hume ii

Utility, Justice, and the Judicious Spectator

§1. Remarks on the Principle of Utility

As I was saying in the last lecture, the really substantive question between Hume and Locke which comes to me as I read the essay “Of the Original Contract” is whether Locke’s social contract doctrine, when applied as a criterion to the form of a political regime, will select the very same family of constitutions or regimes as legitimate or just that would be selected by Hume’s principle of utility.1 Hume, as I mentioned, never discusses, and in fact, never seems to be aware of this fundamental matter. Furthermore, his account of utility is extremely loose in that essay; it means simply the general interests and necessities of society.

Now, in some sense, Locke’s criterion would include that principle. That is, if people proceed from the state of nature to political society by consent, with no coercion, etc., then one would suppose that those agreements, freely entered into, would include the general principle of Hume’s and promote the general interests of society. Therefore, one might want to ask, “Well, what is the difference?”

Recall that in Locke’s system of institutional change, beginning from the state of nature, there is a series of undertakings to which rational persons consent freely and voluntarily. Each of these changes, in Locke’s view, would be collectively rational, barring accident and catastrophe and so on. So, we are assuming a sort of idealized process of such contractual agreements. Locke plainly is assuming that it was collectively rational for everyone to consent to, say, the introduction of money and the many other changes that take place. So, beginning with the state of nature, a well or-

1. [Transcription of March 11, 1983, lecture, with additions from Rawls’s handwritten lecture notes. —Ed.]

[ 174 ]

Copyright © The President and Fellows of Harvard College

Utility, Justice, and the Judicious Spectator

dered commonwealth with a legitimate regime must then, in Locke’s view, improve everyone’s situation with respect first to the state of nature, and then to each of the subsequent stages. Therefore, Locke’s regime would seem to satisfy Hume’s condition of answering to the general interests and necessities of society. So, both principles, both Hume’s and Locke’s, are stated in a sufficiently loose and general way that its hard to tell whether and in what respects they are going to differ. Although, as I mentioned before, they surely don’t mean the same thing, and you might say their basic foundation is very different.

Suppose we give a stricter sense to the principle of utility, and take it to mean that a regime is legitimate if, and only if, of all the forms of regime that might be possible, or that are available at some moment, or at some time historically, it is that regime which is most likely to lead to, or most likely to produce, the greatest net sum of social advantages (we might also use the term “social utility”) at least in the long run.

We are imagining that you can in some way define the notion of the “sum of social advantages.” Instead of talking about Hume’s “general interest and necessities of society,” we’ve introduced the notion of the greatest net sum of advantages, both now and in the future. Would this be the same as Locke’s view or not? Again, it doesn’t sound the same. Take the case that most concerns Locke in the Second Treatise, that is, the case of royal absolutism, or arbitrary rule of the Crown within a mixed monarchy. Locke intended always to exclude such a regime as legitimate, and his argument is set up for that purpose. He argues that that form of regime cannot be contracted into. Does the principle of utility as we have now stated it allow for royal absolutism or not? One might say that it may in fact do so, but it would require a lot of argument. It would depend on circumstances and various contingencies, and it is not at all obvious that royal absolutism would either be excluded or allowed.

I mentioned last time that in arguing against Locke towards the end of the essay “Of the Original Contract,” and in assuming that Locke’s appeal to promises is unnecessary, Hume simply denies what Locke asserts. He doesn’t face up to the possibility of using the social contract as a test of the form of a regime. In much the same way, Locke in turn simply denies what Filmer asserts. (See Locke Lecture I on Filmer.) Locke is assuming that the notions of contract, promise, and other notions are not to be derived, or at least he doesn’t make any attempt to derive them, from the notion of the

[ 175 ]

Copyright © The President and Fellows of Harvard College

h u m e

fundamental law of nature. So, we get a case again where there is not really a confrontation of the two views at the most basic level.

