Your complimentary articles
You’ve read one of your four complimentary articles for this month.
You can read four articles free per month. To have complete access to the thousands of philosophy articles on this site, please
Immanuel Kant
Did Kant Solve Skepticism?
Thomas Morrison asks just what Kant learned from his Critique of Pure Reason.
Can we know how things really are regardless of how they appear to us? Throughout history, many philosophers – known as skeptics – have argued that we can not. Part of what Immanuel Kant is attempting in his Critique of Pure Reason (1781) is to show that arguments for skepticism are unsound since they rest on a misuse or misunderstanding of concepts such as ‘experience’, ‘things’, and ‘knowledge’. I hope to show that in this specific sense Kant does solve skepticism, but that by changing the meaning of crucial terms he leaves us with a far more puzzling problem.
Knowing Skepticism
How did we get ourselves into the problem of skepticism to begin with?
There are two assumptions operating in the skeptic’s question. The first is that there is a distinction between appearances and reality – between objects as they appear to us and objects as they are in themselves, independent of us. The second is that there is a distinction between direct and indirect knowledge – between our own immediate experiences and what we can come to know only indirectly, by inferring it from experience. These two assumptions together lead to what I will call the ‘mind-world gap’: the gap between the inner and outer worlds; that is, between appearances and reality; or between thought, and the world we think about.
David Hume could not see beyond appearances to the reality of causation
To highlight this gap between appearance and reality, consider some common experiences: optical illusions and hallucinations; or that the taste and smell of familiar things seem to change when you’re ill; or that the colors of objects look different in abnormal lighting, although there is no change in the objects themselves. Or, cross your eyes and what was previously one image is now two. Even the common perception of a three-dimensional object in space is only ever of one or a few sides – try to see all the sides of this magazine at once! That’s impossible; yet you will not deny that the thing you’re perceiving is three-dimensional and solid.
All these examples and more illustrate that the way things appear to us are not how they are in themselves, in objective reality, beyond experience (what would that even mean?). But we still want to know how far our appearances correspond to the things in themselves. To do this, we must be able to cross the gap. How do we get to knowledge of things as they are in themselves? Our senses can be mistaken and we can be fooled in a myriad of other ways too. Perhaps we simply have no way of knowing the difference between an erroneous experience and one that reveals (part of) the truth?
Finding a reliable method of avoiding error was the sweetheart project of René Descartes (1596-1650). He begins his Meditations (1641) by stating that he wants to “avoid believing things that are not entirely certain and indubitable.” So for Descartes, one mark of knowledge is certainty, which is holding a belief without any doubt. However, the above cases show that our senses can mislead us, can give us false ideas. So we can doubt that things are as our senses report (and of course that is just what physicists are saying when they tell you that the table you’re sitting at is a swarm of incredibly tiny force-carrying wave-particles, rather than the solid thing it appears to be).
Does this mean that we can never have knowledge about the outside world – about things in themselves? Even cursory readers of the Meditations can see that Descartes does not end up a skeptic. Rather, through his cogito (“I think, therefore I am”), and a sweeping rendition of Anselm’s arguments for God filtered through Descartes’ own pious fidelity, Descartes claims to show that we can have certainty, and hence knowledge, of the actual world. However, without the trustworthiness of God, Descartes’ rationalist criteria of knowledge gain us nothing more than the cogito – one’s own thinking and being – in other words, only knowledge of mental acts, or the inner world.
How else might we gain knowledge of things in themselves? David Hume (1711-1776) claims that such knowledge must be based on causal inference: inferring from effect to cause – from appearances to the thing itself (Treatise 1.4.2.46, 1738). He divides knowledge into relations of ideas and matters of fact. We may come to know something simply from considering ideas (or concepts) – such as knowledge that a bachelor is an unmarried adult male, or that ‘any two things equal to a third thing are equal to each other’. Or we may come to know by turning to experience, to establish facts such as that New York is north of Miami or that all ravens are black. Simply thinking about the abstract concepts ‘New York’ and ‘raven’ won’t give us these facts.
Hume goes on to argue that causal knowledge cannot be gained from the pure relation of ideas, because when we consider the cause we do not thereby intuit the effect. In other word, there is no apparent logical necessity between a cause and its effect. A three-sided square is intrinsically absurd, since it involves a contradiction between the ideas. But striking a match in normal conditions without the match igniting is, however unlikely, not absurd; it involves no contradiction. We can also think of a cause without having any thought about its effect. Consider someone wholly ignorant of computers. Would they be able to tell simply from looking at a keyboard that pushing the buttons will cause letters to appear on the screen? There is no purely conceptual link between causes and effects, then. We can’t know what caused an event just by thinking about that event by itself. Rather, you test a causal link by having different experiences. So knowledge of causes must be a knowledge of matters of fact. We have to turn to experience to get it.
