The writings of David Hume (1711–1776) are a treasure trove for those eager to find pithy, polished memorable quotes to bolster their arguments in favor of freedom, justice, and against the arrogance and follies of governments. It is difficult to resist the youthful élan of his major philosophical work, A Treatise of Human Nature (1739–1740), his provocative ironic style, witticisms, irreverence, and occasional sarcasm, which made him an international celebrity, the darling of Parisian salons, and, even now, a reader’s delight.
The English philosopher Sir Isaiah Berlin wrote of Hume:
No man has influenced the history of philosophy to a deeper and more disturbing degree.” And there’s the problem: Between Hume’s time and our own, philosophy, to the extent that it is part of public life, has not been particularly supportive of the case for freedom as classical liberals and libertarians understand it. If Hume’s influence was so profound and disturbing as it is said to be, then maybe we should wonder whether his philosophy had anything to do with this. Did it undermine precisely the sort of freedom that his age was beginning to see as the shape of things to come?
Hume wrote when the ideas of freedom, justice, and the rule of law were at their apogee in public discussion, and “English liberty” was the envy of many intellectuals on the Continent, where royal absolutism was still the rule. It was not his philosophy that had put these ideas in the spotlight but the philosophy of natural law. Natural law refers to the principles of order (law) in the human world that the mind can discover, no matter how great the conflicts of interest and opinion among persons. The natural-law philosophy held that those principles ought to be respected by all because they are true principles. The advocates of natural law readily acknowledged that people might not always perceive a personal advantage in conforming their actions to the law. However, they were confident that enough people were sufficiently open to reason to accept the validity of the natural-law principles and to use them in their active life.
Human nature was the key concept. It comprised the notion that persons, having the capacity to reason, also had the capacity to modify the perception of their interests so as to include due respect for the dictates of reason. The natural-law philosophy recognized that mastery of these capacities is unevenly distributed among individuals, but it did not see that as a reason to give up on the search for true principles and for ways to increase their weight in practical deliberations. Over the centuries it had erected an edifice of thought in which arguments about human affairs would have to be judged by objective principles—not particular subjective opinions, interests, sentiments, or sympathies, no matter how powerful or influential these might be. This is not to say that every writer who used the term “natural law” was as scrupulous in his reasoning as one might have hoped. Many could not resist the temptation to advance their pet schemes for organizing a particular society, its government, armed forces, educational institutions, and so on as “dictates of reason.” There was junk and pseudo-natural-law theorizing then as there is junk and pseudoscience now.
Disturbingly, Hume was a prime agent in the modern intellectuals’ fight against the philosophy of natural law, not only against its caricatures but also and primarily against its fundamental tenet: that reason ought to guide the actions of men. Hume’s moral philosophy was an explicit denial of that proposition. Reason, it said, is and ought only to be the slave of the passions. Man is motivated by, and only by, self-interest—only considerations of utility can sway man to act one way or another.
Following in the footsteps of Thomas Hobbes, Hume asserted that morality has nothing to do with reason because reason is either philosophic (and then merely a system of definitions) or practical (and then merely a slave of the passions, an instrument without a finality of its own). That distinction replaced the one the natural-law tradition had made between theoretical or speculative reason and its application in particular circumstances. This substitution implied a radical redefinition of the role of the philosopher and scientist in public life, from the role of “conscience of the Prince” (holding up the mirror in which the rulers can see themselves as they are, not as they imagine themselves) to the role of “counselor of the Prince” (advising rulers on how to implement their designs most efficiently). More generally, and far more importantly, it redefined education, from a preparation for a life of reason to a program of socialization—preparing the young to function in their society by instructing them in the opinions and interests that define it.
It is true that Hume tried to explain some of the natural-law principles regarding justice, property, and contracts. However, to him they were not discoveries of the mind reflecting on the human condition, but inventions that were found to be “useful” and retained on account of their “utility”; their perceived utility was the only basis for taking them seriously. He may have thought that they were more useful in more circumstances than other institutions, say tipping waiters or separating the legislative from the executive power, but he did not think that they were of a different kind.
In his time, those natural-law principles, formulated before they were more or less systematically applied, were in their ascendancy as far as enlightened public opinion was concerned. Thus, as applied to his time, his philosophy seemed to support those principles, even if it was an attack on their traditional intellectual foundations. However, if the young Hume were writing today, when the winds of public opinion are blowing from another direction, he probably would not give the principles much attention. I suppose we’d find him attacking John Rawls and other rationalistic apologists for the welfare state with the argument that the welfare state, although not a discovery of reason, nevertheless is necessary for “polished society” and would not exist as a source of “moral obligation” if it was not perceived to be in “man’s self-interest.” For it is a corollary of his philosophy that in the affairs of man opinion, not reason, is the touchstone of truth.
