Abstract
The naturalistic fallacy and Hume's ‘law’ are frequently appealed to for the purpose of drawing limits around the scope of scientific inquiry into ethics and morality. These two objections are shown to be without force. Thus two highly influential obstacles are removed from naturalizing ethics. The relative merits of moral skepticism and moral realism are compared. Moral skepticism and some forms of moral realism are shown to make similar recommendations for developing a science of moral psychology.
Keywords
Introduction
Evolutionists who study the biological foundations of moral and ethical behavior in humans frequently claim that they are obeying “Hume's Law” not to violate the fact/value distinction. That is, they assure the reader that they will not commit what Moore (1903) called the “naturalistic fallacy” by jumping from “is” to “ought”. They accept the premise, which they believe was advanced by Hume (1739; 1752), that statements concerning facts and statements concerning values are two different types of statement, and that these cannot be related in a meaningful way because moral values and natural facts are of a different kind.
Philosophers in the British analytic tradition such as Flew (1967) have long acted as the guardians of this dogma, but more recently philosophers of biology have raised their voices in assent. In their Sex and Death: An Introduction to Philosophy of Biology, Sterelny and Griffiths (1999) announce on page 4 that the naturalistic fallacy is valid and congratulate Kitcher (1994) for demonstrating it decisively. Mention of the fact/value distinction does not appear again in the book until page 317 where a “Keep Out!” sign is draped across it. Generally, philosophers of biology such as Ruse (1979; 1982; 1986; 1999), McShea and McShea (1999), Woolcock (1999), and Singer (1981; 2002) have supported that injunction. What are evolutionary scientists to do except wisely concur? Examples of concurrence in the scientific literature, popular and not so popular, are ubiquitous and include works as impressive as Symons (1979), Alexander (1987), Alcock (2001), Nesse (1994), Pinker (1997), and Ridley (1996). As one might expect, assertion of the fact/value distinction also appears in the work of biologists who are opposed to sociobiology such as Gould (1999). It is odd that Gould is perfectly willing to hand ethics over to a religious tradition that he rejects in all other contexts of biological inquiry.
Does anyone, in fact, disagree with the fact/value distinction? Has anyone proposed a philosophically interesting challenge to the received view on the naturalistic fallacy? Although Ruse (1979) castigated E.O. Wilson (1975) for committing the naturalistic fallacy in the infamous last chapter of Sociobiology: the New Synthesis, in a co-authored paper Ruse and Wilson (1986) proposed to treat the study of ethics as an applied science and they cast the naturalistic fallacy aside without further ado. The question is why have the majority of philosophers and scientists continued to follow the status quo instead of the Ruse-Wilson lead? I think part of the answer is because not much of a philosophical argument was constructed in defense of the move from ethical objectivism to naturalism. Most of the article is devoted to describing advances in behavioral biology and psychology and arguing that moral psychology was now ripe for treatment. We might expect someone like Wilson to forge ahead beyond the fact/value “Keep Out!” sign, but why would Ruse cross that prohibited Rubicon with him? I think part of the answer is that Ruse, even in his 1979 book, correctly understood that Hume's intent was to argue in favor of a naturalistic moral psychology. What was different in the 1986 paper with Wilson was that Ruse now understood that the fact/value distinction was not valid and that it is possible to treat moral values as natural facts. But in 1999, Ruse once again criticized Wilson (1992) for committing the ‘naturalistic fallacy’ (Ruse, 1999, p. 208).
Teehan and DiCarlo (2004) advance the curious thesis that there are also metaphysical facts but that Moore also raised a fundamental objection to treating these as identifiable with the ‘good’, and they thus propose to recognize an additional fallacy that they call the ‘metaphysical fallacy’. After accepting the existence of both natural and metaphysical fallacies, Teehan and DiCarlo go on with the project of developing a scientific approach to ethics. I intend to argue, that even if Moore believes there exist metaphysical entities that are not identifiable with the ‘good’, it is nevertheless the case that Moore believes there is one metaphysical entity that does qualify as the ‘good’. The issue then is that Moore conceives of the ‘good’ as a non-natural, hence metaphysical object. I believe that once evolutionists understand that proponents of the naturalistic fallacy are committed to ethical objectivism — which entails that values are supernatural facts, they will have no more fear of the ethical relevance of brute facts than they fear that creationists will successfully argue that the universe was created by divine providence.
