Why Isn’t There More Progress in Philosophy? Now that’s an interesting topic, and one that touches my explorations in more than one way. I’ve been to David Chalmers’ talk on the subject and thought that it would be a good idea to summarise it and add my own reflections at the end. This should be a good way to check the solidity of my ideas, borrowing the knowledge and experience of an eminent figure of our times.
The Annual Lecture at the Royal Institute of Philosophy this year was delivered by David Chalmers (13/12/2013). His talk was called, “Why Isn’t There More Progress in Philosophy?”, it was filmed and has been uploaded on the Institute YouTube page. I was very curious, because Chalmers is a very important figure in the field of consciousness and because he is a puzzle to me: I disagree with the direction he’s following on philosophy of mind, but even more, I think that his famous definition of the “Hard Problem” has been and still is a major obstacle in the field. By this I mean that it’s preventing lots of people from taking courage and tackle the problem of consciousness with a fresh and optimistic attitude. This happens because Chalmers argument is very sound and thus acts as drag for those that are not brave enough to challenge it directly. You can probably appreciate that this feeling of mine relates directly to the subject of his lecture, so I was eager to hear his own views.
The lecture was excellent and Chalmers was able to address almost all of the ideas that crowded my own mind, plus plenty more, with clarity, wit and remarkable rigour: I had a jolly good time. In this post I will summarise his main points, and since he did leave some questions open, I will propose my own answer, in the light of what I’ve discussed so far (as it happens to fit quite well). I also have one minor criticism that will include in the next section.
When addressing the title question, the first thing to do is to define “Progress”, Chalmers admitted that there are many possible definitions, but proposed one that has the advantage that it can (to some extent) be verified empirically. The definition is something like this:
Progress can be seen as large collective convergence towards the truth.
[Note: I wasn’t taking notes, so all of my quotes are either approximate reconstructions from memory or taken from this paper that follows the lecture contents closely]
Collective convergence is needed because we want the progress to be appreciated by most, if one solitary philosopher finds some important truth but nobody knows about it or most disagree, the progress will have no or very limited effect. Collective convergence is a necessary and not sufficient condition to fruitfully proceed. Admittedly, the “towards the truth” condition is much trickier, not because it’s a controversial condition, but because it’s hard to verify (and this has a lot to do with my own comment below). Most of the lecture was dedicated to the lack of collective convergence: this can be defined in measurable terms (with caveats) and verified empirically. Once this lack is accepted one can conclude that indeed Philosophy shows very little progress and can also ask “why does convergence not happen?” and also “should we define progress in other ways?”. Chalmers left both questions open, although he did propose plenty of possible answers.
“Collective convergence” can be defined in relative terms, by looking at other fields, and thus Chalmers asked:
When addressing its own fundamental questions (the mind/body problem, is there a universal good, the nature of reality and so on) why does Philosophy not show the level of agreement analogous to that reached by the hard sciences? [in his own terms, that’s Physics, Chemistry and Biology(!?)]
Before answering this question, he did verify his claim: he sent a questionnaire to a word-wide range of philosophers, got enough replies and could indeed verify that most big questions show very little agreement. [With the exception perhaps of the God question: it turns out that 73% of the respondents are Atheists or leaning towards it]
The survey results are published here and discussed in this paper (What Do Philosophers Believe? David Bourget and David J. Chalmers, 2013). In other words, we can accept the empirical hypothesis and claim that there is very little convergence on the major philosophical questions; this makes the hard question about the convergence towards the truth redundant. In some twisted way, we are lucky and can conclude that, according to our current definition of progress, Philosophy does indeed show very little of it.
