on empiricist theories of ethics

in contrast to this, the rationalists argue for a normative binding force of ethics, not reducible to contingent facts, and which binds actors categorically. i will seek to undermine the theories of the empiricists, showing how certain trends beginning in Hobbes persist, until they offer a near total threat to the theory of David Hume.

Hobbes, somewhat notoriously, redefines many moral terms. for example, a liberty is an absence of external impediments (in a strictly physical sense) [ch xiv.2]. a right of nature is a liberty to do certain things [ch xiv.1]. all later rights in civil society are the remnants of this original right which contractors never relinquish. a right, therefore, is exactly an ability. in this kind of way, Hobbes reduces all such moral terms.

but later, Hobbes will use the moral terms, in ways which sound plausible, but only if we have forgotten the new meaning the words have. for example, the honor of great persons is to be valued [by the sovereign] for their beneficence and the aids they give to men of inferior rank, or not at all. [ch. xxx.16] Hobbes derives this from the obligation of equity, the eleventh natural law, which says that if a man be trusted to judge between man and man...he [is to] deal equally between them. [ch. xv.23]

this holds, because without [equity], the controversies of men cannot be determined but by war. and whoever causes war offends against the fundamental law of nature, which is to seek peace. now, so far, a reader with the most deontological of ethical terms will be happy with the sort of implications Hobbes is using. but what force does this fundamental law of nature have? it is that, in the state of nature, there is great suffering, and it is in each person's interest to avoid that suffering by seeking peace.

all Hobbes' ethical arguments proceed along just this sort of line. so it is opportune for me to pause briefly, and note several important features of his reasoning. first, Hobbes is not using terms in their usual sense. generally, there are already perfectly good terms that match his definitions (like his use of liberty to mean ability).

second, Hobbes presents himself as offering a moral theory. this is a central point, which looks quite obvious. and while he uses moral terms, he gives them new meanings. these meanings look entirely non-moral, in fact. but if we just read Hobbes and substitute the non-moral definitions for the moral terms, we would no longer have a moral theory. more on this later.

third, Hobbes offers only arguments from self-interest for his moral theory. now, while this self-interest is clear at certain stages of his work, he often suggests that one is bound to the laws of nature even if contrary to one's interest in a particular case. Hobbes's arguments at such points amount to claims that to act contrary to the law is inconsistent. and so, taken strictly, all the laws bind only as inconsistencies to be avoided. and so the laws he derives are entirely based upon self-interest.

now recall back where we were in our analysis of Hobbes. we had followed the argument forward, and discovered that a natural right is nothing more and nothing less than an ability. then, following one thread backwards, we discovered that justice is to be equitable because of the sovereign's obligation to seek peace. but the obligation to seek peace only obtains in the state of nature. it derived from conditions of relative equality. once the state is erected, the sovereign is no longer roughly equal to others. as long as the sovereign maintains power, the laws of nature do not obtain. they were mere means to an end, which now the sovereign can obtain by keeping them only often enough, not always. and so the obligation to equity in justice dies, and in a parallel fashion, a great number of other obligations with it.

writing about seventy-five years later, Francis Hutcheson offers a very different theory from Hobbes. no more are ethical commands deduced as logical conclusions from dubious propositions about human nature. no longer do we have the disappointing inconsistencies stretching across the work. but the main points of empiricist ethics, outlined above, are still here, in full force.

Hutcheson needs to redefine moral terms, but he does so in a covert way. for example, Hutcheson uses the word virtue and defines it as approbation. it is a nicely disinterested approbation, but it is a feeling in we who see, not a characteristic of the thing seen. it is accordingly not deontological any longer. consider, for example, that for Hutcheson, my desire to be virtuous is just my desire to be approved of. virtue isn't in me, but in how those who do the approving see me.

Hutcheson's theory is most attractive, because it parallels the previous work of John Locke in epistemology. just as Locke teaches that color is not in the observed, but in the observer, so Hutcheson treats goodness. but we already have terms to describe effects in the observer—the very terms used by Hutcheson: approbation and disapprobation. so the redefinition is not just making the same point about ethics that Locke made about vision—it is also designed to remove deontological content from a deontological term, while retaining the use and connotations of the term.

again, Hutcheson offers only hypothetical motives for action. for example, obligations are either the products of a feeling of benevolence, or else the pleasing feeling we get from acting well, or else obedience to the superior force of God (see p. 520f). Hutcheson knows full well that these feelings are hypothetical and not always present, despite his nasty language about the possibility (if our moral sense be supposed exceedingly weakened and the selfish passions grown strong, p. 520).

finally, Hutcheson presents himself as giving a moral theory. now, more than with Hobbes, the significance of this as a separate point of note becomes apparent. Hutcheson has picked out one (or a complex of several) of our feelings and labelled them moral feelings. what makes these particular feelings the moral ones?