In this paper, I intend to defend the existence of what Casebeer (2003) has previously referred to as ‘The anti-naturalistic fallacy’. 1 Stated simply, we must recognize that while not all natural facts are relevant to ethical or moral discourse, all facts that are relevant to ethical and moral discourse will nonetheless be natural facts. To hold that values are non-natural facts is to commit the anti-naturalistic fallacy. The relevance of this to the scientific and philosophical analysis of the evolutionary foundation of moral psychology will be that people can stop contradicting themselves. Although it is standard to state that one is not going to commit the naturalistic fallacy, it is also standard to turn right around and commit it — that is, if you accept the terms we've been offered at face value. For example, having stated that he will not commit the naturalistic fallacy, Ridley (1996) concludes his book (rashly he admits) by arguing for the dismantling of the British welfare state because it promotes free-riding among the populace. He argues that virtuous behavior will be elicited by following free market principles. Similarly, Singer (2002) based on acceptance of the same biological and psychological facts argues a contrary position that entreats us to treat the whole wide world, that is to say those individuals beyond our immediate kin—whom he grants us some measure of favoritism, to full reciprocal altruism. On the premises that they explicitly accept regarding the fact/value distinction, they ought not to have done that. Instead, even though I think that both positions might be wanting, the point is not to chastise them for going beyond facts to values, but to realize that moral discourse cannot and should not avoid taking into account relevant facts of nature despite the fact that the ‘facts’ can be interpreted in multiple or contradictory ways
The first step towards realizing a fearless evolutionary moral psychology is to return to Hume and reconsider his stratagem for deriving “ought” from “is”.
From Hume's ‘Law’ to Hume's Stratagem
The passage in Hume that is typically cited to validate the fact/value disjunction is to be found in the first chapter of the third volume of his work, A Treatise on Human Nature. It is sometimes referred to as Hume's “Law” (e.g. Mackie, 1977, p. 64):
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 ought not. … For as this ought or ought not expresses some new relation of affirmation, it is necessary that it should be observed and explained; at the same time 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 (Hume quoted in Flew, p. 38).
Hume is not saying here that “ought” cannot be derived from “is”; rather, he is saying that an argument needs to be given that would justify the connection. Moreover, the task that Hume sets himself is to show exactly how this can be done. Hume's stratagem is to develop a science of morals that unequivocally grounds moral values in the facts of human nature. Hume holds that standards of morality are not abstract in the sense that they could exist independently of the facts of human nature. Indeed, such standards are wholly dependent upon it.
Even in his own day, Hume was frustrated that people misunderstood his intent, and in 1752 he published an additional work entitled, An Enquiry Concerning the Principles of Morals, in which he attempted to clarify his purpose.
The hypothesis that we embrace is plain. It maintains that morality is determined by sentiment. It defines virtue to be whatever mental action or quality give to a spectator the pleasing sentiment of approbation; and vice its contrary (Hume, 1953, p. 129).
Thus, Hume argues that moral values do not exist objectively, as things in themselves, but are instead a projection of subjective psychological standards. By “subjective”, Hume means that they are features, or “facts” if you will, of human psychology.
It was with this aim in mind that Hume and his close friend Adam Smith set out to develop a theory of moral sentiments. And it was Smith's treatise on the subject that Darwin specifically consulted when he addressed his attention to explaining the evolution of the moral sense. The rudiments of the Hume-Smith theory devolve on the crucial role of sympathy. Sympathy is defined as the fellow-feeling we have with any passion we encounter in others (Smith, p. 49). We either approve or disapprove of another person's conduct depending on whether or not we find ourselves in sympathy with his or her motives. We judge the merit of an action in accordance with whether or not we, as the recipient of the action, would likely feel gratitude or resentment were we in that person's place. Smith and Hume both argue that we have within us an impartial spectator whose sympathies or antipathies are aroused in accordance with what the appropriate and natural response to a given situation would be. Thus, we are endowed with internal and innate moral sensibilities.
