Language and metaphysics

Bertrand Russell and Ludwig Wittgenstein had many profound disagreements. They were, for example, diametrically opposed in their respective attitudes to science and religion. But they were united in one respect at least. They both rejected the metaphysics of idealism. Metaphysically speaking, Russell was concerned mainly to counter idealist notions and to defend a science-friendly and empirical view of the world. And Wittgenstein, like many of his contemporaries, saw no role for metaphysics as a discipline in its own right.

Early in his life Russell had accepted the general framework of Hegelian metaphysics that dominated English philosophy at the time. But he soon came to see problems with this point of view and felt a sense of excitement and liberation when he finally extricated himself from this way of seeing the world. He talks in his intellectual autobiography about his early rejection of the doctrine of internal relations which was a key feature of Anglo-Hegelian idealism.

Very roughly, this doctrine constituted a form of coherentism or holism according to which everything is related to everything else and nothing can be satisfactorily understood except in terms of the totality of these relations (i.e. ultimately in terms of the Absolute). This kind of idealism fell out of favour, but gained renewed intellectual respectability when Willard Van Orman Quine proposed a form of holism which was (ironically) partly inspired by the writings of Pierre Duhem, an historian of science who was not only a deeply religious man but also an orthodox and militant Catholic.

Wittgenstein, by contrast to Russell (and Quine), was not well-read in Western philosophy. He was blissfully ignorant both of classical and medieval thought as well as of German (and English) idealism, and the “metaphysical stance” which he himself came to identify in his Tractatus Logico-Philosophicus derived from its logical absolutism: the attempt to reduce the intelligibility of the world to pure logical objects in logical space. The Tractatus was an attempt to give definitive expression to the scientific project and by so doing to reveal its limits. The particular understanding of language, logic and mathematics which was at the heart of the Tractatus Wittgenstein gradually came to see not so much as false but rather as unnecessarily narrow.

One of the factors which led him to see this was hearing, in 1928, three lectures by L.E.J. Brouwer. Brouwer’s mathematical intuitionism was focused on numbers rather than geometry and on finite constructions rather than on infinite logical space. Wittgenstein had to face the fact that there were various ways of conceptualizing the basis of logic and mathematics, and he started to develop a philosophy of logic and mathematics which tried to explain these practices in terms of the sorts of common agreements which make social life possible. But most of his later work was focused not on mathematics or logic but on ordinary language and the ordinary social conventions which sustain it.

Friedrich Nietzsche had noted that each natural language is, as it were, pregnant with a metaphysics, the metaphysics of one language being different from the metaphysics of another. Metaphysics (as he saw it) was largely a projection of the structure of a particular language on to the world.

This general way of seeing metaphysics as a function of language can be applied not just to natural languages but also to more formal, constructed languages or logical systems such as those which were developed from the late 19th century onward. The crucial point is that metaphysics is seen as a kind of gratuitous by-product of a language and its use – or misuse. As such it is not something that can be studied in itself as the natural or social worlds may be studied.

The Vienna Circle is well-known for taking such a line and Wittgenstein, as a close friend of Moritz Schlick and an early participant in the deliberations of Schlick’s invitation-only group, played a crucial role in the development of the ideas which would come to be known as logical positivism.

Wittgenstein’s view of language developed beyond the position outlined in the Tractatus but there is a lot of continuity in his thinking and the Tractatus itself can be read as a critique of traditional metaphysics. It is this aspect of it which appealed to Schlick and the Vienna Circle. At no time did Wittgenstein write anything resembling traditional metaphysics (or ethics, for that matter).

But Wittgenstein also came to see the standard scientific view of the world as logically flawed and as incorporating metaphysical assumptions. The law of identity (‘A is A’) has a long history as a basic axiom of Western logic and plays an important foundational role in most modern formal systems. But in the Tractatus Wittgenstein was already moving away from this kind of approach, explicitly calling the law of identity into question. “To say of one thing that it is identical with itself is to say nothing at all,” he wrote. (5.5303)

What’s more, as a Christian primitivist in the tradition of Leo Tolstoy, Wittgenstein was not sympathetic to the anti-religious stance of most logical positivists. Probably on account of his religious commitments (which are often downplayed by philosophers), he was also rather less interested in scientific questions or in articulating a scientific view of the world than his erstwhile empiricist colleagues.

Wittgenstein certainly disappointed his old mentor, Bertrand Russell, by moving away from dealing with the sorts of science-and-logic-related questions which Russell himself was concerned with as a philosopher and focusing instead on an informal approach to language and other matters.

Wittgenstein saw language as something that has the potential to lead us astray, and much of his later work is designed to highlight the pitfalls of language (especially as deployed by philosophers). Metaphysical questions can usefully be approached in this way: in terms, that is, of language (whether natural or constructed). And often apparent problems can be dissolved.

