Long ago I dove into science studies, which includes history, sociology, and philosophy of science. (Got a U. Chicago M.A. in it in 1983.) I concluded at the time that “science” doesn’t really have a coherent meaning, beyond the many diverse practices of many groups that called themselves “science”. But reflecting on my recent foray into astrophysics suggests to me that there may a simple related core concept after all.
Of course, but he’s certainly not an outsider to Econ. I guess I was just wondering if he’s using “fields that let relatively unknown outsiders to publish in their top journals” as a heuristic to judge “argument orientation”.
It's interesting to me that you singled out parts of physics and philosophy for praise, given that those are the two fields where you have published in prestigious venues (Journal of Political Philosophy and Astrophysical Journal) even though you are, relatively speaking, an outsider to those fields. Have you tried publishing in the journals of other fields and failed? Do you think your judgment would be different if you had been able to publish in those other fields?
I'm curious about the conditions that support "argument-focused" groups, touched on only briefly here. Some quick thoughts: (1) When we have information overload we must often defer to the experts / those with status. (2) Math is mentioned. I would think this field is open to argument because of its unique focus on verifiable proofs. Statistics is an adjacent field, but there we see a political-like divide between philosophical groups (Bayesian vs Frequentists). I'd expect less dependence on argument-like due to the ambiguity present here about what is the "right" way to do things -- and anecdotally I can say I've found blog posts from statisticians to have arguments that are much less pure than those discussed by mathematicians. (3) This discussion reminds of the "ethos pathos logos" framework for persuasion. This is 3d while this post discusses only two competing approaches.
Great topic! It always amazes me that folks use the word "science" as-if everyone agrees what that term means, but if you investigate what people mean then argument soon follows. Saying first what you mean about science really helps. I use the original Latin root *scientia* which generally means knowledge. That means methods that result in knowledge or fact, like data collection, interpreted meaning, and verified (or falsified!) result. Methods that result in knowledge (i.e., scientific method) are available to everyone; the physicist and poet, chemist and cook, musician and mathematician, etc. Science is this dynamic knowing.
“ think that practically everyone would claim to already be doing this.”If Herb Gintis is right, I can’t see how most of the people studying human behavior could claim this with a straight face. How could they, when their first cut criterion for whether a given model even gets a serious look is whether its authors drink the same brand of Kool-Aid?
I think that there is space for the concept of intimacy in this idea.
A community may be argument-orientated, but only with respect to its membership, because of their established trust, interpersonal bonds and shared characteristics. In such communities - outsiders, even if equal or high status relative to broader society are forced far more to rely on those credentials to be listened to and engaged with.
I don't think this trait is necessarily something to be admonished. There are costs associated with taking the time to listen to those outside your group... it's harder to communicate. Less efficient. It makes sense from this perspective to concentrate on the counter-arguments arising from within your group.
But it does not always lead to good outcomes either. I thought this - for example, with respect to how the philosophical community responded to the criticisms it received from physicists a while ago. I thought they were pretty dismissive. (just my opinion though).
I like the idea of supporting argument-orientation generally - but I think there is further nuance here wortth exploring.
One thing which might help, would be being explicit about the current hierarchy. When discussing an issue, and disagreement occurs, everyone explicitly states their current model of the dominance hierarchy. That allows everyone to say "so, yes, we might be being biased towards agreeing with Bob because he is more dominant"
Let us try to identify and celebrate the communities and topic areas where it is more common, and perhaps find better ways to shame the others into becoming more argument-oriented
I think that practically everyone would claim to already be doing this.
It's too much like saying "let us try to be more logical", or "scientific", "skeptical", or even "rational" or "bayesian". I fear that any attempt like this will suffer from language-creep, for various social reasons, and just end up another contemptible buzzword.
Not sure how to reconcile this with the fact that I agree with the rest of this post though, including that it is probably the best way to define "science" (though as usual, the norm in favour of doing experiments in scientific fields, as opposed to the field of english literature, helps)
What about careful observation -i.e. experiments and other equivalently careful gathering of information from the world? Isn't that the main indicator of science? Obviously there's also reasoning to go with the observations, but I'd think that without the careful observations, it isn't science.
