2004 Haldane Award Paper: The Posthuman Condition by Kip Werking
The Posthuman Condition
Kip Werking
The University Of Texas At Austin
Homepage: http://www.ece.utexas.com/~werking
“There is no evil I have to accept because ‘there’s nothing I can do about it’. There is no abused child, no oppressed peasant, no starving beggar, no crack-addicted infant, no cancer patient, literally no one that I cannot look squarely in the eye. I’m working to save everybody, heal the planet, solve all the problems of the world.”
— Eliezer Yudkowsky, Singularitarian Principles 1.0
“How could we drink up the sea? Who gave us the sponge to wipe away the entire horizon? What were we doing when we unchained this earth from its sun? Whither is it moving now? Whither are we moving? Away from all suns? Are we not plunging continually? Backward, sideward, forward, in all directions? Is there any up or down left? Are we not straying as if through an infinite nothing? Do we not feel the breath of empty space?”
— Nietzsche, Thus Spoke Zarathustra
I. INTRODUCTION
This article is my effort to identify the next two anthropocentric beliefs to die. One would not expect Copernicus’s defeat of geocentricism and Darwin’s defeat of special Creation to be the last comforting illusions that science will expose. There is an important difference, however, between the third and fourth anthropocentric conceits that I describe. Whereas the transhumanist community has largely abandoned, to their advantage, the third conceit, I will argue that even transhumanists have ignored the fourth.
My attempt at exposing the fourth conceit shows what transhumanism cannot do. In particular, I will show that while future technologies may remedy part of the human condition, they cannot remedy a remaining part, which I will call the posthuman condition. My article is, in this respect, similar to the critiques by Dreyfus, Searle, and Penrose, which claim to demonstrate what artificial intelligence cannot, even in principle, do. My critique of transhumanism is relevantly different from those, however, because my arguments attempt to undermine, rather than erect, distinctions between human beings and the world.
While preparing this article I considered other possible anthropocentric conceits. One notable possibility, which the transhumanist community has perhaps not abandoned, is the threat that future technologies pose to personal identity. For example, James Hughes considers this possibility in a recent column at the “Betterhumans” webzine. Hughes described the threats that future technologies pose to our sense of personhood. He came to believe, as Hume and Buddha did, that there is no self.
II. THE THIRD ANTHROPOCENTRIC CONCEIT
“The third blow, which is psychological in nature, is probably the most wounding.”
—
Freud
The third anthropocentric conceit is the belief that non-defective human intelligence is relatively unique, qualitative, and optimal. In contrast, I agree with most transhumanists that human intelligence is but an arbitrary mark upon the spectrum of intelligence, that on this spectrum we differ from other intelligent minds in quantity but not quality, and that human intelligence is far from optimal. I feel confident that this distinction is an anthropocentric conceit because I identified myself as a transhumanist well before abandoning it. Even after reading the popular introduction to transhumanist ideas, The Age of Spiritual Machines by Raymond Kurzweil (perhaps because Kurzweil himself had yet to fully abandon the conceit), I failed to appreciate the drastic consequences of superintelligence.
Freud famously claimed to have identified the third conceit. He was correct to draw attention from cosmology and biology to psychology. Freud, however, emphasized the suboptimal nature of man’s mind rather than the optimal nature of future minds. As stupid and confused as Freudian agents were, they were nevertheless the supreme intelligence in the universe. To the best of my knowledge, after abandoning the unpromising research in nineteenth century neurology, Freud settled for talking therapies and never returned to the possibility of using technology to correct and enhance the human brain. So even Freud himself failed to abandon the third conceit.
One approach to understanding the non-unique, quantitative and suboptimal nature of human intelligence is to contrast human intelligence with other characteristics. Human beings have always trembled before the threat of a bear. We have always lost the race to the cheetah. Our vision has never been as sharp as an eagle’s. We can, however, shoot the bear, drive past the cheetah, and use binoculars. Future technologies have the potential to implement these improvements within human biology. But against whom or what can we unfavorably compare the human brain? There is nothing like the human brain within the known universe. Furthermore, with what tool can we complement our intelligence? Giving a person a calculator to improve their intelligence is like giving a person roller skates, instead of a car, to improve their speed. Consequently, while we would expect human minds to have well developed concepts for superior beings with respect to attributes such as speed or strength, we have no such concepts for things that are smarter than ourselves. We fail to appreciate the finitude of our minds because nothing ever competes with them.
Intelligence is also unique among human qualities because intelligence enhancement promises to create a positive feedback loop. Building muscles does not help you build muscle faster. Running faster does not accelerate the improvement. Amplifying intelligence, however, does increase the rate of intelligence amplification. Smart brains design even smarter brains faster than dumb brains. More importantly, this positive feedback loop has yet to even begin. Human beings are not crawling along at the very beginning of this loop. Rather, our species has remained relatively stagnant for tens of thousands of years because we are powerless to alter our own biology. Since the birth of our species human beings have only been able to work with the small amount of intelligence that nature gave us. Only once an intelligence begins amplifying itself will the chain reaction finally begin. So not only is human intelligence quantitative and suboptimal but inherently limited — even frustrated. One can see how artificial intelligence researchers are playing with cosmic matches.