Now, before saying more about Hume, I want to try to specify the utilitarian view in such a way as to make its principles appear at least to have more precision than Hume’s general phrase, “the general interest and necessities of society.” To do this I am going to think of utilitarianism in the classical sense of principles associated with Bentham and Edgeworth and Sidgwick.

The basic idea is that one is going to define a notion of the good that is independent from the notion of the right. That is, we introduce a notion of the good, say as pleasure, or absence of pain, or some sort of agreeable feeling, or as the satisfaction of desire, or the fulfillment of interests of individuals. If we care to, we can idealize that and say that the good is the satisfaction of the rational interests, or the rational preferences, of individuals. In saying that that is independent from the notion of the right, I mean that we can explain the notion of pleasure, or absence of pain, or agreeable feeling, or the notion of fulfillment or satisfaction of desire, or the notion of the fulfillment of rational preference—we can introduce and explain all those notions without saying anything about right and wrong. We can introduce them independently of any notions that intuitively would be characterized as having to do with right and wrong. So, if we say we are going to maximize the fulfillment of desires, then that means that we would include evil desires as well as the good ones. There would not be any constraint coming from the notion of right and wrong on what those desires might be.

The first step, then, would be to introduce independently the notion of the good; and the next step would be to define the right as that which maximizes the good. In order to get a traditional utilitarian view, the idea of the good would have to take the form I have indicated: that is, it would have to be pleasure, or the satisfaction of desires, or the satisfaction of rational preference. If we introduce another notion of the good, say that of human perfection, or human excellence, or something of that sort, then we would not get a traditional utilitarian view, but what we might call a perfectionist view.

If we take the principle of utility and apply it to social institutions, we would get something like this: that those institutions and constitutional forms are right and just provided that they maximize the good, understood in the utilitarian sense as either pleasure, or the fulfillment of desire, where

[ 176 ]

Copyright © The President and Fellows of Harvard College

Utility, Justice, and the Judicious Spectator

we are summing the good of all individuals in society, both present and future. We are beginning from the present time, considering what existing institutions might be, and summing the good over all individuals in this way. Observe that, in this way of putting it, there isn’t any principle of equality built in, and there isn’t any principle of distribution included, so there are no constraints on how the good may be distributed, and there are no notions of right involved. One is simply trying to maximize that sum. That is how utilitarianism is understood in what I’m calling the “Bentham-Edge- worth-Sidgwick view” (although one can best describe their view further so that it includes a more hedonistic characterization of the notion of the good). When we later come to Mill, I want to see whether his principle of utility fits this view of utilitarianism, or whether he has some more complicated notion, as I believe he does.

§2. The Artificial Virtue of Justice

Let’s now look very briefly at Hume’s aims in An Enquiry Concerning the Principles of Morals (1751), as stated in Section I, and then turn to Hume’s account of the artificial virtue of justice in Section III and Appendix III, and elsewhere. To outline Section I, in ¶¶1–2 Hume asserts that moral distinctions are real and are made by us in our judgments, and that this is a fact not to be seriously denied. In ¶¶3–7 he states three pairs of alternatives that explain this fact involved in present controversies, and then in ¶8 he foreshadows his own doctrine, which accepts the second alternative in each of the three pairs. Then in ¶9 he discusses his theory of morals as an experimental (or empirical) study (what we today would call a kind of psychology).

Hume’s own view, as foreshadowed in ¶¶3–7 and 8, is as follows: (i) First, moral distinctions are not known and applied to things by reason alone (contra Cudworth and Clarke; cf. Hume’s footnote 12 in Section III, ¶34). Rather, they depend upon a peculiar sentiment. (ii) More specifically, we recognize moral distinctions and have the ability to apply them, not via deductive or inductive or probabilistic arguments, but from an internal sense. Moral judgments express a response of our moral sensibility to the awareness of certain facts from a certain point of view. (iii) Moreover, we concur in our moral judgments, not because as rational and intelligent beings we

[ 177 ]

Copyright © The President and Fellows of Harvard College

h u m e

grasp their truth, as we grasp the truth for example of the axioms of geometry (as Cudworth and Clarke held). Rather, we concur in our moral judgments because we share the same moral sensibility.

