Interpreters have found it exceedingly difficult to understand how Hume could be right in claiming that his two definitions of ‘cause’ are essentially the same. As J. A. Robinson points out, the definitions do not even seem to be extensionally equivalent. Don Garrett offers an influential solution to this interpretative problem, one that attributes to Hume the reliance on an ideal observer. I argue that the theoretical need for an ideal observer stems from an idealized concept of definition, which many interpreters, including Garrett, attribute to Hume. I argue that this idealized concept of definition indeed demands an unlimited or infinite ideal observer. But there is substantial textual evidence indicating that Hume disallows the employment of idealizations in general in the sciences. Thus Hume would reject the idealized concept of definition and its corresponding ideal observer. I then put forward an expert-relative reading of Hume’s definitions of ‘cause’, which also renders both definitions extensionally equivalent. On the expert-relative reading, the meaning of ‘cause’ changes with better observations and experiments, but it also allows Humean definitions to play important roles within our normative practices. Finally, I consider and reject Henry Allison’s argument that idealized definitions and their corresponding infinite minds are necessary for expert reflection on the limitations of current science.
This is a preview of subscription content, access via your institution.
Buy single article
Instant access to the full article PDF.
Tax calculation will be finalised during checkout.
In the Treatise, Hume offers two definitions twice, within a few pages of each other, and in the Enquiry he puts forward a very similar version of the Treatise definitions. In this paper, I assume, as most scholars do, that the variations in the different versions are not significant. Because the Treatise is the central text of this paper, I employ in my discussions the (first) Treatise version of the definitions.
Robinson (1966), p. 131.
The actual definition of ‘cause’, according to Robinson, says that two objects are causally related when they instantiate event types which are universally connected. The second definition, according to Robinson, is merely a statement of an empirical psychological thesis that the observation of pairs included in the first definition is sufficient to induce their association in the mind. For Hume’s explicit endorsement of the first definition see T 126.96.36.199-15; EHU 7.27.
See, for instance, T 188.8.131.52-10.
In the Treatise Hume seems sympathetic to the objection that both definitions are “defective” because “drawn from objects foreign to the cause” (T 184.108.40.206). Both definitions seem also incompatible with Hume’s references to “secret powers” or “ultimate principles [that] bind causes and effects together,” of which we are forever ignorant. See, for instance, T 220.127.116.11-9; T 18.104.22.168-31; T 22.214.171.124-9. For a helpful discussion of all of these different positions, and their problems, see Garrett (1997), pp. 96–117.
Robinson (1966), p. 131.
Garrett (1997), p. 108.
Garrett (1997), pp. 109–110.
Allison (2008), pp. 201–202.
Allison (2008), p. 202.
Allison (2008), pp. 201–202.
Garrett (1997), pp. 102–103.
Garrett (1997), p. 110.
Garrett remarks that Hume “generally treats ‘constant conjunction’ as something that an individual person may or will already have observed at a given time.” Garrett (1997), p. 109.
I offer an interpretation of Hume’s foundational project in Boehm (2013).
For a good discussion on the normative authority of Hume’s rules for judging causes and effects see Martin (1993).
However, the passage above from Treatise 126.96.36.199 continues: “The same principle makes us easily entertain this opinion of the continu’d existence of body.” But Hume also makes clear that “the principle concerning the existence of body” is one that we have to assent to (T 188.8.131.52). Thus despite his skeptical doubts, at the end of Treatise 1.4.2, Hume indicates that he will proceed by “taking for granted” the existence of “both an external and internal world” (T 184.108.40.206). I suspect that this conflict regarding the mechanisms of the imagination at least partly explains Hume’s dissatisfaction at the end of Book 1.
“Imagination” here means the picturing of an object, or the forming of an image.
I discuss this principle in Boehm (2013).
Allison, H. E. (2008). Custom and reason in hume: A kantian reading of the first book of the treatise. Oxford: Oxford University Press.
Boehm, M. (2013). Hume’s foundational project in the treatise. European Journal of Philosophy. doi:10.111/ejpo.12056.
Hume, D. (2011). In D. F. Norton & M. J. Norton (Eds.), A treatise of human nature. New York: Oxford University Press.
Hume, D. (2006). In T. L. Beauchamp (Ed.), An enquiry concerning the human understanding. Oxford: Oxford University Press.
Garrett, D. (1997). Cognition and commitment in hume’s philosophy. Oxford: Oxford University Press.
Martin, M. (1993). The rational warrant for hume’s general rules. Journal of the History of Philosophy, 31(2), 245–257.
Robinson, J. A. (1966). Hume’s two definitions of “Cause”. In V. C. Chappell (Ed.), Hume, a collection of critical essays. New York: Anchor Books, Doubleday & Company.
I would like to thank the two anonymous reviewers at Synthese, in particular reviewer number 1 who was truly exceptional. I am very grateful for your help.
About this article
Cite this article
Boehm, M. Hume’s definitions of ‘Cause’: Without idealizations, within the bounds of science. Synthese 191, 3803–3819 (2014). https://doi.org/10.1007/s11229-014-0499-x
- Ideal observer
- Rule following