Locating Value in the Natural World

by Michael Lopresto

404px-Alice_Humpty_Dumpty

The idea of objective value has come into disrepute in some quarters. We have an image of the natural world, well defined by physics—a world of mostly empty space filled sparsely with unimaginably tiny objects (an umbrella term for particles, fields and waves) that are governed in law-like ways. Indeed, this world, given precise definition and overwhelming empirical support, is often thought to be radically different to the world we know from experience—the world of vibrant colours and sounds, tastes and smells. The fact that our perception of the world seems to be so profoundly impoverished has led many to despair at the prospects of genuine knowledge of the world. So, this line of reasoning goes, the natural world given to us by physics has absolutely no room for objective values, as pure “atoms in the void” exhaust all of reality.

I think this line of reasoning is wrong, and shows the desperate need for philosophers to make sense of the natural world as defined by physics, with our place as human beings firmly as part of that natural world. To use a term from Wilfrid Sellars, it's the job of philosophers to navigate the way between the scientific image and the manifest image of the world. The scientific image is the “atoms in the void” picture of reality, where ordinary objects like tables and chairs are really just near-empty lattice like structures of atoms. The manifest image is what is presented to us in experience, where tables and chairs are solid objects, we have rich conscious experiences of music that touches us deeply, and, as I'll be focusing on in the remainder of this essay, objective values that bind on us whether we like it or not.

In his superb book, From Metaphysics to Ethics: A Defence of Conceptual Analysis (1998), the Australian philosopher Frank Jackson develops some tools for navigating our way between the scientific image and the manifest image.

Firstly, he describes what he calls a “location problem,” where we have something in our manifest image—whether its solidity, consciousness or value—that fits uneasily into our understanding of the scientific image. Solving a location problem is about giving a conceptual explication about how something manifest to us fits in the natural world. For example, solid objects like tables are chairs are manifest to us. There is nothing about solidity in the picture given to us physics. However, near-empty lattice like structures entails that “middle-sized” objects like us, with our kind of perceptual faculties, will only experience table and chairs as being solid. We humans don't have the perceptual faculties to experience the aggregations of matter that constitute tables and chairs, in a way that, when we “zoom-in” is not constituted by anything that is itself solid. Another example that Frank Jackson gives is that of semantic properties. When I write the sentence “Grass is green,” I thereby bring into existence a physical structure that's true, by virtue of the way the world is and equally by the meaning of the sentence. Of course, no semantic properties like truth, reference and meaning, are to be found at the level of fundamental physics. But again, we can find entailment from the physical structures described by physics, to the truth of physical tokens such as “Grass is green,” because such physical tokens derive their meaning from convention, in this case, rules internalised in the minds of English-speakers.

This is hard work; a project very much still in progress. Although my sketch of a location problem here is very rough and quick, I hope it's slightly clear what the project is. Alternatively, some philosophers opt for elimination rather than location. Sometimes principled reasons are given as to why somethingcannot be located in the natural world. All of us experience déjà vu at one time or another, but our conception of the natural world entails that such experiences could never be correct because that would mean that backwards causation would be at play, and there is no room for backwards causation in the natural world—certainly not at the level of human perception and memory. Furthermore, parsimonious debunking explanations can be given of déjà vu, that our brains misfire squirts of warmth that give us that positive affect of familiarity, thereby erring us into thinking that we've perceived a familiar state of affairs when we haven't.

Similar attacks have been leveled at the existence of value and morality. In his influential book, Ethics: Inventing Right and Wrong (1977), another Australian philosopher, John L. Mackie, offered principled reasons for why objective value cannot be located in the natural world, and hence should beeliminated from our ontology (what we are committed to existing). Mackie begins with the conceptual analysis part. Our question is whether or not objective value (which I'll refer to as simply value for brevity) can be located in the natural world. We understand the natural world to be that given to us by physics, the “atoms in the void” picture. But how are we to conceive of value? Mackie takes his cue from G.E. Moore's Principia Ethica (1903), the received view of how to properly conceive of value and moral properties in general at the time Mackie was writing. Incidentally, I think it's no longer the received view precisely because of Mackie's book.

Mackie conceives of values as non-natural properties in the world (by non-natural property, I mean irreducibly normative, both semantically and metaphysically). This is for two main reasons. The first, following Moore, is to account for the semantics of value-talk: value-talk has a different meaning to mere description. To use Moore's example, I can ask, “This is pleasant, but is it good?” (a description followed by a question of value), and that doesn't simply mean the same thing as “This is good, but is it good?”. So, according to Moore, value statements have different meanings to descriptive statements. Moore then concluded that moral properties are not natural properties, which I think was an overreach of his argument. We need more argument to infer the way things are in the world from the mere meaning of some of our terms.

The second reason Mackie conceives of values as non-natural properties is to account for the “objective feel” of our values. We experience morality as binding on us whether we like it or not. I may desire to steal a substance sum of money in circumstances in which I know I'd never be caught, and yet understanding that this is the wrong thing to do is more than enough to motivate me not to do it. In this respect, we may say that values are mind-independentand intrinsically motivating.

