Background: Intuitionism and 18th-Century Common Law

The legal world in which the declaratory theory was originally at home differs profoundly from the contemporary one. Although complete codification has not yet been achieved in any common law jurisdiction, statute law is now regarded as the default type of law, always superseding judge-made law, and judges, lawyers, and legislators accept it as natural and inevitable that the latter should give way. In the 18th century, in contrast, judge-made law was predominant. There was comparatively little statute law, and it was not highly regarded (Lieberman Reference Lieberman1989). Many believed that judges had the power to declare statutes to be invalid (Reference Berger1969).

Judge-made law was regarded as superior to statute law because it could respond flexibly to the specific circumstances of cases. In Omychund v. Barker (1 Atkyns 21, 33 [1744]), the case in which it was decided that Hindus as well as Christians could testify under oath, William Murray, who became Chief Justice Lord Mansfield, declared that “a statute very seldom can take in all cases, therefore the common law, that works itself pure by rules drawn from the fountain of justice, is for this reason superior to an act of parliament.” His point is illustrated by a recent California case (Martinez Reference Martinez2013) in which a retrial was ordered because an accused rapist had pretended to be the accuser’s boyfriend, but the prosecutor relied on an old statute that stipulated that it is rape when a man impersonates a woman’s husband. If the relevant law had been judge made, it could have been modified by the court in light of contemporary social conditions.

The judges of the 18th century denied that they were making it up as they went along. They denied that they were legislating on an ad hoc basis. Instead, as Blackstone (Reference Blackstone and Morrison2001, 69) put it, they were “living oracles” who revealed the preexisting law. Their judgments were taken not to be the law but to be evidence of it. The preexisting law was held to be constituted by morality, or rather those aspects of it that warranted public recognition and enforcement. As Lord Coleridge (The Queen v. Instan [1893] 1 Q.B. 450, 453–54) put it when affirming Kate Instan’s conviction for manslaughter on behalf of a unanimous court, “It would not be correct to say that every moral obligation involves a legal duty; but every legal duty is founded on a moral obligation. A legal common law duty is nothing else than the enforcing by law of that which is a moral obligation without legal enforcement.”

As for how we know what our moral obligations are, most 18th-century judges did affirm ethical intuitionism, which posits a special moral faculty or ability that enables moral agents to perceive or apprehend moral truths. Ethical intuitionism was the predominant school of ethics in England at the time, flourishing in reaction to Hobbes (Hudson Reference Hudson1967). It is not the same as the natural law view of Aquinas that morality can be read off the observable world (because God’s intentions are detectable in its features), which is now used as the standard contrast to positivism in undergraduate courses. The staunchly Anglican judges of 18th-century England did not rely on a quintessentially Catholic philosophical position. There is no need to go into more detail with respect to the metaethical opinions of the judiciary at the time, however. The view to be argued for is that if intuitionism is true, then the judges of that era were literally correct about being able to discover law, no matter what they actually thought about the nature of morality.

Intuitionism maintains that there are objective moral facts and that we can know what they are. It claims that it is a fact that it was morally wrong for Instan to leave her aunt—on whom Instan was dependent and with whom Instan was living—to die of gangrene without helping her, or getting help for her, or even mentioning her condition to the neighbors with whom Instan conversed from time to time over the 10 days it took her aunt to die. It claims that we know it intuitively when we contemplate the situation—as though we could literally see that it is wrong. It denies that we are merely reacting emotionally. It denies that we have to canvas the consequences in order to know it was wrong. It denies that we have to combine the description of Instan’s omissions with a moral principle serving as the major premise in a syllogism to derive the conclusion that she acted wrongly. Intuitionists do not believe that there is a single principle or even a consistent set of principles. For example, the intuitionist W. D. Ross (1930) contended instead that there are a number of prima facie obligations that may conflict and that must be weighed against each other when we are determining what we ought to do.