Well, Hume asks, exactly what do we experience? We experience one thing (the cause: my striking the match) precede another (the effect: the match enflamed). Unfortunately, to know that something caused something else, we have to know more than simply that it preceded the effect. One reason is that what precedes an event is much more than just the putative cause. A complete description of the moment that preceded the match lighting would include everything occurring a moment prior in the entire universe, from flies landing in Africa to satellite fly-bys. So why not say that a satellite passing above us at that moment caused the match to ignite?
We might ‘test’ this hypothesis by finding that not every time a match ignites it follows upon a satellite fly-by. So now we have this criterion: a cause is what precedes an event in every heretofore experienced case. But even this is not specific enough. Even with this new standard of preceding in every past experience, we are left with much more than the actual cause. A moment before the match ignites there are always air currents operating within the room, electromagnetic radiation, gravity pulling downwards, etc. These things always precede any match ignition you’ve ever experienced. But we misunderstand the concept of ‘cause’ if we say that gravitation caused the match to ignite, as opposed to me striking the match. Hume would himself say that we cannot know that my striking the match caused it to ignite, since we only experience distinct events following or preceding one another (‘in constant conjunction’, as he put it), and we do not experience the causal glue between them which is the putative cause itself (Treatise 1.3.2.11).
What more do we need for knowledge of causation? We need to establish that the effect doesn’t merely follow the cause, but follows from it; not just that in the past I have constantly experienced Event 2 following Event 1, but that Event 1 caused Event 2 – that, all other things being equal, if Event 1 occurs then Event 2 will always follow. So the idea that something caused something else rests on a universal hypothetical statement (‘for all x, if x then y’) which asserts a necessary connection between the cause and effect. If causes were not necessary, we would have to say that a cause may or may not produce its effect, and yet there is nothing else – no ‘hidden variable’ – that determines whether it will or will not produce its effect. Likewise, given an effect, we cannot say it was caused unless effects of this type are always preceded by causes of similar general types. The problem Hume is pointing out is that experience cannot give us knowledge of any necessary connection. We only have experience of what has happened, and of what is happening, but a necessary connection involves a projection into the future – it makes a claim about what will happen any time a certain cause is present. As we have seen, experience gives us only knowledge of constant conjunction. Through habits of thought, we come to anticipate the ‘effect’ every time we experience the ‘cause’; but properly speaking, we have no logical or experiential knowledge that this event caused this following one. For Hume this shows that we cannot have knowledge of how things are independent of how they appear to us, so we cannot have knowledge of things in themselves. Skepticism remains.
Beyond Space & Time Dror Rosenski 2022
Kant Get No Thing-In-Itself
Responding to Hume’s ‘causal skepticism’, Kant will say that we do have knowledge of causation, and of other things too. But he argues against skepticism in an entirely different way.
Kant’s first move is to redefine the all-important term ‘experience’. For Hume, experience is either sense impressions, feelings, or reflections (unfortunately, Hume says nothing about the origin and nature of perceptions). For Hume, visual experience is a series of colored sense data in a parade of mosaics simply presented to sight as objects in various relations to one another. For Kant, however, an experience is a combined set of perceptions organized by the mind, and constituted in large part by our minds. Kant asks, if experience were just a mosaic of colors, how do we have our everyday meaningful sort of experiences, of tables and chairs, bricks, houses, and rain showers? Since these meaningful combinations of the sense data cannot come through the senses themselves, they must be rooted in the nature of thought and representation. Therefore, we can gain knowledge about the world we experience by considering the nature of these mental faculties and what belongs to any experience or thought whatsoever: for instance, the fact that they all exist within time.
To criticize skepticism about causes, Kant argues that we don’t simply experience events following or preceding one another; at times, we experience happenings. That is, we experience an event in a specific relation to time: something that did not exist before but does now. Moreover, the event (effect) can only be experienced in one direction: we experience a match striking and then it igniting. Whereas with experiencing objects we can do this in any order or direction we like: I experience the house from the basement up to the roof or I can start looking at it from the roof and move downward (B230). We experience the necessary connection between the two events, cause and effect (the ‘happening’), because the very possibility of experiencing something that happened implies the existence of a cause. Unfortunately, that’s all it implies. We cannot know anything about the character of the cause by simply experiencing the event. That is why specific causal relations are still a matter of experience. It is knowledge of causation itself that is a priori (ie knowable prior to experience).