Although Hume's intent with respect to the is/ought issue is well understood by some contemporary moral philosophers, among them ethical naturalists such as Arnhart (1998a and 1998b) as well as the ‘projectivist’ Simon Blackburn (1998 and 2001) and the ‘expressivist’ Allan Gibbard (1990), it is still possible to find philosophers that remain confused about the issue, including some that believe ethics can be put on a factual foundation such as Ruse (1999) and Casebeer (2003). For example, Casebeer claims that Hume prohibited any and all derivations of ‘ought’ from ‘is’ (2003, pp. 17–18). Ruse (1999, p. 209) also interprets Hume along these lines. Casebeer believes that Hume specifically designed the ‘Law’ to counter those who assume that it is valid to move deductively from ‘is’ to ‘ought’. Casebeer argues that Hume's ‘emotivist’ position leads him to view morals as passions, rather than reasons, and that since reasons have no explanatory power in themselves, no rational deductions are possible. Thus, any and all derivations of ‘ought’ from ‘is’ are invalid.
In order to naturalize ethics, Casebeer relies on Quine's (1953) critique of the analytic/synthetic distinction. In ‘Two dogmas of Empiricism’, Quine strove to demonstrate that all analytic definitions are factual at base, including the laws of logic itself, and that scientific theories formulated as deductive laws were really just well supported generalizations from facts. Although, it is true that Quine's argument can be interpreted to support the view that ethical facts are facts like any other natural fact, it is unlikely that Hume's ‘law’ was designed to preserve the analytic/synthetic distinction. If anything, Hume's commitment to the factual nature of moral facts as psychological facts about our species, demonstrates that he anticipated the hard-nosed empiricism that Quine championed several hundred years later. Based on the clearly stated aims of his Enquiry I would argue that Hume believed that ‘oughts’ were nothing but well supported psychological generalizations based on the sort of ‘ises’ described above.
If ethical facts are facts about human nature, how did they get there? Were they provided by Divine Providence or did they evolve through natural selection? Mackie (1977) suggests that if we reject the former, then the best candidate that remains is the latter. Towards this goal, Casebeer himself enlists the evolutionary ethics of John Dewey (1922) to argue the same position. What did Casebeer find in Dewey that points us in the right direction? This:
a morals based on study of human nature instead of upon disregard for it would find the facts of man continuous with those of the rest of nature and would thereby ally ethics with physics and biology (Dewey (1922, p. 12 quoted in Casebeer, 2003, p. 28).
It is Darwin's theory of evolution that Casebeer claims inspired Dewey's attempt to create a system of evolutionary ethics (Casebeer, p. 29). Before we attempt to evaluate how an evolutionist might commit the naturalistic fallacy, let us briefly survey Darwin's account of the evolution of ethical behavior.
Darwin's Evolutionary Moral Psychology
Darwin addressed the evolution of the moral sense in the last three chapters of The Descent of Man (1871). Darwin explicitly discusses Smith's ‘laws of sympathy’, the moral sentiments of approbation and disapprobation, and Smith's ‘impartial spectator’ which Darwin refers to as an ‘internal monitor’ (pp. 478–481). Darwin set himself the task of explaining how these phenomena could evolve by the mechanism of natural selection. Let's briefly recapitulate his argument in order to ensure that we understand how far Darwin thought we could go towards ‘ought’ from ‘is”, and to see whether or not Hume would raise any objections.
For Darwin, of central importance in the successful adaptation of any social species are the ‘social instincts’.
We can perceive that an instinctive impulse, if it be in any way more beneficial to a species than some other or opposed instinct, would be rendered the more potent of the two through natural selection; for the individuals which had it most strongly developed would survive in larger numbers (p. 480).
Darwin here slides between the species and the individual as the unit of selection. It might be argued that Darwin endorsed what is sometimes referred to as a ‘common fate’ form of group selection. He believed that in many species, including ours, individual survival depends upon group survival.
A tribe including many members who, from possessing in a high degree the spirit of patriotism, fidelity, obedience, courage and sympathy, were always ready to aid one another, and to sacrifice themselves for the common good, would be victorious over most other tribes, and this would be natural selection (p. 500).