True by Definition?
Hume had a predilection for unqualified (and unquantified) general statements, such as “Man does what he does out of self-interest.” Of course, his philosophical undertaking was one of “pure reason,” as he defined it: It was supposed to deal only with definitions and relations of abstract ideas, not with realities, historical facts, and the like. Thus the quoted statement is just a definition within a conceptual system. There is no need to quantify or qualify it. However, as Hume actually used it, his readers may be forgiven for assuming he meant it as a universally true descriptive statement.
Now, if we look at real individuals and specify self-interest, for example, as the possession of wealth or power, then the statement is obviously false. It remains false even if we attenuate it by adding “confined generosity” to “self-interest”—for in Hume’s system acts of generosity are and must be explained in terms of self-interest. Only if we hold that whatever motive men have for doing what they do is “self-interested,” by definition, can the statement be deemed true; but then it is a mere tautology: “Men do what they do.” Likewise, on Hume’s account, a person can be shown to have a moral obligation only if we can show him that its performance is in his self-interest as well as the perceived interest of others, who therefore are sympathetic to it and inclined to approve of it. That too is merely an arbitrary redefinition, this time of the term “obligation.” Yet, again, Hume suggested, and many of his readers accepted, that no other account of obligation makes sense.
Let us take a look at what many regard as the most famous passage in Hume’s Treatise of Human Nature. It is without doubt one of the founding texts of modern philosophy:
In every system of morality, which I have hitherto met with, I have always remarked, that the author proceeds for some time in the ordinary way of reasoning, and establishes the being of a God, or makes observations concerning human affairs; when of a sudden I am surprised to find, that instead of the usual copulations of propositions, is and is not, I meet with no proposition that is not connected with an ought, or an ought not. This change is imperceptible; but is, however, of the last consequence. For as this ought, or ought not, expresses some new relation or affirmation, ’tis necessary that it should be observed and explained; and at the same time that a reason should be given, for what seems altogether inconceivable, how this new relation can be a deduction from others, which are entirely different from it. But as authors do not commonly use this precaution, I shall presume to recommend it to the readers; and am persuaded, that this small attention would subvert all the vulgar systems of morality, and let us see, that the distinction of vice and virtue is not founded merely on the relations of objects. . . . [Last paragraph of the first section of Book III, “Of Morals”]
This is Hume in full swing: A “small attention” gives you sufficient excuse for not reading any other moral philosophy. But how weighty is Hume’s point? Certainly there are many examples that support his observation: one cannot deduce that one ought to believe in God from the fact that one is mortal, no more than one can deduce that one ought to be a socialist from the observation that the poor are human beings, or that one ought to go to the dentist if one has a toothache. However, the critical question, which Hume does not address, is whether there are counterexamples that invalidate his claim that it is “altogether inconceivable” that one could infer ought from is.
When Hume mentions deduction, he must be understood to have in mind formal deduction, a replacement of an expression or sentence by another expression according to a fixed rule of logic. An example would be the rule “Given ‘A is B’ and ‘B is C,’ one may write ‘A is C.’”With this rule, given the sentences “A lion is a mammal” and “A mammal is an animal,” we can formally deduce the sentence “A lion is an animal,” Hume’s point, of course, was that an ought-statement cannot be a merely formal transformation of any is-statements. Therefore, it cannot be formally deduced from any series of given is-statements. If one nevertheless wishes to introduce ought, then one should explain why and how one intends to do so.
So far so good. However, the quotation does not give us the whole of Hume’s argument. You could have guessed that from the ellipsis with which I terminated it. Let us replace that ellipsis with Hume’s actual words: “[the distinction of vice and virtue is not founded merely on the relations of objects], nor is perceived by reason.”
There it is! Hume’s poison, like the scorpion’s, is in the tail. Nowhere has he argued that ought cannot be “perceived by reason.” All he has done is point to the elementary fact that is and ought are different relations. However, from “A is different from B” and “Reason can perceive A” it does not follow that reason cannot perceive B.
