David Hume on the Foundations of Promises
This short essay was written in Professor Margaret Gilbert’s class on the foundations of Rights and Demands, in Spring 2018, with small modifications for posting online. In it, I discuss Hume’s sentimentalist and conventionalist account of promissory obligation. I am very sympathetic to much of Hume’s account, though I recognize that the skeptical, reductio ad absurdum argument is far from air-tight. Most of his proposed account of the origins of promissory obligation does, however, seem more plausible to me than both the average layperson’s common account and the more complex accounts given by later philosophers.
Hume gives a skeptical argument concerning promises in Book 3, Section V of A Treatise of Human Nature. In it, he raises issues with a common description of promises, and then provides an alternative account that avoids the problems. The focus of both the common account and Hume’s alternative is in describing how promises seem to carry with them a moral obligation. That is, how is it that when someone makes a promise, they now seem morally obligated to fulfill that promise? I then explain how Margaret Gilbert believes Hume’s account has influenced following philosophical thought on promises. Finally, I bring up an issue with Hume’s skeptical argument, as well as the possibility of extending of Hume’s conventionalist account to include promissory rights.
Hume (1738, p.516-517) first sets out to describe an account in which promises are not products of social convention, and result in a moral obligation in the promisor to fulfil the promise. To Hume (1738, p.516), this account he describes happens to be “entirely conformable to our common ways of thinking and expressing ourselves,” which is why I will refer to it throughout the essay as the ‘common account of promises’. Hume thinks that by showing that the common account is absurd, the case will simultaneously be made for his alternative account of promissory obligation.
If promises occurred naturally, without established conventions, there would have to be more to them than their basic conventional actions, e.g. saying, “I promise” or communicating an equivalent. To Hume, this additional thing would have to be the state of mind of the promisor, the person making the promise. Our states of mind can naturally affect other things—we resolve to do certain things, our desires create powerful intentions, and we will our body to move—so perhaps we have a state of mind that results in a moral promissory obligation! Hume dispels that promises are purely one of the three aforementioned states of mind: resolutions cannot create moral obligations; promises may obligate us against our desires; and promises obligate us towards a future action, unlike the immediate results of the will (1738, p.516).
Perhaps there is another state of mind that Hume has not argued against, but he proposes that the train of thought of categorizing promises in terms of a ‘state of mind’ necessarily leads one to believe that a promise is willing into existence the moral obligation according to the language of the promise (1738, p.516). That is the common account of promises that Hume will now argue against. Since Hume believes that the common account reasonably follows from the premise that promises may exist apart from social conventions, then if the common account is shown to be incorrect, Hume will have given us good reason to be skeptical of the premise. I will now go into more detail about Hume’s criticisms of the common account of promises.
Hume’s criticism comes from a larger moral framework of sentimentalism, a view that an action is considered morally good/bad if, for when we consider the action, we are pleased/displeased “in a certain manner” (1738, p.517). Moral obligations, like those in promises, are similarly defined: if refraining from an action sentimentally displeases us, then we say that a moral obligation exists to perform that action (1738, p.517).
From the perspective of a sentimentalist, willing into existence the promise’s specific obligation is equivalent with willing a change in our sentiments, which Hume declares impossible and absurd. Gilbert (2018, p.120) likens his point to the contemporary idea that we are unable to will changes in belief. For instance, if I believe it is not raining outside, I cannot just close my eyes and will myself to believe that it is—to form or change a belief, I need to be convinced. Similarly, I cannot simply will a change in my positive or negative sentiments. The common account that we will into existence obligations thus seems impossible, even more so if we imagine that we must also will a change in other people’s sentiments.
Hume makes two points about promises. (1) Since the common account seems absurd and the argument for the common account seems valid, we have reason to doubt the common account’s base premise that the source of a promise’s obligation is independent of social conventions. Hume takes it further: since it is absurd, it seems as though promises are unintelligible without social conventions (1738, p.517). (2) Even if we assumed that making promises is intelligibly an act of the mind, because of the limitations of our will, a promise would not be able to create an obligation (1738, p.518).
The last step in Hume’s skeptical process is to show that there is an alternative account that describes how social conventions might establish what a promise is and how it works. He describes how social conventions can lead to promises, first by stating that it is in the self-interest of individuals to cooperate to stably possess or trade objects (1796, p.519-520). From this, it follows that promises are your personal resolutions to do the actions you have promised, with the additional social convention that you would be ostracized otherwise (1738, p.521-522). After a while, Hume supposes, these conventions play a role in our forming our sentiments, leading us to have moral obligations to carry out promises (1738, p.523). Promises remain resolutions of our own mind but are obligatory insofar as society’s sentimentality surrounding the social conventions of the act. Hume thinks people find the idea that social conventions could imply a moral obligation strange and unnatural, which is why people instead propose the common account of willing an obligation (1738, p.523).
Margaret Gilbert believes that Hume’s argument has led to later philosophers to make two assumptions in the discussion of promises: (A) promises cannot arise as pure acts of the will, and (B) a promissory obligation is a type of moral obligation (2018, p.122,123). Philosophers then either accept or reject Hume’s general framework of sentimentalism. Of the non-conventionalists, Gilbert places them in two groups: (1) philosophers who reject assumption A and argue that promissory obligations actually are purely willed into existence, and (2) philosophers who accept both A and B, but think that promising consists of circumstances that, without relying on social conventions, regardless imply that one is morally obligated to fulfill the promise (2018, p.124). Gilbert’s account of joint commitments lands her in the first category; joint commitments create obligations between people willed into existence when those people will as one body. Thomas Scanlon’s account in What We Owe to Each Other fits into the second category because of his description of how the creation of expectations in another can, in certain cases, create an instance of moral obligation.
A couple comments on Hume’s account follow. Hume’s skeptical argument relies on showing that looking for an account of promissory obligation that is not derived from social conventions must end in absurdity. However, there are several premises that go into his arguing for the common account, so any of those might be thrown into doubt instead of the not-from-social-conventions premise that Hume reasons to be false. For example, Gilbert points out that there might be other states of mind that result in promises that are not simply the promisor willing an obligation into existence, such as her proposed solution of joint commitments (2018, p.123-124).
Gilbert (2018, p.108-111) describes a view which holds that you need at least two people to jointly form a promise—the promisor has to make the promise, and the promisee has to accept the promise. Hume’s common account of promises has it as a premise that the act only lies in the promisor, which is a glaring hole in his argument. Following from the previous paragraph, this might be why Hume’s common account fails—as Gilbert would likely argue (2018, p.160). However, though this is a large blind-spot in Hume’s account, it seems likely that Hume’s conventionalist alternative would be able to extend a conventionalist explanation for the rights of the promisee in a similar fashion as it does the obligations of the promisor.
Bibliography
D. Hume (1738), A Treatise of Human Nature.
M. Gilbert (2018), Rights and Demands.
