Monday, June 23, 2014

The illusion of moral choice

Following Sam Harris's "Moral Landscape Challenge" has provided me with no real surprises. I've talked enough shop with moral philosophers not only to have expected the response given by the winning entrant, Ryan Born (whose winning essay can be read here), but probably to have been able to write it. In fact, and though I'm quite a fan of Russell Blackford and his work generally, I also have come to know enough about the topic to have known Blackford's general soft-spot--or ability to see the accurate rationality, if it applies, to be fair--for Born's argument to have been able to have guessed something like that would win. I almost did write something along those lines in a vaguely Sokalish mood, in fact, but thought better of it because of my agreement with Harris's arguments both in The Moral Landscape and in Lying.

I must say that Harris's response to Born proved mostly predictable as well, as it's similar to what I would have said in his place and more importantly the obvious extension of the thoughts he's published on the topic before. It is an excellent read, though, and it pushes into the philosophical limelight a few concepts that really need to be there, particularly the one about inhabiting a "single epistemic sphere."

Born, on his blog, responded to Harris's response in a way that was, again, utterly predictable given familiarity with his case. Harris will probably have no trouble dealing with it, but the circumstances remind me of Tim Minchin's metaphor about playing tennis on the opposite ends of two different courts and executing well-aimed serves with great technical skill and minimal point.

As I see it

Not to oversimplify the matter--for there twists and turns that are relevant to their discussion that I will not bother with here (or probably anywhere, ever)--Ryan Born's case boils down to the issue with the fact-value distinction, that we have to import at least one value into our conceptual space to get started with the ethical enterprise, and moral philosophy isn't just best suited to do this, it is uniquely suited to do it. Harris's case rejects that claim, sort of. I agree with Harris, perhaps more deeply than he's written so far on this matter, and this essay of mine seeks to outline why.

As a quick disclaimer, one that will prove more germane as this essay develops, I want to make sure it is clear that I am in no way trying to, or even interested in, stealing Harris's thunder. I trust he can, maybe will, make a similar or the same argument, but as I often do best to clarify my own thinking on a matter by trying to spell it out for others, I've elected to do so here. Rather of proof of that fact, I also have no interest whatsoever in engaging with Born's arguments with Harris in any level of detail. That's for Harris, if he chooses to do it, obviously, and not some upstart with a blog.

I see a fundamental error in Ryan Born's case that absolutely no moral philosopher is likely to agree with. I don't think we can choose our moral values, though it's far more accurate for me to say that I don't know that we can choose our moral values mostly because I don't actually know what "choose" means on this level of analysis. Under certain definitions that would follow from compatibilist views on free will (that the fact that we're a phenomenon of nature and its mechanistic, though not always deterministic, laws is compatible with free will, a position held by philosophers including Daniel Dennett, and contrasted against incompatibilist views), perhaps we can and do choose our values, though even in that case I doubt it, but in that I found Harris's Free Will to be as persuasive as his other printed efforts, I don't camp in a compatibilist tent.

The essence of a case like Born's for the utter necessity of moral philosophy boils down to the idea that moral values are arrived at by thinking about them. The Bard, earning his title, put it as well as it has been put when he had it escape the mouth of Hamlet in the Second Act, addressing the hopeless intelligencer Rosencrantz, "There is nothing either good or bad, but thinking makes it so." Goodness and badness, the only meaningful moral currency, are defined by our values, and it is up to us to think about and then choose what we will value. I don't think this is coherent if we really get to the bottom of things, though it certainly makes sense in how we do, and maybe ought to, approach it in the day-to-day sense. "Ought," here, of course, being as Harris describes: an assessment of what we think will return the best result as measured on some metric of well-being and suffering, and "think" being the progressing set of mental states entering our conscious awareness to which we are treated with a uniquely personal experience.

