Moral Constraint and Complaint

Will Wilkinson and Saul Smilansky discuss a fascinating 'paradox of complaint', which basically comes down to a conflict between the following two principles:
(1) When you are wronged, you have grounds for complaint.
(2) If someone wrongs you in the same way that you have wronged others, you have no grounds for complaint.

For example, it doesn't seem like a car thief has the moral grounds to complain when someone else steals his car. But that doesn't mean an eye for an eye is really right. It's presumably wrong to torture a torturer, for example, even if they would have no standing to complain about it.

I'm reminded of Johann Frick's counterexamples to Scanlonian contractualism, defined as follows:

Scanlon’s formula: An act is wrong just when such acts are disallowed by some
principle that no-one could reasonably reject, or when any principle permitting such
acts could be reasonably rejected by at least one individual.

Johann (inspired by G.A. Cohen) asks us to imagine that a guy knocks on our door and credibly threatens to blow his own brains out unless we give him $10. He hardly has grounds for complaint if we shut the door in his face, so it can't be wrong according to Scanlon, but we can arguably fill in the details in such a way as to make it intuitive clear that it would be wrong not to appease the guy (since the sacrifice for us is minimal compared to the benefit to him of avoiding his own unreasonable suicide). Between this case and Smilansky's, I think we have pretty compelling grounds for rejecting principle #1, and distinguishing the conditions for moral complaint from those which call for moral constraint (against wronging another). The alternative, I suppose, is to reject what we might call

(3) The Wrong-to-Wrongs Principle: If it is wrong to do X in virtue of its negatively affecting another person, then your doing X wrongs that person.

If we reject this, we might think both that it is wrong of us to steal (even from a car thief), and yet that it doesn't wrong the car thief for us to do this to him. We did wrong, but not 'to him' -- which is why he has no special standing to complain about it -- even though our action was wrong in virtue of its impact on him.

So should we reject #1 or #3? Will Wilkinson suggests that perhaps the real culprit is #2. Perhaps strictly speaking even the car thief has grounds to complain when his car is stolen. The difference is instead that we have no reason to take his complaint seriously. This is an intriguing idea, but I'm not entirely sure how to make sense of it. Legitimate moral complaint emerges from what Darwall calls 'the second person standpoint', and concerns what one may rightly demand of another. If I step on your foot, you have special standing to tell me to get off your foot. If I don't, I've wronged you -- failed to give you your due, or to acquiesce to the demands you may rightly make of me. Suppose you later return the favour by stepping on my foot. Do I have grounds for complaint? Do I have the moral standing to demand that you get off, and perhaps apologize? I'm not sure it makes sense to say 'yes, but you don't have to listen'. If a complaint has moral standing, then the addressee must listen to it. That's just what it is for the complaint to have moral standing.