Moral intuitions are not infallible insights, however. Blackstone (2001, 31) put the intuitionist denial that intuitions are infallible this way, where “reason” is what enables us to apprehend moral truths: “our reason … [is not] unruffled by passions, unclouded by prejudice, unimpaired by disease or intemperance. … [Instead, our] reason is corrupt and [our] understanding full of ignorance and error.” It is certainly possible for all moral agents to have the same intuitive access to the same moral facts but to disagree when it comes to the specific moral judgments they make. For one thing, as Blackstone indicates, intuitionism is compatible with people reaching different moral conclusions because they have different beliefs about nonmoral matters or because they are ignorant of different nonmoral truths. People who believe in bacteria will sometimes disagree about what morality requires with people who do not. Furthermore, people’s interests and desires can alter or distort their beliefs. A buyer may believe that a seller should have revealed a defect in what he purchased, while the seller may believe that it is enough that he did not explicitly misrepresent what he sold. Finally, indoctrination on the basis of inadequate moral theories can induce people to believe false moral propositions, and indoctrinated individuals can suppress or misinterpret their intuitions and favor instead what they have been indoctrinated to believe.

Ethical intuitionism is not an outdated doctrine. On the contrary, it has been advocated in several recent works (Audi Reference Audi2005; Huemer Reference Huemer2005; Zamulinski Reference Zamulinski2007). These monographs naturally respond to the standard objections to intuitionism. If intuitionism is still defensible as a true account of the nature of ethics, then the declaratory theory is still defensible as a true account of judge-made law. Arguing that the declaratory theory is true does not commit us to an 18th-century worldview. Metaethically, contrary to Brewbaker (Reference Brewbaker2007), it is not necessary to posit God as ethicists then did, because there are thoroughly naturalistic versions of intuitionism (Zamulinski Reference Zamulinski2007). Normatively, we can disagree with the moral conclusions of 18th-century judges and explain the disagreement by adverting to differences in nonmoral beliefs, and so on.

A Thought Experiment

What follows in this section is not a historical account but a thought experiment. It shows that there is a way in which there can develop a practice that resembles the common law of the 18th century when ethical intuitionism is true; when there is no statute, case, or even customary law to begin with; and when there is communal support for a dispute resolution process in which disputes are decided with reference to the moral facts posited by intuitionism. In the course of resolving disputes, the individuals deciding them discover what they can call law, and, as will be argued in the next section, they are justified in calling it that. The upshot of this section and the next is that there is nothing causally or conceptually wrongheaded about the 18th-century judges’ own account.

Obviously, disputes could arise in a world in which there was initially no law and in which intuitionism was true. In such a world, when disputes arose, people including the disputants could and would make moral judgments about them. People with the same background beliefs would tend to make the same moral judgments. The disputants would tend to disagree, however, because their conflicting interests would motivate them to do so. In these circumstances, it is conceivable that the disputants might ask a knowledgeable and disinterested individual to decide the issue rather than live with the tension caused by the unresolved dispute—just as disputants now sometimes voluntarily ask an arbitrator to decide the issue between them. Since communal harmony would be desirable, other members of the community would encourage or pressure the disputants to seek and accept this kind of arbitration when they were unable to settle their differences on their own. The community at large might want an independent judgment with respect to the dispute, even if the disputants did not, because its members might want to know what stance to take in case they had to intervene.

For instance, suppose that someone takes A’s horse, sells it to B, and subsequently disappears. If B were allowed to keep the horse instead of returning it to A, A would obviously lose out. However, if B were required to surrender the horse, he would lose out, because the seller with whom he dealt has vanished with his money. Clearly, A and B would probably be unable to agree on their own as to which of them should possess the horse. Each would assert that he should. Moreover, the dispute could escalate into a feud. In such a case, A and B, and the rest of the community, could very well prefer to ask a neutral party to decide who should have the horse. If there were moral facts and if human beings had intuitive access to them, then the person asked to decide the dispute would try to decide it in accordance with the moral facts and other relevant factors, and the community would tend to approve of his judgment provided he decided the case in line with their view of the moral facts and the other relevant factors. The other relevant factors would include not only the material facts but also factors such as the truths that the law should deal only with more serious issues, that enforcement is difficult and expensive, and that a degree of moral autonomy is desirable.

