In part 1 we looked at how Boyd draws an analogy between scientific realism and moral realism. The first parallel is that there is a progressively recursive relationship between theory and experience. The second is that there is a progressively recursive relationship between theory and method. Check out part 1 if you’d like further elaboration…I’ve got work to do!
The Problem of Natural Properties
One of the bugaboos moral realism has is to account for the claim that moral judgments refer to real properties out there in the world. This is somewhat incompatible with some versions of empirical science that say if a term doesn’t refer to something that is understandable in the language of physics then it’s not real in the sense that moral realist want it to be. For example, if I say something is ‘baaaaaaad’, the moral realist wants to say I’m referring to some property that actually exists beyond mental concepts.
So how does the moral realist deal with this problem? First, they’re going to reject the idea that things (in a general sense), such as states, properties, need be reducible to language of physical substances in order to be considered real. They’re also going to reject the idea that things can be only considered “real” if they’re directly observable. We don’t directly observe sub-atomic particles and their properties but we still say they’re real, so why not extend that leniency to moral properties.
Connected with that they’re going to reject what is called the “constructivist Humean” idea of causation. Hume’s idea was that we don’t directly observe causation but our mind “constructs” a causal story to explain a relationship between two things. The inference is that causal forces aren’t real, they are constructed by our minds, only the physical objects that interact are real. The moral realist refudiates this constructivism and say it’s perfectly fine to speak of non-physical relations as being real (hello friend zone!).
Here Boyd leans on Locke’s distinction between nominal and real essences. Consider gold and pyrite. A few thousand years ago before they knew anything of their respective chemical properties the two were considered indistinguishable, they were both called ‘gold’.
Gold and pyrite have different micro-structures but the naming convention is determined by the observed properties (nominal) which are not necessarily (and often not at all) the same as the micro properties (real). It is the particular micro structure of gold and pyrite that produce in us the impression of the colour gold and the mistaken perception that they are one and the same substance. If we had a super-duper micro scope and looked at the atoms that make up gold and pyrite, they wouldn’t be gold coloured–it’s their structure that produces in us the perception of the colour gold.
So, the goal of science should be to uncover the real essences of substances, not the nominal essences. As we come to know the real essences of substances we will not confuse natural kinds, like we did with gold and pyrite…since it was the discovery of their real essences (micro-structure) that allowed us to correctly identify the kind of thing they are.
Boyd fancies extending Locke’s idea of a distinction between nominal and real essences regarding (physical) substances to properties, relations, and magnitudes. We can see where he’s going with this. He’s going to leverage this difference to account for moral disagreements.
The strategy will be to say that moral disagreements arise because of something analogous to what’s happening with the gold and pyrite example. People are making judgments based on nominal essences of ‘good’ rather than the real essences. A science of morals will commit itself to discovering the real essences of moral properties.
Now, this all sounds fine and dandy but I anticipate some problems. First of all, Locke’s distinction between nominal and real essences applies to substances–physical things. On Locke’s model, properties are ideas caused in our minds by the powers of a substance’s micro-structure. The properties themselves are non-maleable and are not subject to revision–they are basic. It is through the properties that we identify nominal and real essence of substances.
For instance, the colour gold helps us identify gold (nominally) or the micro-structure of gold–i.e., the property of “number of protons and electrons”–helps to identify gold’s real essence. So, all properties are somehow tied to something physical; they inhere in something physical. But I don’t see how it is a property can have a micro-structure, and supposing it did, how a property’s micro-structure would be different from how we perceive it.
How could ‘yellow’ have a micro-structure? I’m not even sure this is intelligible. The only thing that would make sense is to say that the perception ‘yellow’ is a function of the light wavelengths that strike our visual apparatus. But that wavelength is a function of an objects surface properties. Again, the properties is tied to something physical, exactly what the realist hopes to avoid. In this analysis we end up talking about a physical object’s microstructure, we haven’t talked about the real essence of ‘yellow’. It seems yellow is yellow is yellow regardless of whether we are referring to nominal or real essence.
Homeostatic Property-Cluster Definitions (WTF?)
Whoa! Hold on a second. I know what you’re thinking. It’s probably not politically correct for me to use the term ‘homeostatic’. I understand your concern but I contend that since some of my friends are homeostatic, it’s ok. Sometimes I’ll even say it to their face and we have a good chuckle.
Lets forget about the big words for a second and go back to a major problem that moral realism has: if moral values are real and exist in the natural world, how do realists account divergent moral judgments? In order to preempt this criticism Boyd’s going to suggest that some natural kinds are defined by property clusters. But not just any kind of property cluster–homeostatic property clusters!
What the crap does this mean? Consider what we might mean when we refer to something as ‘cozy’. Are we referring to just one property, like when I say ‘large’? No, I’m referring to a cluster of properties, such as warmth, softness, comfort, etc…
Now, is there only one rigid combination of these properties by which we might call something ‘cozy’? Nope. One cozy house might have a different set of these properties than another cozy house. Also something completely different like a blanket can also be ‘cozy’. None of these considerations mean that there’s no such thing as a cozy house or blanket.
He goes on to argue that this type of reference is used in natural science. Consider the term ‘healthy’. This is a relatively scientific notion that involves a cluster of properties that needn’t be in the same quantities for every case that falls under the kind ‘healthy’.