Harris's case in The Moral Landscape is that there's something fundamentally broken with this idea, pushing the idea that "good" and "bad" can all be boiled down to a single overarching value that compares well-being and suffering. It is critical to understand Harris's point to fully appreciate the genius--which is the right word even if he is wrong about this--required to attempt to foist such a revolution on one of the oldest branches of philosophical thought. That point is that the terms "good" and "bad" are only truly intelligible against the standard of well-being and suffering. As those states are conceivably measurable by meaningful, though undefined and surely complex, metrics, moral values fall within the purview of scientific inquiry, broadly construed. While Harris admits we must import moral axioms to start the effort of a moral science, his case is not just that this is a trivial point that doesn't justify the expansive edifice of moral philosophy as the only vanguard to ethical reasoning, it's also that the axiom he proposes is the only one that makes sense at all, which means it would satisfy the usual "either self-evident or incorrigible" definition of "properly basic" that philosophers are often so concerned with.

He imagines a "moral landscape," something like a function on a (perhaps complicated, many-dimensional independent variable space) that defines some reasonable, but hypothetical, metric measured in well-being and suffering. This landscape is likely to have many peaks and valleys, none of which is guaranteed to be the unique best or worst, even if they are maximal. All nadirs in this space, which if they are multiple must return the same value (reasonably a negative value) in the metric, occur at any point, meaning from any moral framework, that generates "the worst possible suffering for everyone." Different sets of human, (or moral, if we must), values will correspond to different values in the space, and almost anything that moves us along a positive gradient can be construed as a form of moral progress. (This point is in reality somewhat complicated because something that moves us upward but to a relatively low peak that we might mistake for a zenith maybe shouldn't be branded "moral progress," but this is a digression from anything constructively meaningful at the moment.)

Born's case accuses Harris of choosing this axiom, taking well-being and suffering as the bedrock moral value, in a way that is decidedly philosophical, not scientific, and not being sufficiently appreciative of this point. Harris's point is that, really, that doesn't matter. Philosophers, be they moral or otherwise, fuss about with axioms, and those underlie every endeavor. Furthermore, some kind of consequentialism is crucial to every values framework in a way that philosophers seem not to fully appreciate. Still, the meat of our ethical work, Harris's case makes out, is not in choosing this axiom with whatever philosophical machinery is needed to do so; it is that once this is done, so is the moral philosophy because moral science can take over.

I actually don't think Harris goes far enough here.

Can we choose?

The moral philosopher's case, as I'm understanding it, depends upon the notion that we can choose to value whatever we want, including valuing certain things simply for their own sakes. I think this is an illusion (and really, I think Harris does too, given his take on free will).

Until recently, I haven't been able to articulate the necessary points that support my suspicion that we're dealing with yet another of our mind's many fantastic illusions--illusions that probably make living life worthwhile, if not possible in the first place. (The details of this point are largely irrelevant, but the fact is that the brain takes our raw sensory material and filters it, makes it coherent, and feeds what we call our conscious mind a fiction that we call our everyday experience, and it is more than just plausible that a successful livelihood as an animal in a competitive world depends upon experiencing only a remastered construction of what our senses are actually telling us.) In this case, the illusion is a case of being, in a manner of speaking, a bit too smart for our own good. It arises from being able to imagine counterfactual possibilities with some degree of clarity and the ability to make predictions using inferences from both our experiences and imagined experiences, including those counter to the real state of the world at any point in our pasts.

In short, I don't think we choose what we value at all; we simply imagine that we could choose to value different things, in principle. In the past, we imagine and judge our behavior against the consequences we experienced and how we imagine the consequences would have come out had we behaved differently. This thought is used to shape our inferences about consequences of future behaviors, and thus the likelihood that those will align with our values and thereby produce a "good" result.

From my incompatibilist perspective, I don't think we choose any of this at all--depending upon what "choose" means in this context--and I don't think that we could have chosen to do something else in the past. In the future, our choices will manifest as whatever combination of factors, mental states, inputs, and whatever else, cause it to seem to us that they have the best chance of achieving our goals, which are inextricably knotted up with our values. But if we cannot choose any of this, it seems fairly absurd to believe that we're choosing our values, or that we could choose different values. The illusion of choice lies in imagining counterfactuals as legitimate possibilities.

