Anti-naturalism in philosophy (II)

Here are a few more notes on the debate between Paul Horwich and Timothy Williamson which I discussed briefly in my previous post on this site.

In the video, Horwich defends his view, based on his reading of Wittgenstein, that much of what goes on under the name of (analytic) philosophy is "irrational". (Williamson calls a spade a bloody shovel and interprets him as saying that it is "rubbish".)

Horwich starts talking about 39 minutes into the video, and a brief statement of his position starts at about the 42-minute mark. His key contention is that the common philosophical supposition that abductive reasoning from intuitive data can reveal fundamental regularities (more or less as science can be seen to reveal fundamental regularities which underlie empirical data) is simply false.

He summarizes the basic argument he makes for his position in five points. Here is my summary of his summary:

1) Our concepts (I assume he means linguistically-based concepts) tend to be messy. They evolved to deal with a wide range of real-world contingencies and are used in a bewildering variety of ways. The intuitive data upon which philosophical reasoning is based is inextricably bound up with such concepts. In fact, many of the key concepts with which philosophers have traditionally been concerned are particularly prone to conflicting understandings and interpretations.

2) Via conjectured theoretical entities, models, etc., the various sciences have developed effective ways of dealing with messy empirical data to arrive at (often) elegant and relatively simple theories which reveal fundamental regularities.

3) The currently dominant strand of analytic philosophy also seeks elegance and simplicity, but any simplicity obtained is generally obtained "on the cheap" by finding a simple regularity that fits most of the (intuitive) data, and counting intuitions or interpretations which do not fit the theory as mistaken or incorrect.

4) This is a distortion of the scientific method.

5) Not all philosophers make this mistake, but a beautifully simple theory will rarely be obtainable. Almost inevitably there will be a profusion of alternative theories and no prospect of convergence. Competing theories of ethics are an obvious example.


My response...

Horwich is basically correct about the problems of philosophy as a theoretical discipline. And I believe his views on this more or less accurately reflect Wittgenstein's views.

But there are a couple of aspects of Horwich's thinking with which I am uncomfortable. Interestingly, these features are also the very features of Wittgenstein's thinking that I reject.

First, the 'scientism' issue. (Horwich uses the word.) Now, it seems to me that the term can used in at least two ways, one quite focussed, the other broader and stronger.

In the narrower sense, the term is used to highlight the inappropriate application of scientific methods or approaches. Horwich, for example, sees philosophy as inappropriately imitating science. I agree with him. This sort of thing happens and it would be better if it didn't. Philosophy (however we understand it) is not science. (Similarly, social science is not physics.)

But 'scientism' also has a broader meaning. On this view, a scientistic outlook involves having a high regard for science coupled with skepticism about other ways of gaining (anything other than commonsense and practical) knowledge.

This broader sense of 'scientism' (which I would embrace) marks a divide between profoundly different views of the world; and, because there is no agreement on basic assumptions across the divide, there is no way of satisfactorily dealing with such differences via philosophical or ordinary reasoning and argument. One set of assumptions or presuppositions will lead to certain philosophical opinions and lines of argument; another set will lead elsewhere.

Closely related to this scientism issue is the question of naturalism. Horwich rejects naturalism, which he sees as the view that every property, object and fact is naturalistic.

Mathematics is often said to involve objects, etc. which exist but do not exist in time and space. If so, then naturalism (as Horwich defines it) is incorrect. This leaves the door open for the similarly real existence of moral properties, for instance.

I haven't looked yet at Horwich's views on ethics. But I am aware of Wittgenstein's (religion-based?) views on these matters.


I am not going to try to deal here with Timothy Williamson's point of view. To do so would mean addressing topics like the dreaded Barcan formula (which was mentioned by Horwich in the video, by the way). The Barcan formula is an axiom (or schema) of quantified modal logic first stated by Ruth Barcan (who became Ruth Barcan Marcus) the acceptance of which apparently enhances and simplifies the workings of formal systems of quantified modal logic. Williamson defends the Barcan formula as being in some sense 'true', and, despite the fact that there is no general agreement on its informal interpretation, seeks to draw metaphysical conclusions from it – something along the lines that everything exists, but some things as actual objects and some as possible objects. This sort of thinking strikes me as being more in line with medieval scholasticism than with modern scientific thinking (and indeed in his writings Williamson refers to Avicenna, claiming that he informally anticipated both the Barcan formula and its converse).

