When I was in grad school my advisor had a group doing interval arithmetic, which is explicitly giving the range of error in a computation given the limited precision of the computer. The Army Corps of Engineers would dump dirt into the Mississippi to keep it from wandering places where it shouldn’t go. Lots of dirt. They showed that the computation they did to determine how much dirt to use was basically producing random numbers because of the precision issue. So it doesn’t have to be a bad model to make the results garbage.
Bumped.
Still more to do:
https://www.cnn.com/2023/12/05/us/stem-gender-interest-gap-reaj/index.html
IME that is a cultural thing, plus blatant sexism does not help. But, as for the social sciences themselves, these days one uses (where appropriate) the same kinds of mathematical models and data analysis used, for example, in high-energy physics.
I’m only encountering this post now, and I have to say it’s a very well-thought-out post with an important point, not just about the limitations of classical STEM approaches to understanding the natural world but, more generically, our overall thought processes.
I have studied linguistics some and I think they are trying, but they are at the level of Aristotalian physics. It is hard. When I see how economists can differ on basic questions (Inflation is still out of control; no it isn’t. We’re in a recession; no we’re not, etc.) I realize it can’t be much of a science. My attitude towards sociology is summarized by the following joke.
The deans of arts and of sciences are having a stroll across campus. The dean of science explains that he has this wonderful department, mathematics. To keep a mathematician happy all you have to give him is a desk, a chair, a pencil, paper, and a wastebasket. The arts dean says that he can top that. He has a department, sociology, and all a sociologist needs is …
I’ll let you finish it. <<Ducks.>>
Thank you very much.
When I did my BA in history, the attitude in our department was that we were in the humanities, not social sciences.
Donald Creighton would have been proud.
The act of studying ourselves is intrinsically complicated. It should neither be approached with the goal of shoving it into the kind of positivist and objective approach that serves the STEM sciences so well nor disparaged for not being as tightly and narrowly channelled.
The relevant and viable criticism that can be leveled is a part of that complexity: that none of us can climb out of our context and study us collectively from the outside. We have vested interests and compelling reasons to act upon any knowledge of the social situation. We also have emotional and psychological investments already in how we perceive ourselves individually and collectively. And all of this tends to permeate our interpretation of any data, and despite occasional protestations to the contrary, science does not consist of raw data alone. Theory and interpretation of data are fundamental in all of science.
It’s not that we can’t do it, nor that we aren’t doing it right; it’s that we have to take all of the above into account, since we can’t dismiss it, and it makes all the conversations complex in nature.
Linguistics is a vast discipline and some of its fields lend themselves very well to quantitative research, testable hypotheses and falsifiable theories. Phonetics, computational linguistics, and neurolinguistics and for example.
Others are less clear-cut. Some of their theoretical concepts may be a bit fuzzy, but they still rely on objective description of real life phenomena, with some real degree of predicability. I’d argue that typology, comparative linguistics, phonology and morphosyntax fall in this category.
Finally, some fields such as semantics, pragmatics, and sociolinguistics seem to me to be much more speculative and interpretation-dependent.
To be honest, linguists can’t agree on what exactly a word is…
The way I remember joke is the mathematics department is cheap to run because all in it needs pencils, paper, and wastebaskets and the sociology (or philosophy) department is even cheaper because they don’t EVEN need wastebaskets.
…a history department from which the sociologist can steal hard facts and ideas and then make up a theory
…and if that history department doesn’t have the ‘facts’ that comport to your theory, just find another history department.
That is, if you can get past the marching genocide supporters.
Moderating:
Palestinian protesters, or whoever you are describing, are totally off topic for this thread. Don’t go there.
Of course that’s the joke. I was just afraid to finish it.
“The rest of the joke is left as an exercise for the reader.”
“Having reduced the joke to a known solution punchline, no further explanation is needed.”
I finished the joke, and it was the funniest joke ever written, but since there are no book margins online, I had to throw it away.
I somehow knew Trump and Pierre de Fermat were the same dude…
I told ChatGPT to rip off Tenacious D and write a poem about it:
The Funniest Joke in the World - Tribute
I once wrote the funniest joke in the world
It was so hilarious, it made me hurl
It had the perfect punchline and the best setup
It was so witty, it could make a grown man wet up
But alas, I forgot to write it down
And now I can’t remember a single sound
It’s gone forever, like a dream in the night
And all I have left is this poem to recite
This is not the funniest joke in the world
This is just a tribute
I wish I could remember the funniest joke in the world
But this is the best I could do
So please don’t judge me too harshly
For I know this poem is not very funny
But maybe you can imagine the joke in your mind
And laugh along with me, if you are so kind
This is not the funniest joke in the world
This is just a tribute
To the greatest joke that ever was
But I forgot it, what a buzz…kill
Thank you for listening to my poem
I hope you enjoyed it, even though it was lame
Maybe someday I will remember the joke
And then I will share it with you, and we’ll both go insane
I would think that the most basic premise when dealing with social science would be that people tend to behave and perform better when they like who they are or even like the direction they are going. This can go a lot of different directions but if we try to start from first cause and work outward from there it becomes easier simply because many of the problems down the line would disappear. For example, liking who we are needs to be relative to something else for comparison. If I had a father who was a drug addict and a player with basically no morals, he might become my model. I would quickly find out that this would not work in society but with nowhere else to go I might simply struggle with this as a model throughout my life. I would say that exploring the concept of modeling with or without religion would be a major step.