Hutcheson even has an answer: the feelings that are feelings of benevolence—they are the moral ones. but at this point it becomes entirely unclear what exactly grounds moral actions. of a large array of feelings within my soul, Hutcheson picks one out and calls that the moral sense. but the basis of the selection is the principle of benevolence. the link of benevolence and morality is thus logically antecedent to the moral sense, because it is this link which enables us to pick out one sense and identify it as the moral sense.

for we all do have various impulses: some benevolent, some malevolent. in my head i call the benevolent impulses good and the malevolent ones bad, but on what basis? my feelings do not come self identified: this is a moral feeling and this is an immoral feeling. if i have an extra feeling, which picks out among the others, and makes such a labelling, then what makes it the moral determinant feeling? in fact, Hutcheson's moral sense is just such an extra feeling—but why is approbation the right standard? is it only because i approve of people whose standard is their own approbation feeling?

Hutcheson's theory thus fails; attempting to link morality to agreement with a certain feeling, Hutcheson offers no coherent criterion to pick out which feeling is the correct one to link morality to.

we then look to David Hume, the greatest of the empiricists, to resolve our quandary. from Hobbes, we had a nearly voluntarist account of ethics, in which we should look to the sovereign to answer ethical questions. but his theory fails to adequately address the sovereign himself. to improve this, Hutcheson tells us to use our own sense of approbation the standard, not the sovereign. this frees us from the problems of Hobbes' theory, but replaces it with the difficult problem: which of my feelings is the correct sovereign? with Hobbes, the sovereign is obvious, because the sovereign can be labelled. but feelings do not come with labels, and so Hutcheson's theory fails.

does Hume get us out of the difficulty? it seems so. Hume offers to consider what makes us esteem a person, but more purely contingently than Hutcheson. the criterion is to be found in section ii of the enquiry where Hume tells us that benevolence enjoys high social esteem because it is highly socially useful. society praises what is useful for society, and, insofar as we are parts of society, we join in that approbation. the rest of Hume's theory seems to be a working out of the implications of this initial insight.

the conundrum we encountered with Hutcheson is absent now. we might ask which of my senses is the moral sense?, and Hume will stare dumbfoundedly. Hume's goal is not to identify the moral sense, nor even to sketch out its boundaries, but rather to explain moral judgments without reference to a sense. instead, we all have various feelings, and the ones which are socially useful are, ipso facto, socially praised. and Hume defines ethical actions as those which are socially useful, as seen by a suitably disinterested viewer.

in my opinion, Hume is vastly more consistent than either Hobbes or Hutcheson. what makes his moral theory wanting is not any failure in its internal construction, but rather its failure to offer a moral theory at all. because of the elegance and consistency of Hume's theory, this central failing stands out more clearly. however, it is not a failure unique to Hume—it is a failure that stems from the common traits of all these empirical philosophers. and so, let us examine why it is that Hume actually presents no ethical theory at all.

the drill is clear by now. Hume redefines terms, in order to offer a moral theory without moral content, and produces an entirely contingent and hypothetical theory. the correct way to read Hume, and by implication, his predecessors, is to back up the terminological redefinition, and read his theory in the entirely non-moral terms in which it is defined, despite the use of moral terminology.

what we have then is a theory of what humans in fact do. this is an excellent theory. Hume's goal of being the Newton of ethics is realized, but only if ethics is abandoned in favor of ethology. the failure is that we do not have an actual ethical theory.

an ethical theory, distinguished from an ethological theory, is one which offers moral imperatives. more specifically, it tells me what i ought to do. this may or may not be the same as what i in fact do; but it must be one which is truly binding, and not merely a means to some end within myself. and so, the sort of redefinition of ethical terms away from deontological referents precludes the construction of an ethical theory.

if ethics is to be distinguished from ethology, then it must consist in its separation from all local interest. Hume does not take his disinterested stance far enough. he tells that moral judgments are those made when free from local interest, but you do not meet that test merely by broadening the locality. you must broaden past all contingent locality whatsoever. and so, any ethical theory which contains contingent interest as a factor will be lacking.

(since all of our empirical observations of human nature are contingent, an ethical theory must therefore eschew empirical observations if it is to achieve the necessary disinterested stance that Hume so rightly mentions.)

we are left with a theory which pretends to be ethical, but is not really. perhaps Hume thinks that a deontological ethical theory is impossible. but, alas, he doesn't quite say so directly, and it is in this that his theory most lacks. to advance the claim that there is no such thing as deontological ethics is a very interesting claim—but the empiricists in question don't ever quite get to that point. perhaps the history of ideas doesn't see it full blown until the psychology of B.F. Skinner.

there are tantalizing passages in which Hume nearly tells us, straight out, that he doesn't believe there can be a deontological ethics (for example, the discussion of obligation in section ix, part ii of the enquiry, or the discussion of the relation of ought and is at the end of book iii, part i, section i of the treatise). but these passages are matched by others which are tantalizingly close to having deontological content. and, amidst it all, is the disturbing use of language. if Hume really wants to admit to an anti-deontological theory, he needs to abandon his use of deontological terms.

i have identified three traits of empiricist ethics, which i believe to be characteristic of the genre, in that very different writers (all empiricist) can disagree about nearly everything, but always share these three traits: the redefinition of ethical terms away from deontological content, the claim nonetheless to be presenting an ethical theory, and the reliance upon contingent facts grounding the theory. i have argued, convincingly, albeit with hubris, that such theories are all merely ethological, and lack the crucial elements that would make them truly theories of ethics.