MP.L1 Moral Realism
Folklore and philosophy are in agreement about esthetics. Beauty, they say, is in the eye of the beholder. De gustibus, non disputandum est. There are no global objective standards for esthetics.
Oh, to be sure, people sometimes propose guidelines or rules of thumb. Symmetry and proportion in the visual arts, the "hero's journey" in film and literature. But these are only patterns that people have noticed or imagined. They are not founded on any kind of transcendent fundamental principles. In esthetics, it is mere subjective opinion all the way down. No philosopher since the Greeks has believed in esthetic realism.
In ethics, though, things seem to be different. Many believe that there are objective standards of right vs wrong, good vs bad. They are in agreement as to the reality of a moral code, even though they are not in agreement on what those moral rules actually are. Regarding morals, disputandum est.
A moral realist believes that if two people disagree about a ground-level moral statement --- for example "Eating meat is immoral." --- then at least one of them must be wrong. This is a strong claim. Yet the odd thing is that this claim is not about a matter of fact. We cannot imagine any experiment or observation that would falsify moral realism. We cannot offer evidence or argument that would compel a skeptic to accept moral realism as a principle. It doesn't take too much thought to convince yourself of this.
It seems that, as philosophers, we are free to either accept or reject moral realism. And many philosophers, in fact, do reject it. They make the counter-assumption that moral claims are purely subjective, that they are mere opinions. But, once having adopted this stance, there is not much more that philosophers can say about ethics. They must abandon the field to the psychologists and opinion pollsters.
But there is something puzzling here. While it appears that there is no evidence one way or the other about moral realism, it also seems that, if moral realism is assumed to be the case, that assumption has consequences. But as a Bayesian, I am supposed to know that evidence for or against a proposition flows from its consequences. Wtf is going on???
Do you see the solution to the paradox? While we can accept that an assumption of moral realism has consequences, they are not observable consequences. In this regard, a belief in moral realism is exactly analogous to a belief in the existence of God.
We are the victims of a cosmic joke. In seeking to penetrate to the foundations of ethics and morality, we have fallen into a kind of philosophical black hole. And we are still playing at metaethics. We haven't even started to consider the real questions like whether we are allowed to eat meat!
Moral Realism Considered Harmful.
Philosophical foundations are problematic in any field. So, as a practical matter, can't we just assume the doctrine of moral realism (without any real evidence) and proceed to think about what the moral code says rather than what it metaphysically is? After all, it certainly feels like morality is real. And people like Aristotle, Kant, and Bentham weren't stymied by metaethical qualms in analyzing morality at the personal level. Locke, Rawls, and Nozick (over the objections of Bentham1) didn't hesitate to analyze human rights at the political or collective level.
I believe that following that path again would be a mistake. All of those thinkers were, IMO, led astray by the assumption of moral realism, and by the even deadlier assumption that we, as a species, are in possession of a moral intuition which gives us direct insight into moral reality. That approach only makes sense if you believe that the cause of our intuitions is somehow correlated with the hypothetical cause of that hypothetical moral reality.
But that is absurd. We know the cause of our moral intuitions. It is the processes of biological and cultural evolution. And the intuitions should be expected to provide us with help in recognizing our own (biological) interests, particularly in those cases where our actions may have indirect consequences. They do not help us to recognize objective morality, at least not as envisaged by the distinguished thinkers mentioned above.
Intersubjective
We should instead reject the assumption that there is a moral reality which is somehow independent of the minds of the moral agents. But we should also reject the stance that morality is purely and simply subjective. Instead, we will follow Hume2 and Binmore3
in assuming that morality is intersubjective, that is that it truly is a matter of opinion, but that it draws its force from the fact that almost everyone (locally) has almost the same shared opinions. A recent PhD dissertation defends this position with more philosophical rigor than I can muster. And I will point out that some other things that philosophers worry about, like the meaning of utterances in natural language, are clearly intersubjective rather than real.
To give a preview of what this stance means in practice, consider our example ground-level moral question: is it ethical to eat meat? Well, look around. If there are butcher shops and they are not being firebombed by angry mobs, then, locally at least, it is morally acceptable to eat a hamburger.
There will be some readers who object, "This is not really what morality is." Such readers have apparently not followed my advice and rejected the doctrine of moral realism. Other readers may object that in this view of morality, there is no role for philosophy in creating and critiquing a moral code. This is a valid objection. But, in this approach, the questions that philosophy needs to address are subtly different. Ask not what the moral code is. Ask rather what the moral code should be.
Morality, in this viewpoint, is a social construct. And it exists to accomplish a social or collective purpose. The task of philosophy (and this blog) is to help articulate that collective purpose and to critique how well that society's moral code functions in pursuit of that purpose. Philosophy's role is to guide the evolution of the intersubjective moral code, so as to "create a more perfect union".
Notice the word "locally" above. The question arises, If morality is a social construct, then what constitutes a 'society'? I will argue that societies exist at all levels. A friendship or marriage imposes a code of conduct upon its members, as does an extended family or clan, as does a workplace, as does a nation state, as does the species as a whole. These distinct moral codes should mesh in a way which does not create conflicts. If they do conflict, we have the raw material for much of literature.
