Bliss Machines: is conscious experience a good in itself?

A hypothetical question suggests itself: If we could, on the other hand, increase the overall amount of pleasure and joy in the universe by flooding it with self-replicating and blissful postbiotic Ego Machines, should we do that?

The assumption that the first generations of artificial Ego Machines will resemble mentally retarded human infants and bring more pain, confusion, and suffering than pleasure, joy, or insight into the universe may be empirically false, for a number of reasons. Such machines conceivably might function better than we thought they would and might enjoy their existence to a much greater extent than we expected. Or, as the agents of mental evolution and the engineers of subjectivity, we could simply take care to make this assumption empirically false, constructing only those conscious systems that were either incapable of having phenomenal states such as suffering or that could enjoy existence to a higher degree than human beings do. Imagine we could ensure that such a machine’s positive states of consciousness outweighed its negative ones — that it experienced its existence as something eminently worth having. Let us call such a machine a Bliss Machine.

If we could colonize the physical universe with Bliss Machines, should we do it? If our new theory of consciousness eventually allowed us to turn ourselves from old-fashioned biological Ego Machines, burdened by the horrors of their biological history, into Bliss Machines — should we do it?

Probably not. There is more to an existence worth having, or a life worth living, than subjective experience. The ethics of multiplying artificial or postbiotic systems cannot be reduced to the question of how reality, or a system’s existence, would consciously appear to the system itself. Delusion can produce bliss. A terminally ill cancer patient on a high dose of morphine and mood-enhancing medications can have a very positive self-image, just as drug addicts may still be able to function in their final stages. Human beings have been trying to turn themselves from Ego Machines into Bliss Machines for centuries — pharmacologically or through adopting metaphysical belief systems and mind-altering practices. Why, in general, have they not succeeded?

In his book Anarchy, State, and Utopia, the late political philosopher Robert Nozick suggested the following thought experiment: You have the option of being hooked up to an “Experience Machine” that keeps you in a state of permanent happiness. Would you do it? Interestingly, Nozick found that most people would not opt to spend the rest of their lives hooked up to such a machine. The reason is that most of us do not value bliss as such, but want it grounded in truth, virtue, artistic achievement, or some sort of higher good. That is, we would want our bliss to be justified. We want to be not deluded Bliss Machines but conscious subjects who are happy for a reason, who consciously experience existence as something worth having. We want an extraordinary insight into reality, into moral value or beauty as objective facts. Nozick took this reaction to be a defeat of hedonism. He insisted that we would not want sheer happiness alone if there were no actual contact with a deeper reality — even though the subjective experience of it can in principle be simulated. That is why most of us, on second thought, would not want to flood the physical universe with blissed-out artificial Ego Machines — at least, not if these machines were in a constant state of self-deception. This leads to another issue: Everything we have learned about the transparency of phenomenal states clearly shows that “actual contact with reality” and “certainty” can be simulated too, and that nature has already done it in our brains by creating the Ego Tunnel. Just think about hallucinated agency or the phenomenon of false awakenings in dream research. Are we in a state of constant self-deception? If we are serious about our happiness, and if we don’t want it to be “sheer” hedonistic happiness, we must be absolutely certain that we are not systematically deceiving ourselves. Wouldn’t it be good if we had a new, empirically informed philosophy of mind and an ethically sensitive neuroscience of consciousness that could help us with that project?

I return to my earlier caveat — that we should refrain from doing anything that could increase the overall amount of suffering and confusion in the universe. I am not claiming as established fact that conscious experience of the human variety is something negative or is ultimately not in the interest of the experiential subject. I believe this is a perfectly meaningful but also an open question. I do claim that we should not create or trigger the evolution of artificial Ego Machines because we have nothing more to go on than the functional structure and example of our own phenomenal minds. Consequently, we are likely to reproduce not only a copy of our own psychological structure but also a suboptimal one. Again, this is ultimately a point about the ethics of risk-taking.

But let’s not evade the deeper question. Is there a case for phenomenological pessimism? The concept may be defined as the thesis that the variety of phenomenal experience generated by the human brain is not a treasure but a burden: Averaged over a lifetime, the balance between joy and suffering is weighted toward the latter in almost all of its bearers. From Buddha to Schopenhauer, there is a long philosophical tradition positing, essentially, that life is not worth living. I will not repeat the arguments of the pessimists here, but let me point out that one new way of looking at the physical universe and the evolution of consciousness is as an expanding ocean of suffering and confusion where previously there was none. Yes, it is true that conscious self-models first brought the experience of pleasure and joy into the physical universe — a universe where no such phenomena existed before. But it is also becoming evident that psychological evolution never optimized us for lasting happiness; on the contrary, it placed us on the hedonic treadmill. We are driven to seek pleasure and joy, to avoid pain and depression. The hedonic treadmill is the motor that nature invented to keep the organism running. We can recognize this structure in ourselves, but we will never be able to escape it. We are this structure.

In the evolution of nervous systems, both the number of individual conscious subjects and the depth of their experiential states (that is, the wealth and variety of sensory and emotional nuances in which subjects could suffer) have been growing continuously, and this process has not yet ended. Evolution as such is not a process to be glorified: It is blind, driven by chance and not by insight. It is merciless and sacrifices individuals. It invented the reward system in the brain; it invented positive and negative feelings to motivate our behavior; it placed us on a hedonic treadmill that constantly forces us to try to be as happy as possible — to feel good — without ever reaching a stable state. But as we can now clearly see, this process has not optimized our brains and minds toward happiness as such. Biological Ego Machines such as Homo sapiens are efficient and elegant, but many empirical data point to the fact that happiness was never an end in itself.

In fact, according to the naturalistic worldview, there are no ends. Strictly speaking, there are not even means — evolution just happened. Subjective preferences of course appeared, but the overall process certainly does not show respect for them in any way. Evolution is no respecter of suffering. If this is true, the logic of psychological evolution mandates concealment of the fact from the Ego Machine caught on the hedonic treadmill. It would be an advantage if insights into the structure of its own mind — insights of the type just sketched — were not reflected in its conscious self-model too strongly. From a traditional evolutionary perspective, philosophical pessimism is a maladaptation. But now things have changed: Science is starting to interfere with the natural mechanisms of repression; it is starting to shed light on this blind spot inside the Ego Machine.⁷

Truth may be at least as valuable as happiness. It is easy to imagine someone living a rather miserable life while at the same time making outstanding philosophical or scientific contributions. Such a person may be plagued by aches and pains, by loneliness and self-doubts, but his life certainly has value because of the contribution he makes to the growth of knowledge. If he, too, believes this, he may even find consciously experienced comfort in it. His happiness will thus be very different from the happiness of our artificial Bliss Machines or of the human subjects hooked up to Robert Nozick’s Experience Machine. Many will agree that this “epistemic” kind of happiness can outweigh a lot of unhappiness of the purely phenomenal type. The same may be said for artistic achievement or moral integrity as sources of happiness. If it makes any sense at all to speak about the value of human existence, we must concede that it depends on more than the conscious experience of happiness.

As long as such questions remain unanswered, we should refrain from trying to create artificial Ego Machines, and not only for ethical reasons. We cannot overlook the irreversibility of certain developments. Any postbiotic system that comes even close to attaining the properties of phenomenal selfhood — any system developing a reasonably robust first-person perspective — will be an autonomous agent. At a certain level of autonomy, we will have to accept these systems as persons in their own right and enter into a dialogue with them. Our criteria for what is an object of moral concern and what should be treated as a person will make it impossible for us simply to turn them off.

Дата добавления: 2015-10-31