Given the assumptions we started with, there would be no alternative basis on which to justify a resolution of the dispute. The judge would have to treat moral facts as laws in order to adjudicate. People could accept the judgment as reflecting what they called the law, where the law is that which sets the standards of behavior that the community recognizes as worthy of public recognition and enforcement. Since trivial moral issues would be excluded, the putative law per se would not be morality as a whole but a proper subset of moral facts. The relationship between judgments and the subset of moral facts that constituted the putative law would be analogous to the relationship between judgments and statutory provisions. In contrast to statutes, however, the law qua subset of moral facts would never be arbitrary and would always be there. People would share their views of it unless some people’s intuitions were distorted by mistaken beliefs or the absence of relevant true beliefs. The putative law under discussion would not be what Hart (Reference Hart2012, 92) calls primitive law, that is, a human creation that “resemble[s] our own rules of etiquette.” Etiquette can be arbitrary in ways that the hypothetical discovered law cannot be: it can forbid slurping soup in Europe but require it in Japan, as it in fact does.

The judge would have to explain his decision. It would not be a matter of his simply making a pronouncement. For instance, in deciding that the horse should be returned to A, he might point out that B knew that the horse had recently been in A’s possession and that he should have questioned the seller’s claims when he bought it. Alternatively, in deciding that B should keep the horse, he might point out that A had permitted the seller to borrow the horse and that B did not know and could not be expected to discover that the horse did not belong to the seller. The explanation would be primarily an attempt to make sure that people had the relevant true beliefs and did not possess relevant false ones, so that people would understand the rationale for the decision. The judge would have to say why plausible alternatives to the chosen decision were rejected. The explanation would refer to or presuppose moral truths and material facts. Some of the moral truths would be derivative. However, there would also be moral bedrock that could be intuited but not justified by an appeal to anything more basic. For instance, while the dispute could be rendered pointless by destroying the horse or returning it to the rogue who sold it, the options would not be seriously considered. It would not have to be explained why the options were rejected because these are not points on which there would be any variation in opinion.

No one would think that the judgment was unfair to one party to the dispute, or the other, on the ground that neither could have known the law beforehand. They would not follow Bentham in thinking that it was “dog law” because it punished people without first publicly and explicitly warning them that they would be punished. Instead, they would agree with the judge whom Bentham is criticizing and whom he quotes as declaring, “Happily for us, we are not bound by any laws but such as every man has the means of knowing” (Bentham Reference Bentham and Bowring1843b, 235). If people had intuitive access to moral facts and if they were aware of the other relevant factors, it would be literally true that everyone could in principle discover what the judges discover and, hence, could know what the judges know. The important thing would not be that a statement of the putative law had been explicitly published in advance but that it was possible to understand a judgment as justified and appropriate once it had been made.

It is predictable that, in addition to explaining their decisions, judges would tend to deal with similar disputes in similar ways, typically appealing to previous cases as precedents, because it is an indisputable principle of formal justice that like cases ought to be treated alike, whatever treating them alike might amount to. A judge who decided similar cases in different ways would cast doubt on both his decisions. Hence, if the point were to settle disputes and not to resurrect or perpetuate them, he would have to try to maintain consistency. If a judge were to make a judgment that differed from ones he had made previously, he would probably distinguish the dispute in question from superficially similar ones, pointing to the relevant material facts that made the cases different. He would keep a record of his judgments, if only in his own head. If there were more than one judge called on to settle disputes, it is likely that they would try to coordinate their judgments and to follow each other’s precedents unless they could distinguish the cases.

