I’d like to return to one of the favourite topics of Pea Soup for a bit, i.e., the Zombies. Couple of years ago we had great discussions about Zombies, well-being, and the notion of ‘good for’ (here, here, and here). We’ve even talked about whether it would be wrong to eat Zombies (here). I’m actually quite fond of the Zombies discussed in the philosophy of mind. For this reason, I’d like to put on the table a slightly more general question about what implications, if any, the conceivability arguments and their conclusions would have for metaethics.
By Zombies, I, of course, mean beings that are physically identical with humans but which do not have qualia. They have no conscious experiences and thus there is nothing it is ever like to be them. On occasion, one finds such beings conceivable. The classic conceivability argument, much developed by Chalmers, is based on them. Conceivability of Zombies is taken to be evidence for their metaphysical possibility. They would not be possible if our phenomenal concepts referred to anything physical, be it brain-states or behavioural dispositions. Therefore, physicalist views about phenomenal concepts would be in trouble and we would be pushed towards some kind of property-dualism. All of this is very familiar.
I’m not always sure about the argument and I know that there are ways in which physicalists can try to reply to it. But, let’s accept it at least for the sake of an argument. Now, if we can imagine individuals Zombies, it seems that we could imagine a whole Zombie world that would be physically identical with the actual world but in which no-one had any conscious experiences.
If we looked at the Zombie world, we couldn’t tell it apart from our world. Even the moral practices would be the same. The Zombies would talk about morality as we do and they would act in identical ways to us. The Zombies would occasionally help others and sometimes they would break their promises and kill one another. Initially, we would take some of these actions to be right and some wrong. Some of the Zombies would seem to be cruel and some of them would seem to be kind. We would also attribute identical moral beliefs to the Zombies.
However, when I think of the Zombie world as a world in which no-one would have any conscious experiences, I seem to get the intuition that this world would not contain any moral qualities. Some of you shared a similar intuition about nothing being good for the Zombies in any more robust sense than how things can be good for a tree.
I’m uncertain about what my intuition is based on. I don’t want to say that it is based on any strong implicit hedonistic commitments. It’s not only that the world lacks pleasure and pain. That is significant but there is more to it I think. In that world, no-one needs to experience being wronged in the many ways in which we can experience this. There is no ‘second-order evil. No-one needs to feel inferior when their legitimate expectations are violated. There are no experiences of the horror of being killed or of having expectations created by someone’s promises having been disappointed. In a similar way, no-one can experience the kindness of others either. The more I think about this, the more the moral properties of this world begin to disappear. It would just be an amoral world of automatons that was ticking along.
I’m not sure how robust this intuition is. But, here is another route to it. In films, when a bad guy is killed, occasionally one has some moral worries about this. Maybe the general good can justify such killings but at least the villains have a moral status that requires some sort of justifiable treatment. With Zombies it’s different. You don’t worry about their treatment. Even if they haven’t done anything, you don’t think that there even could be anything morally suspect in killing them. Maybe this is zombieism. But, if you have this intuition, then it should generalise to the zombie world in which Zombies do things to one another.
We then get to the question of metaethical implications of the Zombie world. First, I do find the initial intuition that the presence of conscious experiences is a necessary condition for moral properties quite interesting. If the conceivability argument goes through and we must resort to sui generis phenomenal properties, then some forms of ethical naturalism also begin to seem problematic. It couldn’t be that moral concepts referred to purely physical properties because the Zombie world would share them with ours.
We can grant that the Zombies have psychological states like desires and beliefs as functional states (in the same way as we do to thermostats and computers). Often by ‘natural’ we mean what is either physical or psychological. In that case, if the Zombie world lacks moral properties, then our moral terms couldn’t refer to purely naturalist properties either. Furthermore, interestingly, this would mean also that moral properties would not even supervene on natural properties. Our world and the Zombie world are identical when it comes to the naturalist properties but they do seem to differ in their moral properties.
Also, one starts to wonder that if there are good reasons to think in the mind case that there are properties that go beyond the purely natural, then there seems to be less pressure to reject sui generis moral properties merely on the grounds that they are natural. Of course in the mind case there is a temptation to include the sui generis phenomenal properties to an extended conception of the natural. If one can do this, then maybe one can do a similar move in the moral case.
There seems to be an interesting difference between the cases too. The property dualists often claim that, even though Zombies are metaphysically possible, they are not naturally possible. There is a basic ‘law of nature’ because of which supervenience holds locally. In our world and in the other worlds in which this law holds, if two beings share the same physical properties they must share the same phenomenal properties. However, in the moral case, local supervenience, the ban of mixed worlds, seems to be too weak. In moral case, it is more difficult to conceive that an action that is wrong in our world could be not wrong in some other world that shared all the natural and phenomenal properties.