Hume: Induction

Interpreters of Hume on causation consider that an advantage of the ‘quasi-realist’ reading is that it does not commit him to scepticism or to an error theory about causal reasoning. It is unique to quasi-realism that it maintains this positive epistemic result together with a rejection of metaphysical realism about causation: the quasi-realist supplies an appropriate semantic theory in order to justify the practice of talking ‘as if’ there were causal powers in the world. In this paper, I problematise the (...) quasi-realist reading of Hume on causation by showing how quasi-realism does not speak to inductive scepticism. I also offer evidence that Hume takes inductive scepticism to result from his theory of causation, and that his scepticism is tied to his rejection of metaphysical causal realism. (shrink)

Across a series of papers and again in her recent book, Graciela De Pierris has argued that Hume is what she calls an inductivist about the methods of science. De Pierris takes Hume to follow Newton in holding that the ultimate aim of science is to seek "assurance concerning objects, which are removed from the present testimony of our memory and senses",1 and its method therefore to consist in the subsumption of observable particulars under inductively-established universal generalizations. As De Pierris (...) puts it, "[t]he central idea of the Newtonian inductive method, as summarized in his Rules, is that exceptionless or nearly exceptionless universal laws are inductively derived from 'manifest... (shrink)

According to Hume, the paradigm type of inductive reasoning involves a constant conjunction. But, as Price points out, Hume misrepresents ordinary induction: we experience very few constant conjunctions. In this paper, I examine several ways of defending Hume’s account of our practice against Price’s objection, and conclude that the theory cannot be upheld.

Philosophers talk routinely of ‘Hume's problem of induction’. But the usual accompanying exegesis is mistaken in a way that has led epistemologists to conceive of ‘Hume's problem’ in needlessly narrow terms. They have overlooked a way of articulating the conceptual problem, along with a potential way of solving it. Indeed, they have overlooked Hume's own way. In explaining this, I will supplement Hume's insights by adapting Ryle's thinking on knowledge-how and knowledge-that. We will also see why Hume's ‘sceptical solution’ was (...) a perfectly appropriate response to his ‘sceptical argument’ — rather than a merely descriptive response patently missing the normative sceptical point so strikingly formulated by Hume. (shrink)

In a recent Philosophy of Science article Gerhard Schurz proposes meta-inductivistic prediction strategies as a new approach to Hume's. This comment examines the limitations of Schurz's approach. It can be proven that the meta-inductivist approach does not work any more if the meta-inductivists have to face an infinite number of alternative predictors. With his limitation it remains doubtful whether the meta-inductivist can provide a full solution to the problem of induction.

Hume fully recognised, and partially explained, the role of inductions from single experiments in human knowledge - something his Scottish critics, and some more recent ones, failed to understand. Those inferences, he maintains, depend on the use of a Newtonian rule and the removal of superfluous circumstances. But that rule is not sufficient, and Hume never stated the exact conditions of this removal. We should distinguish between survey and experience in his philosophy, to understand how experience of conjunctions where inductive (...) knowledge has accumulated, in uniform or regular classes or sets of phenomena, may produce cognitive situations when once is enough - without relying on any mysterious innate principles. (shrink)

A common criticism of Hume’s famous anti-induction argument is that it is vitiated because it fails to foreclose the possibility of an authentically probabilistic justification of induction. I argue that this claim is false, and that on the contrary, the probability calculus itself, in the form of an elementary consequence that I call Hume’s Theorem, fully endorses Hume’s argument. Various objections, including the often-made claim that Hume is defeated by de Finetti’s exchangeability results, are considered and rejected.