There have been big advances in formal logic over the last century or so, and it would not be surprising if such advances allowed us to see certain general ideas in logic and traditional metaphysics in a new light, validating some old approaches as insightful or prescient and undermining others. But what Williamson is doing strikes me as going well beyond such modest, historical commentary. Though he equivocates about naturalism in this discussion with Horwich, backing away from Horwich's claim that he, like Horwich, is opposed to a naturalistic view of the world, he does seem, in effect, to be rejecting a modern, scientific outlook and attempting to resurrect something like traditional metaphysics.

I am not saying his motivations are religious: they may be entirely intellectual. The realm of formal logic, like the realm of mathematics, has its attractions: it can appear very appealing, especially when contrasted with the messiness and general unsatisfactoriness of mundane reality.

Horwich accepts the messiness, however, and here I am very much on his side.

Anti-naturalism in philosophy (I)

[Revised version]

To each his own. It is not for me to adjudicate on other people's activities – so long as they are not harmful or wasteful of public resources.

In these respects, academic philosophy occupies a grey area. It is dependent on public funds, though its requirements are relatively modest compared to some other disciplines. And it can be used (and often is) as a vehicle for promoting certain metaphysical, religious or ideological views, activities which will generally be seen as good by those who share the perspectives in question but harmful by others.

I have been following up recently on certain philosophical thinkers and trying to assess whether or not my initial judgements about the worthwhileness of their work were justified. (Of course, it's not just individual thinkers I am concerned with but also with the discipline.)

There are two issues here which need to be distinguished: worthwhileness (which relates to the wastefulness question raised in the first paragraph), and the issue of using the platform provided by a teaching position in academic philosophy to promote an ideologically-motivated agenda. The latter, I believe, is inappropriate and unprofessional and betrays the trust which society places in its publicly-funded or subsidized teachers. I will ignore here the most common (and objectionable) kind of bias – involving the promotion of partisan political views – and focus on the more subtle question of general metaphysical and religious attitudes.

I have to say that philosophy as it has generally been practised over recent decades does not look attractive to me: the general philosophical culture has been just too hostile to what is seen as 'scientism', and I have felt that the few who shared my views and who participated in the philosophical culture were only serving to give it unwarranted credibility (as token figures, perhaps).

Moreover, I can't imagine that Patricia Churchland's career path, for example, or even that of Daniel Dennett would be possible today. They learned their science along the way, more or less informally. Things are changing, especially with the advent of younger thinkers (such as the neo-empiricist Jesse Prinz or Edouard Machery) who are breaking down the barriers – and blurring the distinction – between philosophy and science.

But amongst those still committed to their discipline as independent and autonomous I am coming across more and more philosophers with a religious agenda. Even many apparently atheistic philosophers seem to be critical of naturalism. I just can't make sense of this except as a quasi-religious or (in some sense) ideological position. It may be that such a position is a necessary (but not a sufficient) condition for seeing philosophy as an autonomous and coherent academic discipline.

Note that both figures in the video debate embedded below share an anti-naturalistic outlook (though Timothy Williamson equivocates somewhat on this question). Note also the dramatic way Williamson sets up the debate – claiming that what Paul Horwich is saying, in effect, is that philosophy as normally practised is essentially just rubbish. Most academic philosophers spend their time addressing pseudo-problems while others spend their time pointing this out. On this reckoning, we should just pull the plug on the whole enterprise.