Darwin goes on to assert “that members of the tribe would approve of conduct which appeared to them to be for the general good, and would reprobate that which appeared to be evil” (p. 500). Hence, Darwin believed that instinctive sympathy and the moral sentiments are evolved behavioral dispositions that help ensure the survival of the individual and the group to which the individual belongs. He further speculates that these adaptations date to the early prehistory of our species and apply to modern man as well. “The virtues which must be practiced, at least generally, by rude men, as they may associate in a body, are those which are still recognized as the most important” (p. 487).
Hume and Smith both concur that human moral sentiments are designed to ensure the cohesion of society, although Hume places more weight than does Smith on the role that the sentiment of ‘benevolence’ plays in ensuring the well-being of the community (Hume, 1953, p. 14). Smith adduces an additional moral sentiment that he calls ‘resentment’. Resentment serves the purpose of defending the community against antisocial acts. Smith claims that unjust (i.e. antisocial) acts inspire the sentiments of resentment and disapproval in the ‘impartial spectator’.
In order to enforce the observation of justice, therefore, nature has implanted in the human breast that consciousness of ill desert, those terrors of merited punishment, which attend upon its violation, as the greatest safeguards of the association of mankind, to protect the weak, to curb the violent, and to chastise the guilty” (Smith, p. 167).
This passage reveals Smith's firm commitment to placing ethics within the domain of human nature theory. What Darwin brings to the table that was unavailable to either Smith or Hume, of course, is a theory that can explain the mechanism that made human nature what it is. I would also submit that not only are Hume, Smith and Darwin describing what humans believe to be good, but that they also believe that the moral sentiments are in fact ‘good’. If contemporary evolutionary scholars are hesitant to take that last step, they obviously cannot cite Hume's ‘law’ as the obstacle. Perhaps there is some other obstacle to deriving ‘ought’ from “is” or that would prevent us from treating values as facts.
G. E. Moore and the Naturalistic Fallacy
The term, “the Naturalistic Fallacy”, was itself coined by the philosopher G. E. Moore in his famous work, Principia Ethica (1903). Although frequently linked to Hume's injunction about not deriving “ought” from “is”, Moore does not mention Hume once in that work. What sort of argument then does Moore advance, and does it proscribe the sorts of analysis we evolutionists strive to offer?
Moore argues that “good” is “good” and not anything else — just as the color yellow is the color yellow and not a desktop or the color red (Moore, pp. 8–10). The essential point for Moore is that for “good” to be a meaningful expression, the objects that we designate as having the property of “goodness” have to be themselves separate from the predicate of goodness. Thus, for example, for “pleasure” to be “good” it is necessary that “pleasure” be distinct from “good”, otherwise the copula would be a meaningless tautology. Moore has no objection to “saying that ‘pleasure is good’ and yet not meaning that ‘pleasure’ is the same thing as ‘good’, that ‘pleasure’ means ‘good’ and ‘good’ means ‘pleasure’” (Moore, p. 13). In fact, his purpose in adducing the ‘naturalistic fallacy’ is simply to prevent this type of conflation of two different things into one.
Far too many philosophers have thought that when they named those other properties they were actually defining good; that these properties, in fact, were simply not ‘other’, but absolutely and entirely the same with goodness. This view I propose to call the naturalistic fallacy… (Moore, p. 10).
If that is all Moore is out to demonstrate, then we evolutionists are free to head back to our desks to pursue development of a science of moral psychology. But dangers lie just around the corner. Moore does, in fact, make prohibitions against ethical naturalism that might impede the project of a science of ethics.
What sort of moves is Moore interested in barring? Moore reveals that the principal target of his criticism is the ‘evolutionistic ethics’ of Herbert Spencer whom he accuses of making two mistakes. First, he claims that Spencer argued that something was better — in an ethical sense—simply because it was more evolved.
All that the Evolution-Hypothesis tells us is that certain kinds of conduct are more evolved than others; and this is, in fact, all that Mr. Spencer has attempted to prove … Yet he tells us that one of the things it has proved is that conduct gains ethical sanction in proportion as it displays certain characteristics (i.e. as it is more or less evolved)…. It is plain, then, that Mr. Spencer identifies the gaining of ethical sanction with the being more evolved (Moore, p. 49).