Despite the desirability of consistency, old judgments would sometimes be overturned. As noted, our intuitive grasp of the moral facts is influenced by nonmoral beliefs. Consequently, as society changed and knowledge advanced, judgments that had seemed right at one time could later appear to be wrong. There could be tension between the need to maintain consistency and the need to respond to changes in culture and worldview. If an old judgment turned out to be indefensible in the light of new social conditions or new knowledge, it would have to be rejected. Whether consistency or change prevailed would depend on the facts of the case as well as the nature and extent of the changes in society or knowledge. Whether consistency or change should prevail would be a matter of debate on a case-by-case basis. Furthermore, the boundaries of what was included as law could change as enforcement capacities altered. In a changing world, with respect to people’s understanding of what they took to be the law, there would probably always be some degree of tension somewhere between the need to maintain consistency and the need to adapt to change.

In short, a practice could develop that operated in much the same way as the common law of Blackstone’s time. Some of the salient features of the common law at that time were that there was no written code; that precedents were typically followed; that if precedents were not followed, then, most frequently, the new cases were distinguished from the old although old judgments were sometimes overturned; that records of cases and decisions were maintained; and that judicial decisions were made on a case-by-case basis with respect to the facts of the case and what justice required.

A possible development is not always a probable one, but the English judges had a number of advantages when they began to declare the law in Tudor times, as case reports replaced the yearbooks (Holdsworth 1934). Parliament did not compete with them by legislating. The judges had authority as established functionaries in a judicial system that had already been in operation for several centuries. People accepted their authority—and lawyers and litigants actually wanted the judges to tell them what the law was. No doubt, these advantages are part of the complete explanation of the development of the 18th-century English legal system. But the advantages are not essential ingredients in the recipe. Although the state appointed and paid the judges, it does not follow that the substantive law in 18th-century England was ultimately the creation of the state. The kings who sent men on expeditions of exploration did not create the continents that were discovered. Similarly, the declaratory theorist maintains, the kings and parliaments who commissioned judges to determine the right did not thereby indirectly create the laws that were discovered.

Discovering Law

In the thought experiment, judges discover what can be called the law, rather than making it—they apprehend the relevant moral facts when considering particular cases. The moral facts and other relevant factors necessitate particular judgments in particular cases. The judge would see to it that either A or B kept the horse or perhaps that the horse was sold and the proceeds divided. As mentioned, there is no way that any judge would have concluded that the horse should be destroyed or returned to the person who took it from A. Thus, they would discover what people could call the law qua moral facts in the context of material facts and other relevant considerations. Their judgments would be creative acts in the same sense that a mathematician’s discovery of a new mathematical formula is one. A mathematician cannot discover any mathematical formula whatsoever and, likewise, a judge could not make any judgment whatsoever, even in cases of first impression. Even if a judgment was later overruled, we could still say that what was called the law was a matter of discovery. Since it is possible to be mistaken about the moral facts or their implications, it makes perfect sense to arrange things so that “if it be found that the former decision is manifestly absurd or unjust, it is declared, not that such a sentence was bad law, but that it was not law” (Blackstone Reference Blackstone and Morrison2001, 52). But the question now is whether what people in the thought experiment call the law really is a form of law.

Positivists do not believe it is. They repeatedly complain that declaratory theorists fail to see that legality is conceptually independent of morality. Austin (Reference Austin1832, 278n) declares that “the existence of a law is one thing: its merits or demerits are another thing,” explaining that “whether a law be, is one inquiry: whether it ought to be … is another and a distinct inquiry.” In addition, Austin notes that Blackstone says that a human law that conflicts with the law of God is no law, and Austin asserts that “to say that a human law which conflicts with the law of God, is therefore not binding, or not valid, is to talk stark nonsense” (279n). As Hart (Reference Hart1958, 599) puts it, “what both Bentham and Austin were anxious to assert [was that] … in the absence of an expressed constitutional or legal provision, it could not follow from the mere fact that a rule violated standards of morality that it was not a rule of law.”

It is not a conceptual error to identify moral facts as law, however. The declaratory theorist takes a view of the law that is analogous to the Platonic view of propositions. If the Platonic conception of propositions is true, a proposition either exists or does not exist, full stop, and it makes no sense to say that there is a difference between the way the proposition is and the way it ought to be. But there can be a difference between the way in which a proposition has been expressed in a human language and the way in which it ought to be expressed. Likewise, the moral facts hypothesized by ethical intuitionism either exist or do not exist—which is not to say that they are Platonic entities—and, therefore, the law as conceived by the declaratory theorist either is or is not. It is with respect to representations of the law in the opinions of judges, not the law itself qua moral facts, that there can be a difference between what is and what ought to be.

