Monday, August 13, 2012

Reframing the moral debate--a complaint about imprecision

Many difficult debates exist in the world for no better reason than that the language being used is very imprecise, in particular that the same word means different things to different people. This is a common thread on this blog, so common and troublesome is the problem. It's also a bigger problem--in the must-read (albeit dry) The Closing of the Western Mind, Charles Freeman makes it abundantly clear that a certain lack of complexity in language specifically made it harder for the Romans to articulate particular philosophical ideas than was faced by, say, the Greeks, leading to a vastly different culture. Though the English language, and most modern languages, are very, very nimble in this regard, it strikes me as a bit arrogant and precariously silly to assume that our language is sufficiently clear for all of our purposes. Indeed, I would assert that many, though not all, of our important debates, particularly philosophical ones, boil down to a certain lack of linguistic exactitude--a claim painfully obvious from my background in mathematics, where all functional terms are exactly as precise as needed, and no more so.

To name just a few examples of problematic lacks of linguistic clarity that I touch on from time to time:
  • The word God itself, distinguishable from god, is exceptionally poorly defined, rendering it a dangerous and bad metaphor. Indeed, one of the weightier chapters of God Doesn't; We Do handles this particular issue at length. (N.B.: One member of the editorial staff refers to this particular chapter, the fourth, "Defining God," as an example of me at my best, so it will be worth referencing for those interested.)
  • The word liberty means at least two very distinct things, causing much social and political issue lately. As this will be a topic for an upcoming post, I'll reserve further comment on it now save to summarize Abraham Lincoln's remark that when a shepherd rescues a sheep from the jaws of a wolf, the sheep calls the shepherd its liberator while the wolf rightly feels that its liberties have been infringed upon by the same act.
  • A conscientious friend has been repeatedly pointing out to me that even the words truth and probablility are also dangerously unclear.
For my purposes here, though, I want to focus on the claim that the word morality is also insufficiently clear in meaning, creating a host of problems, particularly leading to the ongoing debates about what is and is not moral.

I do this specifically to reframe how we approach moral questions, not to attempt to undermine it. Indeed, to label an idea, and too-often by extension the person holding that idea, as immoral or amoral is so fantastically insulting that it is a powerful barrier to effective dialogue, causing heels to be dug in and personal honor to be dragged into a discussion where it simply does not belong. At some level, this sort of shaming is probably appropriate, especially in cases where someone is wilfully and knowingly engaging in some kind of wrongdoing, but often, the charge is not heard as it should be because the listener believes himself, possibly mistakenly, to be acting morally. I would go further than this and even assert that often the problem comes from the fact that one or both people involved are too moral in differing moral frameworks, all caught under the still-nebulous umbrella of "morality."

This is not an appeal to some kind of complete moral relativism in which moral frameworks are all considered equal! That is such utter rubbish of an idea that I refuse to give it credence in elaborating now that it's stated, although it will play into the ensuing discussion.

There is a certain battle being waged recently on the front lines of the moral debate between renowned author, neuroscientist, and moral philosopher Sam Harris, with whom I agree to tremendous degree, and pyschologist Jonathan Haidt, notably over his Moral Foundations Theory.

Haidt proposes that morality, cross culturally, is at least a six-dimensional phenomenon, based upon a variety of axes, expressed as diametrically opposed ends of a continuum:
  1. Care/harm
  2. Fairness/cheating
  3. Liberty/oppression
  4. Loyalty/betrayal
  5. Authority/subversion
  6. Sanctity/degredation
Harris, responding to Haidt, wrote: "Even if Haidt's reading of the literature on morality were correct, and all this manufactured bewilderment proves to be useful in getting certain people to donate time, money, and blood to their neighbors—so what? Is science now in the business of nurturing useful delusions? Surely we can grow in altruism, and refine our ethical intuitions, and even explore the furthest reaches of human happiness, without lying to ourselves about the nature of the universe." It must be noted that Harris is specifically responding to Haidt's defense of religion under the umbrella of morality here, not that he is specifically criticizing Haidt's moral axes.

