Immanuel Kant’s Metaphysics: What are the Limits of Our World?

Immanuel Kant has many modern day detractors, and has inspired a great degree of antipathy. Nonetheless, no matter what side of contemporary philosophical debates you happen to be on, it is difficult to deny that Kant permanently revolutionized philosophy.

This article aims to get at the very core of his philosophy, and introduce his metaphysics. It begins by discussing Kant in relation to the history of philosophy, both in terms of his influence and those prior philosophers which influenced him. We will then consider two distinctions – between the a priori and a posteriori, and between the analytic and synthetic – on which Kant builds his philosophical system.

Lastly, an explanation of the ‘synthetic a priori’, which has a strong claim to be the distinctive principle of Kantian philosophy, is given, along with some analysis of its initial plausibility.

Immanuel Kant, coming as he did at the end of the ‘Early Modern’ period in Western philosophy, can easily be characterized as just one in a sequence of responses to Descartes (Kant’s work responds to Hume, who responds to Leibniz, who responds to Spinoza, who responds to Descartes). However, it is important to stress Kant is the end of this period not just chronologically, but causally. It is a cliché to say that all philosophers after Kant are fundamentally post-Kantian, but it bears repeating nonetheless.

Richard Rorty, a famous self-conceived iconoclast, was attempting to structure a course which followed an alternative history of philosophy, one which left out all of the major figures whilst nonetheless telling a coherent narrative, and complained to a colleague that he couldn’t find a way to leave Kant out.

Kant was, in many ways, not just the end of but the culmination to this ‘Early Modern’ tradition in philosophy. He engaged with all of his predecessors to some extent, and saw fixing many of their mistakes as fundamentally enmeshed in the success of his own project.

Kant and Hume

Kant’s main influence was undoubtedly David Hume, whose philosophical project was – in a similar way to Kant’s – extremely critical of the metaphysics which came before him. In particular, he was convinced that many previous philosophers were involved in projects to answer questions or define concepts which simply did not yield themselves to philosophical inquiry. Descartes, Spinoza and Leibniz’s attempt to characterize substance is the archetypal example of this.

Yet Hume, on Kant’s account, takes too negative a view of the possibility of metaphysics, i.e. the attempt at understanding the nature of reality. To take one important disagreement as an illustrative example, David Hume holds that given that causation appears to be totally regular and yet cannot be justified as what he calls a ‘relation of ideas’ – it is conceivable, for instance, that dropping a pen it will float upwards, and yet it always in fact falls to the ground – causation has to be explained as a habit of our minds.

Yet for Kant this fails to make sense of causal judgments, which are not just about what happens in our minds, but about the nature of reality. Indeed, for Kant causation is a fundamental principle of reality at large, and should be considered as such rather than as an aggregative tendency of our psychology.

However, Kant agrees with Hume that we cannot establish the truths of causation as a relation of ideas. To understand the status of causation is to understand the definitive concept of Kantian metaphysics, and perhaps his philosophy as a whole.

For Kant, knowledge of causation is based in the synthetic a priori. Previously, various philosophers – particularly Leibniz and Hume – tended to follow a distinction between such things as mathematics or tautological truths, and the truths we acquire through experience, perception or observation. Kantian philosophy rests on the complication of this distinction.

Kant, in contrast, holds that we must distinguish between two different variables: whether we know something a priori or a posteriori, and whether we know something as an analytic or synthetic judgment.

To know something a priori is to know it “absolutely independently of all experience and even of all impressions of the senses”, and to know it a posteriori is to know it in any other way. For example, “Every grandmother has a child” is something we can know a priori; we do not need to go out into the world and gather evidence to check if it is true. On the other hand, “Cows love eating peas” is a statement whose truth needs to be established by gathering empirical evidence, and is thus a posteriori.

An analytic judgment has the following form: “if the predicate B belongs to the subject A as something that is (covertly) contained in this concept A, “A is B” is analytic”. Synthetic judgments do not follow this form. For example, “Vixens are female” is an analytic judgment because being female is contained in the concept of “vixen”. “There are 15 million vixens”, on the other hand, is synthetic, as there is nothing contained in the concept of vixens that establishes there being a certain number of them.

The term ‘predicate’ is a technical way of describing the part of a sentence which says something about the subject of that sentence. For instance, in the sentence ‘the red pen’, the predicate is ‘red’. Having established this, it is clear that the distinction Kant is trying to draw by separating the analytic from the a priori is a subtle one, if it exists at all.

In fact, our first instinct might be to deny that there is any difference between the a priori and the analytic, and likewise with the a posteriori and the synthetic. Kant denies this, claiming that whilst all analytic judgments are a priori, not all a priori judgments are analytic. Some a priori judgments, Kant argues, are really synthetic:

“To be sure, one might initially think that the proposition ‘7 + 5 = 12’ is a merely analytic proposition that follows from the concept of a sum of seven and five in accordance with the principle of contradiction. Yet if one considers it more closely, one finds that the concept of the sum of 7 and 5 contains nothing more than the unification of both numbers in a single one…. The concept of twelve is by no means already thought merely by thinking of that unification of seven and five, and no matter how long I analyze my concept of such a possible sum I will still not find twelve in it. One must go beyond these concepts, seeking assistance in … one’s five fingers, say, or … five points…”

The Plausibility of the Synthetic A Priori

Before going any further, it is worth saying something about why this view has some initial plausibility to its credit. First, the risk with anything like ‘Hume’s fork’ between matters of fact (knowledge acquired from experience) and relations of ideas (knowledge acquired from reflection and deduction) is that it creates two, entirely separate kinds of knowledge. One concern with this is that it has a destabilizing effect on the uniformity of the concept of knowledge.

That is, whenever I claim to know a certain thing, I am really making one of two quite different claims, one of which is probabilistic and one of which is certain (or at least, has a greater claim on some kind of epistemic certainty).

Another problem is that such a thorough distinction will struggle to explain how it is that the arenas of experience and deduction have so much to do with one another – how, for instance, mathematicians and physicists have so much to do with one another despite the fact that their respective fields sit on opposite sides of this divide.

There isn’t space here to set out Kant’s argument for the existence of the synthetic a priori thoroughly. However, I will conclude here by gesturing towards his strategy in general, by way of Adrian Moore’s summary.

It is worth stressing before doing so how important a concept the synthetic a priori is for Kant. As per Moore, “The question ‘How is synthetic a priori knowledge possible?’ thus assumes a much wider significance for Kant. It eventually comes to embrace the question, ‘How is knowledge of an independent reality possible?’, or, more broadly, ‘How is representation possible?”. In Kant’s own view, what he is defending is not the principle as such, but the very possibility of metaphysics after the skeptical philosophy of Hume.

Kant was deeply impressed by many of the things which David Hume (a committed empiricist) argued for. When we have knowledge of external reality, this is made possible because of the way it affects us, which is sensory. This is possible, however,

“[…] only because we have certain capacities for reception. Through these we ourselves make a contribution to the form and structure of our experience. It is as though we have native spectacles through which we view things. And because these spectacles are native, we can have a priori knowledge pertaining to them: we can know, a priori, how things must appear through them. Such knowledge is synthetic. For it does not accrue from pure conceptual analysis.”

To this extent, the synthetic a priori permits no strict distinction in all cases. between the component of knowledge which is ‘within us’ and which is ‘out there’. This is the central principle of Kantian philosophy, which almost everything else follows from.