This paper starts from the familiar premise that psychological anti-individualism is incompatible with materialism. It attempts to state more clearly what this incompatibility consists in, and — rather than arguing in detail for any particular resolution — to inquire whether this incompatibility admits any resolution. However, the paper does offer a conditional argument concerning the possibility that the incompatibility is genuine and cannot be resolved. Provided that anti-individualism and materialism cannot be squared, and anti-individualism is correct, it follows that materialism (...) has to be abandoned. If so, the situation is not as disastrous as it might at first seem. We need not, in consequence of our inability to construe a coherent metaphysics of mind, give up on intentional vocabulary any more than we must stop, in consequence of our inability to make sense of induction, anticipating the future. (shrink)

David Hume's famous riddle of induction implies a second problem related to the question of whether the laws and principles of nature might change in the course of time. Claims have been made that modern developments in physics and astrophysics corroborate the translational invariance of the laws of physics in time. However, the appearance of a new general principle of nature, which might not be derivable from the known laws of physics, or that might actually be a non-physical one (this (...) means completely independent of physical science) supports the notion that the course of nature can change in time. Here it is argued that natural selection satisfies the criteria that identify a general principle of nature which so far, appears to be non-derivable from the known laws of physics and therefore, it is likely that it arose in the course of time, thus leaving open again the quest for a true solution to Hume's second problem. (shrink)

Hume follows Newton in replacing the mechanical philosophy’s demonstrative ideal of science by the Principia’s ideal of inductive proof ; in this respect, Hume differs sharply from Locke. Hume is also guided by Newton’s own criticisms of the mechanical philosophers’ hypotheses. The first stage of Hume’s skeptical argument concerning causation targets central tenets of the mechanical philosophers’ conception of causation, all of which rely on the a priori postulation of a hidden configuration of primary qualities. The skeptical argument concerning the (...) causal inductive inference then raises doubts about what Hume himself regards as our very best inductive method. Hume’s own “Rules” further substantiate his reliance on Newton. Finally, Locke’s distinction between “Knowledge” and “Probability” does not leave room for Hume’s Newtonian notion of inductive proof. (shrink)

There is a tradition perhaps as old as philosophy itself which sees the rationality of man – and in particular, the rationality of the philosopher - as both his essential and his redeeming characteristic; it can not unfairly be said that the discipline of philosophy at least is characterised by its dependence on reason. In this context, the philosophy of David Hume presents something of a stark challenge: Although interpretations vary as to the extent and nature of his scepticism, one (...) of the themes of his work is the limits and imperfections of human reason and the sceptical concerns this engenders. On Hume's system, reason is not the overlord of the imagination but at least in part subordinate to it, and can discover little or nothing without recourse to experience. -/- My aim here is to pinpoint the true nature of Hume's scepticism regarding reason, drawing chiefly on the Treatise but also to a lesser extent the Enquiry. After a brief overview of his project in the Treatise in section 1, I go on to give an account of Hume's conception of reason or the understanding, and the epistemological distinction between demonstrative and probable reasoning, in section 2. In section 3 I examine Hume's famous argument concerning induction, arguing that to read Hume as a sceptic regarding inference is a mistake and that to say induction is grounded in principles of the imagination is not to say that it is therefore irrational. In section 4 I examine the argument of Treatise 1.4.1, 'Of scepticism with regard to reason', and in section 5 the 'dangerous dilemma' to which it leads Hume in 1.4.7, arguing that whilst his treatment of the issues it raises is somewhat unsatisfactory in this section of the Treatise a better solution can be fashioned by appealing to a distinction between the general and the trivial properties of the imagination drawn elsewhere in both the Treatise and the Enquiry. Then in section 6 I conclude by offering my own attempt at characterising the extent of Hume's scepticism regarding reason, arguing that his “mitigated” scepticism is a scepticism about knowledge but not about belief, and that its chief aim is to ensure that we philosophise only on those subjects which are suitable for enquiry given the limits and frailties of our faculties; the “reflections of common life, methodized and corrected” (Enquiry 12.25, p.208) rather than “divinity” or “school metaphysics” (Enquiry 12.34, p.211). (shrink)

Review of Bredo Johnsen's "Righting Epistemology: Hume's Revolution" (OUP, 2017).