Now to comment on this: First, in the Treatise of Human Nature (1740), Hume explained the operations of our moral sensibility via a complicated theory of sympathy, set forth in Book II of that work. In the Enquiry he uses however instead the “principle of humanity.” See his explanation of this in Section V, ¶17, the footnote. [We will discuss the principle of humanity later.] Second, in the first instance Hume’s account of our moral sensibility is epistemological. It explains how we know and apply moral distinctions. The explanation of how we are moved to act from, or in compliance with, these distinctions is a separate question. So you need distinguish then the problem of knowledge and how we come to know moral distinctions, from the problem of motivation and what moves us to act on moral distinctions. Hume is mainly concerned with the former question.

Now I want to turn to Hume’s account of justice and say a few things about that and contrast it with Locke’s view. Hume discusses justice in the Treatise of Human Nature, Book III, Part II, “Of Justice and Injustice,” as well as in his later book, Enquiry Concerning the Principles of Morals, Section III, “Of Justice.” The way Hume uses the term “justice” must be carefully understood, because he isn’t using it in a contemporary sense. He is talking about the basic order and structure of civil society, and, in particular, about the principles and rules which specify the right to property. What Hume calls “virtues” are qualities of human character and dispositions of people to behave and to conduct themselves in a certain way. Justice as a virtue is the disposition of persons to behave and to respect those rules which define property and the other rules surrounding the notion of property. He is using the term “justice” in a rather narrow way. It is only one of many virtues, many of them what he calls “natural virtues,” which operate by instinct. Justice is perhaps the most important, along with fidelity and integrity, of what Hume calls the “artificial virtues”: those which “produce pleasure and approbation by means of an artifice or contrivance, which arises from the circumstances and necessity of mankind.”2

Hume’s principles of justice are, in effect, largely principles for the regu-

2. Hume, Treatise of Human Nature, Book III: Of Morals, Part II: “Of Justice and Injustice,” Sec. I.

[ 178 ]

Copyright © The President and Fellows of Harvard College

Utility, Justice, and the Judicious Spectator

lation of economic production and competition between the members of civil society, as they pursue their economic interests. The basic rules of competition, in Hume’s view, turn out to be essentially three:

First is a principle about private property, and this requires very roughly that everyone be left undisturbed in the enjoyment of what they properly possess. In order to define “properly possess” we have to introduce a number of other rules which specify the rights of ownership. In the Treatise, Hume discusses a variety of such rules having to do with present possession, occupation, prescription (or long possession), accession and succession, and these rules come into play under certain circumstances.3 For example, in case the owner of property dies, in order to avoid controversy about who is to come into possession, there have to be rules about inheritance and the like.

The second rule of justice has to do with trade and exchanges of property, and the idea is that there are rights over property that can be transferred under certain conditions.4 The basic idea is that transfer can only take place by consent. Hume thinks of the second principle as necessary so that the holdings of property within society can be continually adjusted over time according to the various interests and abilities of individuals, and the various best uses that they are able to make of them. So we have to allow for the adjustment and transfer of holdings of property over time.

Hume’s third main principle pertains to contracts and the performance of promises.5 He thinks of it as more general and inclusive than the second, which has to do with trade and exchange, although it also covers that in a way. It covers agreements of all kinds, including agreements to future performances.

We now have these three principles, which Hume thinks of as principles of justice. The first one, you might say, views society in the form of an association of owners, the second one views society in the form of a market, and the third sanctions the general principle of contract and promises. Together, Hume thinks of these three principles as regulating and specifying the rules of economic competition and production between the members of society, and they constitute the basic norms of economic relations between the members of society. So, we can then say that a just person, on

3.Ibid., Book III, Part II, Sec. III.

4.Ibid., Book III, Part II, Sec. IV.

5.Ibid., Book III, Part II, Sec. V.