In the absence of a written code, the only plausible alternatives are that the law is constituted by moral facts or that it is constituted by the opinions of the people who judge the issues. Customary law is not an additional alternative: it too must ultimately be constituted by either moral facts or judicial opinions. To insist that there can be no law without a written code would be to insist that 18th-century English law was not law at all and such a stipulation would be arbitrary and absurd. Furthermore, there are good reasons to identify the law with moral facts and to deny that it is the opinions of judges. The reasons are practical ones, but then the law is a practical enterprise.

If all judicial opinions were law, either the law would tend to become incoherent or it would become incapable of change. As for its becoming incoherent, if one opinion conflicted with another, both would have a claim to be law. One might be good law while the other might be bad law, but both would be law. To declare that a law is a bad law is not the same as overruling or repealing it. The mere fact that one opinion rejected another would not show that the rejecting opinion should supplant the rejected one. With no way to eliminate one opinion when two opinions conflicted, the law qua judges’ opinions would become incoherent, both synchronically and diachronically. A rule to the effect that later opinions should always supplant earlier ones would eliminate some but not all incoherence. With such a rule, synchronic coherence might be achieved, but there would still be a tendency toward diachronic incoherence. The diachronic incoherence would still undermine the credibility of the law, particularly because there would be no guarantee that the opinions of later judges would be better than the opinions of earlier ones in similar cases.

As for the law qua judges’ opinions becoming incapable of change, judges could avoid incoherence by refusing to allow any opinion to be rejected as mistaken, provided that two judges of equal rank did not make conflicting rulings simultaneously, of course. This is what Bentham wrongheadedly recommends. However, the consequence would be that the law would be unchangeable. Once an opinion had been expressed, it would stand forever. This approach would maintain both synchronic and diachronic coherence but would make the law inflexible. Since people, including judges, are fallible, the inevitable accumulation of errors would undermine the credibility of the law. The development of appeal procedures, if such a development occurred, would mitigate but not eliminate the problem. Even if the opinions were good ones for their time, times change, and the law would lose credibility if it could not change with them. As noted in the previous section, it is not only social arrangements that alter over time but also our understanding of the world. Since mistaken opinions and lack of information can have an influence on moral judgments, freezing the official view of the law at a point in the past would mean that the opinions of people with inferior knowledge would often prevail over the opinions of people with superior knowledge. The anachronistic nature of the resulting practice would damage and ultimately destroy its credibility.

Limiting the law to true judicial opinions would prevent it from becoming incoherent, but problems would remain. For one thing, the law would not exist until the opinions had been expressed. Consequently, the first person to commit a wrong would not be a lawbreaker. But it would be unsatisfactory for judges to declare that, although a wrong had been committed and everyone instinctively knew it, nothing could or would be done about it until the next time it occurred. If that were the only possible outcome when a new type of case came up, people would be unlikely to seek out a judge when they had a novel dispute to resolve. There would be a decreased probability that a judge would issue an opinion at all, with the result that later wrongdoers would not be lawbreakers either. For another, in the absence of adequate publicity, wrongdoers could reasonably argue that they should not be held accountable because they did not know and could not be expected to know that they were breaking the law. The slogan that ignorance of the law is no excuse would be seen as an assertion of arbitrary authority and not as an expression of the truth that no one can plausibly claim to be completely unaware of the law when it consists of moral facts to which everyone has intuitive access.

In contrast, if the law were identified with the moral facts themselves, none of the problems would arise. Judges could hold all wrongdoers accountable, distinguish opinions that represent the law accurately from ones that fail to do so, and justify rejecting old opinions that failed to formulate an accurate representation of the law. Identification of law with moral facts would thus result in a more powerful, resilient, and authoritative system in which the totality of opinions was as coherent as possible and in which there was enough flexibility to correct mistakes and to allow for the development of legal understanding in response to both changing social conditions and advancing knowledge. The system would flourish just in case it had these properties and judges did the only thing that could ensure that it possessed them, which is to avoid identifying the law with their own opinions and to identify it instead with moral facts.