Sam Harris, of course, was attacked by Haidt, who claimed that the New Atheists are "polluting the scientific study of religion with moralistic dogma and damaging the prestige of science in the process," so he was obliged to respond harshly to this rather egregious accusation. As Harris went on to publish The Moral Landscape a few years later, it is difficult to say for certain that Haidt was attacking one of the central themes of that work, though it seems he was: that the only salient understanding of morality must rest upon the relative well-being and suffering of sentient beings.

This is where we must step back and see how complicated this whole notion is, particularly since Haidt attacks Harris (and the other New Atheists) as waltzing an understanding of morality toward moral relativism instead of away from it! The terms just aren't clear enough!

It seems to me that Haidt's research carries valid insight into explaining the variety of moral systems that have arisen in the world--a complex study, for certain. Supposing that his six-dimensional construction is both necessary and sufficient to describe the foundations of moral systems, even in the severely discretized case of a five-point scale on each axis (e.g. strongly one way, slightly, neutral, slightly the other way, or strongly), there are 15,625 moral positions defined by his framework. Some proportion of those, at least all of those admitting a net balance of values greater than zero (of which there are thousands) could be called "moral" systems with at least some degree of justification!

Pause for a moment to consider that someone who has adopted a moral system, by the above definition, that acts in accordance with the dicta of that system to the best of his knowledge, believes himself to be acting morally and will consider himself to be a moral person. This does not include those systems that are neutral or negative on the axis being considered moral to some people. Even in this grossly discrete, simplified case, taking Haidt's six-dimensional assessment at face value, there are literally thousands of meanings for the word moral! Moral relativism is the undeniably stupid position that all of these systems are "morally" equivalent, unable to be judged by any criterion and thus given equal status under "moral" considerations.

This, essentially, paints the titular theme of Harris's Moral Landscape, then! There is this hypothetical moral landscape in which various moral systems exist. Harris's central claim is captured in his subtitle: How Science Can Determine Human Values, which explicitly assumes that science can cast determination on moral systems. In other words, when Sam Harris claims that the only salient understanding of morality rests upon the relative well-being and suffering of sentient beings, it might be more clear to say that that his moral landscape gets its hills and valleys by saliently measuring the real moral worth of various moral systems using the metric of some hypothetical measure of their impacts on the well-being and suffering of sentient beings, science, of course, being able to shed considerable light upon that metric. Here we might note that neutral or even strictly negative moral frameworks, as I've defined on Haidt's construction, might be morally positive constructions as measured against the net well-being they create, though it seems extremely dubious to expect that any of those would be optimal moral constructions or even remotely close to them.

The center of Harris's critique on Haidt is that Haidt assumes a position that religious moral frameworks may be worth a lot more salt than Harris and the other New Atheists give them credit, and Harris finds it outlandish, repugnant, and clearly incorrect that a moral system based upon useful untruths has a chance of being measured optimally. I agree with this sentiment in full.

My issue comes with the term "moral" then. Under Haidt's construction, even grossly simplified, we have thousands of positions for which people could make a good case that they are being moral, and under Harris's analysis, we see the clearest, most salient way I have encountered to decide if they are right. My thesis here is that an essential problem arises from the assumed right of those acting in a moral system, per Haidt's construction or something analogous, to call themselves moral because they are succeeding in a moral framework. Harris seems to lay that claim to rest with his highly pertinent metric, but the issue arises over his (justified) use of the word "moral" to accomplish it.

Linguistically, this is exceedingly tricky to tease apart. I am tempted to call people who are succeeding in a particular moral framework "relatively moral," but that term already is loaded unfairly against the purpose. Worse, the expected negation, "absolutely moral" treads unfortunately into another usurped space--that of the moral absolutism usually associated with God by religious people (operating in "relatively moral" systems that Harris's work specifically tries to demonstrate are not "moral" at all). Another alternative is to call these people "framework-moral," which I rather like, but it leaves open the same problem for a name that appeals to Harris's conception. (Well-beingness-moral is too awkward for use, and any appeal to synonyms like goodness, righteousness, etc., is fraught with incredible difficulty.)

