critique-of-pure-reason.com

29 June 2011

Schopenhauer’s Key Concepts 0: Transcendental Idealism

Arthur Schopenhauer

The world is my representation.

So begins Schopenhauer’s World as Will and Representation. The statement distills the essence of his philosophical starting point, which is a version of Kant’s transcendental idealism. The rest of this series will concern the terms that are actually used by Schopenhauer in his magnum opus, but this first post is a preliminary sketch – and a very sketchy sketch at that – to provide some background. I’ve tried to avoid the complications of Kantian terminology as far as I can (a wise move for someone who hasn’t yet read Kant), so I can’t claim accuracy in that regard.

The distinction between appearance and reality goes a long way back. In the modern era, Descartes made the quite uncontroversial point that what seems to be so often differs from what is actually so. From here, an argument can be put together from the works of Hume, and the other philosophers who came after Descartes, that now goes by the name of the argument from illusion. From the well-known instances of illusion, such as the appearance of a stick as bent when it’s submerged in water, the argument concludes that what we find in our daily experiences are not the objects themselves but sense-data, which are productions of the senses and the mind (or brain), presumably in response to external stimuli. Therefore contrary to the doctrines of the so-called naive realists, we do not have direct access to empirical reality, but only to internal representations. What we think of as real objects in the world are in fact mental constructions which are presented to us by our own faculties. Furthermore, Descartes had shown that there was no sure-fire way of distinguishing between veridical and illusory appearance, so in the end we cannot prove that things are as they appear, or even that the external world exists at all – that there is a world out there producing these sense-data. We are trapped behind a veil of perception.

Kant’s transcendental idealism is the end point (or, to some, the high point) of this line of thought, and filtered through Schopenhauer’s thinking it goes something like this. An object is an object at all only for a subject. Whatever the thing is in itself, it is objectified only in subjective experience. We can never have access to the thing in itself, so the empirical world – the world of mountains, people, planets, life and death – is how it appears by virtue of its representation by our internal faculties. Empirical reality, which we usually think of as standing independently, is conditioned by the subject of experience, i.e., by the peculiarities of the subjective faculties that make it possible for us to experience at all. These subjective faculties condition the empirical world in certain forms, namely space, time and causality.

It’s still very common to accept the argument from illusion, and what might be called representational realism, which quite reasonably posits an external world which impinges on our sensory apparatus, therby causing our sense-data. However, it’s not so easy to go along with Kant’s claim that not only colours, warmth, sounds and so on, but space, time and causality are similarly the result of the subjective forms of understanding which are imposed on the objects of experience. How does he establish this?

Well, for Hume, we get all our knowledge from sense impressions, and we cannot use reason to establish anything with logical certainty about the external world. In particular, we cannot prove the reality of necessary causal connections between events. Neither can it be proven scientifically, as every science presupposes that events are caused consistently. There is, it turns out, no reasonable justification for the belief that events are connected in such a way, rather than simply following one after the other, for we would need to justify it from experience, and we do not actually experience such connections at all: what we perceive is just the succession of events. Furthermore, the fact that the same events succeed each other again and again in itself does not allow us, with logical certainty, to conclude that this constant conjunction is a necessary connection. But the repeated perception of such constant conjunctions forms in us a habit of mind, and this is all causality amounts to.

This shocking result is the scepticism that Kant said woke him from his dogmatic slumber. It removed the possibility of a secure foundation for knowledge, a great concern for many of the early modern philosophers, keen on the newly burgeoning science as they were. For Kant, there must be something that we have experience of, but given our incarceration behind the veil of perception we cannot know anything about it. This something is the realm of things in themselves. At the same time, there is more to experience than Hume realized. If sensation were the end of the story, then we would have nothing but a meaningless jumble of impressions, but experience is not like this – it is organized according to certain forms and concepts. Experience is absolutely conditioned by our faculties.

And it’s not as if we pick up these forms and concepts directly as sense impressions. To experience anything at all presupposes space, time and causality – there is no sense in which we could have an experience that does not somehow involve these – but they are merely relations between things, and in what sense can a relation be perceived in an external world? Sure, there’s something out there, but we cannot see a metre or feel a minute, meaning that such forms of thought are just that and no more: ways in which we experience objects through our faculties of perception and understanding. There is no justification for attributing space, time and causality to something outside the objects of experience. Where would such a justification come from? All that can possibly be said is that there must be some kind of thing out there – what it’s like we can necessarily never know, because knowledge itself is possible only via the mental faculties that impose structure on the world, that create the world.

Another way of approaching transcendental idealism is through Locke’s distinction of primary and secondary qualities. For Locke, colour, warmth, sound, smell and so on are secondary qualities which are attached to objects by the senses, while the objects in themselves have only the primary qualities of mass, extension in space, and shape, among others. We cannot conceive of objects as lacking these primary qualities, therefore they must belong to them in themselves. Kant, following Berkeley, saw that our inability to conceive objects in this way was no justification for assigning primary qualities to objects independent of us. According to Kant, we cannot conceive of objects without qualities because objects appear to us conditioned by our mental faculties, which are the prerequisites for having any kind of perception or thought at all. In other words, not only the secondary but also the primary qualities are provided by the mind of the knower.

Now what makes this “transcendental”, and what makes it “idealism”? Well, transcendental is a special term denoting investigations into the a priori conditions of the possibility of experience, not to be confused with transcendent, which denotes an actual transgression of the bounds of experience. Secondly, it is a variety of idealism because objects are considered as objects of experience, in which they are constituted as those objects; in more Schopenhauerian and less Kantian terms, there must be some kind of world in itself, but our world is a product of our own brains. Incidentally, it’s important to keep in mind that this doctrine doesn’t claim that the empirical world is just an illusion, but only that it is not entirely mind-independent. Our experiences are as real as can be.

It might be wondered how Kant could have thought this was an adequate response to Hume’s scepticism, and it appears he was conscious that he had not done much to place us squarely in the world beyond the veil, and had in fact transformed it from a veil into an impregnable wall. In the end, there was more idealism in his doctrine than he would have liked. This is an important and much-debated point that I won’t be able to address until I’ve made some attempt to study Kant for real.

The crucial differences in Schopenhauer’s version of transcendental idealism are, as I see it, that, firstly, whatever lies behind appearances (or phenomena) can have no plurality, because for it there is no space and time; and secondly, that we do have a kind of access to this thing-in-itself (singular), namely through our own bodies, which give us an immediate knowledge, different from any other kind. And it is in this knowledge that we glimpse the world as will. But more of that later.

Incidentally, it is interesting to consider how far Schopenhauer can be called an empiricist. He presupposes much from Locke and Hume, and in his talk of the brain he often looks like a very modern Dennett-type of scientifically-oriented empiricist. He certainly follows the empiricists in taking experience as primary and in their characterization of dogmatic metaphysics, but against the sceptical Hume he is with Kant in recognizing the extent to which the empirical world is in fact perfectly legitimately explained in terms of causality (as well as space and time), i.e., on the strong basis of the universal a priori mental conditions of experience. The empirical world is as it appears to be, and causality is not only an unjustified habit of mind formed through repeated experience of succession, but is rather an innate faculty that makes experience possible in the first place. Causal connections are necessary and objective, because to experience anything at all just is to understand things in terms of necessary causal connections. This is what makes him a transcendental idealist rather than an empirical one.

The importance of transcendental idealism for Schopenhauer’s philosophy should become clearer in the next part, when I look more closely at the notion of representation.

Part 1: Representation

comments powered by Disqus