= 27d567bc2d48d9f40e9ccc20ea370b15)
johnlawrenceaspden comments on The Thyroid Madness : Core Argument, Evidence, Probabilities and Predictions - Less Wrong Discussion
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
Comments (132)
So, my current goals are to:
(a) sharpen this argument until I either believe it or not (I am dithering)
Conditional on (a) being possible, and the 'multiple-hormone-peripheral-resistance-cochran-cause' thing looking solid:
(b) make the idea widespread enough to get people capable of working out the details interested.
(c) put such a rocket up 'medical science' that it will turn into a science
(d) wipe frequentist statistics and the associated false inference techniques off the face of the planet (except as a discipline of pure mathematics, where they are as deeply and eternally true as everything else)
You can tell that medical science isn't a science because it's got the word "science" in the name. We'll have to think of a new name for it. In Ancient Greek, obvs.
and minor goals:
(e) Nevertheless to save the reputations of RA Fisher and Archie Cochrane, great heroes of philosophy both.
(f) Establish the reputations of Barnes and Lowe, serious geniuses who spent their lives on this.
(g) Try to stop the blame falling on the people who introduced the TSH test. They were trying their best on a very hard problem. It is not their fault that their discipline was not strong enough to catch them when they fell.
You should believe (or not) a claim based on how well it matches empirical reality (aka "evidence"), not on how sharp the argument is. Sharpening is useful insofar it narrows down the range of relevant evidence, but it still does not replace it.
Yeah, well...
Misuse is not evidence of uselessness.
I'm not sure about this. I think this was always the problem I had with quantum mechanics at university. You can't catch them out, but they still don't make sense. That was what first drew me to less wrong, years ago.
I mean, could many-worlds and copenhagen both be true at the same time? I think not, but I don't know why, and I think I'm making a claim that I don't understand, and which might be meaningless, and I don't know how to tell.
At any rate, at the moment, with this particular idea, I seem to be able to explain too much. So I'd like to add extra constraints, and see how complicated my ideas have to become to match what I know.
Oh dear, is that what I sound like? It's hard to tell from the inside. The thing is, I think they've got a few interesting things to say, but it's so badly written that I literally can't be bothered to read it closely enough to find out. I think their argument needs sharpening.
The post-modernist claim of 'I have nothing to say and wish to communicate this' seems both relevant and reversed.
Some of the frequentist tricks have their place. But they do seem very prone to misuse.
Not quite, actually the opposite -- you expressed desire for medicine to become more like a science and my link showed you some reactions to this idea.
Yeah, and I should totally read the damned thing. Your enemies' arguments are the most important source of ideas. I need a microfascist to stand on my left shoulder and point a gun to my head.
I don't know about needing to read it. The relevant point is its existence, the details of arguments might be amusing but hardly informative. It's not a paper about medicine, it's a paper about fighting oppression.
And actually, if there's this widespread attitude in medicine that only PCRTs and long-held-expert-opinion matter, and every other form of evidence should be discounted in the name of science, then maybe what they conceive of as 'science' is acting oppressively.
PCRTs rock for solidity, but they are hella blunt tool.
long-held-expert-opinion also rocks, but it does not come from PCRTs!
Anecdotes and clever ideas are tricky-as-all-hell, but that doesn't mean you should ignore them.
I don't suppose you've read that wretched paper and can summarise it?
I've paged through the paper.
Blah blah microfascism blah legitimacy of truth blah Cochrane hierarchy blah Jacques Derrida blah hegemonic discourses blah subjugated knowledge blah 20th century totalitarianism blah Orwell’s totalitarian vision blah we must resist the totalitarian program blah blah blah.
Money quote: "Those who are wedded to the idea of ‘evidence’ in the health sciences maintain what is essentially a Newtonian, mechanistic world view: they tend to believe that reality is objective, which is to say that it exists, ‘out there’..."
They likely don't make a good case for their opposition to the focus of objectivity, but the focus on objectivity is not without problems.
On example would be placebo-controlled drugs, in standard trials it doesn't matter whether people who they get a placebo or an active drug know what they are getting because that's subjective knowledge. The thing that matters is whether an objective third-party thinks the experiement tried to control for placebo.
As far as microfascism goes, the law that's in effect that forbids me from healing other people was passed under Hitler.
Sense make not.
Double-blind trials are double-blind for a reason. And subjective knowledge can certain affect objective outcomes, that's not an issue.
...and so what?
Thanks, maybe I'll save it until after I'm dead.