The Empirical and the Probable: Clarence Irving Lewis’ Response to Hume

I find Trin’s notion of causal minimalism quite interesting.  At first brush, it has reminded me of Husserl, insofar as Husserl metaphysically “brackets” existence from phenomenology and “transcendentalizes” psychology, thereby leaving Husserl only with an epistemological investigation by which to study consciousness, mental states, and appearance on the level of inference and a priori relations.  Actual features of psychology and any relation to reality are left uninvestigated.  Many people remain quite uncomfortable with Husserl on this, as it “radicalizes” Kant by taking not only the forms of experience but the very content of experience and objectifying the whole of subjectivity.  This was in essence Husserl’s solution to Hume, while simultaneously rescuing Kant from Hegel.  The underlying problem with a purely epistemic minimalism that arises in Husserl is the problem of solipsism qua intersubjectivity.  If we take the Cartesian standpoint of objectifying consciousness and generating a phenomenological investigation of appearance and analyzing all things as logical and conceptual relations how do we ensure that our investigations of the transcendental ego match up with real egos in practice?  How do you test for mental state vs. brain state agreement?  This may or may not be a problem with a solution but it is a problem worth remembering.

One of the advantages it seems that epistemological minimalism commits us to is not having to take into account what could arguably be a matter of other investigations to consider, psychology being left to the psychologists—reality being left to anyone but philosophers.  I don’t think causality itself in this light would be susceptible to Husserl’s problem as such.  However, I have been considering the issue Josh has raised. I appreciate what he is saying about Hume and whether or not there is a “gap” that exists that can eventually be filled in by someone.

Relativity is Not Arbitrary!

After thinking about this a bit I want to ring up the second person that came to my mind, which is C. I. Lewis. I have gone back to Mind and the World Order to examine the chapters on experience and probability, which I will now explicate.

One of the most powerful arguments Lewis provided against relativism was to take the relativity of knowledge and objectify it logically in an absolutely certain way.

“[R]elational truth may nevertheless be absolute.  To put the matter in general terms: If relative to R, A is X, and relative to S, A is Y, neither X nor Y is an absolute predicate of A. But ‘A is X relative to R’ and “A is Y relative to S’ are absolute truths…..These relative (or relational) characters, X and Y, are partial but absolutely valid revelations of the nature of A.” (MWO, 168)

The outstanding question then remains, how far can this be taken with probability-judgments?

Hume, Probability, and Infinite Regress

I will come back to this logical point later.  In the mean time, Lewis makes an interesting move by enlarging the field of Hume’s skepticism.  Lewis contends that if we take Hume to his logical conclusion, probability suffers an infinite regress argument.  Whereas Hume stipulates that all matters of fact are constant conjunctions based on probability of habituated association, the grounds for probability are themselves probable, insofar as ultimately the probability-judgments of matters of fact are themselves probably true.

“The validity of probability-judgments rests upon antecedent truths which must be certain.  If all knowledge should be empirical and such principles therefore mere generalizations from experiences, then these principles would be only probable; with the result that the knowledge which depends on them and is ordinarily called ‘probable’ would be only probably probable…and knowledge would disappear in an infinite regress of such qualification.” (MWO, 311-2)

The central problem for Lewis is how do we get from the problem of necessary connections in casual relations to genuine knowledge when all empirical facts and states of affairs of the world are preconditioned upon being probability-judgments?  Lewis believes the only way around Hume’s Fork (see previous post) is to establish an absolute certainty as the foundation for an aspect of experience by which probability-judgments conceptually (epistemologically) interpret the world.

Lewis provides an interesting solution though it has come under attack by the two dominating thoughts of post-analytic philosophy, one by Sellars one by Quine.  One might argue that Lewis’ epistemology, dominant through the 1950’s is the very end of modern philosophy, and epistemologically, of early analytic philosophy.  The position Lewis takes is that the givenness of experience, as qualia, by its very immediacy, presents itself as a type of unanalyzed certainty by which becomes logically validated through organizing the givenness of experience into the analytic-synthetic distinction, whereby the immediate content becomes instantenously mediated conceptually through our mind’s organizing experience into varying categories of purported truths, definitions, identities, and then empirically generalizing these to form probability-judgments.  Here Sellars attacks Lewis in the Myth of the Given in terms of foundationalism epistemology qua non-inferential aspects of givenness and Quine attacks Lewis in terms of the analytic-synthetic distinction with the upshot being that Quine does not believe empiricism has a sort of rational foundation that Lewis presents by melding Peirce and Kant in his conceptual pragmatic epistemology.

Recognition is Causal!

Putting Lewis’ solution aside and whether we wish to follow it I will leave aside.  I want to bring up Lewis’ thoughts on Hume.  Interestingly, Lewis contends that Hume was not logically consistent with his views on necessary connection.  Recall that Hume conceives of three types of relations: resemblance, continuity, and causality.  “A picture naturally leads our thoughts to the original (resemblance = picture): The mention of one apartment in a building naturally introduces an enquiry or discourse concerning the others (continuity = time or place relationship): And if we think of a wound, we can scarcely forebear reflecting on the pain which follows it (causality).  The question for Lewis is simple: what is identity?  Hume assumes no difficulty epistemologically in the recognition of objects as what they are.  In fact, for Hume this does not involve causality as much as an “intuitive relation of the mind”.  But as Lewis asserts, this is impossible, a world without law is a world without recognizable things. (MWO, 320)  One cannot undermine the nature of necessary connection of causation without subsequently undermining the brain’s psychological ability to absolutely connect with the objects that resemble what we believe.