Fictions & Artifices: David Hume on Following a Rule
Donald Ainslie (University of Toronto)

March 16, 2018, 2:00pm - 3:30pm
Department of Philosophy, Western University

Stevenson Hall 1145
1151 Richmond Street
London N6A 5B8


David Hume’s core empiricist principle says that all thoughts (“ideas”) ultimately acquire their content from prior experiences (“impressions”), and thus his account of the mind is often said to be reductively naturalistic and individualistic.  Recent work on Hume, however, by Garrett (2014), Schafer (2013, 2017), and myself (2015) has emphasized the role of language in his account of representation.  In the case of general ideas, for example, Hume indicates that every impression is completely determinate and thus every idea, being copied from impressions, should also be completely determinate.  Yet we can think general thoughts.  His resolution of this paradox is to appeal to the imagination’s associations.  We think a general thought by thinking of a particular and being disposed to think of other particulars in the same class, where a linguistic term triggers and controls this disposition.  The rules of language thus dictate whether it is right to associate, say, an idea of a particular poodle with the idea of a particular Airedale (but not with the idea of a particular shoe or stone) when thinking of dogs-in-general.  For Hume, the social and normative structures of language give our concepts a social and normative dimension.

I point to Hume’s claim that “without a fiction” ideas can only represent the content they inherit from impressions (Treatise to argue that general ideas qualify as fictions and that the core Humean fictions – of the self and of external objects – can be understood through language-involving mechanisms similar to those at work in general ideas.  I conclude that, for Hume, all of our core thoughts are social and normative in structure.

Hume, however, does not give a full explanation of language.  The most significant discussion occurs in Book 3 of the Treatise where Hume analogizes the “artifices” or conventions that establish property to those that establish a language.  Unfortunately, Hume’s account of the obligations involved in obeying the rules of property are far from clear, and thus his appeal to language to explain our fundamental beliefs might seem similarly unclear.  I investigate Darwall’s (1995), Cohon’s (1997, 2008), and Garrett’s (2007) attempts to explain how Hume understands the artificiality of justice, and suggest that each fails for a common reason.  They overlook the fact that, for Hume, “just acts” are not a discrete category in need of a particular motivation.  Rather, justice ultimately concerns our motivations simply excluding violations of others’ property rights without our ever thinking in those terms.  I suggest that a similar analysis explains our obedience to the rules of language.

Finally, I compare my account of our following linguistic rules with Schafer’s (2017) argument that Hume ultimately uses moral concepts to account for linguistic propriety.  While I agree that some language-involving traits (such as glibness or eloquence) will count as virtues or vices in Hume’s broad conception of morality, the basic rule-following character of everyday linguistic practices is not moralized.

Supporting material

Add supporting material (slides, programs, etc.)




Who is attending?

1 person is attending:

Karanveer Singh

See all

Will you attend this event?

Let us know so we can notify you of any change of plan.

Custom tags:


This event has been submitted and is maintained by:

(University of Western Ontario)

You should login and contact this user if you believe the information on this page needs updating.

If you judge that this event entry is inappropriate, please login and report it.