I understand that we can show intolerance of certain religions via criticism, as Sam Harris or Richard Dawkins do. So that is not my premise.

There are probably a few premises I'm invoking in this article.

One premise is that many people have muddled thinking when it comes to tolerance, and that that thinking both resulted in - and has subsequently been encouraged by - the specific wording in the First Amendment.

Another premise is that, whenever a conflict arises within the legal code, where a method for resolving that conflict is not provided somewhere within the legal code, that conflict will have to be settled by the courts. And when conflicts are settled by the courts, that is ultimately a relatively undemocratic and unpredictable way of resolving those conflicts.

It's probably impossible to write a constitution that forsees all potential conflicts. But I think one best practice in this regard is for a good constitution to try to explicitly claim what principle has first precedence. That way, the courts will have a democratically chosen standard to go against, rather than merely whatever precedents have been decided within the courts.

In this section that you highlighted, I provided the "pursuit of happiness" as the principle of first precedence. It could be something else, like "freedom", or the "pursuit of food, water, shelter, and community".

But whatever we might choose as our first principle, I don't think the Constitution, as it is currently written, has done a good job of defining a first principle. And it's also not done a good job of defining what is meant by "religion" - regardless of what Jefferson may have intended (which again, leads to relatively undemocratic precedents being set by the courts). And I think the public's muddled thinking about tolerance and religion make it hard to even have this conversation in a very useful way - which is ultimately why I wrote this article.

Regarding the "paradox of tolerance", I read your article, and think I am in agreement with you. I'm not sure how you think I've overplayed it. Really what I've done here is view the question through a consequentialist lens, and from that lens, deemed the whole framework as simply not that useful.