Michael Williams on Popper

Michael Williams Problems of Knowledge: A Critical Introduction to Epistemology, Oxford Uni Press, 2001. At Amazon.

Ernest Sosa, of Brown University, writes

A masterly introduction to epistemology and an original contribution, this book succeeds on both levels. Those who know Michael Williams’s earlier work will not be surprised by the rich texture of his writing and by how well it conveys the history and geography of the land of epistemology, while staking out a position of his own within it. Without piling on references, never woodenly didactic, Williams’s manuscript still shows his mastery of the subject, in both its historical length and its contemporary breadth.

This is one of a genre of books where the central issue is the justification of beliefs with scarcely a  hint of the CR or Popperian approach. Williams is better than most because he actually mentions Popper several times, sometimes in passing and once with a few lines of dismissive criticism. The book received a number of five star reviews on the Amazon site and the one that is billed as the most helpful can be read to get a fair impression of the content.

After a diagnostic treatment of the foundational assumptions of Philosophical Skepticism, the epistemology (theory of knowledge) for which Williams finally argues is the so-called Contextual theory. While Contextualism rejects the assumptions of Foundationalism and its quarreling cousins, it allows, within a “default and challenge” framework, for: immediate knowledge, a methodology of fallibilism (i.e., falsification), and epistemological risk-taking. A deflationist approach to knowledge, contextualism is neither atomistic nor strictly holistic. It is critical to notice that Contextualism is not mere epistemological Relativism, as Williams says, “the relativist, like the sceptic, is a disappointed foundationalist.”

Incidentally, if “immediate knowledge” is required for contextualism to fly, it is dead in the water.

In the Introduction Williams approvingly refers to Popper’s “broadly rationalistic approach to understanding the world…as a kind of ‘second-order tradition…a practice of critically examining current ideas so as to retain only those that pass muster”. In a section on Conjectural Knowledge he refers to various forms of ‘inductive’ (non-demonstrative) inference that are acceptable in different contexts, then notes that in Popper’s view “neither science nor rational thinking generally has any connection with inductive procedures.”

He refers to “an elaborate epistemological structure” built by Popper, where the aim of testing is not to establish truth but to eliminate errors. Science proceeds by trial and error, not induction from observations. Theories are prior to observations.

Aligning Popper’s views with more conventional epistemological stances is not altogether straightforward…Poper is much more interested in methodological and demarcational problems than he is in ‘refuting the sceptic’. His aim is not so much to reply to scepticism in a conventional way as to alter our sense of what epistemology ought to be about…

In several ways Popper’s ideas represent a significant advance over empiricist foundationalism. There is much in Popper that a contextualist ought to welcome. But…his own emancipation from standard forms of sceptical argumentation is far from complete.

The first thing to notice is that falsification is nowhere near as simple as Popper’s early views imply (and as Popper himself came to recognize). We don’t automatically abandon a successful theory because one result runs counter to our expectations…Morevoer, it is not clear that the idea of falsification is detachable from our entertaining inductive expectations. After all, we expect refuted theories to stay refuted…

In taking Hume’s inductive scepticism at face value, and thus completely dismissing the idea of justification, Popper in effect contrasts a rigidly foundationalist conception of  justification with a contextualist notion of falsification. However, once the move to contextualism is made, there is no need to dismiss justification.

Popper’s problems stem from his failure to  make crucial distinctions. He does not distinguish two aspects of justification: entitlement and grounding. He therefore fails to see that the Prior Grounding conception of justification is not the only possible conception. In fact, he swallows it hook, line and sinker. As a result he does not see that his ‘conjecture and refutation’ model of inquiry can be recast as a model of justification – the Default and Challenge model – in which certain unchallenged beliefs represent default entitlements. Nor does he distinguish between radical and non-radical scepticism. Accordingly he endorses radically sceptical views like Hume’s, where his real interest is only in fallibilism.

This is a rather peremptory dismissal of Popper’s idieas, without any further explanation of the crucial distinctions that Popper fails to make. The distinction between entitlement and grounding looks like the difference that Popper draws between justifying a preference (given the evidence and state of the argument at the time) and the kind of justification that justificationists really want. That is a very important distinction and it can be defended without reference to default entitlements, and without suggesting that Popper has (foolishly?) swallowed some defective propositions “hook, line and sinker”.

The Amazon site now permits access to the Table of Contents and the Index of books so it is possibly to check very quickly whether books refer to our favorite people. Bartely never appears. Popper has moved from the status that he ejoyed last century when you could usually find a mangled version of his ideas in the philosophy of science. Since 2000 it is hard to find any reference to Popper at all in philosphy texts.

In this book by Williams the bibliograph lists Conjectures and Refutations and Objective Knowledge but there is nothing in the text to encourage a reader to proceed to read more. There is no serious engagement with the arguments, no quotes to give the flavour of the original. The best that you can say for contexualism is that it mimics the theory of conjectural knowledge,. In this book it is  wrapped up in 250 pages of very elaborate (scholastic?) argumentation without providing any helpful connection with live scientific issues or with the work of Popper and others such as Miller, Agassi and Bartley who have so much to offer.