Williamson is an apparently skeptical but metaphysically-inclined logician. I respect his formal logical expertise but not his judgement in applying formal logical methods to theoretical and metaphysical questions. This general tradition of thinking, which owes not a little to Saul Kripke's work, has always struck me (I may be wrong) as profoundly misguided when it moves in a metaphysical direction, though it does have useful applications in semantics and other areas. I suspect that much of it – to the extent that it is not just game-playing – is ultimately motivated by a desire to undermine a naturalistic (or to develop a non-naturalistic) view of the world. Kripke, for example, has strong religious beliefs, and the orientation of his theoretical work is arguably related to these. It is no secret that many of today's logicians and philosophers (religiously inclined or not) are antagonistic to science.

The philosophy of language (so-called) seems to be more firmly rooted in 19th-century metaphysics than in modern linguistics. Furthermore, Wittgenstein's and other philosophers' compelling insights into what natural language is, how it works and how it can lead us astray seem to have been ignored or forgotten (except by the likes of Horwich, definitely in a minority these days).

Formal languages and approaches are valuable for their practical applications and for what they can teach us both about the nature of natural language and about thinking. But, like ordinary-language-based approaches, they can be deployed in very unproductive ways.


I will look more closely at what Williamson and Horwich are saying in my next post.

Anti-metaphysical musings

I have been looking recently at some material relating to "the metaphysics wars", and thought it worthwhile to jot down a few notes.

No doubt, my general position would be characterized by those with other views as scientistic. It is also anti-metaphysical in that I don't see the traditional philosophical discipline of metaphysics as having much point these days.

I don't deny that there are very interesting questions in the philosophy of physics, the philosophy of mathematics and the philosophy of logic which may be characterized as metaphysical. The meta-thinking that goes on at the margins of physics, other sciences and mathematics, etc. is necessary and valuable.

But somehow, when such thinking moves away from the discipline in question and becomes more generally philosophical, problems arise.

Timothy Williamson is perhaps the most powerful and impressive advocate for this broader kind of metaphysics (and analytic philosophy generally). As an avowedly non-religious person, he can't be dismissed as having ulterior motives of a religious nature; and, being at home with formal – and specifically modal – logic, he can't be dismissed as natural language-bound or as being daunted in any way by technical rigor.

Some of the points he makes in this interview are good ones – such as noting the light that modal logic can undoubtedly throw on the workings and nature of natural language (via Montague grammar, for example), and perhaps also on the foundations of set theory – but I have to say that I am strongly inclined to reject the basic thrust of his argument in defense of metaphysics, and, by extension, philosophy (as he understands it).

Essentially the questions he seems most interested in are reminiscent of medieval scholasticism. I too have great respect for thinkers such as Avicenna (to whom he refers approvingly) and respect also for more recent – and more mathematically sophisticated – exponents of that general tradition of thought (such as Bolzano, to whom he also refers), but it seems to me that it is now incumbent upon any thinkers who aspire to deal with questions of what there is in a fundamental sense to base their accounts – at least in large part – on contemporary physics; or on mathematics if they are restricting their focus to mathematical realities.

Williamson seeks to defend the relative independence of his core preoccupations from science by invoking the old shibboleths, scientism and reductionism, and rejecting naturalism as a confused and inadequate concept.

I grant that mathematics does pose problems for advocates of strong forms of naturalism and empiricism, and there are real unresolved issues in the philosophy of mathematics. But my preference is to address these issues in a broadly scientific and mathematical context rather than in a purely logical or philosophical one, or – worse – not to address them at all and instead merely to use them as a kind of justification or license for logical excess and metaphysical self-indulgence.

Williamson cites Quine as an example of scientistic naturalism.

"Quine privileged natural science, and in particular physics, over all other forms of inquiry, to the point of not taking very seriously any theory that couldn't be reduced to part of natural science."

Williamson's view, by contrast, more or less allows the analytic metaphysician carte blanche, and Williamson's own approach to analytic metaphysics is clearly – in my view at any rate – insufficiently constrained and guided by science.

