I never metaphor, but I met a 7 once.
2 Likes
That one is pretty simple:
Criticism need not offer solutions, but it’s better when it does.
(I happen to agree with this.)
The units of certainty aren’t “entirely” subjective. The entire point of science is to give us a common language where we can decide together what we are certain of, rather than in our own heads. While there is some element of philosophy or politics in this, it’s not equally right to believe statements with p < 0.001 and to believe those with p > 0.999.
If ten years from now the weather reverts to a 100 year mean and all signs of temperature change and extreme weather events disappear, we ought to collectively lose faith in climate science, at least in the short term.
2 Likes
A p-value is a statement with respect to a given model; that is, it is a model/metaphor. The decision to respect that model as representative of reality is wholly subjective (as evidenced by our constant context switching with respect to the models we are currently employing, e.g., saying “the sun rises” instead of computing a Lorentz coefficient for the Sun’s mass and velocity). The point of common departure isn’t science, it’s language, which lets us share our construction. There is no common point of departure that represents “science”.
Shorter: “it’s not equally right” is a subjective valuation from your squishy meat-brain.
The decision to respect a model is wholly subjective. The respectability of a model is not.
You appear to be arguing that there is no underlying reality, or that if there is one, it doesn’t matter that there is one (i.e. we can’t access anything about it to tell whether something is true or not). I tend to assume there is one and it matters whether there is or not.
Science lets us share our construction by carving out a language to share it in, a language that is dedicated to precisely communicating about what we know about reality.
““it’s not equally right” is a subjective valuation from your squishy meat-brain” is a subjective evaluation from your squishy meat-brain. Or is that just what I have subjectively chosen to believe?
4 Likes
I’d be curious to know if this ‘data isn’t ideological’ thing is a relatively new invention/justification for laziness among data collectors; or whether it hearkens back to older ideas.
Off the cuff, I’m reminded very strongly both of the (discredited; but attractive) notions of “Baconian Method”, which was very optimistic about the possibility of more or less dispassionately collecting observational evidence without prior theory; and once you had done so, applying inductive reasoning from it; and of the endless and rather gruesome saga of people who just want Christianity to be based on a ‘plain reading’ of the gospels, without obfuscating interpretation.
One can see why the idea is appealing; but you don’t need a lit-crit PhD to see how tricky the process of obtaining a text’s ‘plain meaning’ without ‘interpretation’ is; and the number of people with different non-interpreted plain readings killing one another over those differences speaks to the difficulty in practice.
As much as it is sometimes used as a craven, or even overtly dishonest, cop-out “Don’t blame me, it’s just the algorithms; can’t argue with the numbers, man”; there is a long history of (doomed; but sincere, even fervent) desire for some kind of unmediated objectivity; whether it be through an idealized Scientific Method; esoteric mysticism, pinning your hope on direct revelation, and so on.
This doesn’t make me any more likely to think well of the smirking ‘big data’ kiddies selling consumer analytics solutions; but it does make me inclined to suspect that, even if they are just making excuses, some of their heavier math people may well be sincere, if misguided, in their hopes of a glorious dataset of pure objectivity. The tools are new; but the desire is old indeed.
2 Likes
Meditate a bit on what precision is. Even agreeing on what ‘precise’ is (or, if you will, when a p-value is ‘significant’) is subjective. Agreeing to be bound by a common set of precepts on what is good and what is the best way to understand the world is subjective. This doesn’t mean reality does not exist, it means the only organ we have for apprehending it is subjective. Objectivity is a temporarily shared delusion.
So. I believe only and entirely correct and moral things.
Naturally.
(Everyone believes this because, of course, if they thought their beliefs were incorrect and immoral they’d change them)
Is it the right thing for me to do to seek power by lying about my preferences and then, once I am in power, to act according to them because, of course, The Greater Good.
Before or after they are elected?
This attitude is what gave us what Mark Blyth calls Global Trumpism. We’ve decided that the people are dumb and don’t know what’s good for them and have decided that what they really need is a stiff dose of neoliberal restructuring and globalization. And you can’t vote against it because no matter what they might say during election season, every option offered is for it: the political caste, being our collective betters, have made their decision and ours is to shut up and do as we are told.
Except of course people don’t want to shut up and do as they are told, and so they vote in a way that says, nay, screams from the rooftops that they are sick and tired of the soi-disant elites. Hence: Trump. Brexit. Renzi’s failure. And in due course the ascendancy of AfD and Front National, more likely that not.
You misunderstand, I think, the function of democracy. Voting, really, is substitute for political activism via rifle, and it is absolutely meant to change things. Vox populi is meant to dictate to the political caste, no matter what they believe, because, in the long run, the people will have their say. It is the miracle of the modern age that the worst this produces is the odd brexit or Trump. Without this vital mechanism, it would produce civil war. And, historically, it has.
The modern day neoliberal nomenklatura has forgotten this function, forgotten the purpose of it all, and thinks it can just take choice away by applied paternalism. I hope they come to their senses before we all get a refresher course on popular revolt.
You might want to meditate a bit on whether someone who disagrees with you has necessarily thought things through less than you.
As I said, the agreement is subjective, the end result of the epistemology you subjectively choose is not. If you choose to determine truth by inhaling vapours from a vent in the Earth then you’ll have some interesting dreams. If you choose to determine truth by the scientific method then in a few hundred years you’ll go to the moon. [Though listening to oracles might have been an important step towards better methods of getting at truth.]
Or, using an epistemology that results in the technology we are using to discuss this matter, subjectivity is built out of the underlying objective reality (weren’t you the one talking about meat-brains a moment ago?).
2 Likes
A huge problem in any kind of social science, therefore any interview-based polling, is that people are lying shitbags. I’m mostly thinking of political polling right now, if you see the raw data before pollsters massage them into something more realistic, that’s just insane. So the situation in other fields of which I’m less familier is probably just as bad.
I hate to put my origin into every post noways, but it’s a relevant example. In the Viennese election last year, the far right FPÖ was expected to get something in lower thirties. End results about 32%. Raw data: 10%. If you can’t trust people - and you can’t - overly relying on the “data” becomes very dangerous.
Most pollster are weirdly good at their job though.
It’s a problem that people who actually do social science are very aware of, but I think we rightly doubt that companies using data to improve their products have hired on some sociologists and psychologists with lots of experience conducting studies on humans to help them navigate the pitfalls. And even then, the #1 thing any respectable expert would tell you is that even if you do it right, the data you get might just end up being nonsense.
2 Likes
This topic was automatically closed after 5 days. New replies are no longer allowed.