Some potential positivist objections are worth rebutting. Hart (2012) contends that what judges decide is law because there is a rule of recognition that says that it is, where a rule of recognition is a secondary rule that states the conditions for being a valid law. It might therefore appear that Hart has managed to trump the declaratory theory. In fact, however, judge-made law qua moral facts neither has nor needs a rule of recognition. There are no counterfeits that can appear to be law when they are actually not law, and there is no substitute for determining, through an examination of its content and its consequences in particular cases, whether a proposed statement of the law actually reflects the law.

Hart (Reference Hart2012, 254) also maintains that there would be no difference between discovering and making law when it comes to deciding particular cases. Even so, law that is discoverable still differs significantly from law that is made, in that it cannot be repealed. The fact that it cannot be repealed means that a procedurally proper statute that mandated or licensed immoral acts would be invalid in a common law system in which there was both statute law and judge-made law—unless the immorality was trivial enough to be ignored. The law that can be discovered constrains the law that can be made because enacting a seriously immoral statute would imply the repeal of the law qua moral facts, just as enacting a later statutory provision impliedly repeals an earlier one when the two are incompatible. Since the law qua moral facts cannot be repealed, if an immoral statute were valid, the law would contradict itself. But the law as a whole must be consistent. Since consistency cannot be maintained unless immoral statutes give way, they must give way. Therefore, immoral statutes are void just as “manifestly absurd or unjust [judicial decisions are] … not law” (Blackstone 2001, 52). Of course, the orthodox view is that a sovereign parliament can pass any (constitutional) law at all. But being compatible with morality is a necessary condition for being law in a system that includes moral facts as law, and the power to pass any law at all is not a power to pass off anything at all as a law.

The same difference distinguishes the declaratory theory from interpretivism, which is presented as an alternative to both created and discovered law and which identifies the law with the theory of political morality that provides the best justification for judicial decisions (Dworkin 1986). Judges and legislators contribute to the development of the law by creating the data that must be justified by the theory. Resembling an attempt to achieve reflective equilibrium in ethics, interpretivism does not entail that immoral statutes are void but that either statutes must conform to the theory or the theory must be modified to accommodate the statutes.

Hart (2012, 204) acknowledges that “in some systems … the ultimate criteria of legal validity explicitly incorporate … substantive moral values.” But it is not the case they do so in any common law jurisdictions, and it could not be the case in any jurisdiction with a sovereign parliament as currently understood. Moreover, if judges made law in accordance with morality instead of discovering it, the law could be coherent even if it included immoral statutes. The fact that judges could have made a law that conflicted with a statute does not entail that they did make such a law because they may never have dealt with a relevant case—establishing criteria of legal validity is not the same as making law in accordance with the criteria. Even if they had dealt with a relevant case, there could be a secondary rule that says that statute law always outranks case law and that would therefore prevent them from making an incompatible judgment. If judges discover law, however, the secondary rule would have to give way just as immoral statutes must.

The declaratory theory does not explain statute law. It is obvious that legislators make and do not discover law. Declaratory theorists do not have to hold that what is true of the one kind of law is true of the other as well. They can also view legislation as perfectly good as long as it is compatible with judge-discovered law in systems that include both kinds. Moreover, they do not have to, and should not, agree with Blackstone (2001, 31) that “no human laws are of any validity, if contrary to [ethics].” The declaratory theorist’s claim is not that there is a conceptual connection between law and morality of the sort that natural law theorists propose but that moral facts can literally be substantive law in systems that include judge-made law. The invalidity of immoral statutes in such a system does not depend on a definition of law but on the fact that moral facts have been incorporated as substantive law and the fact that the substantive law must be consistent. If all law were statute law, as it is in many jurisdictions, no moral facts would be incorporated, and the declaratory theorist would have no grounds for concluding that immoral statutes are void.