[ 179 ]

Copyright © The President and Fellows of Harvard College

h u m e

Hume’s view (because he thinks of virtue as a quality of people—it transfers back, you might say, from the institutional structure to the person) is one who is disposed to honor these basic rules. Hume assumes throughout his discussion that social institutions actually satisfy his principle of utility, however broad and general that is. In other words, on the presumption that institutions do in fact fulfill that principle, then Hume thinks of the just person as one who is disposed to honor these basic rules. He goes on to say, “justice is an artificial virtue and based on convention” in a sense that I will explain later.

In Section III of An Enquiry Concerning the Principles of Morals, where Hume talks about justice, his thesis is that public utility (that’s another term, I take it, for the general interests of society—he uses a lot of different terms, and his language is very loose)—public utility is the sole origin of justice, and reflection on its consequences is the sole basis of its merits. That is supposed to be put in contrast to the case of the natural virtues, where public utility is perhaps one basis of their merit, but certainly not the sole one.

What this thesis means to Hume is that the institutions of justice (I will abbreviate them as property, transfer, and contract) would not exist, or be adhered to, unless people recognized their public utility, and unless people had a sense that these institutions were in the general interest. I take Hume to be saying that we would not approve of these institutions unless we recognized that as general systems of rules publicly recognized and generally acted upon by all, or at any rate by most persons, these institutions have beneficial social consequences and serve the public good.

As I mentioned in the last lecture, Hume calls justice an “artificial virtue” because it is a disposition to adhere to a general system of rules recognized to be for the public good. This system of rules is itself, so to speak, an artifice of reason, and that’s what the term “artificial” meant at that time. An artifice of reason was something that could be understood by reason and in no other way. Moreover, the recognition that this general system of rules has these consequences for the general good is itself something that requires the use of reason.

Let me make a further point, contrasting the artificial virtue of justice with a natural virtue like benevolence. The idea is that an individual act of benevolence—being kind to someone, being kind to children, say, or to people who need our help—does not require the conception of a general

[ 180 ]

Copyright © The President and Fellows of Harvard College

Utility, Justice, and the Judicious Spectator

system of rules. It is something that we are prompted to do because we recognize that an individual person needs our help. It doesn’t involve in the same way that the rules of property do some conception of social good, or depend on an idea of how general systems of rules are required to produce the social good.

A further point is that in contrast to the natural virtues like benevolence, the public good that results from rules regarding property, transfer and contract, all regarded as systems of public rules, depends essentially, in Hume’s view, on their being adhered to even when, in an individual case, complying with the rules may seem to do more harm than good. This would not be so in the case of a natural virtue like benevolence. The rules of property are distinctive in that we are to adhere to them as public systems of rules even when, despite their being as well designed as they can be, they are still going to require us, in certain particular cases, to do things that may seem to us to be harmful. For example, rules of property may require that misers who are, perhaps, not able or willing to use their property productively, nevertheless have the right to keep it. Or in the case of inheritance, the rules specify who is going to inherit property, even though it may seem to us that that person who does inherit it cannot or will not use it productively; or perhaps we think that they’re bad or unworthy and ought not to have it. Nevertheless, in Hume’s view, the benefits of the system of property can be attained only if these general rules are mutually recognized as applying to everyone, and only if we adhere to them more or less inflexibly, even when, in particular cases, our actions seem to do more harm than good.

So, the general social background, in Hume’s view, for an artificial virtue is roughly the following: First, that there exists a system of general institutional rules which define property transfer and contracts, which system of rules he regards as an artifice of reason. The second feature is that this system of rules is publicly recognized by the members of society as promoting the public good and the general interest and necessities of society, and that this recognition by the members of society is itself a work of reason. By “publicly recognize,” I mean that each person recognizes that the system of rules is for the general benefit of society, and each recognizes that the other also recognizes that, and so on.

