Here it is, a rundown of interesting articles I’ve come across recently, to consider for your weekend readings:
As science advances, does Ockham’s Razor still apply? The notion that the simpler explanation is usually right has been useful for centuries. New science may change that. (Salon)
14 common Sophistical tricks Aristotle already “called”, still with us today. (Medium)
Wilfrid Sellars and twentieth-Century philosophy. (Notre Dame Philosophical Reviews) [For background, see here.]
The inheritance of nations. To what extent does a work of art belong to the people of the world? (The American Scholar)
Archaeology, architecture, and “Romanizing” Athens. (OUP Blog)
Hi, Dr. Pigliucci.
The first article you shared reminded me of something i wanted to ask you.
I read – and really enjoyed – your book “Nonsense on Stilts’. It got me interested in baysesian reasoning. I was surprised to discover though, that people like David Deutsch doesn’t like it very much, to say the least.
But i’m having a hard time understanding what the critique really is (maybe because english is not my first language, and i’m not a philosopher and formal student of it). Anyway, you’re someone whose opinions i came to trust and respect, given how insightful what you write is. So, i would like to ask your opinion on this issue. I mean, why are some philosophers so against bayesian reasoning? Is there any weakness in it you hadn’t space to deal with in your book and of which i should be aware? I appreciate any thoughts you may share.
Thank so much for your work.
Liked by 1 person
Natã, ah, that’s a long story, maybe I’ll write a separate essay about it. There are known issues with Bayesianism, especially when we are talking about subject priors. And then there are people who don’t like the approach because it doesn’t yield results favorable to their point of view.
A good and balanced discussion can be found here: https://plato.stanford.edu/entries/epistemology-bayesian/
Like
Re Bayesianism, and subject priors and related issues, I think the big issue is one of subjectivity. I think 6.2.A and even more, 6.2.F, on your SEP link, the second even talking specifically about “subjective Bayesians” vs “objective Bayesians,” gets at this.
==
Plus, per one of your former blog incarnations, and a certain Brit named Coel who regularly cited the Bayesian handwaving genius of mythicist Mark Carrier … and, of course, Carrier’s actual handwaving via Bayesian claims … keeps me skeptical about the whole project.
Liked by 1 person
Wiki’s entry on Bayesian probability, first paragraph, perhaps gets at this even better:
“Bayesian probability is an interpretation of the concept of probability, in which, instead of frequency or propensity of some phenomenon, probability is interpreted as reasonable expectation[1] representing a state of knowledge[2] or as quantification of a personal belief.[3]”
==
Better yet, look at Wiki’s piece on Bayesian statistics:
“Bayesian statistics is a theory in the field of statistics based on the Bayesian interpretation of probability where probability expresses a degree of belief in an event. The degree of belief may be based on prior knowledge about the event, such as the results of previous experiments, or on personal beliefs about the event.”
>>>Where probability expresses a degree of belief in an event.<<<
Or, re Carrier the mythicist, a degree of non-belief in an event.
https://en.wikipedia.org/wiki/Bayesian_statistics
Liked by 1 person
Right, but that’s one of the things I actually like about Bayesianism: it takes serious that the frequentist approach, where probability is defined objectively, is highly problematic, and bites the bullet directly.
Liked by 1 person
Third comment for tonight, before I forget.
Wiki’s piece on Bayesian probability notes that Laplace did work that would today be considered Bayesian probabilities.
Well, he’s an astronomer.
Carrier is someone pretending to be a theologian or something.
Even if he were actually one, though?
That’s a humanities area.
Humanities and social sciences — sociology, psychology, anthropology, linguistics, theology and philosophy and more — have a variety of assumptions, some of which are currently not empirically verifiable, along with other things like demarcation problems, that Laplace’s astronomy and other natural sciences don’t really have.
String theory, even, can illustrate this. It’s generally not empirically verifiable, but it is based on a mathematics-like structure. String theorists propose 10 dimensions, or is it maybe 11 in some cases. But, as far as I know, there’s nobody proposing 194 dimensions.
Whereas, to take languages, linguistics and language families. What all constitutes a language family? How big or small are specific families? “Clumpers” like Joseph Greenberg vs. “splitters” show this is in part a demarcation problem. And, that spills into anthropology. If all American Indian languages except the Na-Dene and the Inuit-Aleut are one family, does that mean that their progenitors were all, essentially, one migration? That can then tie to archaeology, when new ruins are discovered or whatever. How one reframes one’s priors is going to be “belief” indeed.
Liked by 1 person
Thank you for your comments, Dr. Pigliucci.
It would require much gullibility of me to think i could find an idea against which would not exist objections. And i agree we all should be careful before asigning exact percentages to abstract propositions. Furthermore, as any other tool, we will find people using it inapropriately.
Nonetheless, it provides a clear way in which to frame rational thinking (the ideia that the likelyhood of finding the same evidence in other hypotheses is part of the equation, for example).
Liked by 1 person
I agree with the idea that objective probability is problematic. That said, that means that different persons, in viewing the same situation, can have different takes on exactly how subjective a specific prior probability estimate is. I think this partially connects to my humanities and social sciences vs natural sciences, where the former have less concrete of “anchors,” if that makes sense?
Liked by 1 person
It does make sense. However, one thing critics of subject Bayesianism tend to forget is that subject priors, no matter where they start, will converge to the same posteriors, once enough iterations have been done. In other words, for a Bayesian algorithm, truth will emerge with sufficient iterations of empirical input.
Liked by 1 person
Per your last comment, Massimo and per the SEP entry, this gets back to Ye Olde Problem of Induction … and SEP mentions Goodman and the New Problem of Induction, too.
A further thoughts:
1. What is “enough iterations”?
2. What about scenarios that (since neither you nor I are Dan Dennett) can be run algorithmically?
QED: Is this not, at least tangentially, related to Ye Olde Division Between Academics and Pyrrhonists?
Liked by 1 person
“Enough iterations” is actually defined rigorously, so it’s not as wishy-washy as it may appear. But the specific number depends on both how far off the initial priors were and on the quality of the incoming empirical updates.
I’m not sure what you are asking at point 2.
Like
That should be “cannot be run algorithmically” on No. 2!!!! Makes no sense as originally written.
Like
Hi Massimo,
Interesting reading selection last week, and I always enjoy articles like the one on the 12 “sophistical” tricks. But let me vent a moment …
>The sophists themselves were of course defined by Plato by their alleged intention to say “truthy” but deceptive things, in order to win friends and influence people. Aristotle agrees with his teacher here, opening the SR by defining “the art of the sophist” as “one who makes money from an apparent but unreal wisdom.” He’s in it for the cash, the glory, and the advancement. And for this, the mere semblance of truth or rightness is enough.
… I find it ironic all sophists it seems are getting painted here with the same brush.
—–
By the way, I’m looking forward to reading Answers for Aristotle! I just bought it on Kindle.
Liked by 1 person
Marc, yes the Sophists do get a bit too much of a bad rap. Here is my take on it: https://www.patreon.com/posts/protagoras-we-21959417
I hope you’ll enjoy Answers for Aristotle!
Liked by 1 person
Thanks for your link on Sophists Massimo!
Liked by 1 person