The third conceit will surely, as long as civilization progresses, confront our species. Strong artificial intelligence alone could be the existence proof of superintelligence. We can no longer afford to ignore the quantitative and suboptimal nature of human brains. Our species will be forced to answer questions about how to design and control the first minds that are more intelligent than our own. How will we ensure that these persons are moral? How will we ensure that these programs are not riddled with errors? If we are to survive the twenty-first century, we must answer these questions.
While answering these questions, we must acknowledge that our brains will be as inferior to future minds as the brains of mice are to our own. This is a not a trivial acknowledgement. Truly appreciating the inferiority of our intelligence — the gross limits that our wetware places upon the speed and content of our thoughts — requires an almost religious humility. This explains Freud’s claim that the third is probably the most wounding of anthropocentric conceits. The prospect of superintelligence renders human history into a comedy.
Although Freud did not perhaps fully appreciate the third conceit, Nietzsche did:
“Zarathustra, however, looked at the people and wondered. Then he spoke thus:
Man is a rope stretched between the animal and the Superman — a rope over an abyss.
A dangerous crossing, a dangerous wayfaring, a dangerous looking-back, a dangerous trembling and halting.
What is great in man is that he is a bridge and not a goal.”
III. THE FOURTH ANTHROPOCENTRIC CONCEIT
One might very well expect that the third anthropocentric conceit would be the last. My thesis is that this cannot be the case. Indeed, I intend to show that a fourth anthropocentric conceit awaits any posthuman. Posthumans should inevitably confront the fourth conceit and the consequences of abandoning its comfort are nontrivial. Although every problem that I describe in this section is two and a half millennia old, few have emphasized how future technologies will aggravate them. Many alarmists, such as Bill Joy, are quick to note the dangers that such technologies pose to our lives and health. These technologies, however, also pose more subtle threats.
The fourth anthropocentric conceit is not as well defined as the others. The definition that I will use for this essay is “the naive answers to the collective hard problems of philosophy”. These problems are simple questions such as “what exists?”, “what should I do?”, and “am I free to do so?” The naive answers would be similar to “the exterior world and I obviously exist”, “I should maximize happiness in the world”, and “Of course I am free to do so.” I do not wish, however, to limit the fourth conceit to these three questions at the expense of other questions that may also be relevant.
The fourth anthropocentric conceit is quite different than the third because the transhumanist community largely still clings to these comforting illusions. Most transhumanist authors, to the best of my knowledge, write about how to move beyond the third conceit, given the fourth. Assuming that the exterior world exists, that happiness should be maximized, and that free will exists, how can we freely use technology to change the world such that happiness is maximized? The prospect of nihilism, hard determinism, or immaterialism never seems to threaten or disturb them. For Singularity enthusiasts, the assumption that the Singularity will solve all of our problems, including philosophical ones, is omnipresent. Indeed, this essay is largely inspired by the tension I find between Singularity activist Eliezer Yudkowsky’s cursory treatment of these questions in his FAQ about the Meaning of Life and the seriousness with which he considers the threat of “philosophical crises” in his Creating Friendly AI.
This essay is also largely inspired by another transhumanist who, unlike the contemporary community, did appreciate the fourth anthropocentric conceit. Friedrich Nietzsche denied the truth of any moral facts, asserted that the real world is a lie, and denied that people possess free will. In stark contrast to Nietzsche is another philosopher who, although she did not consider herself a transhumanist, has strongly influenced the transhumanist community. This influence can be seen in the Principles of Extropy and the writings of Robert Ettinger. The philosophies of Ayn Rand and Friedrich Nietzsche, in the context of transhumanism, create a fascinating dialogue. Whereas many transhumanists subscribe to Rand’s philosophy of Objectivism, Nietzsche’s contribution has been comparatively minor. While the transhumanist community shares Nietzsche’s enthusiasm for the übermensch, a more likely source for this inspiration is Ettinger’s Man Into Superman. Nor would any transhumanist, despite the fears of Luddites such as Francis Fukuyama, claim that ordinary men ought to be the slaves of supermen.
There is a subtle tension here. This tension is further shown by Rand’s fierce commitment to the fourth anthropocentric conceit. She passionately argued that each of the naive answers to the hard problems of philosophy is correct. For example, Rand was once asked if she could present the essence of Objectivism while standing on one foot. To the questions of metaphysics and ethics, she replied without hesitation “objective reality” and “self-interest”. Likewise, Rand insisted that man possesses free will:
“Because man has free will, no human choice — and no phenomenon which is a product of human choice — is metaphysically necessary. In regard to any man-made fact, it is valid to claim that man has chosen thus, but it was not inherent in the nature of existence for him to have done so: he could have done otherwise.”
Rand’s uncritical belief in these comforting ideas largely explain why academic philosophers, unlike transhumanists, have treasured Nietzsche but ignored her. Yet Nietzsche’s influence upon Rand is patently obvious. Her love of great individuals and her hatred for dualism testify to the impact Nietzsche’s philosophy had upon her. So Rand adopted Nietzsche’s philosophy but modified it. She reinserted what Nietzsche had removed: the fourth conceit. Nietzsche purchased transhumanism at a price Rand was unwilling to pay. This tension brings into question the consistency of transhumanists today. They adopt both Nietzsche’s transhumanism and Rand’s fourth conceit.
III.i ETHICS
“Morality is merely an interpretation of certain phenomena — more precisely, a misinterpretation.”
