I had the exact same thought. I guess I'm glad I didn't post it as a top-level comment, because it looks like you have unintentionally become an example of how HN is prone to downvote swarms from those same conservatives who feel so victimized on Twitter.

One key here is the limitations of a "neutrally applied" standard. As others have pointed out, an alternative hypothesis is that the liberal majority is more likely to report violations. This can happen through sheer weight of numbers, it can mean completely neutral application of the standard but non-neutral invocation of it, and the author does nothing to address it.

Another confounding factor has to do with the very nature of conservatism and liberalism. One could argue, and conservatives do argue in other contexts, that the very definition of "hate speech" and the degree to which it is suppressed have a very liberal bias. Those who believe in those standards are more likely to invoke them, even if the two groups are equal in number. Those who do not believe in those standards are more likely to transgress. This is a difference in both invocation of the standard (by users) and its construction, but still not its application (by Twitter's mod team). Again, the author does nothing to address it.

AFAICT, what the author is demanding is equal outcomes despite unequal factors leading to those outcomes. It's an odd position for a conservative to take. Does Richard Hanania believe such affirmative action is justified in other contexts where a particular neutrally applied decision process yields unequal outcomes because of discrimination skewing its inputs? As always on deliberately "contrarian" Quillette, it seems like there's a bit of inconsistency here.