While discussing all this, Chalmers frequently pointed out that to be absolutely sure, one should go back in time and ask philosophers what they see as the main philosophical questions, ask the same to scientists, and then compare the convergence 10, 100, 200 years on. In his words (taken from the summary paper):
To properly address this issue, we would need analogs of the PhilPapers Survey in (for example) 1611, 1711, 1811, 1911, and 2011, asking members of the community of philosophers at each point first, what they take to be the big questions of philosophy, and second, what they take to be the answers to those questions (and also the answers to any big questions from past surveys). We would also need to have analogous longitudinal surveys in other fields: the MathPapers Survey, the PhysPapers survey, the ChemPapers Survey, the BioPapers Survey, and so on. And we would need a reasonable measure of convergence to agreement over time. I predict that if we had such surveys and measures, we would find much less convergence on answers to the big questions suggested by past surveys of philosophers than we would find for corresponding answers in other fields.
This seems very logical, but does not consider a pretty obvious consideration: science as we know it it’s very young. Even 150 years ago you would find it difficult to identify people who defined themselves as practitioners of one science branch alone, and if we go further back also the distinction between science and philosophy becomes more and more blurred, to the point that I’m inclined to believe that it would be impossible to get separate answers for philosophy and physics if one goes back to 1611 and beyond. This may well be a pointless technicality (after all, we can’t go back and ask), but it is actually relevant, for reasons that I hope to clarify in my own comment (below).
Anyway, Chalmers did conclude that there is little progress (as defined) and that therefore, if one wants to find a more positive outlook, it may be a good idea to alter the initial definition of progress, and see what happens if one drops one or more parts of it (specifically, the “large” attribute of convergence, and/or the “collective” one, and/or the “towards truth” constraint). If I recall this correctly, Chalmers admitted that this is possible, but also that it is not satisfactory, at least for him: he would like to see large collective convergence towards the truth, that’s why he does Philosophy.
The consequence is disappointing: one can only try to understand why the progress is so scarce and see if this understanding can allow us to remove some obstacles. What are the possible reasons? Chalmers offered a wide range, explained in full in the corresponding paper linked above. For brevity, I will only list them (quoting directly):
Powerless method. There is no convergence because philosophy does not have a method with the power to produce convergence (while science, maths and other disciplines all have their own tools to settle controversies).
Anti-realism. There is no convergence to the truth because there are no objective truths to be had in the relevant domains.
Verbal disputes. There is no convergence to the truth because participants are talking past each other.
Sociological factors. There is no convergence to the truth because although the problem has been solved and some know the answers, sociological factors have kept others from agreeing.
Unknowability. The problems are unsolvable in principle.
Human inadequacy. There is no convergence to the truth because while the problems are solvable in principle, they are too hard to be solved by us.
Current nonideality. There is no convergence to the truth because we are not yet reasoning as well as we can, or because we have not yet had insights that are within human grasp.
As far as I can tell, this list covers all possibilities, although I would probably collapse points 3 and 4 into one single entry. In the talk he also included another possibility:
Self selection. The big problems in philosophy remain unanswered because philosophy is about questions that can’t be answered in a definitive way.
I think this option does not appear in the paper excerpt because it can be seen as an overarching, meta-option: one that, for different reasons, may apply to most or all the primary possibilities listed above.
In conclusion, I am not entirely sure what the intended take home message was, and maybe this happened on purpose. Chalmers did leave the discussion open on two questions:
a) Are the options above both necessary and comprehensive? Can we exclude some or add some more?
b) Assuming we have identified the causes of the problem, how do we work around them?
This kicked off the Q&A session: it was interesting, but I got the impression that we were all circling around the main point without hitting the (obvious and quite large) nail on the head. I timidly lifted my arm to offer my opinion, but far too late, and lost my opportunity. This post tries to remedy my mistake, or at least to clarify my thoughts before they fade away.
The summary above is of course incomplete, but I hope it will be enough to make my point clear. I think that Chalmers did an excellent work, his approach is sound and promising and the final questions are well worth some crowdsourced answering effort, so here are my 5p.
I think that self-selection is the key, especially if applied to professional philosophy in academia, and to the set of “big questions”. Specifically, as I’ve said before, one can define philosophy as:
The activity of trying to figure out things without the aid of empirical verification.