Since I agree with Harris's assessment that the only salient metrics of morality measure the resultant well-being and suffering of sentient beings as a consequence of some framework-moral system, I will opt to give favor to his work and call his conception of morality--a set of ethics that provide for optimization of well-being and a functional minimization of suffering for sentient beings--by its usual name: morality.

By this measure, we have no problem whatsoever in recognizing that an Islamic fundamentalist who beats or murders his own unmarried daughter over a matter of "honor" arising from an accusation of being seen alone with a boy or one that throws acid in her face for daring to learn to read might be very framework-moral while being abjectly immoral due to the extremely low likelihood that this particular framework-morality scores well on any salient metric. Notably, under Haidt's conceptualization, this particular Muslim may subscribe to and utterly fulfill a framework-morality that scores strongly in loyalty, authority, and sanctity--and may even be net-positive despite abysmal scores in oppression and harm--but he is not acting morally. This can, in fact, be pointed out without drawing direct criticism to the person, who may mistakenly think he is behaving morally, adding some sliver of hope that the accusation of immorality can fall squarely on the motivating moral framework instead of on a person who may actually only be severely miscalibrated in his moral thinking.

Now, much becomes clearer, but it requires us to talk about morality vastly differently if we wish to be precise. Field-testing of this concept in some recent discussions I've had seems to indicate that it is a game-changer for conversations about morally charged topics. I'll illustrate presently.

With this understanding, many moral frameworks can be evaluated. For example, political conservatism and liberalism can be seen as moral frameworks, and a particular conservative or liberal person can be very framework-moral, and yet that implies absolutely nothing about his level of morality. This, indeed, with a Randian Objectivist (of course, strongly conservative and starkly capital-L Libertarian), was one of the more successful conversations of this kind that I had, convincing me that reframing the discussion this way can work positively by keeping the matter firmly entrenched in the realm of abstract ideas, at least slightly lowering the chances of someone taking it personally and digging in for battle.

The conversation proceeded as so many do with Randian Objectivists, where anything but strict Objectivist ideas are doused with the black paint of being "immoral." It is easy enough to rise to this kind of talk and point out that many of the precepts of Randian Objectivism hardly qualify as moral, perhaps the utter exaltation of the self, the rejection of regulation on anything, the utter rejection of altrusim, or the recasting of anyone who isn't a "producer" as a vile parasite on everyone else--or generally a "lesser" that is worth nothing and should get even less. In doing so, though, the usual and imprecise language might read "it's flatly immoral to exalt the self and reject altruism entirely," which can very easily be taken as a personal shot at the Objectivist. (Objectivism, of course, is hardly unique in this regard--any moral framework that ties its morality to a set of fictions, especially idealized ones, instead of to the real-world notions of well-being and suffering of living, breathing sentient beings, is very likely to fail morality, in Harris's sense.) Cosmic battles of moral overture usually result, along with enmity and people who have hardly changed their positions. This is probably suboptimal a result.

A delicate change in thinking allowed me to say, though, "The problem isn't that you aren't a moral person. Indeed, you're very moral, perhaps even too moral. The issue is that you are 'very moral' under a moral system that, from my perspective, is immoral, and you see me the same way. Since this creates nothing but an impasse, we've got to look at it another way. We've got to somehow compare our moral systems, and see what falls out of them. My own particular framework aside, I would suggest that a concept of morality has to be measured against its consequences in the real world, compared against the well-being and suffering of all those subjected to those following that framework. Would you agree?"