Here, for example, is an extract from an old interview in which he explains his developing views:

"My work on vagueness and ontology doesn’t really concern ontology. Probably my most distinctive ontological commitment comes from my defence of a controversial principle in logic known as the Barcan formula, named after the American logician Ruth Barcan Marcus, who first stated it. An application of this principle is that since Marilyn Monroe and John F. Kennedy could have had a child (although they actually didn’t), there is something that could have been a child of Marilyn Monroe and John F. Kennedy. On my view, it is neither a child nor a collection of atoms, but rather something that merely could have been a child, made out of atoms, but actually has no location in space and time. The argument can be multiplied, so there actually are infinitely many things that could have been located in space and time but aren’t. It takes quite a bit of work to show that the Barcan formula is better than the alternatives! That’s what my next book will be on. The working title is Ontological Rigidity."

The book was actually called Modal Logic as Metaphysics, and this is how he recently stated its main point:

"I am ... saying that it is necessary what there is. Necessarily everything is necessarily something. There could not have been more or fewer things than there actually are, and which particular things there are could not have been different. What is contingent is only what properties those things have, and what relations they have to each other. I call that view necessitism. Its denial is contingentism. Who knows how far back necessitism goes? Maybe Parmenides was some sort of necessitist..."

On the face of it, talking about (apparently countable) things (minus their properties and relations!) as given strikes me as breathtakingly naïve in the context of a physics-based understanding of reality. I can only imagine that Williamson is – like the medieval scholastics – implicitly asserting a privileged role for logic.

Quine's assertion of a privileged role for physics makes a lot more sense to me.

Admittedly I haven't looked at Williamson's ideas in any depth, but what I have seen so far – and what he says in this latest interview – really makes me question whether it would be worth the effort. I am intrigued, however, by what is driving such thinkers.

Strangely, Williamson appears not to be quite sure whether his latest work is meaningful or not – or at least seems unwilling to commit himself on the matter. There is (don't you think?) just a touch of arrogance in this passage (from Chapter One of Modal Logic as Metaphysics)?

"This book compares necessitism and contingentism. Which is true? Of course the question has a false presupposition if the definitions of 'necessitism' and 'contingentism' lack meaning or content. But if every enquiry must first establish its own meaningfulness we are on an infinite regress, since the enquiry into the meaningfulness of the previous enquiry must first enquire into its own meaningfulness, and so on. Better to act on the assumption of intelligibility: readers can decide for themselves whether they understand the book as they go along, and recycle it if they don't."

This passage is a combination of facile reasoning and rhetorical sleight of hand. By using the word 'understand' in the final sentence, he subtly shifts the focus to the reader's possible inadequacy and away from the original question concerning the work's meaningfulness.

In fact, I am tempted to see Williamson's work as emblematic of a broader trend. On the basis of my (admittedly limited) knowledge of the history of the relevant intellectual cultures, I discern, since the middle years of the 20th century, a disturbing falling off in intellectual seriousness in secular circles accompanied by an equally disturbing rise in anti-scientific name-calling and credulity amongst those thinkers who remain favorably disposed towards religion.

I'll finish here with a few comments about Paul Horwich, Williamson's great philosophical antagonist, whose deflationary views on truth I have referred to favorably in the past.

Horwich is opposed to the sort of traditional theoretical philosophy ('T-philosophy') which Williamson defends. I have made the point that, though I broadly accepted Horwich's account of truth, I doubted that his Wittgensteinian view of philosophy was compatible with a continuation of philosophy as an academic discipline. And, interestingly, Williamson makes a similar point in the recent interview.

"...Horwich didn’t explicitly call for T-philosophy not to be funded. I pointed out that if the picture of philosophy in his book were accurate, philosophy should be abolished. The reader encounters just two sorts of philosophy: irrational T-philosophy, and level-headed Wittgensteinian debunkers of T-philosophers. Philosophy is presented as an activity in which some people make a mess and others clear it up. Why on earth should taxpayers fund that? It looks as though we’d be better off simply abolishing the activity altogether."

Finally, I was surprised (and a bit disappointed) to learn recently that Horwich rejects naturalism, and even more unequivocally than Williamson does. He cites not only mathematical but also moral claims as a basis for his view.

Horwich is more thoroughly Wittgensteinian than I had previously thought.