— Nietzsche,
Twilight of the Idols
To demonstrate the threat of nihilism one can consider the following dilemma:
The diagram shows how neo-Darwinism has exposed an inconsistency between human intuition and knowledge about values. Throughout history humans have tended to regard happiness as the supergoal of life. If one inspects human behavior, human beings pursue happiness and avoid misery all or most of the time. Some variant of psychological egoism seems to rule human behavior. Furthermore, happiness does not feel to us to be a means to another goal. Rather, we wish to happy for the sake of happiness alone.
The neo-Darwinian synthesis, however, confronts us with a reality that disagrees with our perception about morality. In this century science has shown that happiness, which had the impression of being a supergoal, is only a subgoal to another end: the propagation of genes. Humans pursue happiness because happiness inducing behaviors, such as copulating and eating, are adaptive. The worthlessness of a supergoal, however, renders its subgoals worthless too. So human beings face an inconsistency between two paradigms which they wish to reconcile.
Two alternatives present themselves. In response A, human beings recognize that happiness is a subgoal of gene propogation and choose to henceforth value happiness only to the extent that happiness helps propagate genes. In response B, human beings abandon the apparent purpose of their reward mechanisms, propagating genes, and value happiness for its own sake. I will argue that mankind is split between these two alternatives and that neither solution is satisfactory. We are torn between utility and fitness.
Consider response A. I doubt that any person has ever truly considered adopting response A. However, from a distanced, cerebral perspective, I think that response A is at least not obviously wrong. So response A deserves philosophical attention. Response A is motivated by our commitment to a simple principle. This is the principle that if something was only created to serve as a means to an end then this something only has value to the extent that it furthers this end. If we could obtain the end without the means, the means would become worthless.
Consider this analogy. Imagine a world that is just like this world, except for one small difference. Suppose that in this world, human beings discovered hammers in the earth. Suppose further that, for whatever reason, these human beings valued these hammers. Indeed, they treasured them. Academic philosopher wrote long and complicated articles defending the value of hammers in professional journals. They loved these hammers — even though nails do not exist. Now suppose that an alien race, perhaps our posthuman selves, arrived in a saucer and showed these primitive human beings nails. Suppose that we played for them a humiliating documentary showing how we once made hammers for the sole purpose of hammering nails. How should these primitive human beings feel about their behavior? I would expect them to feel quite embarrassed. But, according to neo-Darwinism, this is exactly the dilemma that our own species faces. We value behaviors that induce happiness but we have no particular commitment to propagating genes — the only reason these behaviors induce happiness. Our intuitions are so strong against the purely intellectual response A, however, that nobody would adopt it.
Alternatively, consider response B. Please note that the ethical system I challenge is utilitarianism. While some will note that utilitarianism is an easy target, transhumanist sympathies belong more to utilitarianism than any Kantian system. Response B is much more attractive and deserves serious consideration. However, response B must survive inspection of its limiting cases. I will show that future technologies involving direct manipulation of the brain’s reward mechanism (BRM) strongly suggest that response B cannot be satisfactory. The brain’s mesolimbic dopamine system (a set of nerve cells that originate in the central tegmental area) is a strong candidate for the BRM but perhaps not the entire explanation. This essay will not inspect the details of the BRM but simply assume that one exists inside of the brain.
Consider the case of wireheading, as described in the transhumanist classic The Hedonistic Imperative [1], in which people “subvert” the happiness treadmill and short circuit the reward mechanisms in their brains. Wireheading surely does not inspire within us feelings of admiration for such moral behavior, but rather, if anything, feelings of anxiety. Yet wireheading is an exemplar of morality according to the utilitarian response B. People who wirehead themselves and others maximize happiness better than those who do not. Transhumanism threatens our utilitarian sensibilities further in the limiting case of “universal orgasm.” If morality requires the maximization of happiness, the final success of moral agents will be the rendering of all matter in the universe similar to that in the reward centers of happy brains. A universe brimming with cheery gray matter, however, does not strike us as a particularly moral and meaningful goal to pursue.
Why does response B fail in the limiting case? Again, neo-Darwinism illuminates the dilemma. Natural selection built the BRM within the context of the ancestral environment and under metabolic constraints. The BRM is not an infallible guide to gene propagation and fails to distinguish between adaptive and nonadaptive behaviors when it can afford to do so. Moreover, the further removed humans are from the ancestral environment the more irrelevant the BRM becomes. Today, the BRM would encourage adolescent boys to sleep with a Playboy Playmate whether or not she is fertile. Likewise, the BRM cannot distinguish between champagne that is poisoned and champagne that is not. The BRM, however, can afford these mistakes. These situations at least resemble adaptive behaviors. Enough adolescent boys sleep with fertile women and enough people drink non-poisoned champagne to justify the BRM’s laziness. In the future, however, the BRM will make mistakes that it cannot afford. In particular, because in the ancestral environment the BRM never needed to distinguish between stimulation by wireheading and stimulation by adaptive behaviors, the BRM will strongly encourage wireheading in the future. Wireheading, however, subverts the entire purpose and intent of the BRM, which is to propagate genes. So BRM manipulation threatens to magnify trivial mistakes into grave errors. Human behavior will become, in the terminology of Stephen Jay Gould, an exaptation.