Rudolf Carnap was a major 20th-thinker thinker who followed this general approach and sought to downplay the significance of ontological claims, characterizing philosophically-based metaphysical – and, specifically ontological – claims as being either trivial or problematic.

Carnap saw ontology, understood as “the study of what there is”, as being misguided. Questions about the existence of things which are assumed to be in a given linguistic or conceptual framework are trivially true. In his paper, “Empiricism, Semantics and Ontology” (first published in 1950), Carnap writes: “A question like: ‘Are there (really) space-time points?’ is ambiguous. It may be meant as an internal question; then the affirmative answer is, of course, analytic and trivial…” But if the question about existence is seen as general and unrestricted it becomes very problematic.

Thomas Hofweber (writing on Language and ontology in The Stanford Encyclopedia of Philosophy) summarizes Carnap’s view:

"Ontology, the philosophical discipline that tries to answer hard questions about what there really is is based on a mistake. The questions it tries to answer are meaningless questions, and this enterprise should be abandoned. The words ‘Are there numbers?’ thus can be used in two ways: as an internal question, in which case the answer is trivially ‘yes’, but this has nothing to do with metaphysics or ontology, or as an external question, which is the one the philosophers are trying to ask, but which is meaningless."

I am inclined to agree with this general position though I shy away from the word ‘meaningless’. Carnap termed such (external) questions “pseudo-questions”, and characterized the ontological pursuits to which they lead as “useless” and “futile”.

There is no question that metaphysical and logical ideas are related. The logical framework which one chooses has metaphysical implications. For example, if you reject the law of non-contradiction (as Hegel did, for example) this will have implications not just for what you see as valid forms of argument but also for how you see the world more generally.

I want to say something here – by way of clarification – about the various meanings and connotations of the word ‘metaphysics’. Sometimes it is used to refer to an intellectual discipline, sometimes more broadly to refer to a general view of the world. In the latter case, sometimes (but not always) there is a connotation to the effect that the view in question is akin to a religious view.

This can be confusing. For example, in his later writings, Martin Heidegger is often (and quite rightly, I think) seen as moving away from specifically philosophical and metaphysical discourse, at least as traditionally understood in the context of the Western academic tradition. But in another (quite valid) sense what he is doing is very metaphysical. When Graham Priest applies the term ‘metaphysical’ specifically to Heidegger’s later work (as he does around the 13-minute mark in this interview) he is using the term in my latter sense (and with the religious connotations, I think). I want to make it clear that this is not the kind of metaphysics which I am implicitly criticizing here. Nor (as I see it) is it the kind of metaphysics which (in their different ways) Nietzsche, Wittgenstein and Carnap were attacking. (In fact, a good case can be made that that Nietzsche, (the later) Heidegger and (the later) Wittgenstein have a lot in common.)

The sort of metaphysics that Wittgenstein, Carnap et al. were concerned to counter was the traditional scholarly kind which (on their view) is based on pseudo-questions arising from a misreading or misuse of language. They both rejected the view that there is a deep “ontological” sense in which the implicitly projected objects can be said to exist. Various kinds of objects exist, but only in an ordinary sense. And a keen sense of what language is and how it works – such as Nietzsche (as a philologist) certainly had, and as Wittgenstein and many of his philosophical contemporaries also had – helps to make this clearer.

Nouns are useful abstractions. The objects that concrete nouns describe exist individually in a practical or pragmatic sense (this dog, that fork…); or not, in the case of unicorns, etc.. Useful abstractions like nation states or agreements can also be said to exist in a practical and pragmatic sense. They are social realities. But all too often, and especially in the context of philosophical discourse, useful – or not so useful – abstractions are taken to be real in a metaphysical sense, or something real or substantive is seen to lie behind an abstract noun which is merely a convenient tool facilitating concise expression.

One thing which is particularly interesting, as I see it, is the relationship between metaphysics (as a discipline) and religion. Western metaphysics – from Plato to medieval and through to modern times – grew out of what came to be called natural theology and was usually associated with a particular kind of (intellectualized) religion.

But other religious tendencies existed within the Judeo-Christian West which were hostile to metaphysics and which saw metaphysicalized religion as a betrayal of the more direct and intuitive form of religion to which they were committed. Blaise Pascal typified this approach. He rejected the “God of the philosophers” entirely, and embraced the God of Abraham, Isaac and Jacob. Tolstoy and Wittgenstein were decidedly within this fideist tradition.

A commitment to metaphysics is often associated with a commitment to religion. But a hostility to metaphysics can also be driven by religious commitments.


[This is a slightly edited version of an essay first published at The Electric Agora.]

Anti-naturalism in philosophy (III)

The role and raison d'être of academic philosophy in today's world is far from clear: there is certainly no consensus amongst philosophers regarding what philosophy is or is for. I personally would like to see the discipline become less insular and more engaged with the sciences, and there have been some encouraging moves in this direction. On the whole, however, I think that you would have to say that the discipline has become less rather than more science-friendly or science-oriented over the last fifty or sixty years.

