UPDATE: I've changed the term used to describe the fourth category in the taxonomy below from 'conceptual analysis' to 'conceptual reflection'. I hope the new term is better able to cover the many approaches suggested by commenters which did not seem to fit the original description in a straighforward way.
In light of the very interesting methodological discussions we’ve been having here at New APPS on the relations between physics and metaphysics, I’d like to put forward a tentative taxonomy of different strands within philosophical methodology. I suspect it can also be useful for discussions on the analytic vs. continental divide and its overcoming, which is also a recurrent theme in this blog.
Indeed, looking at past and present work in philosophy (and trying to be as encompassing as possible), it would seem that we can identify four main strands of methods used for philosophical analysis:
- Formal methods – these correspond to applications of mathematical and logical tools for the investigation of philosophical issues. As examples one could cite the development of possible world semantics for the analysis of the concepts of necessity and possibility, applications of the Bayesian framework to issues in epistemology (giving rise to so-called formal epistemology), Carnapian explication, and many others.
- Historical methods – they rely on the assumption that, to attain a better understanding of a given philosophical concept/problem, it is useful (or even indispensable) to trace its historical origins in philosophical theorizing. Of course, the study of the history of philosophy has intrinsic value as such (emphasis on ‘history’) but at this point I’m interested in what Eric Schliesser has once described as ‘instrumental history of philosophy’ (emphasis on ‘philosophy’).
- Empirical methods – these are the methodological approaches that systematically bring in elements from empirical sciences, such as the sciences of the mind (particularly relevant for philosophy of mind, epistemology, but to my mind also for philosophy of logic and mathematics), physics (possibly relevant for metaphysics), biology (arguably relevant for ethics, and everywhere else where evolutionary concepts come into play) etc. Sometimes this approach is described as ‘naturalistic’, but as we know there are (too?) many variations of the concept of naturalistic philosophy (many self-described naturalistic approaches are not sufficiently empirically-informed to my taste).
- Conceptual reflection – arguably the most traditional philosophical method, consisting in unpacking concepts and drawing implications, introducing new and hopefully useful concepts, problems, conceptual frameworks etc.
So we seem to have a plurality of methods actually being used for philosophical theorizing. Are they all equally legitimate and adequate, both in general and in specific cases? I submit that the correct response to this plurality is methodological pluralism.
But even methodological pluralism can come in at least two forms, which I will describe as the disjunctive and the conjunctive forms. Disjunctive methodological pluralism in philosophy would be the idea that different methodologies can coexist and be applied in different instantiations of philosophical theorizing, i.e. that there is no unique methodological canon that philosophical practice should adhere to. Conjunctive methodological pluralism, by contrast, is the idea that these different methodologies should not only coexist but should actually be combined in philosophical analysis. It should come as no surprise for those who follow my posts that I am a vocal supporter of the latter, which I also like to describe as an integrative approach. Indeed, I am convinced that an extremely wide range of philosophical issues can only be treated adequately with a combo methodology, precisely because each strand brings in a distinctive contribution to the investigation.
Thus, formal methods can not only add precision and clarity, but they also seem to allow us to draw non-trivial conclusions from assumptions, which would be hard or perhaps even impossible to attain with purely conceptual means (as I’ve argued here). Historical methods can have the very beneficial effect of bringing to the fore where our assumptions and ‘philosophical intuitions’ come from, and of isolating the other (possibly contentious) assumptions that they rely on (as I’ve argued here). Empirical methods help us keep our confabulations ‘in check’ and may increase the explanatory value of philosophical theories, taking them beyond the status of ‘just-so stories’. (John Protevi has an excellent phrase: philosophical theorizing should be empirically-responsible!) Finally, conceptual reflection plays the fundamental role of articulating these different elements in a (hopefully!) coherent and compelling whole.
Naturally, not all philosophical analysis will require the engagement of these four different perspectives at a given time, but I believe that most if not all philosophical issues would benefit from a combination of at least two of them. Conceptual reflection will always be present, but by itself it is a methodology that tends to lead to doxastic conservativeness and other ‘theoretical vices’.
As a final note, let me say a few words on how this proposed taxonomy could shed some light on the analytic vs. continental divide. So here’s an oversimplified but hopefully illuminating characterization of the methods used in each tradition: while analytic philosophy is essentially based on a combination of conceptual reflection and formal methods (arising from the confluence of Russellian formal methodology with Moorean conceptual analysis, as I’ve argued here), continental philosophy has as its two main pillars conceptual reflection (but of a very different kind!) and focus on the relevance of the history of philosophy for philosophical theorizing. Accordingly (?), analytic philosophers tend to be suspicious of the value of history of philosophy for strictly philosophical analysis (naturally, with many exceptions, but I take it that it is still the predominant position), and continental philosophers tend to be skeptical concerning the explanatory value of many of the formal methods consistently used by analytic philosophers. From this point of view, it might seem that there is indeed not much hope for these two traditions to be in a fruitful conversation with one another.
But here is where the fourth strand comes in, i.e. the empirical methods. In both traditions, focus on empirical methods is a rather recent and still non-mainstream tendency, but the good news is that in both traditions there is a wide range of philosophers seriously engaged with empirical sciences such as biology, psychology, cognitive sciences etc. So this new tendency could actually represent a bridge between analytic and continental philosophy, a common ground which philosophers in both traditions feel comfortable working on. Just to illustrate: while I find it difficult to enter the Deleuzian conceptual framework as such, being exposed to how my co-blogger John Protevi relates the framework to empirical findings that I am indeed familiar with (and use for my own ‘analytic’ philosophical reflections) makes it much easier to assimilate some of the Deleuzian concepts. Similar considerations can be applied to how someone such as Julian Kiverstein relates cutting-edge cognitive science to a Heideggerian conceptual framework, and how others (names don’t come to mind right now) do the same with Merleau-Ponty. I am not sure to what extent this generalizes, and certainly do not wish to imply that focus on empirical methods will erase once and for all the differences in methodology and style between analytic and continental philosophy. But I can at least offer the personal testimony that sharing a common ground of empirical results has made it much easier for me to converse with many of my ‘continental’ colleagues. (In fact, I find it much easier to talk to them than to the traditional analytic philosopher who rejects the relevance of empirical data for philosophical theorizing – but ok, I am myself a rather idiosyncratic, weird analytic philosopher!)
I’d be very interested in hearing what others may have to say on all this. How inclusive is my proposed taxonomy of methodologies? Is there a significant body of work that does not fit any of the four proposed strands? What about the distinction between disjunctive and conjunctive methodological pluralism?