One interesting fact to note is that weak technologies which already attempt to subvert the BRM are often regarded as immoral. Pornography and birth control have long been regarded as immoral by religious groups. These technologies allow human beings to stimulate the erotic reward mechanisms in their brain without the prospect of gene propagation. Indeed, masturbation, which ameliorates the suffering of adolescent boys, has also been considered immoral and many groups hold that opinion today. Likewise, drugs, which more closely resemble wireheading technologies, are also widely condemned. Indeed, more pleasurable drugs, such as heroin, are illegal around the world, whereas less pleasurable drugs, such as nicotine and caffeine, are not. This is so partly because drugs such as heroin can be more dangerous. However, even considering the hypothetical case of a drug as powerful as heroin but perfectly safe, the consensus that using such a drug is immoral would remain. Humans would feel anxious about such a safe drug because it nevertheless subverts the intent of the BRM. People who take drugs feel rewarded for doing nothing and this strikes us as wrong.
So we can regard utilitarianism as a heuristic that works within a narrow, local scale but fails at a wider, global scale. Most everyone will agree that pursuing happiness for ourselves and others in the near future without manipulating the BRM is relatively moral. However, once we adopt the wider, global scale and consider the limiting cases involving BRM manipulation, response B fails. If both responses A and B fail, however, the ethical dilemma is insoluble.
To further illustrate the threat that future technologies pose, consider the following diagram:
In this diagram CMB stands for Conventional Moral Behavior, X1 and X2 are arbitrary events and X1 causes the respective species to feel happiness. CMB does not necessarily signify the sort of courageous acts that ethicists discuss in moral dilemmas, such as saving drowning strangers, but rather ordinary behavior that people take for granted as moral. That a person wakes up, eats breakfast, goes to work, helps children with homework, has sex with the spouse, and goes to sleep again does not immediately strike us as moral. And yet a rational agent who values being a destitute, anorexic, insomniac, sexually abstinent child murderer is a moral monster. We might say that ordinary people are so moral that we forget to mention it.
I will show that the utilitarian arguments that ethicists use to justify human behavior would just as well justify the behavior of HS2, HS3, and HS4. Yet the behavior of these others is intuitively wrong. First we might consider Homo Sapiens 2. HS2 is exactly like HS1 except that instead of gene propagation, CMB is conducive to some arbitrary event X2. We might suppose that X2 is the collection of a gold coin within some bank account. X2s engage in sexual intercourse but possess no gametes. Newborn X2s happen to grow out of the ground at about the death rate. Every time HS2s would behave as HS1s would to propagate genes, by raising children and grandchildren, one gold coin is deposited. Our first impression is that the change from HS1 to HS2 is trivial and non-threatening. That CMB accumulates gold coins instead of propagates genes does not seem to undermine our intuitions about right and wrong. The saints of HS1 would be saints in HS2.
Next, consider HS3. HS3 is exactly like HS1 except that HS3 do not need to exhibit CMB to propagate genes. Although CMB or similar behavior is perhaps the only method natural selection could discover to propagate genes, one can conjure up examples in which almost any behavior propagates genes. Suppose, for example, that a demon controller has compiled a genetic database of every HS3. Suppose further that the demon outfitted every HS3 with a sensor that detected X1 events and sent a signal to a cloning factory that created a long line of successful descendents for that person. HS3 becomes interesting when we consider the spectrum of possibilities for X1. X1 might be CMB or similar behavior. Alternatively, X1 could be positively maladaptive behaviors at the local scale. For example, the BRM of HS3s might be such that HS3s feel rewarded not for CMB but for being destitute, anorexic, insomniac, sexually abstinent child murderers. HS3s might delight in setting themselves on fire and laugh while their families burn.
At first blush HS3 might strike as extremely immoral people. We must remember, however, that their psychological reward and genetic propogation systems are quite different from our own. We are observing HS3s through the tinted spectacles of our own brain reward mechanism. Although being destitute, starving ourselves, killing children, and setting ourselves on fire would not at all contribute to our utility, it surely contributes to their own. Indeed, utilitarianism would require that HS3s live in poverty, starve themselves, butcher their children, and set their families on fire. According to utilitarianism, the better and more frequently HS3s do so, the more saintly they become. Moreover, if any skeptic were to challenge this conclusion by claiming that happiness is not an end in and of itself, but is only a subgoal to some supergoal, neo-Darwinism has identified this supergoal — gene propagation — and we can remind this person that while the saints of HS3 are kindly respecting their children’s wish to be set ablaze, cloning factories are creating entire new crops of HS3 to colonize the cosmos. The universe will soon be brimming with HS3 DNA. So, although the behavior of HS3 strikes us as profoundly immoral, from a utilitarian perspective their actions are perfectly moral.
If the changes from HS1 to HS2 and HS3 did not invalidate the morality of each species’ respective actions, we must also conclude that a combination of these changes leaves morality untouched. HS4, however, is a species of people who exhibit arbitrary behavior to feel happy and this behavior is conducive to an arbitrary end. Instead of child murdering HS3s sending a signal to the cloning factory, these saints send signals to the bank account which deposits a gold coin. X2 need not be a gold coin deposit. It may just as well be a dirt deposit. The conclusion is that morality, the idea of what actions people ought to take, is itself arbitrary. To support this conclusion, one need only ask how a member of one species could distinguish between others that are moral and others that are not. Each species has its own BRM and none, including our own, would seem to be privileged over the rest. Furthermore, the ethical systems of each species are threatened by the prospect of BRM manipulation and universal orgasm.