Reciprocity
In our analysis of what the moral code should be, one of the first steps is to determine what the word "should" means in this context. (Perhaps, it is now clear why a series of essays on ethics is titled "Collective Rationality"). Our conceptual framework will be Game Theory as it has been developed over the past three quarters of a century by economists. The main theme will be reciprocity. The slogan is "Do unto others as others do unto you". But, as our analysis will show, the human species is sufficiently united by kinship relations that the net consequential result (assuming almost everyone is almost 'rational') will be close to the (rule-) utilitarian ideal under the slogan "The greatest good for the greatest number".
Before we begin on this moral construction main project, it will be necessary to devote several postings to the site clearance and demolition subproject. We need to clearly and convincingly analyze and reject the work of Kant, Bentham, Aristotle, Rawls, et al. This will occupy our attention for several postings.
But even before that, we need to take a brief look at whether our conclusions will continue to apply as new kinds of moral agents are added to the mix. What changes if people are able to procreate by cloning? What if they can become virtually immortal in a simulated environment after being scanned and uploaded? What if we encounter intelligent aliens who, like ourselves, came into existence by a process of (xeno)biological and cultural evolution? In all of these cases, there is also reason to reject the hypothesis of moral realism, and to expect that morality can be (and must be) based on a framework of reciprocity and societies at nested levels.
But what about AGIs? We can certainly imagine creating an AGI which, like ourselves, is guided by some kind of utility function. Can we safely interact in a framework of reciprocity with such entities? Or will the power differential be too great? But we can also imagine creating an AGI with a built-in moral code4. How would such a real moral code interact with our own intersubjective one? We will take a preliminary look at these questions in some of our next postings.
Bentham famously rejected the moral realist view of human rights. In his pamphlet "Anarchical Fallacies" criticizing the French Revolution, he wrote:
... there are no such things as natural rights -- no such things as rights anterior to the establishment of government ...
In proportion to the want of happiness resulting from the want of rights, a reason exists for wishing that there were such things as rights. But reasons for wishing there were such things as rights, are not rights; -- a reason for wishing that a certain right were established, is not that right -- want is not supply -- hunger is not bread.
That which has no existence cannot be destroyed -- that which cannot be destroyed cannot require anything to preserve it from destruction. Natural rights is simple nonsense: natural and imprescriptible rights, rhetorical nonsense, -- nonsense upon stilts.
Hume writes [AN ENQUIRY CONCERNING THE PRINCIPLES OF MORALS Section IX. Part II]
Having explained the moral APPROBATION attending merit or virtue, there remains nothing but briefly to consider our interested OBLIGATION to it, and to inquire whether every man, who has any regard to his own happiness and welfare, will not best find his account in the practice of every moral duty.
...
And, indeed, to drop all figurative expression, what hopes can we ever have of engaging mankind to a practice which we confess full of austerity and rigour? Or what theory of morals can ever serve any useful purpose, unless it can show, by a particular detail, that all the duties which it recommends, are also the true interest of each individual? The peculiar advantage of the foregoing system seems to be, that it furnishes proper mediums for that purpose.
John Rawls offers a theory that reduces our notions of justice to those of fairness. I think our traditional personification of justice as a blindfolded maiden bearing a pair of scales in one hand and a sword in the other provides some support for this reduction. Her blindfold can be identified with Rawls’ veil of ignorance. She needs her scales behind the veil of ignorance to weigh up the relative well-being of different people in different situations.
The issue of how interpersonal comparisons are to be made is often treated as a side issue of no great importance by traditional moral philosophers, but it is clearly necessary for people to be able to make such comparisons in order for it to be possible for them to use the original position to make fairness judgements....
Finally, attention needs to be drawn to the sword carried by our blindfolded maiden. The enforcement question is often neglected altogether by traditional moral philosophers, who commonly take for granted that fairness exists to trump the unbridled use of power that they think would otherwise reign supreme.
However, I shall be be arguing that fairness evolved as a means of balancing power rather than as a substitute for power. Without power being somehow exercised in her support, our blindfolded maiden would be no more than a utopian fancy. As Thomas Hobbes put it: “Covenants without the sword are but words”.
When I argue against traditional moral philosophers, I have in mind the metaphysical tradition that begins with Plato, and continues through Descartes and Kant to modern times, where it is firmly established as the reigning orthodoxy. Even John Harsanyi and John Rawls, from whom I draw much of my inspiration, regarded themselves as Kantians.
However, the naturalistic tradition is just as venerable. It begins with Aristotle, and continues through Epicurus, Hobbes and Hume, to the present day. Its leading modern exponent was John Mackie, whose "Inventing Right and Wrong" seems to me to offer a devastating critique of the orthodox view that morality somehow has an absolute status unconnected with the biological and social history of the human species. Instead of imagining that it is adequate in studying human morality to adopt the pose of Rodin’s Thinker and await inspiration, he tells us to read the works of anthropologists and game theorists.
It is to this project that this paper and the books from which it is derived are devoted. It is particularly important to understand that the project requires disavowing Immanuel Kant on moral questions. If his categorical imperative implies anything specific, it surely calls for cooperation in the one-shot Prisoners’ Dilemma, but his claim that such behaviour is rational seems absurd to game theorists. Our philosophical hero is David Hume, who was preaching our creed to an uncomprehending audience two hundred years before the first game theorist was born.
For example, Asimov’s Three Laws of Robotics