values, not instrumental goals.
I.
This statement doesn't really fit with the philosophy of morality. (At least as I read it.)
Consequentialism distinguishes itself from other moral theories by emphasizing terminal values more than other approaches to morality do. A consequentialist can have "No murder" as a terminal value, but that's different from a deontologist believing that murder is wrong or a Virtue Ethicist believing that virtuous people don't commit murder. A true consequentialist seeking to minimize the amount of murder that happens would be willing to commit murder to prevent more murder, but neither a deontologist nor a virtue ethicist would.
Contractualism is a framework for thinking about morality that presupposes that people have terminal values and their values sometimes conflict with each other's terminal values. It's a description of morality as a negotiated system of adopting/avoiding certain instrumental goals so that the people who implicitly negotiate the contract for their mutual benefit at attaining their terminal values. It says nothing about what kind of terminal values people should have.
II.
Discussions of morality focus on what people "should" do and what people "should" think, etc. The general idea of terminal values is that you have them and they don't change in response to other considerations. They're the fixed points that affect the way you think about what you want to accomplish with you instrumental goals. There's no point to discussing what kind of terminal values people "should" have. But in practice, people agree that there is a point to discussing what sorts of moral beliefs people should have.
III.
The psychological conditions that cause people to become immoral by most other people's standards have a lot to do with terminal values, but not anything to do with the kinds of terminal values that people talk about when they discuss morality.
Sociopaths are people who don't experience empathy or remorse. Psychopaths are people who don't experience empathy, remorse, or fear. Being able to feel fear is not the sort of thing that seems relevant to a discussion about morality... But that's not the same thing as saying that being able to feel fear is not relevant to a discussion about morality. Maybe it is.
Maybe what we mean by morality, is having the terminal values that arise from experiencing empathy, remorse, and fear the way most people experience these things in relation to the people they care about. That sounds like a really odd thing to say to me... but it also sounds pretty empirically accurate for nailing down what people typically mean when they talk about morality.
Contractualism is a framework for thinking about morality that presupposes that people have terminal values and their values sometimes conflict with each other's terminal value
Instrumental values can clash too. The instrumental-terminal axis is pretty well orthogonal to the morally relevant/irrelevant axis.
This post is a half-baked idea that I'm posting here in order to get feedback and further brainstorming. There seem to be some interesting parallels between epistemology and ethics.
Part 1: Moral Anti-Epistemology
"Anti-Epistemology" refers to bad rules of reasoning that exist not because they are useful/truth-tracking, but because they are good at preserving people's cherished beliefs about the world. But cherished beliefs don't just concern factual questions, they also very much concern moral issues. Therefore, we should expect there to be a lot of moral anti-epistemology.
Tradition as a moral argument, tu quoque, opposition to the use of thought experiments, the noncentral fallacy, slogans like "morality is from humans for humans" – all these are instances of the same general phenomenon. This is trivial and doesn't add much to the already well-known fact that humans often rationalize, but it does add the memetic perspective: Moral rationalizations sometimes concern more than a singular instance, they can affect the entire way people reason about morality. And like with religion or pseudoscience in epistemology about factual claims, there could be entire memeplexes centered around moral anti-epistemology.
A complication is that metaethics is complicated; it is unclear what exactly moral reasoning is, and whether everyone is trying to do the same thing when they engage in what they think of as moral reasoning. Labelling something "moral anti-epistemology" would suggest that there is a correct way to think about morality. Is there? As long as we always make sure to clarify what it is that we're trying to accomplish, it would seem possible to differentiate between valid and invalid arguments in regard to the specified goal. And this is where moral anti-epistemology might cause troubles.
Are there reasons to assume that certain popular ethical beliefs are a result of moral anti-epistemology? Deontology comes to mind (mostly because it's my usual suspect when it comes to odd reasoning in ethics), but what is it about deontology that relies on "faulty moral reasoning", if indeed there is something about it that does? How much of it relies on the noncentral fallacy, for instance? Is Yvain's personal opinion that "much of deontology is just an attempt to formalize and justify this fallacy" correct? The perspective of moral anti-epistemology would suggest that it is the other way around: Deontology might be the by-product of people applying the noncentral fallacy, which is done because it helps protect cherished beliefs. Which beliefs would that be? Perhaps the strongly felt intuition that "Some things are JUST WRONG?", which doesn't handle fuzzy concepts/boundaries well and therefore has to be combined with a dogmatic approach. It sounds somewhat plausible, but also really speculative.
Part 2: Memetics
A lot of people are skeptical towards these memetical just-so stories. They argue that the points made are either too trivial, or too speculative. I have the intuition that a memetic perspective often helps clarify things, and my thoughts about applying the concept of anti-epistemology to ethics seemed like an insight, but I have a hard time coming up with how my expectations about the world have changed because of it. What, if anything, is the value of the idea I just presented? Can I now form a prediction to test whether deontologists want to primarily formalize and justify the noncentral fallacy, or whether they instead want to justify something else by making use of the noncentral fallacy?
Anti-epistemology is a more general model of what is going on in the world than rationalizations are, so it should all reduce to rationalizations in the end. So it shouldn't be worrying that I don't magically find more stuff. Perhaps my expectations were too high and I should be content with having found a way to categorize moral rationalizations, the knowledge of which will make me slightly quicker at spotting or predicting them.
Thoughts?