Now, this objection devolves on Spencer's mistaken commitment to progressivism. That it is a problem to guard against is a genuine possibility. Ruse (1999) argues that Julian Huxley was guilty of it when he suggested that the human species had evolved to the point where it could exert control over and independence from nature and that the existence of this accomplishment was in itself good. It was the move from ‘control and independence exists’ to ‘control and independence is good’ that involves Huxley in the Spencerian version of the naturalistic fallacy according to Ruse. Similarly, it is over this type of claim that Ruse felt compelled to again criticize E. O. Wilson. Wilson (1992) argued the biodiversity was itself ‘good’. Ruse claims the identification of one with the other here commits the ‘naturalistic fallacy’ (Ruse, 1999, p. 208).
One route out of this apparent difficulty has been proposed by Simon Blackburn (1998). Blackburn draws a distinction between ‘concepts’ and ‘properties’. Extending the philosophical insights of Putnam (1983) from epistemology to ethics, he argues that two different concepts, say ‘good’ and ‘happiness’ might possess identical properties, and if they do, then it would be no fallacy to claim that the concept ‘good’ is identical to the concept ‘happiness’ (Blackburn, 1998, p. 119). 2 The same might be argued about ‘control and independence’ or ‘biodiversity’ as long as one took care to distinguish the properties from the concepts and as long as one could establish an identity of the properties of the two concepts.
Moore thus criticizes the idea that because something is natural that it by that very reason, and for that reason alone, is “good”.
It is obvious … that we cannot say that everything natural is good … If everything natural is equally good, then certainly ethics, as it is ordinarily understood, disappears: for nothing is more certain, from an ethical point of view, than that some things are bad and others good… (Moore, p. 42).
Moore then goes on a tirade against those who would have us imitate beasts and savages or to get down on all fours and run around if it was our natural inclination do to so.
What then can evolution tell us in connection with ethics? Moore grants three ways in which evolutionary analysis might contribute to the understanding of ethics. First, it might turn out to be the case that the “more evolved” trait actually is better in a moral sense. The criterion of goodness, however, is independent of the evolutionary fact of being more evolved. Second, Moore grants that although being more evolved is not the same thing as being good, being more evolved might be concomitant with that which is good, and therefore being more evolved can be useful as a guide by which we can identify that which is good. Third, although evolution cannot determine what is good, it might still be useful in determining how much is possible for humans to attain and the means by which they can best go about attaining it (Moore, p. 55–56). On all three of these possible scenarios, the “goodness” of a thing is kept separate from any of its other properties.
Even if we accept the usefulness of Blackburn's referential innovation, I would still locate a second problem that stems from Moore's anti-naturalism. To keep properties of a concept distinct is one thing, but to place the concept itself outside the realm of nature is another. Moore characterizes “good” as an indefinable, non-natural property that objects possess that is independent of the natural properties that they possess (Moore, pp. 8–10). The soft underbelly of his larger metaphysical position is exposed. He here commits the anti-naturalistic fallacy. Although we can readily agree that not all behavior is good simply because it has evolved, it would be a mistake to assume that because some behavior is behavior that we deem “good”, that it is therefore not natural (i.e. not a product or property of the evolution of our species). By placing the “good” in the realm of the non-natural, Moore demonstrates his commitment to the supernatural realm. 3 Moore's perspective can be characterized as a variant of ethical objectivism. This is the view that ethical objects are real things in and of themselves, separate from things that exist in nature. Although, this type of metaphysical commitment will appeal to those who believe in divine providence, evolutionary biologists might be expected to reject it out of hand with no further consideration — just as they would reject creationist “science”.