But values change

Of course, our values aren't static. It strikes me as increasingly likely that our values are best construed as certain states of mind, these having been created by the various inputs--experiences, genetics, and what-have-us--that have defined who we are at the present moment. Various inputs anywhere along the line can change the state of our brains, and this is happening literally all the time. This ongoing process leads us to value somethings other than what we did previously. Imagining that things might have been different satisfies part of the illusion that we somehow chose those values, the rest of it being supplied by the usual mistake of believing that we are the conscious author of our own thoughts. None of this, though, is the same as choosing our values, or choosing to value something other than we did, unless, as I've suggested is possibly reasonable, this is how we're defining "choose."

Since we, construed as our conscious minds, only become aware of our choice after it happens, via a change in the state of our brains, it seems very odd to suggest that we are choosing our choices in the usual sense, that of a mind driving itself. This, of course, is not how it feels to think, to live, to experience, or to choose, but how it feels isn't necessarily how it is, even if it does happen to be the best way to act. If we know anything at all about ourselves, we should certainly include that fact amongst the known.

What are values?

This leads to a surprising difficulty with the idea that we could choose to value something other than what we value, and with that thought, I hope to address what I've left out until now: where our values come from, or, rather, what they are. Indulge yourself for a few minutes in the following exercise. Attempt to explain why you value anything that you value without referencing perceived consequences of acting in accordance with those values. My bet is that the only examples you can come up with are ones that somehow seem biologically hard-wired but that drive us in the opposite direction of what we feel like we truly value. Note that those too are consequences-based, but they may be at odds with other more refined notions that we hold for surprisingly similar reasons.

This brings me to a discussion that will seem admittedly a little weird, but I do hope that my reasoning becomes clear as I develop it. As it turns out, some, perhaps most, plants emit certain chemicals from their leaves in response to various circumstances. For instance, the tomato vine is known to emit a compound from its leaves when attacked by certain kinds of insects, and that compound has the effect of causing any tomato plant it happens to land upon, other parts of itself or its neighbors, to exude foul-tasting chemicals into the leaves and stems that make them unpalatable to the bugs. This is action as a result of biological hard-wiring.

Now pretend for an instant that plants can think. In this thought experiment, it isn't simply particular enzymes being released in the damaged leaves that trigger a response of emitting other compounds that engender the same effects in other nearby plants, it is a decision on the part of the plant to defend itself and its neighbors (which are likely to be kin) in this way. This decision, based upon the plant's imaginary ability to imagine that things would go worse (consequences) if it didn't perform this action (based upon evaluating a counterfactual), is almost impossible to construe as anything but a value not to be eaten by bugs, itself based upon values related to what all living things do, which is produce more entities of roughly the same kind. (And "roughly" here is important because it is a consequence of evolution, and thus that it is a successful strategy for biological things to employ, not to simply copy themselves but rather to make genetically distinct facsimiles.)

This is going to sound extraordinarily controversial, but the only real difference between ourselves and a tomato plant in these kinds of circumstances is that we would never say that tomato vines value not being eaten by bugs because we don't imagine them as being able to think about it. Tomato vines are our cousins. Somewhere on the phylogenetic tree, admittedly way back near the bottom, there was a common ancestor that gave rise both to tomatoes and to us. If we imagine our parallel evolutionary paths, the notion that we can value not being eaten by bugs but that a tomato plant can't starts to lose a little of its force. This is a feeling that we humans are critically different in a fundamental way as most or all other animals, although the only real critical difference might be thinking so. Thinking this way is a form of stealth dualism that creeps upon us frequently, even when trying to avoid anthropocentric solipsism, and I think its endemic to most of moral philosophical thought.

In our evolutionary past, we had "values" in the same way that tomato vines have them today, as a pure consequence of our biological system reacting to its environment. My thought is that nothing has changed in this regard. We are still biological systems reacting to our environments in ways that we perceive, broadly construed, to be to our benefits. It's simply that our ability to imagine counterfactuals and to make predictions, along with a few other factors intimately related to our general measure of intelligence, like sociality, make these systems intractably complex (even in a way we don't attribute to tomato vines despite the fact that we didn't even know about these messenger chemicals until quite recently). Our predictive capacity is still a reaction to our environment, as is the ability to imagine various possibilities that can be weighed, projected both into the future and into the past.