Kant argues we can get further knowledge of nature itself in the same sort of way – by thinking about the implications of our experience being the way it is. On this way of thinking, what is an object, for example? The eighteenth century philosophers, aware of Descartes, say that a physical object is an extended being, meaning that it has magnitude in space. However, Kant first asks, is space something we know from experience ? Rather than experiencing space itself (whatever that might mean), every sensory experience we have is always already in space, or we could say, is always of something in space – it is over here, and not over there, it is next to this, behind that, etc. Neither is space something we can know about through abstract reasoning about the concept, since our abstract idea of space is of an infinite magnitude, and we can’t conceive of infinite magnitude. Experiences themselves are things with instances, not infinite magnitudes. So whence our idea of space? It must be gained from the a priori features of experience, specifically, the nature of our ‘outer intuition’ – our experience of the external world. For Kant, space is a subjective feature of all our outer intuition: he calls it the form of appearances (B42). It is something that’s necessary for us to have any experience of the world at all. Time works the same way, as a form presupposed by all sensory experience.
Now if space, as a pure form of outer intuition, is a function of experience itself, then so too must be objects, since a physical object is always and only perceived as being in space. In fact, all we know about objects is their appearances. As Kant writes, “Only through representation is it possible to know anything as an object” (B125). There is therefore no sense to saying anything about the nature of an object independent of our experience. Indeed, it seems that for Kant there is no such thing as a thought-independent individual object.
Can we avoid this line of thought simply by changing the discussion from the real nature of objects to being about how ‘things in general’ really are? Since the only idea of reality we have stems from our experience of the world, asking ‘How are things in general?’ in one sense is just asking about our experience. The direct objects of knowledge of the world are appearances. By showing how the only knowledge of objects is had through looking to our experiences of them, Kant ‘solves’ the problem of skepticism by dissolving it. What is the nature of Nature? Nature for Kant means the causally-connected world we experience. And as he says, Nature is just what we experience of it.
But can’t we also ask about what lies behind or outside our experiences? For Kant, experience of the world involves sensation; so we might ask, from where do these sensations come? Kant argues that the idea that our sensations have a source beyond them is a problematic concept – a concept that does not involve a contradiction but which is impossible to either affirm or deny. Any possible source of sensations that you might suggest (our minds? physical objects in themselves? God?) is impossible to affirm or deny because our experience is of the sensations and nothing else, and logical speculations about the ultimate nature of reality end up in contradictions which we cannot resolve, such as that space is both infinite and non-infinite.
So Kant might ‘solve’ skepticism about experience to some extent by redefining what we mean by ‘experience’, ‘object’, and ‘reality’. But doesn’t he thereby create a whale of a new problem? Kant calls things in themselves the noumena, as opposed to the phenomena of sensory experience (B297). (‘Phenomena’ literally means ‘things of the senses’. ‘Noumena’ means ‘things of the intellect’.) So doesn’t Kant’s method just mean that we cannot have knowledge of things in themselves, the noumena?
This is where this whale of a metaphysical problem delights in the murkiness of its depths. We should remember that Nature – the world of objects – is a feature of outer intuition or appearances. So by this definition, noumena are not part of Nature. Also remember that for Kant an object is only ever an appearance or set of appearances, and whatever we can reliably infer from that. So noumena are not objects, either. Furthermore, space is the form of our outer intuition, or one prerequisite of our experience of phenomena; so for something to even be in space it must be an appearance. Noumena are therefore not in space. This also means they’re not outside or beyond anything in a physical sense. Finally, we should not forget (although Kant sometimes seems to) that causation is likewise a concept required for experience by the nature of sensory representation. Causation is specifically, a relation between two appearances: the cause and the effect. So what could Kant possibly mean if he were to say that noumena cause the appearances of the world? Noumena cannot cause anything in Kant’s sense of cause. So although Kant argues that there is sense to the idea of something existing independent of our thought – a world as it is in itself – the noumena is nevertheless a problematic concept, in that we cannot know anything about it.
The world as it is in itself isn’t a part of nature, or the cause of anything, and we cannot know anything positive about it. So did Kant solve skepticism?
© Thomas Morrison 2022
Thomas Morrison is a graduate student at the University of Massachusetts Amherst with a passion for public philosophy.