Skip to content

Projects and Drafts

As of now, I mainly worked on consensus formation, more precisely on the question "On which conditions is it impossible for two subjects to rationally disagree?". (See question 1 from theoretical philosophy in my Research Interests.) In particular, I consider two methodological approaches: the Bayesian approach and the method of reflective equilibrium. In the future, I want to broaden the field, though. Suggestions are welcome!

Bayesianism / Accuracy-first Epistemology

Bayesianism gives general constraints for rational degrees of belief. Especially the more objective variants significantly narrow the room for rational disagreement. However, despite Bayesianism's many applications, its normative foundations have always been shaky. Accuracy-first epistemology is a growing research programme trying to remedy this by arguing that Bayesian degrees of belief are systematically more accurate than others. To get an excellent overview on this, click here for the associated SEP entry.

My work on accuracy-first epistemology has been mostly critical. Even though the programme is very attractive, I think it has been unsuccessful so far and it is doubtful that it can be successful at all. The main problem, I believe, is its inability to justify constraints on accuracy measures that are powerful enough to deliver the desired results. Also, there may be some vicious dialectical circles when it comes to arguing for the special epistemic status of accuracy. I will soon upload drafts of papers explaining these worries in detail.

Reflective Equilibrium

The method of reflective equilibrium goes back to Nelson Goodman, who wanted to use it for justifying theories of deductive and inductive inference. It is by some considered to be the method of philosophy (e.g. David Lewis) and by others the method of scientific justification in general (e.g. Catherine Elgin).

The notion of reflective equilibrium is tightly connected to the notion of coherence. In fact, I believe, the former is simply one way of spelling out the latter. This may lead some to think that the method of reflective equilibrium presupposes a coherentist view on justification and, thus, that it falls prey to the objections to such views (Elgin disagrees). I am undecided on this matter. However, regardless of whether coherence is sufficient for justification or rationality, it certainly is necessary.

Thus, coherence gives (more or less strong) constraints for rationality, thereby narrowing the room for rational disagreement. However, reflective equilibrium as a very promising way of spelling out coherence is most often left rather vague. I am part of a research project trying to make both reflective equilibrium and the associated method formally precise to study its properties and power. The project is based in Karlsruhe (Germany) and Bern (Switzerland). For the website of the project, click here. For a detailed informal description of the method of reflective equilibrium (with which I agree for the most part), see Baumberger and Brun, 2020, "Reflective equilibrium and understanding", Synthese, DOI 10.1007/s11229-020-02556-9.