To appreciate the fallacious nature of Hume’s argument, we should step back from his implied notion of deduction-within-a-formal-system and look at its relation to logical thinking or reasoning. We should not confuse any of the many proposed formalizations of logic with logic itself. (“Logic,” from the Greek “logos,” refers to the same thing as “reason,” from the Latin “ratio.”) Any system of formal logic is a construction, a product of a human mind. As such it is to be evaluated by human minds, specifically with respect to the claim that it is a formalization of logic. In other words, we need logic to construct a system of formal logic, and we need logic to evaluate the extent to which it captures the properties of logical thinking without bringing in elements that might lead our thinking to illogical conclusions. Without logic we cannot have formal logic. However, a formal system, whether or not we prefer to call it a formal logic, is merely a product of our logical thinking. It is not the same as the logic or reason that produced it. That logic, or reason, is one of our active capacities. (To grasp this point is to see how little Hume’s “inert reason” has to do with human rationality.)
Thus the mere fact that we cannot “deduce” an ought-statement in a formal system that accepts as primary data only is-statements tells us nothing definitive about the logical relations between is and ought.
There are two obvious ways in which one could proceed to deduce ought from is: 1) one might look for an empirically observable ought, as it were, a fact that can be described by means of an ought-statement; or 2) one might look for a rule of logical inference that permits moving from is to ought. Hume’s contention is that there is no such rule of inference. Therefore he can only proceed with his moral philosophy by turning to the first option: finding an observable ought.
However, Hume is wrong to say that no rule of logical inference permits us to move from one relation to another. Immanuel Kant suggested “Ought implies can” or its variant, familiar to lawyers, “Nobody ought to do the impossible.” (However, Hume would not have been impressed: a collection of is-statements no more implies a statement such as “A cannot possibly be B” than it implies an ought-statement. He had famously argued that causal statements [“A causally necessitates B”] have no deductive cognitive basis. They are psychologically, not logically, related to the facts—just as our moral judgments are.)
Another example is: “If A is B then you ought to believe that A is B.” What logical objection is there against assuming the validity of that rule? Suppose someone says to you, “John is ill but you ought to believe that that is not true.” His proposition clearly is indefensible. Suppose you were to ask him why you should not believe what is true. He could begin to answer the question only by giving his “reasons” for not conceding that John is ill while conceding that he is. In short, he would have to concede that his “reasons” carry no weight—they are not reasons at all. Note that the sentence “John is ill” does not formally contradict the sentence “You ought not to believe that John is ill.” Nevertheless, a human speaker—a rational or logical mind—would contradict himself if he were to commit himself to what these sentences express. The fact that “ought” and “believe” are not in “John is ill” may seem to block the inference that one ought to believe that John is ill—but only if we think of logic, as Hume did, in terms of relations and definitions encapsulated in an inert formal system. However, if we consider logic as an active capacity of the human mind, then that fact is utterly irrelevant.
Here we see where Hume was pulling the wool over our eyes. He had mistaken a product of reason for reason itself and jumped to the conclusion that the inertness of the product meant that its producer was equally inert. However, for an active mind, bridging the gap between is and ought, far from being “altogether inconceivable,” is a common and often proper logical move. That gap, which is so obvious in a system of formal logic, actually means little as far as the critique of “vulgar systems of morality” is concerned. Few of these ask us to move from “A is B” to “A ought to be B.” In some cases they make the Kantian move from “A cannot be but B” to “A ought to be B.”
In other cases they move from is-statements to ought-statements, not by way of a formal deduction of the latter from the former, but by invoking logically irrefutable normative principles. The inference may be fairly direct, as in “One ought to believe what is true,” or more indirect, as in “One ought not to act on beliefs, theories, or principles that have been shown to be false,” or “You and I are separate beings of the same rational kind, therefore we ought to respect and treat one another equally as separate beings of the same rational kind.” Obviously, it would be quite legitimate to query whether some theory or belief is false, or whether we are separate beings of the same rational kind, but it would be illogical to concede the facts and deny their normative consequences. More exactly, it would be illogical to do so if the normative principles on which the inferences are based are irrefutable, indefeasible, and without an equally irrefutable or indefeasible alternative.
The Humean Dismissal of Freedom
Arguments to the effect that the case for freedom, justice, and property derives from reason have been made in all ages. True, they were never beyond criticism, often confusingly intertwined with other arguments, but despite the differences in their details and qualities, they all pointed in the same direction, to the same conclusion. Isn’t it a tragic irony that the philosophy of the young Hume is so often invoked to dismiss them out of hand because they did what he had defined to be “altogether inconceivable”? The truth is that there is no logical basis for Hume’s “revolutionary” claim that the case for justice, property, and contractual obligation cannot be derived from reason, hence must be derived from a murky amalgam of self-interest, public interest, confined generosity, sympathy, and approbation.