Murky complexity from abundantly clear simplicity

There are reasons we value what we value, and they have everything to do with our evolution, however complex our set of inputs or our capability for attempting to process them. The reasons we originally valued precious stones, and thus largely still do, for instance, are murky. It is probable that almost all of us have picked up a shiny or peculiar rock at some point in our lives because we thought it looked cool (aesthetics, which has psychological impact), that it would be neat to show to our friends (social, including possibly status), or it might well be worth holding onto in case it was in some way able to be used or traded. All of these are consequences related to the meeting the challenges we are evolutionarily adapted to handle.

Likewise, it is probable that almost everyone who has picked up rocks in that manner, geological rock-hound or not, has at one point or another reflected on the "objective" worthlessness of the stone, especially if it is not precious or even semi-precious. And yet we picked it up and held onto it, maybe for a considerable amount of time, because we valued it, even if for inscrutable reasons. Of note, this behavior is not unique to humans, and our reasons may ultimately be quite similar to why magpies and ravens collect shiny things as well, though it feels like a bit unlike how moral philosophers use the term to suggest that the birds value their treasures.

Our variety of values, all of which can be construed in terms of "the good life," which is well-being or suffering, arises from our evolutionary past, including its social aspects, and is probably too complicated to describe in satisfactory detail. For simplicity, then, I will focus only on a generic term that is definitely the mental result of our evolution and all that came with it, dubbing them "psychological pathways." There are reasons for our values that may be quite simple, as in the valuation of adequate nutrition to sustain our lives, or quite complicated, as with valuing abstract art, but in these cases, a common, underlying theme is that we are affecting our psychological pathways in ways we subjectively perceive as being positive or negative, producing or precluding well-being, causing or ameliorating suffering.

But we don't choose our psychological pathways, at least not really, as we discussed before. It is more accurate to say that we consciously acquiesce to particular thoughts that crop up in our minds, those being particular patterns of neurological activity, chemistry, and related processes. We don't even choose to acquiesce, though; our acquiescence is yet another expression of the same kind of physical phenomena. Tautological as it sounds, we value what we value because we value it, and those reasons have everything to do with the wide set of inputs that have shaped our mental and psychological pathways and nothing to do with choices consciously made. Instead of rendering this position a pointless tautology, though, what this perspective provides is the ability to see values as certain kinds of facts about the world.

It comes down to need

This, of course, only partially answers the question of where our values come from. The more direct answer to their genesis is from needs--many of these complicated psychosocial phenomena that are nearly intractable and often inscrutable. But the chain doesn't need to end there because our needs aren't a stealth-dualistic concept either. If we compare ourselves, or a tomato plant, or a virus for that matter, to a rock, it seems clear where our "needs" come from: goal-directed behavior. And in this case, the goal isn't chosen either. The goal is self-replication.

In order to self-replicate, as all genetic material does by definition, there are two obvious requirements. First there is the need for the raw materials and paraphernalia (e.g. enzymes) needed to create a copy, and second there is the need to avoid destruction long enough to replicate. These are not intentional goals. They are fundamental physical requirements that apply equally to intentional biological organisms and utterly mindless single-molecule chemical self-replicators. At this level, though we don't usually call it "good" or "bad," it is "good" to succeed at self-replication (for a self-replicating molecule is then doing what it does) and "bad" to fail at it (because it is not doing what it does). There is no evaluative process involved here, just simple statements of whether or not a kind of thing meets a definition (self-replicating).

