copyright Robert M. Ellis 2011

Common but unhelpful assumptions 4: The acceptability of pure analysis not applied to concrete contexts

If Western thought is really a set of footnotes to Plato, as Whitehead claimed, then Plato is answerable for a good deal - a whole 'abstracted turn' in Western thought. For Plato applied sceptical arguments selectively, only to support distrust of information from the senses, whilst at the same time asserting the absolute status of the Forms known through reason alone. Even today, when most philosophers would not accept Plato's metaphysics of the Forms, the results of a priori analysis, establishing what must always necessarily be the case, are still often given a status which in practice is metaphysical. For example, the status of mathematics (whether it is treated as analytic or synthetic) and of the laws of logic are often treated as non-negotiable truths. In the jargon of analytic philosophy, 2+2=4 and a bachelor is an unmarried man in all possible worlds.

However, the sceptical arguments that can be turned against these claims of reason were put forward effectively by Hume and other empiricist philosophers. We can establish the consistency of an a priori framework, such as that of mathematics, but consistency is not adequate to show absolute truth. We may not be able to imagine using any other framework to replace mathematics, but it is nevertheless just a framework. Mathematics is a filter that we apply to make sense of the universe, but we can have no idea whether mathematics is true of the universe itself in any sense. Nor can we establish the converse, that mathematics is definitely not true of the universe. Consistent analytic claims provide us with more information about what we already assume, not with new information beyond this sphere. To assume otherwise is no less of a metaphysical assumption than the existence of God. A mathematician has experience of mathematical consistency which she takes to be evidence of the laws of the universe, but she has as little ground for doing so as a person having an overwhelming religious experience jumping to the conclusion that God is known to exist with certainty through that experience. In both cases cultural norms and group expectation play a large part.

This point can be seen more clearly if we consider a priori thinking in its concrete context. The tendency in analytic philosophy is both to abstract out of the concrete context and to also conveniently forget the limitations of that concrete context and treat it as absolute. Mathematical reasoning is always done by a person with a body in a particular place, for a particular purpose: whether this is a purpose that has further application (such as engineering) or the academic purpose of solving a problem or making a calculation in 'pure' maths. The meaning of the mathematical reasoning depends, not just on the abstract significance of the symbols, but also the motives and physical context of the reasoning. For that reasoning to have anything like the same significance for others, at least some of the contextual meaning must be shared. For example, a person whose brain has not been sufficiently trained to have the right synaptic connections to understand mathematics will not feel the mathematical reasoning to have the same significance as its author. At the same time, for those who do share its significance, the reasoning may have great importance.

A priori thinking is particularly limited because with its timeless necessities it is static. However, everything in our concrete experience is changing. Not only our motives, but our sense of significance and our beliefs change from moment to moment. For this reason a priori reasoning is limited in its application to our experience: it deals with knowledge, not learning (or, for that matter, forgetting), and its attempts at timeless truth bear no relationship to our experience of development.

The limitations of a priori reasoning are conveniently forgotten when it is applied to ethics. It is often assumed in Western thought that analysis is the only kind of justification we can provide for ethics: whether that analysis is assumed to be universal in a naturalistic ethics, or (more convincingly) taken to be unavoidably relative. But analysis only tells us about what we already assume, whilst the whole purpose of ethics is to stretch and challenge our moral thinking into greater objectivity. It is assumed, without the slightest justification, that the only possible accounts of moral objectivity that could be given are either metaphysical or analytic. However, there are other possible accounts of moral objectivity, such as the epistemological and psychological account offered on this website. These are accounts that think dynamically about the concrete process by which our limited moral understanding at one time can be expanded at another, instead of in the static dualism of relative vs absolute knowledge at one time. The dead hand of decontextualised analysis has just closed off whole avenues of thought for us, and quite unnecessarily made relativism an intractable problem.


Links to related pages on this website

Truth on the Edge

Plato (from thesis)

Hume (from thesis)

Analytic philosophy (from thesis)


Return to home page

StumbleuponCiteULikedel.icio.usRedditDigg itFacebook
Social Bookmarking Service



Seven common but unhelpful assumptions in Western thought:

1. The negative implications of scepticism

2. The need to accept or reject metaphysical claims

3. The identification of objectivity with absolute claims

4. The acceptability of pure analysis not applied to concrete contexts

5. An account of meaning confined to representation or expression

6. The fact-value distinction

7. The identification of ego with self

Concepts index

Discussion forum

Return to home page