Sumários

Lent Break

1 Março 2022, 09:30 • António José Teiga Zilhão

Lent break

Reason, Agency and Obligation

22 Fevereiro 2022, 09:30 • António José Teiga Zilhão

Lecture by Prof. David Owens (King's College London - UK) titled "Reason, Agency and Obligation".

The Problem of Induction - 3

15 Fevereiro 2022, 09:30 • António José Teiga Zilhão

I. Kant's Response to Hume's Problem

1. The 'Law of Universal Causality' (LUC) as Kant's alternative to Hume's 'Principle of the Uniformity of Nature' (PUN). 
2.  LUC has the status of an a priori principle of organization of our subjective perceptions; from Kant's standpoint, what distinguishes simple streams of subjective perceptions from temporally ordered objective experience is the fact that the latter is itself the result of the understanding's unification of the former under the universal and necessary connexion provided by LUC.
3. Kant's re-description of Hume's PUN in terms of his own LUC provides, at the same time, Kant's response to Hume's problem of justification: the reason why the principle of uniformity admits of no non-circular justification lies in the fact that it is this a priori principle that constitutes experience, rather than being derived from it. 
4. Thus, from Kant's standpoint, there is no need to appeal to habit or custom to justify induction; the validity of induction is to be traced back to the a priori truth of LUC.  

II. Some Critical Comments on Kant's Response to Hume's Problem

II.1. Some Critical Comments on Kant's Response to the Descriptive Aspect of Hume's Problem

If we were to organize in a causally coherent way our stream of subjective perceptions by means of a priori knowledge, in the way Kant says we do, we would need to bring to the fore much more than just an abstract and unspecific law of universal causality; in order to accomplish such a job, we would need to take into consideration many specific causal expectations. But we find no clue in Kant's writings on how to unravel LUC in such an usefully specified manner. More important, however, is the fact such specific causal expectations would have to contain a lot of empirical information; but then, of course, the threat of circularity would loom large.

II.2. Some Critical Comments on Kant's Response to the Justification Aspect of Hume's Problem

Even if we assume that Kantian thinking might be able to solve, in some way or other, LUC's causal specification problem, and, therefore, the descriptive difficulties brought about by Hume's formulation of the PUN, the issue of justification would not go away. As a matter of fact, under such imaginary circumstances, Kant would have shown that we cannot help seeing the world the way we do (something Hume had already done, by the way), but he would not have shown that we are in the least justified in seeing the world that way.  

The Problem of Induction - 2

8 Fevereiro 2022, 09:30 • António José Teiga Zilhão

Goodman's New Riddle of Induction

The two dimensions involved in Hume's problem of induction: the dimension of justification and the dimension of description. The descriptive problem associated with Hume's use of his Principle of Uniformity of Nature (PUN): what are the notions of 'sameness' and 'resemblance' involved in the assertions that "the course of Nature continues always uniformly the same" and ""instances, of which we have had no experience, must resemble those, of which we have had experience"? There is no answer to these questions to be found in Hume. The introduction of the 'New Riddle of Induction' (NRI) by Nelson Goodman as a means to highlight this problem. The conclusion of the NRI: the regularities there to be found in Nature depend upon the systems of classification we use to describe it; different systems of classification generate different regularities (different criteria of 'sameness' and 'resemblance') and may give rise to contradictory predictions, equally justifiable according to standard inductive rules; but how are we to choose among them? Goodman's notions of 'projectible' and 'non-projectible' predicates and of 'entrenchment'. Discussion of these notions. 

The Problem of Induction - 1

1 Fevereiro 2022, 09:30 • António José Teiga Zilhão

Hume's Problem of Induction. 


Sound combinatorial ways of acquiring uncertain knowledge - statistical syllogisms and statistical generalizations. Notion of a Humean induction: inductions or generalizations projecting present observations into the future or onto unexperienced parts of the world. Humean inductions are not sound statistical generalizations. As Hume showed, any attempts at justifying them reveal themselves to be circular. Hume's appeal to the Principle of Uniformity of Nature (PUN). Humean inductions supplemented with the PUN as an extra premise are indeed valid. But how is the PUN itself to be justified? As Hume also showed, attempts at justifying the PUN are themselves circular. Hume's conclusion: there is no rational way of justifying inductive reasoning; it is a habit we acquired from Nature; it remains with us because of its practical usefulness.