What concerns me here is the general question of metaphysical or religious motivation. Generally the logical empiricists, who dominated analytic philosophy in the mid-20th century, were not well disposed towards traditional metaphysics or religion. But the general climate of opinion within academic philosophy has changed quite a bit since then.

There are (and always have been) overtly religious philosophers, of course. But I am not concerned with them so much as with apparently non-religious (or even atheistic) philosophers who nonetheless seem very uncomfortable with a straightforwardly naturalistic view of the world.

Take John McDowell. I am not aware that he has any particular religious affiliation, and yet there is something about his approach which seems almost hostile to science and I have trouble trying to figure out what on earth is driving it. It must be more than just a desire to defend his professional turf, I feel sure.

Starting about the 24 minute mark in this interview, McDowell defends himself against accusations by Tyler Burge (who is not featured here) that his approach betrays ignorance of the relevant science, but his defense is not entirely convincing.

Note his curiously unscientific discussion about recognizing Bill Clinton's face (43-45 min.): no reference at all is made to brain processes (and much is known about these sorts of processes). Note also some fairly dismissive remarks about science in general and cognitive science in particular. "What [cognitive science] does," he says (more than once), "is just fine" – but it has little to contribute to telling us about how we can know things, apparently! *

Though he concedes that in order for his views to be acceptable they must be compatible with the relevant science (of perception and cognition, etc.), McDowell's view of the world is clearly not driven by science (which he shows little interest in). He certainly does not have what the logical positivists dubbed a scientific worldview. In fact, he seems to espouse a form of Kantian idealism. He himself notes that he came over time to recognize the importance of Kant's transcendental perspective; and, though this topic is not raised here, McDowell is a defender of moral realism.

The only way I can make sense of an intelligent and obviously serious thinker going on at such length and in such detail about the distinction between the perceived and the seemingly perceived etc. is that something more than just an apparently trivial semantic point about the meaning of the word 'know' and its cognates is (perceived to be) at stake here. What is being obliquely presented and defended in all of this is a particular metaphysic, a fundamentally anti-physicalist, anti-empiricist and possibly anti-naturalist way of seeing the world.

Terms like physicalism and naturalism are awkward to define, but what this philosopher and many like him are claiming is that there is more to fundamental reality than what the various sciences and mathematics can describe or access – and that the philosopher (through pure thought, apparently) has access to it.

This sort of thinking strikes me as being very much in the same category as priestly mystification. And indeed many of today's philosophers, whose discipline in the Western tradition grew out of schools of divinity, could be seen to be competing to fill the same general intellectual, social and economic space as was once occupied (almost exclusively) by clerics.

One last point. When I talk of priestly mystification I don't mean to imply that it necessarily involved or involves deliberate deception or hypocrisy. In many (perhaps in most) cases, the perpetrators are 'true believers' in the doctrines they are promoting.



* In order to understand this discussion fully one would need to read certain texts, by Wilfred Sellars in particular. Sellars was a 20th-century philosopher who is often appealed to by those who seek to make a space for philosophy, a philosophy which is respectful of but distinct from science, in today's intellectual landscape. Daniel Kaufman, with whom I have had some dealings (at Scientia Salon, Apophenia and, more recently, at The Electric Agora), is a great promoter of Sellars' work. The interviewer on this video is also a Sellarsian. But not the interviewee, I would say. For McDowell, Sellars' view of the world is too reductionistic or scientistic apparently and, though he appeals to some of Sellars' ideas (like 'the Myth of the Given'), he "averts his gaze" (a strange – and revealing – phrase) from other aspects of Sellars' thought.

Mathematics; panpsychism

Last month I listed a number of topics (mainly linguistic, psychological and historical) I have been thinking about. The last item on the list was the most philosophical, relating to the challenge that mathematics can be seen to pose for empiricism.

Mathematics is often presented as a deep and interesting area of knowledge which is somehow independent of the empirical world. Well, it certainly is a deep and interesting area of knowledge, but it is also very much a product of physical brains interacting with the wider physical (and cultural) world. Sure, it operates at a very high level of abstraction; and, sure, many mathematicians are mathematical realists (or Platonists) who feel themselves to be exploring (and discovering things within the context of) an independently-existing and non-empirical realm.

But the old idea of a realm of pure mathematics without applications (promoted by mathematical Platonists such as G.H. Hardy) is looking increasingly forced and dated (i.e. tied to a particular cultural tradition). It's well known that ideas from pure mathematics often find subsequent – and unexpected – applications. Non-Euclidean geometries, for example, were originally developed in the 19th century as pure mathematics but subsequently found applications in cosmological theories.