One interesting thing about Harris's conception of morality is that once it is put to someone very plainly in terms of well-being and suffering, it is intensely difficult to dismiss. Perhaps only the likes of apologist William Lane Craig can even attempt to, and in doing so those folks are outed for their overwhelming preference of some set of fictions (the Bible, the Quran, Atlas Shrugged, etc.) to grounded, real-world concerns. While this doesn't exactly dismiss them, it certainly strikes them a heavy blow toward marginalizing their perspective.

The result, in this case, was that my debate partner had to agree with Harris's metric, complaining rightly that it is exceptionally vague and still ill-defined (of course!). In the process, while I certainly did not change her mind in real time, which is incredibly hard, I was able to get her to agree that it is "possible" that her moral framework only has evidence for working successfully in a collection of romantic fictions by Ayn Rand, even if she held to the claim that we don't know enough about "well-being" and "suffering" to say for sure about the real world--which perhaps now she will look at with increased interest and clarity.

If you've ever debated morals, or really anything else, with a Randian Objectivist, then you will realize the non-triviality of this accomplishment. I expect it was only possible because I directly started out by accusing her of being very (framework)-moral instead of leaving the matter where she could take it as an assault on her character. To me, the result was striking, and I was able to use this reformulation in another conversation to diffuse someone who was very socially conservative (telling him that he was very moral in the conservative moral framework, a commitment that was commendable in a sense) and reframe the conversation around the moral frameworks themselves instead of angry people calling each other names.

By increasing the clarity this way, perhaps with better terminology, we could tell someone that they are being very successfully, perhaps too framework-moral while getting them to re-evaluate the moral framework itself and perhaps see things from a different perspective.


If you enjoy my writing, you can read more of it in my first book, God Doesn't; We Do: Only Humans Can Solve Human Challenges. If you choose to pick it up, I thank you for your support of myself, my family, and indie authors in general.


  1. I think your agreement with Harris stems from a fundamental misunderstanding of morality. As you say, it's difficult to define, especially if you insist on incorporating intangibles such as "well-being and suffering of sentient beings." It might be easier to grasp if you look at it from a sociological or ethological perspective, in which it becomes clear that the point of morality is to bind people together in a group.

    You can't declare any particular permutation of Haidt's six-axis system as being "moral" unless there is a group out there whose prevailing norms and values are consistent with that permutation. If you insist that the only "true" morality is the morality that promotes altruism, you're missing the point. ANY morality is valid if it promotes social cohesion.

    In this context, the human sacrifice of the Aztecs and the headhunting of lowland New Guineans are perfectly moral, despite the "suffering of sentient beings" they cause. When Christian missionaries forbade the New Guineans from collecting heads, those lowland societies completely collapsed. The missionaries' efforts intruded upon and broke the social cohesion fostered by those shared norms and values.

    And no, I'm not talking about "moral relativism." Morality isn't relative, but *normative*. Any one group's morality is the sum total of that group's norms and values. Morality, in other words, is imposed on the individual, and serves as the initiation into the group. If you cannot meet the group's moral standards, you risk penalty, including ejection from the group.

  2. I think you're missing the very point Harris is making, the one that I am supporting here: social cohesion is not an adequate or informed take on morality, and calling it normative or otherwise doesn't change the fact that you are engaging in the support of a form of moral relativism here. Harris's construction is salient, even if it has to appeal to "intangibles," and your claims about the breakups of social cohesion have to play a role (as they induce suffering, don't they?).

    The thing to note about your example where societies were damaged by a change of moral frameworks is that it wasn't the new framework but rather the transition that caused the problem for those individuals--the unstable period while the stable framework that was being used was being changed to another. Your argument implicitly assumes that it is more appropriate to maintain and protect a variety of moral frameworks because of the (local) social cohesion they might create, regardless of net suffering--which is as measurable, in principle, as health. This is a problem on its own, in the local sense, because the folks could be enduring more suffering as a society than they would if they looked at things differently. This is a bigger problem in that societies are not islands, and as you pointed out clearly: the interaction of different systems causes instability that leads to more suffering.