Much of what Hume calls probable reasoning is deliberate and reflective. Since there are aspects to Hume’s psychology that tempt some commentators to think, on the contrary, that for Hume all such reasoning is simple and immediate, I will be concerned to emphasize Hume’s recognition of the sophisticated sort of probable reasoning (section I). Though some of the details of my case may be new, the overall point of this section should not be news to recent scholarship. But once we (...) recognize that this reflective and deliberate reasoning constitutes a significant portion of all probable reasoning, it becomes legitimate to ask how Hume accommodates this reasoning in his psychology, his ‘science of man.’ I believe that .. (shrink)

In this paper, I discuss Eino Kaila's (1890-1958) understanding of David Hume. Kaila was one of the leading Finnish philosophers of the 20th century and a correspondent of the Vienna Circle. He introduced logical empiricism into Finland and taught Georg Henrik von Wright. Final draft.

Many solutions of the Goodman paradox have been proposed but so far no agreement has been reached about which is the correct solution. However, I will not contribute here to the discussion with a new solution. Rather, I will argue that a solution has been in front of us for more than two hundred years because a careful reading of Hume's account of inductive inferences shows that, contrary to Goodman's opinion, it embodies a correct solution of the paradox. Moreover, the (...) account even includes a correct answer to Mill's question of why in some cases a single instance is sufficient for a complete induction, since Hume gives a well-supported explanation of this reliability phenomenon. The discussion also suggests that Bayesian theory by itself cannot explain this phenomenon. Finally, we will see that Hume's explanation of the reliability phenomenon is surprisingly similar to the explanation given lately by a number of naturalistic philosophers in their discussion of the Goodman paradox. (shrink)

In this revised and updated edition of The Secret Connexion, Galen Strawson explores one of the most discussed subjects in all philosophy: David Hume's work on causation. Strawson challenges the standard view of Hume, according to which he thinks that there is no such thing as causal influence, and that there is nothing more to causation than things of one kind regularly following things things of another kind. He argues that Hume does believe in causal influence, but insists that we (...) cannot know its nature. The regularity theory of causation is indefensible, and Hume never adopted it in any case. (shrink)

Hume fully recognised, and partially explained, the role of inductions from single experiments in human knowledge - something his Scottish critics, and some more recent ones, failed to understand. Those inferences, he maintains, depend on the use of a Newtonian rule and the removal of superfluous circumstances. But that rule is not sufficient, and Hume never stated the exact conditions of this removal. We should distinguish between survey and experience in his philosophy, to understand how experience of conjunctions where inductive (...) knowledge has accumulated, in uniform or regular classes or sets of phenomena, may produce cognitive situations when once is enough - without relying on any mysterious innate principles. (shrink)

I outline Hume’s views about conceivability evidence. Then I critically scrutinize two threats to conceivability-based modal epistemology. Both arise from Hume’s criticism of claims to knowing necessary causal relationships: Firstly, a sceptical stance towards causal necessity may carry over to necessity claims in general. Secondly, since – according to a sceptical realist reading – Hume grants the eventuality of causal powers grounded in essential features of objects, conceivability-based claims to comprehensive metaphysical possibilities seem endangered. I argue that although normal conceivability-based (...) claims are defeasible, they are prima facie vindicated. (shrink)

A common criticism of Hume’s famous anti-induction argument is that it is vitiated because it fails to foreclose the possibility of an authentically probabilistic justification of induction. I argue that this claim is false, and that on the contrary, the probability calculus itself, in the form of an elementary consequence that I call Hume’s Theorem, fully endorses Hume’s argument. Various objections, including the often-made claim that Hume is defeated by de Finetti’s exchangeability results, are considered and rejected.

The principle of uniformity of nature has sometimes been invoked for the purpose of justifying induction. This principle cannot be established "a priori", And in the absence of a justification of induction, It cannot be established "a posteriori". There is no justification for assuming it as a postulate of science. Use of such a principle is, However, Neither sufficient nor necessary for a justification of induction. In any plausible form, It is too weak for that purpose, And hence, It is (...) insufficient. Since a justification which does not rely upon this principle can be given, It is not necessary. (shrink)