In his latest latest Mindcape podcast, Sean Carroll hosts Herb Gintis who spends a decent chunk of his time arguing for a related claim: that the disciplines which concern themselves with human behavior (including psych, econ, bio, anthropology, and sociology) seem to have developed entirely incompatible models to try to explain it, and “don’t put nearly enough effort into talking to each other to try to reconcile their views.”(The quotes are from Carroll’s intro to this entertaining interview).
I understand what you are trying to get at but I don't think it's anywhere near that clear cut. In particular, the real difference between many of these communities isn't the extent to which status influences which position people believe but much more about whether the system as a whole faces strong pressure to converge on the right answer (eg high status folks risking rep if they don't accept certain arguments).
For instance, what differentiates continental and analytic philosophy isn't how much status matters or wether arguments matter more relative to speakers as much as the fact that structure of analytic phil forces the high status members to drill down in ways which systemically push towards truth while contenental phils can all just jump on the fashionable answer with no way to force them all to have same reasons or parcel that answer out into pieces small enough for their incentive not to lose status by looking dumb to outweigh how much they dislike that conclusion .
---
It's not like those of us in math read most papers. Rather, we believe claims about what has been proved almost entirely based on what the high status folks say.
Moreover, I suspect math would converge on the truth much more slowly if the subject didn't allow proofs to be broken up into pieces and the norms mean status can be gained by identifying which pieces work. Even if all the high status prod agree some claimed proof is false the low status prover can create situations where high status individuals risk status if they won't opine on whether lemma X is correct (peer review most clearly but also informal). If the high status folks disagree about which part is wrong now you've got high status on both sides. If they keep agreeing you can kinda do a binary search on the proof until you either identify the flaw or have such a short argument high status folks would lose status to deny it.
I didn't include this in my post yesterday, but Robert Trivers noted that the more social content there is in a field, the more the field is afflicted with self-deception. He highlights social psychology as being particularly afflicted with this compared to the rest of psychology, and sure enough that's where the replication crisis arose from.
He studied science/philosophy before econ, and as a social science implicitly doesn't seem to consider it to be as scientific.
Of course, but he’s certainly not an outsider to Econ. I guess I was just wondering if he’s using “fields that let relatively unknown outsiders to publish in their top journals” as a heuristic to judge “argument orientation”.
He has, of course, also published in economics journals.
It's interesting to me that you singled out parts of physics and philosophy for praise, given that those are the two fields where you have published in prestigious venues (Journal of Political Philosophy and Astrophysical Journal) even though you are, relatively speaking, an outsider to those fields. Have you tried publishing in the journals of other fields and failed? Do you think your judgment would be different if you had been able to publish in those other fields?
I'm curious about the conditions that support "argument-focused" groups, touched on only briefly here. Some quick thoughts: (1) When we have information overload we must often defer to the experts / those with status. (2) Math is mentioned. I would think this field is open to argument because of its unique focus on verifiable proofs. Statistics is an adjacent field, but there we see a political-like divide between philosophical groups (Bayesian vs Frequentists). I'd expect less dependence on argument-like due to the ambiguity present here about what is the "right" way to do things -- and anecdotally I can say I've found blog posts from statisticians to have arguments that are much less pure than those discussed by mathematicians. (3) This discussion reminds of the "ethos pathos logos" framework for persuasion. This is 3d while this post discusses only two competing approaches.
Great topic! It always amazes me that folks use the word "science" as-if everyone agrees what that term means, but if you investigate what people mean then argument soon follows. Saying first what you mean about science really helps. I use the original Latin root *scientia* which generally means knowledge. That means methods that result in knowledge or fact, like data collection, interpreted meaning, and verified (or falsified!) result. Methods that result in knowledge (i.e., scientific method) are available to everyone; the physicist and poet, chemist and cook, musician and mathematician, etc. Science is this dynamic knowing.