From these examples one can generalize about certain ethical crisis that any such species must confront. Our own species has already noted the first crisis, which is the discovery of the supergoal and the realization that happiness is a subgoal. In particular, since Darwin human beings have known that happiness inducing behaviors were selected during evolution not because they promote human happiness but rather because they promote genes. Curiously enough, our species has duly noted this discovery and not felt much anxiety at all. We have yet, however, to confront the second crisis, which is the manipulation of the BRM. Although weak technologies such as illicit drugs already exist, human beings are not yet faced with the prospect of divorcing happiness from adaptive behavior. If technology continues to progress unchecked, we will soon have the ability to feel rewarded for any behavior. Which behaviors will we choose?
One possible objection to the previous argument is that human nature is not necessarily adaptive. The correlation between rewarding behaviors and adaptive behaviors is not perfect. My reply to this objection is not to deny that human beings can exhibit maladaptive behavior. Indeed, human being often delight in being positively maladaptive. Rather, such behavior only signifies noise. Neutral or maladaptive behaviors are not immoral so much as they are confused. Human beings want to found their moral norms upon good reasons, whatever those may be. Noise, however, has no reason. If adaptive behaviors cannot be the source of morality, surely noise can do no better.
Consider the possibilities available to any species that confronts the second crisis. Such a species could alter their BRM to reward them for maladaptive, rather than adaptive, behavior. Our own species could alter our BRM to be like that of child murdering HS3s and HS3s could alter their BRM to be like that of ourselves. HS3 could begin to resemble saints, from our perspective, but at the cost of their supergoal. Alternatively, such a species could fine tune the BRM so that the correlation between rewarding and adaptive behavior approaches one. The adolescent boys who would so readily sleep with an infertile Playboy Playmate now would then find her to be as repulsive as the sickly ogres of today. HS3 could alter their BRM to reward the successful running of cloning factories. So HS3s might abandon their self destructive pyromania and feel delight at raising the next crop of HS3. The possibilities available to species in the second crisis are endless.
Another objection to the previous argument is that objective morality has never existed and no sophisticated, intelligent person claims otherwise. On this view, the previous argument attacked a straw man. My reply to this objection is that these persons admit too much. Indeed, except for a semantic quibble, I feel that I am in complete agreement with the moral relativists. The semantic quibble concerns what the word “moral” signifies. I have used the word “moral” to denote what human beings ought to do. The word “moral” as used by the moral relativists, however, denotes something much closer to what human beings feel they ought to do. I think moral relativists would resist the naturalistic fallacy of arguing that a person ought to do something simply because she or he feels that a person ought to do so. We hesitate to found our moral norms upon things as capricious and uncontrollable as feelings. Moral relativism has shifted the focus from normative to descriptive claims. The semantic ambiguity need not undermine my argument against moral realism, however. Moral relatives and I largely agree about the respective concepts that such ambiguous language denotes.
If, however, moral relativists and I largely agree, what are the consequences? I have shown that human beings do not swear allegiance to either end of the ethical spectrum between subgoal and supergoal. Indeed, although we have the power to know how we feel people ought to act, we can never know how people really ought to act. We can never know, in other words, whether or not our intuitions are correct. The consequence of this discovery is existential paralysis. This paralysis, however, is not simple inaction. The person who refrains from taking any action nevertheless does choose the action to take, namely inaction. Rather, the consequence of moral non-realism is total paralysis. Those agents who suffer from hard paralysis cannot even, in the midst of their doubts, commit to inaction. Their actions, if any, can only signify noise. Hard paralysis, from which the current inability to alter our BRM protects us, awaits posthumans. Morality will die as paradise is born.
III.ii FREEDOM
“The causa sui is the best self-contradiction that has been conceived so far, it is a sort of rape and perversion of logic; but the extravagant pride of man has managed to entangle itself profoundly and frightfully with just this nonsense. The desire for “freedom of the will” in the superlative metaphysical sense, which still holds sway, unfortunately, in the minds of the half-educated; the desire to bear the entire and ultimate responsibility for one’s actions oneself, and to absolve God, the world, ancestors, chance, and society involves nothing less than to be precisely this causa sui and, with more than Munchhausen’s audacity, to pull oneself up into existence by the hair, out of the swamps of nothingness.”
— Nietzsche, Beyond Good and Evil
Just as modifying the brain reward mechanism threatens to undermine our sense of morality, so too do other future technologies threaten to undermine our sense of freedom. Following Robert Kane, I will define free will as the ability to be the creator and sustainer of one’s own ends or purposes. Free will is a necessary condition of moral responsibility and therefore, if people do not possess free will, they cannot be morally responsible for their actions. According to one view, morality is a hereditary or environmental program that causes an agent to do certain behaviors such as propagate genes. Although this position, which has been labeled hard determinism, was popular among intellectuals throughout history, including Spinoza, Nietzsche, Darwin, and Einstein, today those who deny the existence of free will are a minority. Derk Pereboom and Galen Strawson are two contemporary philosophers noted for their denial that free will exists.