These components outline the semantic premises of Mackie's argument, as pursued through a little bit of conceptual analysis. Now for Mackie's main argument. Our conception of the natural world entails that it cannot be that values exist. Our “atoms in the void” picture of reality rules out the possibility of non-natural properties. Mackie gives two arguments for this entailment. The first is the argument from queerness. Non-natural properties are wholly distinct from anything that we're familiar with in physics; they are so strange, in fact, that they couldn't exist. That is, they can't be located in the natural world, and hence have to be eliminated from our ontology. The second is the argument from relativity (perhaps more aptly called the argument from disagreement). The argument takes the form of an inference to the best explanation: the best explanation for such deep and widespread disagreement about morality between different people in the world is not that some people are latching onto a reality and some people aren't; it's that there is no moral reality and everyone is in error.

I think Mackie's argument for elimination, as opposed to location, of value is a good illustration of the difficulty of location problems. The conceptual analysis involved is difficult, and it's easy to go awry. If Mackie's conception of value is the correct one, then I'd agree that it follows that value can't be located in the natural world. However, I reject Mackie's argument because I think his (and Moore's) conception of value is wrong. It's hard to pinpoint exactly where it goes wrong. After all, it's not completely arbitrary, and it does capture much of our “folk” conception of morality. It's important not to completely redefine a concept such as value, or give it your own idiosyncratic definition, because it may turn out rather trivial that your concept can be located in the natural world. Frank Jackson eloquently puts it thus (p. 118):

As Lewis Carroll said through the character of Humpty Dumpty, we are entitled to mean what we like by our words. But if we wish to address the concerns of our fellows when we discuss the matter—and if we don't, we will not have much of an audience—we had better mean what they mean. We had better, that is, identify our subject via the folk theory of rightness, wrongness, goodness, badness and so on.

It seems to me the best way in which we can conceive of value (identify our subject) is by examining our actual moral judgments, and then seeing if there is any mapping between our moral judgments and the natural world. Are moral judgments just arbitrary and haphazard, or do they systematically covary non-moral differences in the natural world? The former would suggest that Mackie is right, and value has to go from our ontology, and the latter would suggest that moral realists such as Jackson and myself are vindicated in thinking that value can be located in the natural world.

Let's consider the famous trolley problem, first discussed by moral philosophers Phillippa Foot and Judith Jarvis Thomson. A runaway trolley is hurtling down its track, on course to kill five people. There's nothing you can do to stop the trolley, but you can pull a lever that will divert it on to another track, only killing one person. Do you have a moral obligation to pull the lever? That is, do you have to pull the lever and divert the trolley whether you like it or not? It seems perfectly obvious to me that you do. The interesting thing about this hypothetical case, for present purposes, is that the entire scenario is described in purely non-moral terms: number of people who live or die, a diversion of a trolley. We take these purely natural, non-moral facts, and then make inferences about what's right or wrong. We don't perceive there to be any difference in moral facts without a difference in natural, non-moral facts. The moralsupervenes on the natural, as philosophers say.

There another widely discussed variation on the trolley problem just described, which we can call the fat man variation. There's a runaway trolley on its way to kill five people. But in this case, you're standing on a footbridge above the track. There's nothing you can do to stop the trolley, except push the fat man standing in front of you onto the track, thus halting the trolley but killing the one fat man, and saving the five. As social creatures, we naturally hesitate when confronted with cases like this. But is our moral judgment that pushing the fat man is obligatory or forbidden? Various people come to different judgments on this case, and I suspect it's simply because this case is so much more emotionally salient than the first case. However, in both cases the natural, non-moral facts seem to be the same: five human deaths as opposed to one human death.

Various attempts have been made to vindicate the seemingly inconsistent tendency that people have to pull the lever in the first case but not push the fat man in the second case. Some people have suggested a morally salient distinction between doing and allowing (or between commission and omission): in the original case, pulling the lever amounts to allowing one person to die and five people to live; whereas in the second case you're not merely allowing the fat man to die, but causing him to die. Some people have invoked the “doctrine of double effect”: in the original case, you intend to save the five people, and one person dies as a known by-product of this intention being carried out; whereas in the case of fat man, pushing him off the footbridge is no mere known by-product of your intention to save the five people, you actually intend to kill him.

It seems to me that in both of these responses, people have been searching for moral properties where there are none to be found. In the original trolley case, there seems to be a non-arbitrary covariation between our moral judgments and the non-moral facts of the case. Our judgment goes with the most number of lives saved. In the fat man case, drawing distinctions between doing and allowing, and so forth do seem to be arbitrary, because there is no covariation between non-moral facts and moral facts. In response to the doing/allowing distinction, we can ask why causation by omission isn't the correct way to describe our action? In response to the doctrine of double effect, we can ask why isn't it correct to say that in the original trolley problem that we do in fact intend for the one person to die?

It seems to me that this suggests an answer to why Mackie and others conceive of value the way they do. Sometimes our concepts of value get so confused or muddied that it seems like we commit ourselves to non-natural properties that can't be located in the natural world. We often forget to ground our concepts of value and morality in the natural world that we can go looking for moral properties that aren't there. Sometimes we need to stop and examine our concepts of value and make sure that they map onto natural properties, such as pain and pleasure, happiness and suffering, autonomy, cooperation with one another, and in the case of the trolley problem, number of lives saved, so that the values we manifestly experience can be located in the scientific image of the world.