Rebutting the Positivist Case against Intuitionism

If ethical intuitionism is true, there are moral facts, and people have intuitive access to them. If the moral facts are not the only source of normativity, they are the dominant source. It is possible for a legal system to develop in such a way that judges can literally discover what they can justifiably call law. It is implausible that a rationalization could be as readily supported as a true account. Therefore, it appears that the best explanation for the 18th-century English judges’ belief in the declaratory theory is that the theory is true, provided that ethical intuitionism is true. But part of the positivist attack on the declaratory theory was an attack on the metaethics that was used to support and explain it. Bentham and Austin not only substituted positivism for the declaratory theory but also substituted utilitarianism for intuitionism. The preceding sections show that they did not justify rejecting the declaratory theory—provided intuitionism is true. The purpose of this section is to show that their attack on intuitionism also failed.

Bentham and Austin assumed that moral intuitions would be immediate and clear and objected that “the respective moral sentiments of different ages and nations, and of different men in the same age and nation, have differed to infinity” (Austin 1832, 107). If judges do not strictly follow precedent, legislation, or rules of evidence, either they have intuitive access to objective moral facts that can inform their judgments or they impose their subjective moral views on those who come before them. In the view of the positivists, the observable variation in moral opinion precludes the possibility that anyone actually has intuitive access to objective moral facts. Hence, they inferred that judges who “make law” impose their personal opinions on those who come before them. In conjunction with this inference, Bentham’s (1843a, 3) despotism/anarchy dilemma implies that they are despotic.

There are good reasons to doubt the positivist argument from variation against ethical intuitionism. First, it is not the case that the positivist stance possesses an intrinsic plausibility that the intuitionist stance lacks. On the one hand, if human beings share a common basis for their moral opinions, it does become necessary to explain the observable differences and disagreements. Intuitionists have provided an explanation in terms of ignorance and error. On the other hand, if there is no common basis, it is necessary to explain why moral codes are found in every human society. Positivists have not tried to explain the observation, and it may not be easy for them to do so. It would be necessary not only to show that a moral code is always in the interests of individuals but also to establish that people always have the insight to know that it is in their interest, that they are not faced with a prisoner’s dilemma that prevents them from initiating development of a moral code, that they have the ability to develop one, that they have the ability to maintain it once they have developed it, and that their efforts will not be undone by defectors or undermined by free riders. Moreover, it must be shown that the development cannot fail to happen in a wide range of circumstances, which means that it is necessary to explain why people are so strongly predisposed to develop moral codes. We all react emotionally to a variety of acts and omissions, but if emotions are basic, the question becomes why they are always manifested as a moral code. The difficulty of accounting for the universality of morality in the absence of a shared foundation means that we cannot take it for granted that the default situation is one in which there is a variety of brute subjective moral opinions.

Second, the observable variation in moral opinion can definitely be explained by a shared basis for morality combined with ignorance and error, which can be motivated by factors such as self-interest and group loyalty. As noted, Blackstone appealed to this explanation. In view of this alternative explanation for the variation, of which they should have been aware, what Bentham and Austin needed to do in order to justify rejecting intuitionism was to establish that equally knowledgeable and disinterested people could have different intuitions. Since they failed to do that, they failed to show that intuitionism is false. Pointing out the surface variation in moral opinions is simply not enough to show that intuitionism is false and that there is nothing but brute subjective opinion.