    What I've attempted to do here is to make sense of two competing views on morals. Haidt presents something that you can call normative or local or whatever, but it fundamentally defines moral frameworks and measures how moral someone is based upon how well they conform to that framework. This could be measured in terms of social cohesion or whatever you want. Harris takes a more global view and says all that really matters is net well-being versus suffering. Though difficult to quantify and measure, these are not, in principle, impossible to examine, particularly if we are putting our focus on it.

    So, I think that your disagreement with me and, really, with Harris, stems from a fundamental misunderstanding of what makes morality salient, whether that forms the current definition of the term or not.

  3. "Harris takes a more global view and says all that really matters is net well-being versus suffering."

    That's actually a more narrow view. He's reducing morality to mere ethics. There is a big difference between the two. Ethics is about reciprocity, and as such entails a certain requirement to prevent suffering. Morality, beyond the system of ethics, is still about belonging to the group, about accepting the norms and values it imposes; it is about obeying the rules. It goes well beyond actual direct harm, and incorporates such things as outward behavior, mode of dress, mode of speech, etc. Every person who has ever belonged to a clique or a social club has encountered this kind of morality.

  4. You're using the word "moral" in a particularly narrow way and then trying to use that semantic shift to prove your point, but I'm not going to squabble with you over the fine-graded meanings of words like "morals" and "ethics." Honestly, we have better things to do with our time.

    If I concede that, then, that Harris's view is not more "global," but rather that it's more *salient*, I still hold my point, and you still fail to make yours.

  5. I'm not using moral in any "particularly narrow way." I'm adhering to the way morality is defined by the ethological and sociological sources I've been studying. I also don't think I'm failing to make any point. It's just a point that doesn't sink in where ideology blocks it.

    "Harris's construction is salient, even if it has to appeal to "intangibles," and your claims about the breakups of social cohesion have to play a role (as they induce suffering, don't they?)."

    Only if you insist that morality is about "suffering" rather than social cohesion. I think you are overgeneralizing the ethical aspect of morality. Doing unto others is part of fitting in, but only part. Your mode of speaking, mode of dress, facial expressions, courtesy, and obedience to the rules are all also part of it.

    "The thing to note about your example where societies were damaged by a change of moral frameworks is that it wasn't the new framework but rather the transition that caused the problem for those individuals--the unstable period while the stable framework that was being used was being changed to another."

    What caused the problem for those individuals was the fact that the fabric of their society was being forcibly changed. Their norms were being overruled by an outside agency, and this was fatally damaging to group cohesion. Whether you focus on the "new framework" or the "transition," the same principle applies.

    "Your argument implicitly assumes that it is more appropriate to maintain and protect a variety of moral frameworks because of the (local) social cohesion they might create, regardless of net suffering--which is as measurable, in principle, as health."

    Indeed it does. Your insistence that morality is somehow only about "suffering" is the impediment to reaching a better understanding here. It's precisely the issue that compelled Haidt to write on morality in the first place.

    Is there any kind of objective or absolute morality? If so, whence does it derive? You're going to have a hard time making a case that doesn't appeal to magical thinking. Most societies have injunctions against "murder," but they don't all define it the same way. Killing in war isn't murder, in most. Human sacrifice isn't murder, in others. In other societies, rape and slavery were (and still are) perfectly acceptable.

    "Haidt presents something that you can call normative or local or whatever, but it fundamentally defines moral frameworks and measures how moral someone is based upon how well they conform to that framework."

    He's not the first. The work of Konrad Lorenz and Niko Tinbergen, on which the field of ethology is based, begins with observations of morality in animal populations. In all cases, it's about how well they conform to local norms.

    If you don't like ethology, consider psychiatry. Freud made the case that morality is rooted in the sense of guilt, which is expressed whenever the individual performs an act not approved of by his peers. The ego, the self-censoring part of the personality, is concerned with fitting in, and as such is a keen observer of those norms.