A third point is that the benefits of this general system of institutional rules depend, as I have just said, on its being inflexibly followed, even in

[ 181 ]

Copyright © The President and Fellows of Harvard College

h u m e

particular cases where it may appear harmful to do so, or where there appear to be better alternatives than complying with the existing rules. I take it to be Hume’s view that not to follow them, or to regard them in too flexible a fashion would undermine legitimate expectations—it would undermine the reliability of being able to count on what other people are going to do. For social behavior to be reliable and foreseeable, it is necessary to have certain general systems of rules that can be counted on to be inflexibly followed. One can allow for certain kinds of exceptions (e.g. to prevent imminent disaster) and allow for complicated rules to some degree. But, in Hume’s view, there is a limit to the extent to which one can do that.

Finally, the fourth point is that the disposition to be just is a quality of character to adhere to these rules with the appropriate degree of inflexibility, provided that others in society have a manifest intention likewise to comply with them. And Hume believes that once we understand the background of these rules, then it is a normal fact about people, given the laws of human psychology and the like, that they will have this disposition to be just.

I call your attention to the fact that towards the end of the last section of the Enquiry Concerning the Principles of Morals (Section IX, Part II) Hume talks about a “sensible knave” who, for his own profit, may allow himself exceptions to these rules. Hume doesn’t actually address any argument to him in terms of his own interests. He just regards such a person as one who is not motivated as most of us are, who isn’t offended at the thought of himself, say, acting unfairly, or unjustly, or free-riding, as we might say, on this system of rules.

I would urge you to read Appendix III of the Enquiry, “Some Further Considerations With Regard To Justice.” It is very instructive on what Hume’s notion of artificial virtue is. Pay attention there to the sense in which he says that justice is based upon “convention,” understood as “a sense of common interest.”6 He uses the example of two men rowing a boat, each relying on the other to pull his oar, without the need of promises or contract, to illustrate what he has in mind when he says that justice is based on convention. Now the four points I have just gone over touch all the things that Hume has in mind there.

6. See Hume, Enquiries Concerning the Human Understanding and the Principles of Morals, ed. L. A. Selby-Bigge (Oxford: Oxford University Press, 2nd ed., 1902), p. 306.

[ 182 ]

Copyright © The President and Fellows of Harvard College

Utility, Justice, and the Judicious Spectator

There are two further points I want to make in regard to this account. Hume talked as though the general interests of society alone account for the institution of property and transfer and contract, and also account for how these institutions provide a background for the artificial virtues of justice and fidelity and integrity and the like. But he does not seem to allow for the possibility that it may actually be the case that it’s not the general interests of society that account for private property and this particular specification or arrangement of it. Rather, there may be some other interests involved that account for property—perhaps the interests of the more powerful, or maybe the interests of those who have the most property. He just does not seem to allow for that. Now I don’t think that one should say that Hume is not aware of that possibility. One would have to assume that he is. I interpret him as giving a kind of idealized account of how the institution of property and the virtues of justice, integrity, and so forth could come about, and as setting out the general features and general factors that actually explain the natural roots, the psychological basis of our moral behavior.

In other words, I think it is important to understand about Hume that he is trying to give an account of why it is that we have the virtues that we have, and why it is that we are motivated to act in accordance with those virtues. And this is intended to be, for the most part, an actual psychological account. It is not like Locke’s view, a normative doctrine beginning from the fundamental law of nature and other natural laws, and saying what our rights and duties are, and then giving an account of the form of regime that could legitimately come about. That is not what Hume is doing, or at least it is not what I take Hume to be doing. I see him as explaining why we have some virtues—why they exist, why they are praised, why we are motivated to act in accordance with them—as one might do in psychology, or, one might say more broadly, a science of human nature. So, for his purposes I think it’s adequate for him to give this more or less idealized account, leaving aside certain other possible interests, and seeing how the institutions of property and the virtues that are associated with them might come about, and how they would be different from other virtues, for example, the natural virtues.

This would mean that on Locke’s account, a system of property would appear to be derivative from the fundamental law of nature, and it would include certain rights to property which would have to be respected in cer-

[ 183 ]

Copyright © The President and Fellows of Harvard College