Third, there is a version of intuitionism that predicts that it will appear as though cultural relativism were true (Zamulinski 2007). This version of intuitionism holds that morality itself—and not merely our belief in morality—is an evolutionary by-product. The evolutionary explanation for morality as a by-product is that there was selection for the adaptations (of which it is a by-product and to which it is inextricably linked) because they first improve our ability to complete longer-range projects that are in our interest and then enable us to cooperate on such projects. Since consensus among potential cooperators matters more than truth when it comes to enabling cooperation, since it is often easier to reach agreement than to discover the objective truth, since different societies are liable to make different errors and to be ignorant of different things, and since biasing factors like self-interest and group loyalty are liable to have different effects in different societies, evolutionary intuitionism predicts that there will be differences in the moral codes of different societies despite the fact that all moral agents share the same moral substructure. There will be differences within societies as well because people differ in knowledge and interests. Since there is at least one intuitionist theory that predicts the observable variation in moral opinion, the variation is not necessarily evidence against intuitionism but could, on the contrary, be evidence for it. It may be added that, if morality is an evolutionary by-product, we have a naturalistic explanation not only for the variation but also for the intuitionist contention that moral agents share a common basis for their moral codes.

Fourth, there is now empirical research that indicates it is an exaggeration to say that people’s intuitions “differ to infinity,” at least with respect to cases in which opinions are unlikely to have been influenced by interests or indoctrination. Many convictions are widespread, and if none are shared by all moral agents, the range of variation is limited (Hauser 2006). In Foot’s (1978) trolley cases, for instance, most people’s intuitions are that they should alter the course of the trolley so that one rather than 10 will die but that they should not save the 10 by throwing a bystander in front of the trolley.

Finally, the positivists assumed that utilitarianism was evidentially in a better position than intuitionism. They were wrong. Bentham’s method was “to disregard the presumptions arising from moral approbation … to get at the naked utility … [and then to use utility] to regulate our moral approbation for the future” (Jeffrey 1804, 12). Bentham’s claim that utilitarianism is obvious has always been questionable, and the development of competing varieties of utilitarianism makes it impossible to contend that any of them is self-evident. Consequently, in the undeniable absence of empirical evidence, a logical proof, or conceptual considerations, the only possible support for any variety is that it can subsume some of our moral intuitions under a principle. The principle of utility is at best a generalization and not, contra Bentham, an external standard.

But, at the same time as utilitarianism has to rely on its capacity to subsume intuitions, it has to “disregard the presumptions arising from moral approbation.” In other words, utilitarianism must discount intuitions that it cannot subsume. But it is question-begging to rely on confirming data while discounting disconfirming data of the same type for no reason other than that it is disconfirming. While there can sometimes be good reasons to reject putative data, it is not the case that avoiding falsification is among them. Ironically, one good reason for utilitarians to reject recalcitrant intuitions is the very same one to which intuitionists appeal. In the case of moral reasoning that involves a moral principle, including any variant of the principle of utility, the moral conclusions that people reach depend not only on the principle but also on what they believe about the situation in which they are applying it. Consequently, utilitarians could reject intuitions that their principle cannot subsume when the intuitions are the result of ignorance or error. Given their own need for the auxiliary hypothesis, they cannot consistently ignore or criticize intuitionists who use it.

There is no discounting some recalcitrant intuitions, however. While our intuitions tell us that we should redirect a trolley to reduce the number of deaths but that we should not throw a bystander in front of the trolley to achieve the same end, utilitarianism says that there is no relevant difference between the two situations. In fact, utilitarianism entails the extraordinarily counterintuitive proposition that, if necessary, we should sacrifice ourselves in the same situation. Not only is it counterintuitive, but also there is an argument that we do not have such an obligation: since we normally cannot bring ourselves knowingly to give our lives for strangers (for powerful evolutionary reasons) and since “ought” implies “can,” we do not normally have an obligation to sacrifice ourselves for strangers. Since we do not but utilitarianism implies that we do, utilitarianism is false—it takes only a single counterexample to falsify a universal generalization. As for rule utilitarianism, it must either concede that the disparity between losses and gains can become so great that self-sacrifice becomes mandatory or admit that it is not a form of utilitarianism at all.

Not all of the points made in this section could have been made 2 centuries ago. Nonetheless, it could have been noted that Bentham and Austin were attacking a straw man, that intuitionists had an explanation for the observable differences in intuitions, that the positivists ignored it despite themselves requiring it, and that the principle of utility more closely resembles a dogmatic opinion than a rational principle. There never has been an adequate justification for the rejection of intuitionism.