See my earlier comment to make sense of the Gintis reference
“ think that practically everyone would claim to already be doing this.”If Herb Gintis is right, I can’t see how most of the people studying human behavior could claim this with a straight face. How could they, when their first cut criterion for whether a given model even gets a serious look is whether its authors drink the same brand of Kool-Aid?
I think that there is space for the concept of intimacy in this idea.
A community may be argument-orientated, but only with respect to its membership, because of their established trust, interpersonal bonds and shared characteristics. In such communities - outsiders, even if equal or high status relative to broader society are forced far more to rely on those credentials to be listened to and engaged with.
I don't think this trait is necessarily something to be admonished. There are costs associated with taking the time to listen to those outside your group... it's harder to communicate. Less efficient. It makes sense from this perspective to concentrate on the counter-arguments arising from within your group.
But it does not always lead to good outcomes either. I thought this - for example, with respect to how the philosophical community responded to the criticisms it received from physicists a while ago. I thought they were pretty dismissive. (just my opinion though).
I like the idea of supporting argument-orientation generally - but I think there is further nuance here wortth exploring.
Sounds like a good application of status apps: https://www.overcomingbias....
One thing which might help, would be being explicit about the current hierarchy. When discussing an issue, and disagreement occurs, everyone explicitly states their current model of the dominance hierarchy. That allows everyone to say "so, yes, we might be being biased towards agreeing with Bob because he is more dominant"
Let us try to identify and celebrate the communities and topic areas where it is more common, and perhaps find better ways to shame the others into becoming more argument-oriented
I think that practically everyone would claim to already be doing this.
It's too much like saying "let us try to be more logical", or "scientific", "skeptical", or even "rational" or "bayesian". I fear that any attempt like this will suffer from language-creep, for various social reasons, and just end up another contemptible buzzword.
Not sure how to reconcile this with the fact that I agree with the rest of this post though, including that it is probably the best way to define "science" (though as usual, the norm in favour of doing experiments in scientific fields, as opposed to the field of english literature, helps)
What about careful observation -i.e. experiments and other equivalently careful gathering of information from the world? Isn't that the main indicator of science? Obviously there's also reasoning to go with the observations, but I'd think that without the careful observations, it isn't science.
In his latest latest Mindcape podcast, Sean Carroll hosts Herb Gintis who spends a decent chunk of his time arguing for a related claim: that the disciplines which concern themselves with human behavior (including psych, econ, bio, anthropology, and sociology) seem to have developed entirely incompatible models to try to explain it, and “don’t put nearly enough effort into talking to each other to try to reconcile their views.”(The quotes are from Carroll’s intro to this entertaining interview).
I understand what you are trying to get at but I don't think it's anywhere near that clear cut. In particular, the real difference between many of these communities isn't the extent to which status influences which position people believe but much more about whether the system as a whole faces strong pressure to converge on the right answer (eg high status folks risking rep if they don't accept certain arguments).
For instance, what differentiates continental and analytic philosophy isn't how much status matters or wether arguments matter more relative to speakers as much as the fact that structure of analytic phil forces the high status members to drill down in ways which systemically push towards truth while contenental phils can all just jump on the fashionable answer with no way to force them all to have same reasons or parcel that answer out into pieces small enough for their incentive not to lose status by looking dumb to outweigh how much they dislike that conclusion .
---
It's not like those of us in math read most papers. Rather, we believe claims about what has been proved almost entirely based on what the high status folks say.
Moreover, I suspect math would converge on the truth much more slowly if the subject didn't allow proofs to be broken up into pieces and the norms mean status can be gained by identifying which pieces work. Even if all the high status prod agree some claimed proof is false the low status prover can create situations where high status individuals risk status if they won't opine on whether lemma X is correct (peer review most clearly but also informal). If the high status folks disagree about which part is wrong now you've got high status on both sides. If they keep agreeing you can kinda do a binary search on the proof until you either identify the flaw or have such a short argument high status folks would lose status to deny it.
I didn't include this in my post yesterday, but Robert Trivers noted that the more social content there is in a field, the more the field is afflicted with self-deception. He highlights social psychology as being particularly afflicted with this compared to the rest of psychology, and sure enough that's where the replication crisis arose from.