What Hardy shied away from particularly, however, were technological applications. He would not have been happy that his own area, number theory, which he loved for its purity and uselessness, turned out to have important applications in computer science.

Cantor, of course, was also a Platonist. I was looking again at his 'diagonal argument' which shows that the set of real numbers is not countable (denumerable) – so that some infinite sets (as George Orwell might have put it) are more infinite than others.

But I always feel uneasy when infinities (even common or garden variety infinities like the sequence of natural numbers or the expansion of pi) are built into arguments. Cantor's argument is clever and convincing in a sense, but the (infinite) matrix on which it is based is merely imagined (or postulated, or projected).

I think I'm okay with mathematical procedures which involve an unending series of steps (potential infinity); but not with mathematical objects which contain an infinite number of elements (actual infinity).

So it seems that I am an intuitionist, but I can't really say at this stage whether I'm a finitist also. (The latter only recognize mathematical objects which can be constructed from the natural numbers in a finite number of steps).


Another topic I've been thinking about is panpsychism. What prompted my (renewed) interest was coming across a philosophically-oriented blogger with a PhD in theoretical physics whose nom de plume happens to be 'Panpsychist'. He had made some intriguing comments on a post by Massimo Pigliucci on reductionism in science and invited people to continue the discussion on his site. (Massimo has set limits on comments at Scientia Salon.)

I followed quite closely but didn't participate in the reductionism debate which was characterized by a certain degree of terminological confusion, specifically about the meaning and application of certain terms used in the philosophical literature, e.g. "token physicalism" (which is associated with "supervenience") and "type physicalism" (which is associated with "strong emergence").

Panpsychism is relevant to the topic of reductionism in that it can be seen as a way of getting around the problem of reducing mental properties to physical properties.

'Panpsychist' rejects dualism and also rejects the idea of emergence, the idea, as he puts it, "that mental properties emerge from certain configurations of regular, non-mental matter." He says that reading David Chalmers convinced him that the standard idea of emergence was wrong because it failed to address the 'hard problem' of consciousness. He also argues against the view that panpsychism is an essentially religious position.

Some time ago I considered and rejected David Chalmers' take on the so-called hard problem of consciousness as unconvincing and unscientific. Basic to his approach is imagining beings that look and behave just like humans but lack conscious awareness; logically coherent perhaps but utterly implausible both in terms of common sense and in terms of science.* Biological creatures have various levels or degrees of consciousness or awareness or sentience: that's just how things work. And imagining a world in which quite arbitrary – not driven by scientifically-based reasoning – differences apply is merely idle speculation. Also, Chalmers-inspired approaches tend (as I see it) to be too much focused on human consciousness rather than on more primitive – and basic – forms of awareness from which the former ultimately derives. The sentience of simple life-forms is where the real (philosophical) interest lies, in my opinion.

I have also in the past seriously considered and rejected panpsychism, but I do acknowledge that the curious spectacle of an apparently inanimate universe producing sentient and ultimately conscious organisms – and so in a sense becoming conscious of itself – does give one reason to ponder the possibility that consciousness (in some form or other) is fundamental (in some sense or other).

Looking again at these issues, I note that the revival in panpsychism in philosophical circles (prompted in part by Chalmers' work in the 1990s) is being driven largely by 'process' thinkers. (It used to be called 'process theology' but the 'theology' is generally dropped these days and replaced with 'thought' or 'thinkers' or nothing at all.) It all goes back to Whitehead – and ultimately to neo-Platonism, I suppose.

I tried to read Whitehead a couple of times, but found him rather vague and wordy and (unnecessarily?) obscure. It's not just that he had grown up when 19th-century philosophical idealism was at its zenith and had internalized old idealist assumptions and ways of speaking because I've read and found interesting the work of F.H. Bradley who was not only an idealist but had far less mathematical and scientific knowledge than Whitehead. I think perhaps Bradley had keener insights into human psychology than Whitehead and so was more grounded. He also had a better prose style, which is often a sign of groundedness.**

Despite not warming to Whitehead's work (or that of his followers), I do like the idea of seeing fundamental reality as process rather than 'stuff'. (The fact that matter and energy are functions of one another makes old-fashioned materialism unviable.)

This view fits in nicely with the idea of computation, and with seeing the cosmos as some kind of computational (or computation-like) process.

And finally, returning to mathematics, I see the natural numbers also in terms of process: namely iteration.***



* As I see it, logic derives from and is intimately related to mundane real-world and scientific reasoning. Logic may be an independent discipline but this does not entail that the subject of the discipline constitutes or forms the basis of some kind of alternative reality.

** Heidegger comes to mind here also: despite the excesses and idiosyncrasies, his language meshes with reality somehow (at least some of the time!).

*** As they are expressed, for example, in Church's lambda calculus. On the whole, Church is a bit too abstract (and Platonic) for me however.