In the current context, this definition is illuminating. Controversy is inevitable, because “by definition”, whatever conclusion one reaches can’t be verified post-hoc. Therefore we are dealing with a very specific and circumscribed case of self-selection. Of the other points, the first seems to be a variation of the self-selection one: philosophy is concerned with topics where no controversy-killing method is available. The others will apply in turn, hopefully not falling into case 6 too often.
To solidify my position I need one more argument about the “obvious consideration” that I’ve added in the description of the talk. If I’m right, and philosophy and science used to be one single discipline, and they got separated only gradually and following the refinement of scientific method(s), does this tell us something about self-selection? I would believe that yes, it does. In fact, I see it as a strong indication that self-selection does indeed happen, and is a gradual process where, as new empirical methods emerge, they gradually erode the philosophical enquiry space and redirect one or more “big questions” in the area of one or the other sciences.
I could support my statement with plenty of historical examples, but instead, I will use a current topic that is dear to both Chalmers and myself. Philosophy of Mind, and more specifically the question: is the mind separated from the body? Or, is our mind actually one and the same as our brain? Suppose that you make a philosophical breakthrough that states: “the mind is definitely not the brain”. I would disagree, but can I disproof this claim? No, not at the moment. Why not? Because I would need to be able to create a functioning brain from matter: if I could do so without adding some special immaterial ingredient, I could kill the controversy. Conversely, if another philosopher makes the opposite breakthrough, and concludes “the mind is definitely the brain”, some philosophers will strongly disagree, but will be able to offer exactly zero definitive evidence to support their claim. Hence, the question remains open, and it does because (as I’ve argued over and over) pure thought alone can’t provide definitive answers, and instead is often able to move us further away from truth (as I claim about cognitive attraction and moral reasoning). To ensure that our thoughts won’t mislead us, we can only use empirical evidence, and that’s all.
My conclusion however is quite optimistic: considering where it stands, philosophy is doing very well. I can say this because the consequence of self-selection, combined with the (arguably sad) need to accept that rationality alone can’t reach any definitive truth about reality (it can reach some definitive truth only on abstract matters), is that philosophy does not aim to reach definitive answers (in my own language, I’d say that it is condemned to remain entirely volatile, because if/when it moves toward solidity it is already science), and that therefore we need another way to measure its progress. I have one, but this piece is already long enough, so I won’t discuss it in full.
What is philosophy good for?
One very important thing: it’s a good, very efficient, and very useful front-runner. It explores the space of possible realities, takes some premises (pretty much any premise that may be right but can’t be verified yet) and computes the consequences (note that Chalmers did mention this view in his talk). By doing so, it generates hypotheses and their consequences, opening the way for science (and other disciplines, if one wants to see science in a restrictive, ‘classic’ way). In this view, science plays the role of the late-comer dispute-arbiter. In its own time, when the right tools of enquiry are made available, science can have a look at philosophical ideas, see if the new tools can settle one or the other controversy, and stupidly do the empirical verification. But in fact, the role of philosophy is even better: with some luck, it may also generate conceptual tools that can be used scientifically. Take my example on the body-mind problem: if someone will be able to generate a complete theory of how matter could generate a mind (or be moulded to become a mind), in the absence of experimental verification (e.g. when this theory is only an idea), it would be a purely philosophical effort. But the result is a testable hypothesis, and surely it would be tested empirically. In this sense, philosophy is the twin mathematics, able to generate new scientific tools and hypotheses. Hence, to measure the progress made, one can only look backwards, and ask two questions:
– How many times has Philosophy generated ideas that were later scientifically proven to be either right or wrong? Scientifically, being able to demonstrate that one hypothesis is wrong is important, maybe even more important than proving that something is right (assuming the latter is possible, and in my opinion, it isn’t).
– How many times has Philosophy generated the conceptual tools that made a subject empirically explorable?
My guess is that the answer to both questions is: many times indeed. But I’m not qualified to answer: I know very little of the history of philosophy, so I should probably end it here… Bye.