The biological process, even in its most rudimentary form operating only on simple RNA molecules, inherently--this being critical--has the goal of self-replication that requires access to the appropriate materials and allotment of time to succeed. Those materials, including the space of time, constitute needs. Needs lay the basis for all simple values, whether we conceive of them broadly, as with tomato plants and viruses, or rather narrowly, as with intelligent animals like human beings. Complex values are servants to the simple ones, which have nothing directly to do with our conscious experiences, all their intricacy being due to the long and ruthlessly competitive operation of natural selection, which pits self-replicating chemical structures against each other in a battle for restricted resources. That means not only do the most successfully prolific self-replicating molecules possess the best mechanisms for preservation and replication, they create for themselves mechanisms that are best at serving self-replication. (And that variation in the expression of the replicating thing is one such mechanism is as incidental to the fact as is that it produced sentient animals that can marvel at such things.)

Why do we value what we value? Because we have needs. Why do we have needs? Because we are biological, which is ultimately to say because we are self-replicating chemistry in action--and given our position a few billion years into the evolutionary process, it's a reasonable guess that our needs, thus values, are attendant to extraordinarily successful self-replicators. All of our values, simple or complex, are reflections of this fact, and we certainly do not choose them. They all exist because successful self-replicating molecules in long-term competitive restricted-resource environments must be extraordinarily successful self-replicators, one consequence of which seems to be the ability to modify the environment around oneself to meet those utterly basic, physical needs of chemical processes). The complicated evaluative systems that are our brains drive thoughts and actions in the directions of guesses that it estimates are most likely to satisfy our labyrinthine collection of needs, all of which are slaves to self-replicating molecules that have only two simple values: more raw materials and enough time to use them.

But, but, but...

What about the fact that people value very odd things that are hard to make sense of? What, though, about the facts that some people don't want to replicate themselves? What about the fact that some people don't value replicating themselves, or anyone doing so at all? What about the notion that it's possible to "value" that all life be eliminated (supposing that's even possible)?

A digression into pica

Why do we value odd things that are hard to make sense of? Psychological pressures that are the evolutionary product of the two simple needs of self-replication. The psychological disorder--meaning deviation from the norm--called pica creates a good example. A person who suffers pica has an appetite for non-nutritive substances like ice and dirt and acts upon it, eating those things, often to deleterious effects (particularly on one's dentition).

While the fact that it is a psychological disorder (somehow related to the obsessive-compulsive spectrum) indicates that people may not actively value behaviors like eating dirt, but if we understand action upon a psychological pressure as a means of reducing psychological distress and achieving some modicum of relief from suffering, it can be seen as a kind of value indeed. The specific behavior doesn't define the value; the reduction of psychological distress does. But what is psychological distress for? It is for helping the organism pursue what it has deemed necessary, perhaps not intentionally, to maintain itself.

Pica makes an interesting example as well because despite its recognition with OCD (and in some cases cultural factors--a psychosocial phenomenon integral to the evolutionary product we call human beings), it is also recognized to be tied in certain ways in many cases to particular mineral deficiencies--a case where the self-replicating bits register a shortage of a necessary raw material to their overall self-replication machine (here: a human organism).

Values that are hard to make sense of are easier to understand in light of the concept of psychological pressures, which are evolutionary products of self-replicating molecules no matter how complicated. That these, as with pica or other psychological disorders, can go awry or be hijacked is not surprising nor any reason to believe that we need to assume that our realm of thoughts is somehow a privileged domain for the existence of values, accessible in principle only by philosophers and empirically tractable in principle only by normative means, however useful those approaches are in practice.

The other objections

The objections are only superficially meaningful. Part of the consequence of our evolutionary heritage is having developed the capacity to imagine potential futures and to evaluate the consequences of those. That capacity brought with it psychology, which brings upon each individual an experience that may not press individual replication in the form of reproduction to the forefront of importance but still has everything to do with self-replication at the level of the thoughtless clockwork of chemicals that make up a huge proportion of the cells in their bodies. It's simply misleading to think of this problem in terms of individual human beings seeking to produce human offspring, however many of their social activities and attendant psychological pressures are slaves to that drive anyway.