The view that free will is incompatible with determinism but that determinism is false and free will exists is labeled libertarianism. Today several philosophers defend libertarianism with very sophisticated theories which appeal to quantum mechanics. For the purposes of this essay, however, I will assume that Hume was correct to note that any indeterminism within the universe cannot help the libertarian.
One future technology that threatens to undermine our sense of freedom is intelligence enhancement. To see how such enhancement will do violence to belief in free will, consider this telling quote from On Free Will by Albert Einstein:
“If the moon, in the act of completing its eternal way around the earth, were gifted with self-consciousness, it would feel thoroughly convinced that it was traveling its way of its own accord on the strength of a resolution taken once and for all. So would a Being, endowed with higher insight and more perfect intelligence, watching man and his doings, smile about man’s illusion that he was acting according to his own free will.” [italics added]
To those who would assert that free will exists, Einstein’s remark smacks of arrogance. He seems to beg the question by assuming that more intelligent persons would agree with him. There is, however, a very good reason for thinking that such mentally enhanced beings would agree with him. This reason is that intelligence enhancement would allow these beings to better predict human behavior. This premise is uncontroversial. Yet predictability strongly offends against our sense of human freedom.
Suppose that you received a letter in the mail today. You open the letter and discover another, smaller letter. This note was self addressed and mailed two centuries ago. You open the note and read the contents which are a precise description of your activities today. Given the premise of intelligence amplification, we live in a world in which such letters can arrive in our mailboxes. Wiser beings could have predicted whether or not you would fall in love — or commit a crime — two hundred years before you were born. A simple letter can abolish our fragile sense of free will. This is so because the decision seems to have been made before you were born. If this were not the case, how could the more intelligent being have predicted your behavior? The simple fact that the being was more intelligence could not suffice. The being would have required evidence. This evidence was written, before you were born, into the fabric of the universe. You are not the author, you are the message.
Note that Einstein had no interest in transhumanism that would explain his focus upon intelligence amplification. Rather, his interest is in the fourth anthropocentric conceit. The fourth conceit led him to consider a future technology, intelligence amplification, which strengthened his case against human freedom. Intelligence amplification just happens to also be one of those future technologies which transhumanists anticipate.
Other future technologies which threaten to undermine our sense of freedom are alternatives to punishment. Although academic philosophers tend to disregard the importance of these technologies today, the psychologist Burrhus Skinner drew attention to them. Skinner described his idea of a utopian community in Walden Two. Punishment is never used in this community. Rather, people modify the behavior of others through therapy and behavior engineering. Crime, on this view, is only a symptom of illness. In principle, these illnesses can be cured without punishment. Today, the best treatment we have for crime is punishment. If we did not punish criminals, there is no alternative which could protect society. The idea that people possess free will, however, and therefore deserve punishment for crime, rests upon this present inability. In the distant future, alternatives to punishment might be available. Perhaps one alternative will be as simple as a pill. When a person commits a crime in the future, society and philosophy will be faced with a choice between two responses. According to the orthodox response, the criminal was sane, freely committed the crime, and consequently deserves to suffer. Surely, this response would continue to work. But according to the alternative response, we would give the criminal a pill and therefore reduce or eliminate the chance of future crime. Given the transhumanist hypothesis that future technologies will radically improve our understanding and control of the brain, this response would also work.
An important distinction exists between the two options, however. Since the birth of our species, we have relied upon the orthodox response and therefore caused tremendous suffering. The alternative solution has no such cost. To maintain the orthodox response despite the availability of such an alternative would be cruelty.
Against this new response to crime, one might object that we are violating the criminal’s autonomy. We are no longer regarding the person as a person but rather as some meat machine to be modified at society’s leisure. The defender of this new response, however, can note that the criminal has already violated the autonomy of another by committing a crime. He has therefore sacrificed a proportional measure of autonomy and this violation is required to restore order. A better reply, however, is that punishment already violates the criminal’s autonomy. Indeed, punishment already alters the criminal’s brain. This is the reason why we punish criminals. If punishment was motivated solely by retribution, and had no deterrent effect, only the cruel would continue to make criminals suffer. The question is not whether or not we should regard a person as a person, but whether or not we should restrict our manipulation of others’ brains, against their will, to that involving the archaic interface of sense perception or, instead, progress into direct manipulation that avoids the cost in suffering.
Note that the future technology of behavior engineering will have its own costs too. Sadists who have a vested interest in the orthodox response to crime, punishment, will have to forego that pleasure. This is a change that transhumanists should welcome. A more disturbing cost, however, is the belief that people are free and responsible for their actions. In our ordinary lives we feel that criminals are not ill. In the future we must note that some people go ill and other do not. Yet we do not feel that illness is something that a person can control in the way that a person can control whether or not to commit a crime. This illusion of control must go.
The future technology of intelligence amplification also compliments this idea that criminals need therapy and not punishment. If we can predict the behavior of others, we can also predict their crimes. The ability to predict crime, however, also helps to prevent crime. Law enforcement officials, instead of only being able to implement behavior modification after a crime has been committed, would be able to do so before the crime ever takes place. This suggests a future world in which not only punishment, but also crime, is absent. In a world without punishment, crime, or responsibility, the antiquated idea of human freedom will become obsolete.