Alternatives to Ethical Objectivism
Two highly promising alternatives of ethical objectivism are moral realism and moral skepticism. Although proponents of both schools frequently claim Hume as an ancestor, and both usually claim to be hospitable to ethical naturalism, one school argues that moral facts are real while the other argues that moral facts are not. Let's begin with moral realism and the prospect that moral facts really do exist. (It should be noted at the outset that different kinds of moral realist exist and they differ considerably as to what they regard as a real moral fact.) David Brink offers a quick synopsis of the realist position:
Moral realism is roughly the view that there are moral facts and true moral claims whose existence and nature are independent of our beliefs about what is right and wrong.(1989. p. 7)
Although Brink believes that ethical naturalism is one of the possible forms that moral realism can take — and although he suggests that it is the most plausible one, he makes it clear that it is not the only one. In fact the other versions of possible realism turn out to be Moore's type of non-natural ethical objectivism and outright religious/supernatural views (Brink, 1989, p. 22). Hence the emphasis is on the claim that moral facts are real more than on their natural status. Brink opposes realism of any sort to the skeptical position that doubts the true existence of real moral facts. This position places the sort of realist Brink is between a rock and a hard place. On the one hand, he holds that being committed to metaphysical, or scientific, realism entails that you be committed to moral realism because moral facts are part of the same universe as non-moral facts. On the other hand, Brink admits that even if moral realism is inconsistent with physical realism, it should be retained because it is “a small price to pay to preserve our existing normative practices and beliefs” (p. Brink, 1989, p.173). This easy capitulation to the violation of nature's laws makes it clear that Brink holds an a priori view of the moral universe and is a super-naturalist (i.e. an ethical objectivist) in disguise.
Ethical objectivism has been criticized by moral skeptics such Mackie (1977) on two grounds. The first is called the argument from relativity; the second is called the argument from queerness (Mackie, pp. 32–42). Although the multiplicity of moral systems found around the world presents a challenge to evolutionary approaches as well, the second argument is more pertinent to our purposes here, and the evolutionist wins a decisive victory. To put the point in a nutshell: If ethical values are not part of nature, then they are queer things understood by very queer means, including the queerest of non-natural sensibilities. Although Brink supplies an argument against Mackie's argument from queerness, the argument is flawed by Brink's attempt to accommodate supernatural reality within metaphysical (i.e. scientific) realism. What Brink actually does is to erase the distinction between the two. The result is Brink informs us that we can use our natural sensibilities to perceive both natural facts and non-natural, but nonetheless real, moral facts. 4 We should ask what happens, however, when the moral universe and the physical universe do not agree within the sphere of one's cognitive system? One does not have to be a moral skeptic to wonder if violations of physical realism by instances of non-compliant moral realism are highly suspect. Moreover, when push comes to shove, non-compliant moral cognition is likely to do some special metaphysical pleading to retain the status of its moral claims — a move Brink indicates he would support (Brink, 1989, p.22).
If we reject the view that ethical values are objective non-natural facts existing in a transcendental realm, the alternative is to view them as subjective facts of our evolved psychology. One view that operates from this premise is called moral skepticism but is also sometimes referred to as subjectivism (Mackie, 1977, pp. 15–17.) As noted above, Mackie reminds us that the most likely candidate to explain our moral sensibilities once we have discarded non-natural or supernatural alternatives is Darwinism. Thus, the skeptic accepts the proposition that the moral sense is evolved. It is a part of nature, but moral ‘facts’ are facts of human nature. And as such, there is no real (i.e. external objective) foundation for normative ethics (cf. Ruse, 1999, p. 217). In this later paper, Ruse here clarifies his position on Hume and the ‘is/ought’ problem. Ruse reveals that he believes Hume denied that ‘ought’ could be derived from ‘is’ because no ‘ought’ is possible. He states that what he said in 1986 was true, namely that morality is an “illusion of the genes” (Ruse, 1999, pp. 218–219). On the skeptical view, the objective reality of meta-ethics is obviated once moral facts have been turned into facts of psychology. This subjectivist stance also enables Ruse to clarify his position on the naturalistic fallacy. Although he castigates Huxley for committing the naturalistic fallacy by arguing from ‘is’ to ought' (see page 13 above), he absolves Gaylord Simpson because the latter approached ethics from an acceptable psychological stance.
The undermining of the meta-ethical foundation is unacceptable to realists of the Brink sort because Brink commits moral realism to the view that moral facts exist independently of our beliefs and other psychological states (Brink, 1989, p. 7). Are there forms of moral realism that comply with the stricture that moral facts be limited to facts about the nature of the human species? 4 I have previously noted that Gibbard (1990) and Blackburn (1998) follow Hume and endorse the latter's position that moral facts are nothing but facts about the sentiments expressed by human beings in regard to the conduct of other people. Blackburn identifies himself as an ethical, or Cornell, realist whom he defines as believing that “ethical predicates refer to real natural properties of things” (Blackburn 1998, p. 119). The first thing one should note is that he did not include supernatural properties or non-natural properties on his list of moral realities as distinct from Brink.
