The debts of Hume's account of Convention to Locke

It has been asserted by some, that justice arises from HUMAN CONVENTIONS, and proceeds from the voluntary choice, consent, or combination of mankind. If by convention be here meant a promise (which is the most usual sense of the word) nothing can be more absurd than this position. The observance of promises is itself one of the most considerable parts of justice; and we are not surely bound to keep our word, because we have given our word to keep it. But if by convention be meant a sense of common interest; which sense each man feels in his own breast, which he remarks in his fellows, and which carries him, in concurrence with others, into a general plan or system of actions, which tends to public utility; it must be owned, that, in this sense, justice arises from human conventions. For if it be allowed (what is, indeed, evident) that the particular consequences of a particular act of justice may be hurtful to the public as well as to individuals; it follows, that every man, in embracing that virtue, must have an eye to the whole plan or system, and must expect the concurrence of his fellows in the same conduct and behaviour. Did all his views terminate in the consequences of each act of his own, his benevolence and humanity, as well as his self-love, might often prescribe to him measures of conduct very different from those, which are agreeable to the strict rules of right and justice.

Thus two men pull the oars of a boat by common convention, for common interest, without any promise or contract: Thus gold and silver are made the measures of exchange; thus speech and words and language are fixed, by human convention and agreement. Whatever is advantageous to two or more persons, if all perform their part; but what loses all advantage, if only one perform, can arise from no other principle. There would otherwise be no motive for any one of them to enter into that scheme of conduct-David Hume, An Enquiry concerning the Principles of Morals, Appendix 3.7-8 [italics/bold added to facilitate discussion.--ES]

Before I start: I would be amazed if my underlying claim in this post hasn't been noticed before. But while my search was certainly not at all exhaustive, I couldn't find it in the literature. So, I put it forward here tentatively with the invitation to be pointed to works that have treated it more at length and in scholarly fashion.

It is not uncommon to start discussions of convention with a nod to Hume, and especially the passage quoted in bold/italics. In addition, people often will mention the rowing boat example. (See, for example, this entry by Michael Riscorla in the Stanford Encyclopedia of Philosophy, which mentions Hume throughout.) In my accont, Hume's definition has seven (or eight) parts: (i) a sense of common interest (i*) felt in each person's breast; (ii) and it (that is, (i)) observed in others; (iii), this fact (the existence of (i&ii) creates collaboration; (iv) the collaboration is structured in non-trivial ways; (v) and this has good consequences or positive externalities for society. (I avoid the language of 'utility' to avoid issues pertaining to utilitarianism.) And (vi) a Humean convention is contrasted with practices founded in promises and in practice regulated by formal governmental law. In addition, (vii), the process (i-iii) need not be verbalized at all. It can be entirely tacit.

Now notice that Hume's beloved two rowers example actually doesn't fit his definition all that well. It is by no means obvious that when two thieves row away from a heist that this rebounds to society's benefit (so (v) is absent). Nor is it obvious that cases of rowing really instantiate a general plan or a system of actions (iv). I don't deny that the example is an attractive one to show that promises are not necessary to create joint action even in the absence of good enforcement mechanism.

But I do want to suggest that Hume's definition is more naturally read as being about major social institutions (like language, money, and property (which he calls 'justice' here.) In larger context justice is the main subject, after all. And that lurking in Hume's discussion there are really two kinds of conventions: (a) large scale social institutions that exist over time, fit a wider social context, and have social utility (like language, money, property); (b) apparently spontaneous behavior that benefits those that directly participate in it and that can draw on, perhaps by re-inventing it each time anew, previously existing patterns of behavior (like rowing). I don't want to insist that there is a very sharp distinction here -- I usually don't bet against Hume (whatever the Bayesians might say) --, queuing may well be thought to be an instance of (a) or (b).

My reason for suggesting the distinction between (a) and (b) is that I want to convince you Locke actually presents a version of (a), and is plausibly Hume's inspiration. (That's the thing I suspect has been noticed before.) My claim is a bit counter-intuitive at first because it is pretty clear that Hume is criticizing the social contract tradition here (which he calls 'absurd'), and since Locke is a social contract theorist, one might naturally take him to be the target here. But the view he is criticizing is Hobbes'. 

