Philosophy by Counterexamples

In comments to a recent post, Ken Aizawa raised the following question:

      How is the development of a counterexample [to a philosophical theory] different than the 
      development of a falsifiable/falsified prediction of a scientific theory? 

I think this is a fascinating question that gets at the heart of contemporaty philosophical methodology, and I’d be curious to see what other people think.

It seems to me that there are three main kinds of philosophical enterprise:

1. Some philosophers attempt to analyze or explicate “ordinary” language and concepts (e.g., “folk psychology”).  For the most part, the resulting philosophical accounts should respect “ordinary intuitions” about cases.  That is, examples that intuitively go against the theory count against it.  Perhaps this is analogous to the way empirical evidence may disconfirm scientific theories.  Of course, there are many questions to be asked about the nature and stability of the intuitions in questions, but hopefully they can be answered.  And of course, such theories may tell us little about how things in fact are; they tell us something about how we ordinarily think about things, but it remains to be seen whether we think correctly about them.

2. Other philosophers attempt to analyze or explicate scientific language and concepts (e.g., the notion of space, or gene, or mechanism, used within certain sciences).  In such cases, presumably it is still possible to find counterexamples to the philosophical theories, but the source of relevant counterexamples is different.  In this case, the source of counterexamples may be found in certain legitimate scientific assumptions and practices that are not properly accounted for by the philosophical theory in question.

3. Yet other philosophers attempt to engage in theory construction more or less analogous to, or continuous with, scientific theorizing (e.g., Fodor on the modularity of mind).  In such cases, I think it’s pointless to argue that the theory is counterintuitive on the basis of “ordinary intuitions”.  For the only evidence that can truly disconfirm such theories is empirical evidence.

Ken’s question arose in response to a post on philosophical theories of consciousness.  Which of the above kinds of enterprise are philosophers of consciousness engaged in?  What kind of enterprise should they be engaged in?  Depending on the answer to this question, we get a different answer to the question of which evidence truly counts against the theories.


  1. I agree with Gualtiero’s 3-fold distinction.

    I add that the distinction between 1 and 3 corresponds to Clark Glymour’s distinction between two traditions in philosophy–the socratic tradition (which corresponds to 1) and the aristotelian tradition (which corresponds to 3).

    Additionally, experimental philosophy is obviously relevant for 1. It might be useful to spell out what the intuitions are (see, e.g., Knobe’s work on intentionality) or to cast some doubts on the philosophical interest of intuition-driven conceptual analysis (see, e.g., Stich et al.’s papers on the concept of knowledge).

  2. Ken Aizawa


    I agree that there is this rough kind of divide in philosophical methodology. There is also sometime some antagonism between philosophers of one stripe and those of another.

    What I was driving at was one respect in which it seems to me that they might have some kind of move in common.

    Philosophy concerned with developing intuitions and philosophy concerned with understanding science often both explore the consequences of intuitions/theories in an effort to refine/refute those intuitions theories.

    This was relevant to your earlier post, since you suggested that developing counterexamples to intuitions is a suspect business, where I would have guessed that you think aiming for scientific refutations is not suspect, but has a similar kind of structure.

  3. Doesn’t philosophy need to be more like math, and less like sciences?
    In my thinking, if philosophy searches for transcendental truths, it should work with something like conjectures (proposals for theorems) in math, and not theories.
    And it seems to me in such picture intuitions are the basic thing. But intuitions based on understanding (so, intuitions in Kantian sense), and not intuitions in the “I’m pretty sure that…” or “I feel that…” sense.
    So, in that sense, I would say counterexamples in philosophy, are like the pointing to a special case in math which doesn’t satisfy some conjecture. It works as showing that conjecture is false, without showing exactly (proving) why is it false.

  4. I like Gualtiero’s way of framing these questions and like also the threefold distinction. Weighing in on the “how should philosophers of consciousness proceed?” question, I’d say that only 2 and 3 are of any value. I think there are very few domains in which option 1 is worth much, and consciousness seems a poor candidate. Echoing a point that I think Lycan made in an earlier thread, the target domain is just too weird to expect folk intuitions to be anything other than a guide to further confusion. Asking the folk about what is or isn’t grammatical is ok, but asking them about what counts as consicousness is like asking them to explain how gyroscopes work.

  5. gualtiero piccinini

    Great.  So here is a way to express my worry about some of the literature on consciousness, including the discussion of some of the counterexamples:  it sounds like it’s engaged in something close to 1, while what I would be interested it is something more like 2 or 3 or a combination of the two.

  6. kenneth aizawa

    I think that it is common enough for philosophers to have a preference for one or another of these ways of doing philosophy.  What I am hounding you about is your resistance to the practice of exchanging counterexamples.  Prima facie giving counterexamples is at some important level like drawing attention to falsified predications of a theory.  It looks like counterexamples could be common to all three “modes” of philosophy. 

    Of course, I have not argued that giving counterexamples is methodologically the same as drawing attention to a falsified prediction.  That’s why I said something like, “Here’s a thought …”  and then asked a question.

  7. gualtiero piccinini

    I agree that exchanging counterexamples can be legitimate and fruitful in philosophy, and it bears an analogy with testing the predictions of a theory.  I just think that some exchanges of counterexamples are not fruitful, for instance, if one attempts to use a counterexample based on “ordinary intuitions” to refute a theory that aims at theory construction.  For it may well be that the ordinary intuitions are just prejudices.