Skip to Navigation Skip to Content Skip to Search Skip to Site Map

ESP is best

Peter Anstey writes…

There are two ways to carve up 17th and 18th century philosophy: the traditional way is to divide it into rationalist versus empiricist philosophy (REP); a new way is to divide it into experimental versus speculative philosophy (ESP). We argue that the ESP way is far better than the traditional terms of reference.

Let’s start the comparison by pointing out the fact that the ESP distinction provided the actual historical terms of reference that many philosophers and natural philosophers used from the 1660s until late into the 18th century. There are literally scores of books from the period that use these terms and deploy this distinction (They are used by Boyle, Hooke, Sprat, Glanvill, Cavendish, Locke and Newton). By contrast, the terms ‘Rationalism’ and ‘Empiricism’ (and their non-English cognates) were introduced by Kant and his followers in the late 18th century. One can find very occasional uses of the terms in the earlier period, but they have completely different meanings. For example, ‘empiricism’ in Johnson’s Dictionary (1768) means ‘Dependence on experience without knowledge or art; quackery’.

Second, on the standard view, REP is largely about epistemology, that is the origins of ideas and sources of knowledge. The REP view has it that rationalists claimed that there are innate ideas and that these are the foundations of knowledge: the empiricists claimed that all ideas originate in the senses and that knowledge is built upon experience and not innate ideas or principles. By contrast, ESP is largely about methodology, about how to proceed in acquiring knowledge, especially knowledge of nature. It includes questions about the sources of knowledge and ideas, but it also includes views on the nature of hypotheses, principles, theory, mathematics, experiment and natural history.

So ESP has more explanatory range than REP and allows a more nuanced understanding of individual philosophical positions and debates. One important example is Newton’s rejection of hypotheses. This is very nicely explained by ESP, but is largely irrelevant to REP and has therefore posed a problem for scholars who approach Newton from the REP framework.

Third, you’ll notice that I highlighted ‘standard view’ above. This is because nowadays it is pretty difficult to settle on exactly what ‘Empiricism’ and ‘Rationalism’ mean. The Hume scholar Don Garrett reckons that there are 5 types of Empiricism. The Locke scholars Jonathan Lowe and Michael Ayers give us 3 and 2 types respectively. Paul Feyerabend reckoned that there were 3 types and Ernan McMullin has a few more. This proliferation of empiricisms is an indicator that the term no longer earns its keep.

By contrast, the term ‘experimental philosophy’, while it admits of some latitude of application, its widespread use by the philosophers themselves means that it’s pretty easy to work out whether someone is an experimental philosopher or not and whether they are sympathetic to it.

Fourth and finally, there’s the issue of demarcation. According to REP the leading empiricists are normally thought to be Locke, Berkeley and Hume and the leading rationalists are Descartes, Leibniz and Spinoza. But even this is often contested: Des Clarke has Descartes as an empiricist; Richard Aaron had Locke as a rationalist; Nicholas Rescher has Leibniz as an empiricist! And where does Robert Boyle, the archetypal experimental philosopher sit? He promoted observation by the senses, but he was partial to innate ideas. ESP provides a far more natural line of demarcation and one that explains this vacillation on the part of modern scholars.

All in all ESP is better. So what should we do with the Rationalism and Empiricism distinction? Are there any good reasons to retain it?

2 thoughts on “ESP is best

  1. This is a good and welcome posting, promoting much needed reconsideration of the debates within early modern philosophy. Only one point I would like to correct and this is the notion that Kant was responsible for the “empiricist”/”rationalist” division. In fact, his account is considerably more facetted than that. The division he uses is between “empiricists” and “dogmatists”, he does not refer to “rationalists”. Newton’s views on space and time are cast by him as “dogmatic” whilst Leibniz is taken by Kant to have the “empiricist” view on this area. Later, Kant also classifies both Locke and Berkeley as “dogmatic” albeit for different reasons. So the standard view does not come from Kant!

  2. Hi Gary. Thanks for your comment. I agree that Kant does not normally contrast or distinguish empiricism and rationalism. As you note, he contrasts empiricism and dogmatism (in the Antinomies). Yet he also contrasts empiricists and noologists in the “History of Pure Reason” (A854/B882). This is only one of several alternative classifications of previous philosophers that Kant outlines, and certainly not the most important. The most important is probably the distinction between dogmatism, scepticism, and criticism — the three stages of the necessary historical progress of human reason.

    I also agree that Kant the canonical lists of empiricists and rationalists cannot be found in Kant (though he classes Locke as an empiricist and Leibniz as a noologist in B882).

    A few years later, in 1792, Reinhold calls Leibniz a rationalist (not a noologist) and Locke an empiricist. According to Reinhold, Locke’s empiricism, Leibniz’s rationalism, and Hume’s scepticism are important historical stages of the historical progress towards Kant’s Critical philosophy. I haven’t thoroughly studied Reinhold’s views on this topic yet. However, it seems to me that he is developing Kantian ideas: in particular, the distinction between noologism and empiricism and the idea that there are certain stages in the necessary historical development of human reason.

    If my guess is right, the canonical classifications of empiricists and rationalists will come later, with Tennemann and/or Kuno Fischer. These historians of philosophy, like Reinhold, were influenced by Kant, and developed some Kantian ideas, together with original views. So I agree with you that Kant is not the father of the standard view. He should also not be blamed for the simplistic views of some later historians. However, I would still regard Kant as the grandfather of the standard view. How does this sound?