After all, for Locke, property precedes the social contract. Now, from this it does not follow that Locke has a conventionalist account of the origin of property. He famously doesn't. His is, as is well known, a weird labor mixing account. I am not suggesting that Hume has modelled his account of convention (in the sense of (a)) on Locke's account on the origin of property. He is a serious critic of Locke's account of property (I have published on that here). Rather, I am suggesting Hume is modeling his account of convention on Locke's account of the origin of money. And notice that 'gold and silver' are 'the measures of exchange' is one of the conventions in the sense of (a) that Hume mentions here.

Okay, with that in place let's turn (recall) to the key passage that I have in mind in Locke, which is, in fact at the end of the account of property (which arguably is one of Hume's targets.

Sect. 50. But since gold and silver, being little useful to the life of man in proportion to food, raiment, and carriage, has its value only from the consent of men, whereof labour yet makes, in great part, the measure, it is plain, that men have agreed to a disproportionate and unequal possession of the earth, they having, by a tacit and voluntary consent, found out, a way how a man may fairly possess more land than he himself can use the product of, by receiving in exchange for the overplus gold and silver, which may be hoarded up without injury to any one; these metals not spoiling or decaying in the hands of the possessor. This partage of things in an inequality of private possessions, men have made practicable out of the bounds of society, and without compact, only by putting a value on gold and silver, and tacitly agreeing in the use of money: for in governments, the laws regulate the right of property, and the possession of land is determined by positive constitutions.--John Locke Second Treatise of Government

Again, I am not suggesting Hume accepts Locke's account of money. I am interested in their underlying views about  convention. 

Notice, first, in the last sentence, that Locke contrasts the use of money with both promises ("compacts") as well as formal laws and positive constitutions. So, I claim Locke embraces (vi). Second, Locke insists that the practice is socially useful ("men have made practicable"). Even though it precedes the existence of what he calls 'society' (which is the effect of the social contract). So, Locke adopts (V). In fact, I think the underlying point here is that whether it is useful is something discovered over time ("found out.") And this keeps the convention in place for Locke.

In addition, Locke emphasizes that the convention can be entirely tacit. So, I think it's clear he insists on (vii), too. Fourth, the example of money is meant to illustrate what Hume calls a 'system of actions.' Money is an especially durable convention. For Locke it precedes society and shapes the development of society. So, I claim Locke is the source of (iv).

So, that leaves (i-iii). These are not wholly explicit and disambiguated in Locke. And in my view, Hume's distinctive contribution is to make these features of the nature of a convention explicit. For, in Locke (i-iii) are generally treated in terms of consent. And for Locke consent is the glue that makes human associations of all kinds of sorts possible (including ones that are not conventions at all). And if I were only to emphasize the similarities between Locke and Hume, I would suggest that this account of (tacit) consent anticipates Hume's (iii). But I also recognize that if one is thinking about consent in terms of some other institutions that Locke discusses (churches, political society) then consent pulls one away from seeing that he is also talking about Humean conventions.

So my proposal is this. Hume liked Locke's account of conventions, and saw how he could use it against Locke's account of property. (Amusingly, Smith likes Hume's account of conventions and uses it against Hume's account of property to make it in some sense more Lockean.) But what Hume noticed is that Locke misses a mechanism of how one would arrive at such tacit consent. And (i-ii) supply that mechanism.

And it is important to see that appealing to 'common interest' is Kosher in Locke's approach. Locke appeals to it at least twice (as an explanatory concept) in the Second Treatise see sect 82 & 216. In both cases Locke takes for granted that there is such a thing and that can be known by individuals (see especially 82). So, Hume's methodological individualism in (i-ii) is consistent with Locke's philosophy, but not exploited by Locke in his account of the nature of convention. 

 

 

Comments

[Josef Moural]

Very interesting (I suspect the grammar of the following sentence is out of order: "Rather, what I am interested in is the structure of Locke's underlying argument that money is valuable is a convention.")

[ERIC SCHLIESSER]

Thank you, Josef. I corrected the passage in light of your suspicion.

[Matt]

It's been a long time since I read it (more than 15 years, I think) so I don't remember if Locke is discussed at all, and I don't have easy access to it now, but you might find something of interest in this paper by Samuel Freeman:

"Property as an Institutional Convention in Hume's Account of Justice," Archiv für
Geschichte der Philosophie, vol.73, no.1, 1991, pp.20-49.