Other examples used in the (harder) science of biology are the notions of species, class, phylum, etc…Biology relies on the “imperfectly shared and homeostatically related morphological, physiological, and behavioural features which characterize its members. Just because we can’t nail down those biological terms to something specific and immutable, doesn’t prevent people in lab coats from using them.
What Boyd is getting at with this idea of homeostatic property clusters is that there’s flexibility in how we define certain traits, or kinds. Obviously, he’s going to apply this idea to moral kinds, like ‘good’ and ‘baaaaaaaaaad’. For something to be good it can have varying degrees of a cluster of properties, it needn’t be rigidly defined by definite amounts of each member of a set of properties.
I think there is something to this idea but it’s hard to see why his approach isn’t evidence for a constructivist picture of how we define ‘good’ rather than a realist approach. For example, he gives a list of possible elements that, in varying combinations and amounts, could constitute what we reference by ‘good’. They are things that satisfy different human needs, such as physical and medical needs, social and psychological needs like the need for love and friendship, the need to engage in cooperative effort, the need to exercise control over one’s life, the need for intellectual and artistic appreciation and expression, the need for physical recreation, etc…
So, why do I say this sounds constructivist? Well, he begins by looking at facts about humans, then reasons from there to what properties might encompass moral goodness. If moral values exist in the natural world, then I’m not sure why we need to reason from human needs to figure them out.
Why shouldn’t we be able to detect them, independent of prior knowledge about what’s good for humans? If ‘good’ is an objective value, it shouldn’t matter how it relates to humans, that just might be a happy coincidence. And besides, what if actual natural absolute moral values in fact better accord with the needs of spider monkeys?
Furthermore, I think Boyd has to walk a fine line with his homeostatic cluster model because if he allows too much flexibility the kind he’s defining gets too vague and becomes useless.
Ok, so a few paragraphs back I laid out some of the human needs whose satisfaction relate to our notion of moral ‘good’. The next point is that there are many possible different social arrangements that could satisfy these needs. That is, there is a wide variety of psychological and social mechanisms such as political democracy, mutual respect, rules of courtesy, etc…that can contribute to the homeostasis of the different human goods. I can get jiggy with that.
From there, Boyd says that we can define moral goodness in terms of the cluster of goods that satisfy human needs combined with the homeostatic mechanisms which unify them. So, action, policies, character traits, etc…are good to the extent that they bring about the goods and reinforce the mechanisms that bring them about. I can get jiggy with that too, although I’m concerned that this is starting to sound like the liberal brainwashing by professors that Santorum warned me about.
This next part is a little harder to get jiggy with because it sounds a little too much like wishful thinking. The hypothesis is that “in actual practice, a concern for moral goodness can be a guide to action for the morally concerned” because the fact the that the different moral goods are “mutually reinforcing”. Aaand that they are mutually reinforcing will help mitigate conflict between them.
Ok, anyone who follows US–or any country for that matter–politics will note that “in actual practice” there is plenty of conflict when it comes to choosing between moral goods. Although I absolutely loath Rick Santorum, if I presented to him the list of human goods (satisfaction of medical, psychological, social needs, etc…) I doubt he’d deny any of the members of the list. He’ll probably also agree to the list of mechanisms: attitude of mutual respect, political democracy, egalitarian social relations, customs, etc…Whether he actually practices them is another matter, but I don’t think he’d deny any of them as having value.
So, the problem for the realists is that we have a fairly clear counter example to the idea that the concern for moral goodness, in actual practice, would mitigate conflicts between members of these lists. I see very little mitigation going on where the extreme right is concerned. Of course, the moral realist could simply refudiate this example by arguing that Santorum isn’t committed to this particular set of goods and mechanisms, but a different set. And they may be correct. But then we’re back to the problem of explaining divergent conceptions of a good that is supposedly objective.
There is another reply suggested by Boyd, and that relates back to the reciprocal relationship between advances in theory and method. As our theory of moral goodness approaches truth, our psychological and social mechanism for balancing demands for the cluster of goods will allow us to make better decisions. This, I think is entirely plausible.
We can explain away the Santorum counter-example as evidence that in our journey to discovering the moral truth, we are still at a point where our choice-making mechanisms are still imperfect. We aren’t close enough to the truth to accurately know how to reconcile tensions between satisfying different human needs in such a way that there can be a net gain. I guess I can dig that.
One more issue that moral realism is going to have comes back to is the problem of proving morality is real and a natural property rather than a construction we impose after the fact. To try to solve this Boyd will rely on the similarity of how we acquire moral knowledge and how we acquire scientific knowledge. In both cases knowledge will arise out of empirical investigation.
This is a clever line to account for barbaric moral codes throughout history. Ostensibly if moral facts exist in nature how is it that we had slavery for so long? The answer is that certain empirical facts only become accessible to us under the correct conditions. In the case of slavery, its injustice didn’t become apparent until limited democracies developed. In wasn’t until we had the structure of democracy that we could see the psychological and social human good in equal social and political relations. Once those facts about the good were observed it was a matter of time before they were logically extended to include everyone.
This seems plausible enough, but I don’t see why we can’t explain the same phenomena using a constructivist story. Were we discovering moral properties of actions and policies or were we constructing them? How do we know that once we discovered the empirical facts of the conditions under which humans thrive, we didn’t ex post facto ascribe to them the notion of ‘goodness’? (using Latin makes me feel smrt) Are we human or are we dancer?