Another perspective that threatens to undermine our belief in free will is the global view of HS4. The focus upon the BRM shows how a controller could manipulate people. Indeed, we do not regard the members of HS4 as free at all. They appear to be slaves of their appetites who unwittingly pursue both base and more sophisticated rewards offered by the BRM. By choosing which behaviors X2 the BRM rewards, the controller, rather than HS4, is free and responsible for their actions. But according to the previous argument, we cannot distinguish between HS4 and ourselves with respect to morality.
To support these claims with evidence and add urgency to the dilemma, consider this famous trial. In 1924 two young men, Richard Loeb and Nathan Leopold, attempted to commit the perfect murder. While Ayn Rand studied Nietzche at the University of Leningrad, Leopold studied Nietzsche at the University of Chicago. He came to agree with Nietzsche that moral codes do not apply to supermen. Loepold considered Loeb such a superman. After killing a boy named Bobby Franks, the two were caught and Clarence Darrow defended them. He admitted that the two did kill the victim, but argued that they did not possess free will. Leopold and Loeb were victims just as much as the boy they killed. Darrow succeeded in bringing tears to the judge’s eyes and saving the boys’ lives. Thus Nietzsche’s argument against moral facts inspired the murderers and his argument against free will vindicated them.
I am perfectly willing to admit that Clarence Darrow might have argued for hard determinism to save the life of clients whether or not he believed they were morally responsible. My understanding, however, is that he believed they were not. Nor is Darrow’s argument particularly novel — the ancient Greeks appreciated the problem of incompatibilism two millennia ago. Furthermore, Darrow’s argument is confused. If Darrow was willing to deny the death penalty, why not deny them any penalty at all? The boys’ lack of free will would only argue against retributive punishment and not punishment intended to deter. Many have claimed that the death penalty does have such a deterrent effect. Darrow did not address this further point. Finally, to my knowledge no defense attorney has repeated Darrow’s appeal to hard determinism and the early twentieth century is rather removed from contemporary transhumanism. Nevertheless, transhumanism was born during the early twentieth century in the thoughts of Haldane and Schrödinger. Darrow posited a moral revolution no less astounding than the cosmic revolution of Copernicus:
“No one knows what will be the fate of the child he gets or the child she bears; the fate of the child is the last thing they consider. This weary old world goes on, begetting, with birth and with living and with death; and all of it is blind from the beginning to the end. I do not know what it was that made these boys do this mad act, but I do know there is a reason for it. I know they did not beget themselves. I know that any one of an infinite number of causes reaching back to the beginning might be working out in these boys’ minds, whom you are asked to hang in malice and in hatred and injustice, because someone in the past has sinned against them.”
Posthumans will be victims of fate as much as Leopold and Loeb ever were. As Einstein predicted, when posthumans amplify their intelligence they will discover how futile attempts to deviate from their original programming will be. Self-modifying programs are still programs. The transhumanist community must eventually return and confront Darrow’s challenge. At Cambridge on May 21, 1924—just sixteen months before Leopold and Loeb murdered Bobby Franks with a chisel —John Burdon Sanderson Haldane presented to the Heretics society his paper titled “Daedalus, or, Science and the Future.”
III.iii REALITY
“The apparant world is the only one: the real world is merely a lie.”
— Nietzsche, Twilight of the
Idols
A third problem I wish to explore in this essay is the question of external reality. Perhaps the first description of this problem is Plato’s allegory of the cave. In this allegory, Plato considers the hypothetical world in which human beings are locked inside of a cave and believe shadows to be real things. The founder of modern philosophy, Descartes, described the same problem by imagining an evil demon who deceives him. Descartes asked how a person can distinguish between the real world and the dream world. Only a few years ago, the character Morpheus from the Hollywood film The Matrix, inspired by Plato and Descartes, asked the same question. The movie might not have been so successful if it had been released earlier than the end of the twentieth century. Technology has progressed so far that the public was willing to entertain the transhumanist future depicted in the Matrix.
The Matrix presents an explicit image of the problem that Plato and Descartes identified. This image is that of a brain in a vat:
The brain in a vat image shows how vulnerable all of our beliefs about the exterior world are. Light might bounce off of objects in the exterior world and into our eyes conveying information about reality. However, it is also possible that wires stimulate my visual cortex to give this illusion. Both interpretations are perfectly consistent with our perceptions. Furthermore, there is no way to distinguish between the two.
In 2001, two years after the Matrix arrived in theatres, Nick Bostrom contributed another image, together with Plato’s Cave and Descartes’ Demon, to this subject. He calls this work The Simulation Argument [3] The argument attempts to show that:
“at least one of the following propositions is true: (1) the human species is very likely to go extinct before reaching a “posthuman” stage; (2) any posthuman civilization is extremely unlikely to run a significant number of simulations of their evolutionary history (or variations thereof); (3) we are almost certainly living in a computer simulation. It follows that the belief that there is a significant chance that we will one day become posthumans who run ancestor-simulations is false, unless we are currently living in a simulation.”
Note how the image has evolved throughout history. Plato only required the technology of fire. The image of people chained in one place and deceived by shadows strikes one as too implausible to be compelling. Descartes drew attention to dreams. We feel confident, however, that we can distinguish between dreams and reality. Descartes increased the plausibility of his account by considering a global supernatural controller — an evil demon — to deceive us. This deception could be more convincing than that involving shadows or dreams. The brain in a vat image, however, increases the plausibility even further by being naturalistic. The argument that we could be such brains invokes only computers — not demons. Finally, the Simulation Argument presents a reason for thinking not just that such a naturalistic scenario is possible but that in fact one is probable. Perhaps the most telling fact about The Simulation Argument is that the author, Bostrom, founded the World Transhumanism Association together with David Pearce, author of The Hedonistic Imperative.
