[shrugs]. You construed riots in a sweepingly negative way as "blackmail". The fact that I do not agree does not mean I am construing them in a sweepingly positive way. This is as a pattern you have repeated throughout this discussion, and it illustrates how politics mindkills.

In this post JulianMorrison was, at least partially, trying to inform you that there is in fact something like a Bayesian Conspiracy within the Social Sciences - that there are social truths that are understood from within the discipline (or at least, from within parts of the discipline) that can't be discussed with outsiders, because non-rational people will use the knowledge in ways with a highly negative net utility. He was also trying to test you to see if you could be trusted with initiation into that Bayesian Conspiracy.

How do you and Julian know that you are indeed in the "inner ring" of this conspiracy and/or that it's actual purpose is what you think it is? How sure are you that this conspiracy even has any clue what it's doing and hasn't started to believe its own lies? Do you have an answer to the questions I asked here?

This sounds like an overly convenient excuse to avoid having to confront the implications of said truths. Restrict them to only people who won't ask awkward questions and tell everyone else pious lies.

In this post JulianMorrison was, at least partially, trying to inform you that there is in fact something like a Bayesian Conspiracy within the Social Sciences - that there are social truths that are understood from within the discipline (or at least, from within parts of the discipline) that can't be discussed with outsiders, because non-rational people will use the knowledge in ways with a highly negative net utility.

I don't think I'd identify the activist subculture with the social sciences, at least in the case JulianMorrison was talking about. If there's an academic community whose members publish relatively unfiltered research within their fields but don't usually talk to the public unless they are also activists, and also an activist community whose members are much more interested in spreading the word but aren't always too interested in spreading up-to-date science (charitably, because they believe some avenues of research to be suffering from bias or otherwise suspect), then we get the same results without having to invoke a conspiracy. This also has the advantage of explaining why it's possible to read about ostensibly forbidden social truths by, e.g., querying the right Wikipedia page.

Whether this accurately models any particular controversial subject is probably best left as an exercise.

Hold on a moment. I think the labels are accurate descriptions of the phenomena. There's hostility to this kind of discussion, so sometimes the only winning move is not to play. But if the labels (heternormativity, privilege, social construction, rape culture) are not describing social phenomena, then we should find accurate labels.

And if experts use the labels right, but [Edit: sympathetic] laypeople do not, then we should chide the laypeople until they use them right. Agreement with my preferred policies does not make you wise, because arguments are not soldiers.

In short, I think I win on the merits, so let's not get caught up in procedural machinations.

is more concerned with politics than truth...is more concerned with politics than truth.

And that's a bad thing?

If your goal is having an accurate model of the world, yes. If you're goal is something else, you're still better of with an accurate model of them world.

Edit: If you want to do politics, that's also important, just don't pretend you're doing science even "soft science".

I didn't say the humanities shouldn't be taken seriously (I hate the term "soft sciences"). The humanities study fields where scientific evidence is generally not available, thus they have to rely on other kinds of evidence. Unfortunately, this makes it easier to get away with sloppy work, BS, or even outright lies in those fields. This should be taken into account when assigning probability values to various statements.

I would add than a narrow focus on quantifiable data can be limiting, especially when you are researching culture and doing content analysis. Coding is a way to convert that content into numbers - counting mentions of words or themes - but that requires a lot of qualitative analysis to begin with and certain aspects are often lost in translation. Any social scientist worth their salt will take into account as many biases as they can and devise an experimental design to control for them as much as possible. But again, you have to be prepared to defend why you did what you did and human nature is complicated.

Having said that, I think a mixed qual/ quant designs are pretty great, especially when you have meta analyses to back you up.

If people insist on drawing inferences from what was never intended as a hint...what can you do?

'On hearing of the death of a Turkish ambassador, Talleyrand is supposed to have said: "I wonder what he meant by that?"'