Initially, Blackburn had defined his position as being ‘projectivist’ but he later adopted Gibbard's ‘expressivist’ nomenclature as being better suited to describe their form of ethical realism. The ‘expressivist’ views ethical knowledge more as a matter of knowing how to behave rather than knowing that something × is good or bad (Blackburn, 1998, p. 149). The goal of ‘expressivism’ is distinctly anthropological in that it is descriptive and not prescriptive:
So the expressivist thinks we can say interestingly what is involved for a subject S to think that X is good. It is for S to value it, and this can be explained in natural terms (Blackburn, 1998, p. 50).
The aim of ethics should be “to express and systematize our actual values” (Blackburn 1998, p. 50). Blackburn takes care to emphasize the pragmatic rather than cognitive nature of ethical discourse. He states that ‘expressivism’ does not look for the actual state of mind involved in making an ethical evaluation but rather towards determining the goals and aims of those that place ethical demands on others. Blackburn argues that moral rules are justified by their impact on general, i.e. social, happiness. This makes his type of realism a form of Utilitarian realism, and he claims that Hume himself was an indirect Utilitarian. Ethical structures are created for the purpose of encouraging social cooperation because social cooperation is necessary for the protection of the ‘good of society’ (Blackburn, 2001, p. 91). These concerns are the same ones that underlie Darwin's account of the evolution of our moral sense.
Realism, Skepticism and Evolutionary Moral Psychology
Moral skepticism and (some) forms of moral realism refocus our attention as to what is the proper object of ethics. Both traditions recognize that moral facts are psychological facts of human nature and conduct and are not an external transcendental or supernatural objectivity. What appears to distinguish the two is that for the skeptic, placing moral facts on a psychological or biological basis undermines the claim to any real foundation for ethics, while for the utilitarian (or Cornell) realist there is a foundation that can be located in optimizing the ‘good’ of society. I don't think these two positions are very far apart in terms of their practical consequences for the study of ethics.
The skeptic proposes that a practical morality must be one that is negotiated between the members of a polity or between members of different polities (Mackie, 1977, p. 170). That is, we must recognize and accept that egoistic self-interest exists and that people can be expected to pursue their own self-interests and that as a result, individuals or groups will possibly compete with one another and that conflicts are likely to arise. The purpose of ethical discourse is to ameliorate these conflicts by limiting selfish behavior for the good of the larger group or community. The ‘ethical realist’ Blackburn advocates very much the same directives about the purposes of ethics.
Amongst the activities involved in ethics are these: valuing, grading, forbidding, permitting, forming resolves, backing off, communicating emotion such as anger or resentment, embarrassment or shame, voicing attitudes such as admiration, or disdain or contempt, or even disgust, querying conduct, pressing attack, warding it off (Blackburn, 1998, p. 51).
Similarly, Mackie refers to Hume's notion of “confined generosity” to illustrate the centrality of the task of constraining selfish acts for the common good (Mackie, 1977, p. 170).
Although the subtitle of Mackie's book is “Inventing Right and Wrong”, a better subtitle might have been “Negotiating Right and Wrong”. Given the large number of considerations that need to be taken into account in any given case, the development of ethical strictures is inevitably dependent on contextual or situational factors with considerable individual and/or group variation. Dennett lists a number of principles and maxims, which, on a unitary approach to the construction of moral systems, would put contradictory items into the same toolbox (Dennett, 1996, p. 507). But that is Dennett's point: the various maxims simply are practical tools that might come in handy in achieving moral decisions. Although Dennett regards them primarily as short-cuts, the moral skeptic would regard them as required negotiating tools. If you want a formal moniker, Mackie refers to the position as “rule-right-duty-disposition egoism/utilitarianism” to try to encompass the full range of tools in the moral psychologist's toolbox (Mackie, 1977, p. 200). At base, Mackie's approach is pragmatically indistinguishable from the Utilitarian realism offered by Blackburn (see p. 19 above).