If the gap between is and ought goes overboard then so goes the idea that science can only be science if it confines itself to Humean empiricism assisted by a purely formal logic of is-statements. That idea, together with the distinction between facts and values, is the basis for the notion that there can only be a science of ethics if ethical statements are reinterpreted as empirical statements—say, about psychological phenomena. Values (“what is desirable”) must be reduced to facts (“what is actually desired”); ethical truths must be reduced to subjective opinions—otherwise, or so it is alleged, science would not be “value-free” and then would not be science at all. However, that notion clearly is incoherent.
Yes, science should be value-free in the sense that it should proceed without prejudice from well-established facts to any conclusion it can reach by means of well-tested methods of fact-finding and logical reasoning. However, it should not be value-free if being that means representing values as something other than values or an “ought” as a twisted “is,” or if it means disregarding the validity of norms that no one logically can deny merely because they are norms.
Let us turn to Hume’s famous dictum that “Reason is, and ought only to be the slave of the passions.” It implies that reason’s role in guiding actions is, and ought to be, limited to its utility in aiding the fulfillment of desire. On Hume’s account, the passions, not reason, insert the “ought” in our thinking: one ought to do what best serves one’s passions or desires. Now, on Hume’s definitions, that is perhaps the only general ought-statement that makes sense. However, it is not an insight into our moral life. It is not a logically irrefutable rule of inference. It is not an observable fact. It makes sense only as a peculiar definition—Hume’s definition—of the word “ought.” If we mistook it for a normative principle, you could ask me, “Why ought I do what best serves my interests?” In some cases, depending on the circumstances and your particular interests, I might be able to give you a satisfactory answer but in other cases I would not. However, if I were a Humean, I would have to reply that your question shows that you have not understood the meaning of the word “ought.” I would have to say that your question simply means, “Why does what best serves my interests best serve my interests?”—or, what is the same, “Why ought I do what I ought to do?”
In fact, the idea that the role of reason in guiding actions is limited to its utility in satisfying desire is not really coherent. Granted that the role of reason is limited, just where are the limits? One might be tempted to say that the passions set the ends and leave it to reason to find means and methods for “best” achieving them. However, the distinction between means and ends is not immediately given. One of the main functions of active reason is to see goals or ends as means in the perspective of an ulterior, maybe even an ultimate end. “What purpose would there be to achieving that particular goal?” is not a meaningless question. However, that is a question only reason can raise. For every passion presents itself as an ultimate end—unless we assume that there is reason in the passions and thereby place ourselves far beyond the pale of Hume’s system of definitions.
We might say: Modern decision-theory vindicates Hume. Does not a formal representation of a decision problem, with neat lists of goals, means, and cost-and-benefit valuations of various strategies, leave no other role for reason than to calculate which strategy is “best”? However, it would be a serious mistake to assume that the properties of such a representation match those of the decision problems a person experiences in real life when the goals, means, strategies or their valuations are not “given.” Only reason can judge whether such a formal representation is adequate as an aid to solving a real problem. Again, we find that Hume is mistaking his “reason”—an inert formal system—with the reason that gets us through the day.
Reason the Judge
Which will be the judge of what serves our desires best: our passions or our reason? Only reason can make that judgment, in particular, of course, on those frequent occasions when our passions conflict. What is the best way for the slave (reason) to serve conflicting masters? Get out of the way—which here means: go mad? Work out a compromise—as if reason could talk to the passions when they cannot talk to one another? Take sides—which here means: decide which passions will be served and which will be left in the cold? Anyway, what passions are we talking about? If we specify some particular passions, then Hume’s dictum is obviously false. If we do not specify the passions then what keeps us from postulating a passion for reason, self-control, science, or whatever?
We see many people do things that are motivated by other passions than a passion for reason. Nevertheless, we can and often do discuss with them the wisdom or reasonableness of giving in to such passions. Few of them even try to cut the discussion short by objecting that it is pointless because reason, being their mere slave, has no command over the passions, and hence no command over us, as regards the question whether to satisfy a particular passion or not. Many, it is true, will exit from such a discussion with the ultimate copout: “You’re right, of course, but I cannot help it; I’m so hopelessly weak-willed.” Still others will not budge until we have answered their question “What’s in it for me?” However, there are many other cases where argument, not incentive, intimidation, force, threat, or treatment, makes the difference. The basic presumption of civilized life is indeed that one does not resort to incentives, threats, force, or treatments until one has seriously but in vain tried to reason and to argue.