When it comes to valuing destruction of sentience, that is, some kind of destructive nihilistic ambition, there are still reasons for those values that reduce to psychological pathways. Self-destruction can follow from the need to ease a pathological psychological pathway, or as we commonly put it, suffering. To imagine that all sentience, all life, or all self-replicating molecules are bad and should be destroyed is simply to commit a gross error in believing that suffering is a universal feature of sentience, life, or mindless self-replicating molecules in a way so complete as to warrant its destruction as the only real positive gradient on the Moral Landscape.

If one conceives of the nadirs on the Moral Landscape as being below sea level, negative values in which the balance of experience is tipped toward the suffering, a need to reduce suffering can make self-destruction appear to be a moral positive. Valuing something extreme like the utter destruction of life is to have come to believe that the only achievable nonnegative point in this space under any metric rests at exactly zero by means of flattening the Landscape by rendering it moot.

(Note: This matter of ethical responsibility to handling well-being states that land irrevocably in the negative, at the level of individuals, is more challenging a question than it appears on the surface. The entire ethical debate about euthanasia--the right to die--is centered upon this question, and it can be construed as a question of whether or not permanently negative states of well-being can actually exist for someone. This important discussion is unnecessarily and damagingly obfuscated by inhumane, dogmatic religious taboos.)

Anyone who seems to be valuing these things must be understood to be, like everyone else who values anything, in thrall their psychological circumstances, which themselves are an evolutionary byproduct of the simple fact that there exist self-replicating molecules, some better at the game than others. A self-replicating chemical system that can bring itself more raw materials and prevent its own destruction for long enough to replicate itself one or many times has a natural advantage in the self-replication game over others and will proliferate. Apparently, brains that exhibit psychology is a sufficiently successful strategy to have led us to our present state of affairs, in which those amazingly clever brains can convince themselves of fantastic things. One such fantasy is that our values arise purely from thoughts that can be chosen and are distinct from the satisfaction of needs that all, ultimately, exist to play their role in the self-replication of certain molecules that do what they do simply because it's what they do.

We have values, period.

It isn't, then, that we choose even our first value. We simply have values as a fact of nature, and as a fact of nature, those values are in principle discoverable by the means and methods that are collectively known as science. Construing them as well-being and suffering puts primacy on sentience, but why shouldn't we? There seems to be something central about sentience to moral values (and, obviously, to the "human values" Harris wrote his book about), and well-being and suffering aren't anything like an arbitrary choice. The very meanings of those terms--which arise from and cannot be freed from our empirical experiences, a point that often seems lost on philosophers--are expressions of our psychological pathways. Those, though, are just another expression of our evolutionary heritage as chemical systems consisting of and in service to self-replicating molecules. We value what we value because that's what value means. Harris is right, then: it's simply unintelligible to talk about human values in any other way.

But philosophy is important

Moral choice is an illusion, but this need not and should not be the death of moral philosophy or its primacy in the ethical arena. Morality is simply too complicated to be addressed empirically in a clear way in practice at present, and that may always be the case. We haven't even anything like the proper metrics or the tools with which we could measure much by way of various attempts at human values, and even if we did, the data we would find might simply be too unwieldy to wrangle into anything fully sensible. Further, the philosophical matter upon which moral philosophy ultimately hinges, the compatibilist versus incompatibilist interpretation of free will, must continue and seems primarily a job for philosophy. Whether that discussion is settled or not, philosophers should aim to make clear and useful sense of the notion of "choice" under incompatibilism, which also looms over the salience of moral philosophy as a uniquely priviledged endeavor in making sense of human values. For all of these debates, one thing is clear: philosophers interested in the task aren't going to get anywhere meaningful without taking on a lot of neuroscience.

Therefore, this does make a definitive case that to the degree that moral philosophy remains relevant, which should be very significant, it must be informed by and, in the parts where it is relevant to do so, ceded to the nascent moral sciences. "Lead, follow, or get out of the way," the saying goes, and moral philosophy is now justified in the position to do all three as is appropriate, but it no longer has permission to use the illusion of moral choice to justify confusion on which it should do in what circumstance.

No comments:

Post a Comment