Both approaches have the virtue of being consistent with the multiplicity of moral systems found around the planet and hence does not encounter a problem with the ‘argument from relativism’ that undermines the objectivist position. The evolutionary moral psychologist, of course, is interested in describing a limit to moral relativism and searches for this in the evolved facts of human nature. Have we found anything that could serve as ethical home base? Is there anything that is universal to various different moral systems that can be traced to species psychology? A good candidate, according to Mackie, might be Kant's central focus on the fundamental role that goes into our ability or disposition to make and keep promises. We might also include the control of violations of promises, and the disposition to punish those who fail to punish violations as interesting and highly relevant natural facts of moral psychology. Research on reciprocal altruism, especially the game theoretic experiments with prisoner's dilemmas and commons exploitation have provided a solid foundation for the evolutionary analysis of moral psychology (e.g., Skyrms, 1996).
Finally, let us consider what, if any, constraints might be placed on the aims and objectives of evolutionary moral psychology. In 1985, Philip Kitcher criticized the “vaulting ambition” of E. O. Wilson's (1975) sociobiology program on grounds that it committed the naturalistic fallacy. Although, it might be possible to commit that fallacy, placing ethics beyond the realm of natural facts is certain to commit the anti-naturalistic fallacy. Despite his misgivings about the prospects for an evolutionary science of ethics, Kitcher made four assertions as to what he thought an evolutionary account might legitimately be able to accomplish (Kitcher, 1985, pp. 417–418; 1994). Each of the four possibilities he advanced was more ambitious than the one before it. The first of these possibilities was purely descriptive: it gave the scientist license only to investigate how people came to acquire ethical concepts or how they make ethical judgments the make and how they construct ethical systems. These aims are easily attainable within the stated province of both moral realism and moral skepticism as presented above. The second possibility granted that we might be able to derive new ethical principles from a combination of newly discovered facts and already accepted normative principles. The third possibility is that evolutionary theory might be able to explain meta-ethics, i.e. provide the foundation that underlies our ethical beliefs and behavior. The fourth possibility was that evolutionary theory might justify new and different ethical norms. Perhaps as Arnhart (1998a) has argued it might be possible, by happy coincidence, that new facts and old principles might converge, but it would be a mistake to stipulate that already existing moral principles must be true because they privilege the possibility of a sacrosanct non-natural basis. 5 Moral realism of a non-natural sort (e.g. Brink, 1989) leaves us open to making this mistake. If the moral skeptic is right, then possibility three is impossible because there is no legitimate foundation for meta-ethics. If, on the other hand, we wish to form a foundation for meta-ethics at the level of the social ‘good’, then Utilitarian realism might be preferable, but it should be remembered that this is not the sort of foundation that the skeptic is interested in denying. The most far reaching possibility is that the science of evolutionary moral psychology might invent new unforeseen ethical principles. Although expressivists such as Gibbard and Blackburn restrict themselves to a descriptive approach to ethics, if the members of society were to utilize scientific facts in constructing novel ethical principles, they would not and could not disapprove. Thus, there is nothing in either the skeptical view or the ethical realist view that would deny the possibility of the invention of new ethical principles to replace old ones. Therefore, let it be said that evolutionary moral psychology has nothing to fear and much to gain from the investigation of brute facts.
Notes
David Buss (1994, p. 14) also refers in passing to the ‘antinaturalistic fallacy’, but the nature of the fallacy is not formulated.
Blackburn also credits Saul Kripke as one of the innovators but does not supply a citation. The relevant text might be Naming and Necessity (1980).
Brink draws a distinction between supernatural and non-natural moral entities, claiming that Moore was committed to the latter but not the former (Brink, 1989, p. 22). I see no useful distinction between the two since both lie outside nature. Casebeer (2003, p. 21) discovered a passage in a posthumous text authored by Moore that indicated that he might be willing to assimilate the non-natural nature of the ‘good’ to either a natural or a supernatural property. Despite Casebeer's equivocation as to which of the two Moore had in mind, I think it is likely, given the definitive critique of one option, that he had the supernatural one in mind.
The author is grateful to one of the anonymous reviewers for clarifying the central importance of this issue.
Arnhart (1998a, p. 274) finds that the facts of ethical naturalism coincide with the moral teachings of the Judeo-Christian tradition.