An interesting consequence of Hume’s assault on the rationality of man was pointed out by Hume himself (Treatise, Book III, Section 8). The original motive for the establishment of government and the source of its moral approbation are exactly the same as in the case of justice: self-interest and sympathy for the public interest. However, although the obligation to obey the government derives from the same origins as the obligations of justice, it serves an entirely different interest. Thus, apparently, the obligation to obey the government—any government—is separate from and independent of the obligations of justice. “This,” Hume continued, “separates the boundaries of our public and private duties, and shews that the latter are more dependant on the former, than the former on the latter.”
Surely, it is one thing to claim that utility alone can explain the institutions of justice and property (which reason, reflecting on human nature and the human condition, can indeed show to be universally valid); it is another thing to claim that there is an obligation to submit to the rule of other men, provided only they get away with calling themselves the government. Hume must have sensed that he was on the verge of tumbling into an abyss of absurdity. With bold irreverence he extricated himself from the need for philosophical reasoning by appealing to “the universal consent of mankind,” or as he put it, “the sentiments of the rabble.” On that authority, and from its utilitarian premise that “the distinction between moral good and evil is founded on the pleasure or pain, which results from the view of any sentiment, or character,” Hume concluded that “it is certain, that there is a moral obligation to submit to government, because every one thinks so.” On the previous page he had written, “Education and the artifice of politicians concur to bestow a farther morality on loyalty, and to brand all rebellion with a greater degree of guilt and infamy.” He might as well have said that there is a moral obligation to submit to the government because the government has succeeded in making people believe that there is such an obligation.
If it were a deep insight into the truth about human nature, Hume’s dictum that reason is and ought to be the slave of the passions would raise the question, what sort of education could be built on it? At the individual level it translates into: “Your reason is and ought to be the slave of your passions”; do not question your desires, only the efficacy and efficiency of the means to realize them. At the institutional level, it translates into, say, “Schools and universities, the pursuit of scientific knowledge itself, are and ought to be the slaves of politics—or if not of politics then of public opinion.” In either case, we seem to be left with no principle of education at all, for what is the purpose of an education if not to teach people to learn to discern right from wrong?
Why did Hume introduce such concepts as the universal consent of mankind or the sentiments of the rabble in support of his own analysis of human nature? His stated reason is that although he himself was convinced by his exercise in “pure reason,” he had doubts about other philosophers. So why not go over their heads and point out that after 600 pages or so of “pure reason,” he had arrived at what the rabble had known from time immemorial? Why not, indeed, if the opinion he ascribed to the rabble (without actually consulting it, of course) was exactly the thesis with which he hoped to revolutionize moral philosophy: that, where morality is concerned, things are what people think they are because they think so—a colloquial form of the proposition that the distinction of good and evil lies beyond the ken of reason.
To arrive at that conclusion by “an endeavour to establish my system on pure reason” certainly was a remarkable feat. It set the stage for the welter of pure and applied social sciences that thrive on collecting and statistically processing opinions and on exploring ways in which opinions can be formed, influenced, and manipulated. It also denied that there was a basis for serious ethical criticism in the fields of morality, politics, and economics. Why try to find out what is desirable if the desirable is at bottom nothing else than what is desired? And what did the young Hume desire? As he put it in the advertisement for the separately published first two books of his Treatise, he was making his work public “in order to try the taste of the public. . . . The approbation of the public I consider as the greatest reward of my labours.” A curious advertisement for a philosophical work, indeed, but not for the work for which it was written!
Freedom and justice are not served well by appeals to the universal consent of mankind, alias the sentiments of the rabble. If that wasn’t clear in Hume’s time, it most certainly became clear in the nineteenth and especially the twentieth centuries. Hume himself came to realize that his “vast Undertaking, planned before I was one and twenty & composed before twenty five” was seriously defective. “I have repented my Haste a hundred, & a hundred times.” No matter, A Treatise of Human Nature was out and eventually became a, perhaps the, most important founding stone of modern philosophy. It is in many ways a great book, but it is tragic that it achieved its status chiefly on the basis of, rather than despite, its defects. Its cavalier demotion of reason from the principle that defines the human animal’s humanity to an unprincipled technique in the service of anything or anybody that happens to be the effective master, played into the hands of demagogues and manipulators; gave philosophical sanction to the corruption of intellectual life by the rulers of the day; and left education without a compass of its own, at the mercy of the shifting winds of public opinion.
It is one thing to mine Hume’s writings for nice quotes on freedom, justice, and property; it is another thing to assume that the philosophy he helped launch is an asset in the struggle for freedom, justice, and property.