Re: "In psychiatry, one can try to explain depression, say, by doing statistical analysis comparing people with and without depression. [... .] These efforts at explanation would be nomothetic."
This method would produce *empirical generalizations* about depression.
If one wishes to provide *causal explanations* of depression, I would distinguish two methods (and perhaps there are others):
(a) Identify *mechanisms* (ways in which things can happen). This is an abstract causal explanation.
Mechanisms aren't laws. Mechanisms are contingent. We scarcely know what contingencies are necessary and/or sufficient to trigger mechanisms in psychology.
(b) Identify a sequence of facts, events, mechanisms, and actions, each of which substantially increased the probability that the person would experience depression. This is a narrative causal explanation.
Where liberalism (maybe leftism is a better term for what I refer to) seems to most often go wrong, including wokeism, is when it is idiographic without being adequately informed by nomothetic knowledge and understanding. I think the same applies to psychiatry and macroeconomics.
It seems to me that the best approach would be to try to develop nomothetic "templets" that one could try applying the specific parameters of the ideographic episode under investigation seem to suggest.
I don't think it's true that true that textbooks don't apply their theories to analyze historical events. Certainly some textbooks use the Great Depression and inflation in the 1970s as examples of how theory works in practice. I'd agree not enough, but I think use of historical examples is pretty widespread.
Your psychiatry example might not be a good one since issues like depression tend to be specific to one type of society and not “always true” across cultures. I suspect that when social scientists claim they are using a nomothetic approach, they are mostly deceiving themselves or others with spurious claims to objectivity.
Biology uses nomothetic mechanisms to explain idiographic sequences. Social sciences study phenomena emergent out of biology and should work similarly, not drown in physics envy.
Wouldn't your psychology example only be nomothetic if the number of orthogonal variables considered in your thesis also describes other cases? Any function can be approximated (like a sample of random noise) by a series of orthogonal functions: like we do all the time with Fourier transformations-- https://en.wikipedia.org/wiki/Fast_Fourier_transform --where we can map a pattern match on truly random functions.
Creating patterns in random data is what astrologers did when they didn't even know the distances or types of stars (necessary to see the real pattern in our solar system and galaxy). We just codify such imaginary patterns into law as we did with all the "cancer cluster" observations blaming environmental pollution for clusters without a biological mechanism.
The distinction you describe in the two paragraphs beginning with "In psychiatry" seems to be miscast, an apples-to-oranges comparison. The first doctor is merely compiling statistics (and creating "explanations" that are probably useless even if accurate), while the second is trying to solve the practical problems of individual patients. The first is like a theoretical scientist, the second like an engineer.
Apply this to economics and I'd say the first guy is the macroeconomist and the second the microeconomist. And just as with the two psychiatrists, I'd listen only to the second and fire the first guy. I feel sorry for him, but his work is still useless.
Years ago I went to a lecture by Otto Eckstein at Harvard. He was running the big Data Resources macro model. He told how it was made and, as if this were the most natural thing in the world, that whenever it spit out a forecast he thought was wrong, he changed it (I can't remember whether he went in and tweaked the model or just changed the result). In other words, when you paid Data Resources for a prediction, you weren't paying for the fancy computer model. You were paying for Otto Eckstein's experience, his educated guesses and gut feelings. The computer had a nomothetic model but he made it more idiographic.
Re: "In psychiatry, one can try to explain depression, say, by doing statistical analysis comparing people with and without depression. [... .] These efforts at explanation would be nomothetic."
This method would produce *empirical generalizations* about depression.
If one wishes to provide *causal explanations* of depression, I would distinguish two methods (and perhaps there are others):
(a) Identify *mechanisms* (ways in which things can happen). This is an abstract causal explanation.
Mechanisms aren't laws. Mechanisms are contingent. We scarcely know what contingencies are necessary and/or sufficient to trigger mechanisms in psychology.
(b) Identify a sequence of facts, events, mechanisms, and actions, each of which substantially increased the probability that the person would experience depression. This is a narrative causal explanation.
Where liberalism (maybe leftism is a better term for what I refer to) seems to most often go wrong, including wokeism, is when it is idiographic without being adequately informed by nomothetic knowledge and understanding. I think the same applies to psychiatry and macroeconomics.
It seems to me that the best approach would be to try to develop nomothetic "templets" that one could try applying the specific parameters of the ideographic episode under investigation seem to suggest.
I don't think it's true that true that textbooks don't apply their theories to analyze historical events. Certainly some textbooks use the Great Depression and inflation in the 1970s as examples of how theory works in practice. I'd agree not enough, but I think use of historical examples is pretty widespread.
Your psychiatry example might not be a good one since issues like depression tend to be specific to one type of society and not “always true” across cultures. I suspect that when social scientists claim they are using a nomothetic approach, they are mostly deceiving themselves or others with spurious claims to objectivity.
Biology uses nomothetic mechanisms to explain idiographic sequences. Social sciences study phenomena emergent out of biology and should work similarly, not drown in physics envy.
Wouldn't your psychology example only be nomothetic if the number of orthogonal variables considered in your thesis also describes other cases? Any function can be approximated (like a sample of random noise) by a series of orthogonal functions: like we do all the time with Fourier transformations-- https://en.wikipedia.org/wiki/Fast_Fourier_transform --where we can map a pattern match on truly random functions.
Creating patterns in random data is what astrologers did when they didn't even know the distances or types of stars (necessary to see the real pattern in our solar system and galaxy). We just codify such imaginary patterns into law as we did with all the "cancer cluster" observations blaming environmental pollution for clusters without a biological mechanism.
The distinction you describe in the two paragraphs beginning with "In psychiatry" seems to be miscast, an apples-to-oranges comparison. The first doctor is merely compiling statistics (and creating "explanations" that are probably useless even if accurate), while the second is trying to solve the practical problems of individual patients. The first is like a theoretical scientist, the second like an engineer.
Apply this to economics and I'd say the first guy is the macroeconomist and the second the microeconomist. And just as with the two psychiatrists, I'd listen only to the second and fire the first guy. I feel sorry for him, but his work is still useless.
Years ago I went to a lecture by Otto Eckstein at Harvard. He was running the big Data Resources macro model. He told how it was made and, as if this were the most natural thing in the world, that whenever it spit out a forecast he thought was wrong, he changed it (I can't remember whether he went in and tweaked the model or just changed the result). In other words, when you paid Data Resources for a prediction, you weren't paying for the fancy computer model. You were paying for Otto Eckstein's experience, his educated guesses and gut feelings. The computer had a nomothetic model but he made it more idiographic.
Impressive use GWAS and polygenic score for an economist.