If you assume both that the actual action (A) and the possible, but not actual, action (B) you have an inconsistency.
Yes.
But if you assume only B; ie. you assume a world as similar to this one as possible where you are such that you will choose action B, then it is perfectly consistent.
You need to show that this other "similar" world has any relevance to reasoning about the actual world. You are not justified in considering a problem statement that stars a non-actual world unless you explain how that helps in reasoning about the actual world.
Of course we know that intuitively that's how it works, but it's not clear why, and "just so" doesn't help in understanding this mystery. Also, it's not clear how to generally construct those counterfactuals, even if we leave the question of their relevance aside. Where is the "you" that ought to be replaced in the environment? What about the thoughts about you in other people's minds, should they be modified as well? If not you run into pitfalls of CDT.
Until you are willing to grapple with your confusion on this issue, I don't see this conversation being productive.
I'm trying to argue that you should be confused, just as I am confused. Notice your own confusion and all.
You need to show that this other "similar" world has any relevance to reasoning about the actual world.
It allows you to think about cause and effect, and it is a necessity in making rational choices. You cannot make a rational choice without thinking through the consequences of different (counterfactual) possible choices.
You are not justified in considering a problem statement that stars a non-actual world unless you explain how that helps in reasoning about the actual world.
In reasoning with counterfactuals, the counterfactual worlds are...
Keep in mind: Controlling Constant Programs, Notion of Preference in Ambient Control.
There is a reasonable game-theoretic heuristic, "don't respond to blackmail" or "don't negotiate with terrorists". But what is actually meant by the word "blackmail" here? Does it have a place as a fundamental decision-theoretic concept, or is it merely an affective category, a class of situations activating a certain psychological adaptation that expresses disapproval of certain decisions and on the net protects (benefits) you, like those adaptation that respond to "being rude" or "offense"?
We, as humans, have a concept of "default", "do nothing strategy". The other plans can be compared to the moral value of the default. Doing harm would be something worse than the default, doing good something better than the default.
Blackmail is then a situation where by decision of another agent ("blackmailer"), you are presented with two options, both of which are harmful to you (worse than the default), and one of which is better for the blackmailer. The alternative (if the blackmailer decides not to blackmail) is the default.
Compare this with the same scenario, but with the "default" action of the other agent being worse for you than the given options. This would be called normal bargaining, as in trade, where both parties benefit from exchange of goods, but to a different extent depending on which cost is set.
Why is the "default" special here? If bargaining or blackmail did happen, we know that "default" is impossible. How can we tell two situations apart then, from their payoffs (or models of uncertainty about the outcomes) alone? It's necessary to tell these situations apart to manage not responding to threats, but at the same time cooperating in trade (instead of making things as bad as you can for the trade partner, no matter what it costs you). Otherwise, abstaining from doing harm looks exactly like doing good. A charitable gift of not blowing up your car and so on.
My hypothesis is that "blackmail" is what the suggestion of your mind to not cooperate feels like from the inside, the answer to a difficult problem computed by cognitive algorithms you don't understand, and not a simple property of the decision problem itself. By saying "don't respond to blackmail", you are pushing most of the hard work into intuitive categorization of decision problems into "blackmail" and "trade", with only correct interpretation of the results of that categorization left as an explicit exercise.
(A possible direction for formalizing these concepts involves introducing some kind of notion of resources, maybe amount of control, and instrumental vs. terminal spending, so that the "default" corresponds to less instrumental spending of controlled resources, but I don't see it clearly.)
(Let's keep on topic and not refer to powerful AIs or FAI in this thread, only discuss the concept of blackmail in itself, in decision-theoretic context.)