Religious Tolerance: Governments are morally forbidden from (i) enforcing religious tenets on their citizens that are not the religious tenets of those citizens (or requiring of them sworn allegiance to such tenets) and (ii) forcing its citizens to say or do things that contradict their religious tenets (if they have such).
On this view, the truth or falsity of some of our conclusions about permissible government coercion depend on whether or not people have religious beliefs according to which what they would be coerced into doing would be a sin. And the landscape of relevant or good reasons is similarly relativized to such religious belief, at least in this way: that one would be forced to commit something that one views as a sin comes to be a controlling reason against a proposed law, at least generally outweighing what would otherwise — from a neutral or objective or apart-from-what-religious-beliefs-people-have perspective — be sufficient or decisive reasons in favor of the law.
However, since what does the normative work here is the belief-state not the belief-content, I don’t see that false propositions come to be reasons in the relevant logical sense. And I don’t see how the relevant justification or reasoning comes to be governed by reasons being “acceptable,” or not where acceptability has little to do with evidence, logic, truth or otherwise with the reasons being good reasons.
We do get an evaluative standard that cross-cuts goodness in reasons (and justification) if we take Religious Tolerance (or its grounds) to imply some plausible things about how we are to respect each other in debate (in the giving and taking of reasons) about when government coercion is permitted. Maybe if you hold a belief in a dogmatically religious way, I’m in some cases obligated not to simply contradict that belief or “steamroll” you with all of the reasons why what you believe is silly. Of course, in the business of giving and taking of reasons with other persons, there are pragmatic considerations that make it inappropriate or unacceptable morally (or otherwise) for me to press perfectly (evidentially) good reasons on you. Or that make it appropriate or acceptable to give you some (evidentially) bad or phony reasons (maybe ones that will have some chance of swaying you to a position that is in fact conclusively supported by relevant reasons, some of which are not available to you due to ignorance or bias). I’m doubtful, but you might get an acceptability standard with the sort of content Estlund wants (as well as the right form) from the context of political debate and the same broad considerations that support Religious Toleration (and the generalized form of it that we might call Doctrine Toleration).
However, in drawing implications about “acceptability requirements” in political justification in general from Religious Tolerance and Doctrine Tolerance, I don’t think Estlund is appealing to the ethics of discussion and debate. (I’ve changed my interpretation here.) I think he is appealing to the idea that the circumstance of what we take our obligations to be contributes, in the decisive way suggested by Religious Tolerance, to the truth or falsity of relevant propositions about government coercion. And hence to the landscape of reasons (and their weighting) for getting at the truth in such situations. If this is right though, things look bad for acceptability requirements as I think Estlund is thinking of them (DEMOCRATIC AUTHORITY, especially Ch. 3, p. 43 “The Idea of Acceptability”). But maybe there is something that I am overlooking.