Consider the article “How To Live In A Simulation” [4] by another contributor to transhumanist philosophy, Robin Hanson. Hanson suggests that, given Bostrom’s premise that we live in a simulation, being entertaining is in our best interest. If our descendents simulated us we need to amuse them to secure our own safety and future. We do not want our descendents to pull the plug. The important — but often unstated — consequence, however, is that the value and significance of human life is diminished. According to Hanson’s argument, humanity might be nothing more than a court jester performing for our cruel descendents. Transhumanist philosophy and future technologies do violence to the fourth anthropocentric conceit.
The Matrix presents another image of the problem that virtual reality technology poses for our sense of reality. This film suggests a distinction between internally consistent and internally inconsistent illusions. Consider the people who live within the Matrix such as Neo. Neo was surprised and alarmed by Morpheus’ revelation that Neo lives within the Matrix. Neo felt surprised because the world within the Matrix does not suggest that technology is sufficiently advanced to create such an illusion. Before Morpheus demonstrated the illusion to him Neo never felt troubled by the prospect of Plato’s Cave or Descartes’ Demon. Consider, however, the people who live in Zion and regularly enter the Matrix. These people do not find the existence of global illusions surprising at all. The citizens of Zion know many people, specifically those trapped in the Matrix, who are victims of such an illusion. Now suppose that someone suggested that Zion itself is a global illusion — a Matrix within a Matrix (MWAM) — without demonstrating the truth of his claim. The citizens of Zion can no longer afford Neo’s incredulity at such mere possibilities. This dilemma would cause the Zionites no small amount of anxiety. Yet, the film never depicts Zionites as being particularly troubled by this possibility.
We can immediately note, however, that this distinction between internally consistent and inconsistent illusions is unwarranted. This is so because any and every illusion is consistent with the notion of brain in vat scenarios. There is nothing that constrains the scientist who manipulates the brain from creating illusions in which global illusions cannot happen. For example, such an illusion need not be like Zion, in which the notion of global illusions are presented as feasible. The scientist may just as well create the illusion that you live in ancient Greece with Plato or in Holland with Descartes. The impression that some, such as those in the Matrix, are the victims of illusions, whereas others, such as those in Zion, are not, does not protect one from this dilemma. The citizens of Zion may just as well be trapped within a MWAM. These citizens ought to wonder whether or not their world is an illusion. So should we.
Considering the previous examples one can see how the Matrix did not present the strongest case against the fourth anthropocentric conceit. The citizens of Zion are never shown to be themselves the victims of illusions. Moreover, the controllers of the Matrix only manipulate their victims’ sense of the external world. The controllers might just as well alter their moral intuitions, as the creator of HS3 did, or decision processes, as future police might. They could manipulate their memories. The Matrix suggests, but does not present, the possibility of human beings that are victims of multiple layers of virtual reality and repeated memory and moral intuition wipes. Such a posthuman future threatens to create a philosophical crisis for the human race. To show that this crisis is already beginning to happen consider this quote from a recent news story with the headline “Matrix makes its way into courtrooms as defense strategy,” [5].
“It’s taking people a little further into the future. For people who are already confused between fantasy and reality, it gives them a framework to articulate it,” she said. “People who are already on the edge, it can be argued, can be set off by these types of movies.”
The article describes how Hamilton, Ohio resident Tonda Lynn Ansley was found not guilty by citing an insanity defense. Ansley claimed she was inside of the Matrix. Vadim Mieseges from San Francisco also successfully defended himself at his murder trial by claiming he was inside of the Matrix. Joshua Cooke from Oakton, Virginia argued similarly. Even Lee Boyd Malvo, one of the Washington D.C. snipers, wrote “free yourself of the matrix” while in jail.
I am not claiming that these situations are common or that these people are sane. Moreover, I agree that suspects probably find the Matrix defense attractive whether or not they believe what they say. However, I want to note that before future technologies and ideas made the Matrix plausible and popular, nobody cited this defense. I submit that these cases are merely the beginning of a future philosophical crisis. Although we can perhaps afford to ignore this dilemma, posthumans will not have this luxury. They will find The Human Condition by Magritte to be especially moving for that painting depicts the posthuman condition as well.
IV. CONCLUSION
The anthropocentric conceits that Copernicus and Darwin undermined are not the last. At least two more comforting illusions will be challenged by progress in science and technology. The transhumanist community is familiar with the third conceit and enthusiastic about surpassing the limitations upon human intelligence. The fourth conceit, however, has largely been ignored by the transhumanist community. Future technologies, in particular modification of the brain reward mechanism, intelligence amplification, behavior engineering, and virtual reality, threaten to undermine comfortable or naive beliefs about ethics, freedom, and reality. By exploring these potential threats, one can see how multiple facets of the fourth conceit compliment each other. Furthermore, these philosophical dilemmas have already emerged in courtrooms. In the worst case scenario, our ethical intuitions and choices are manipulated by a controller who keeps us trapped within a virtual reality. Posthumans will be forced to acknowledge that they might be such people. One can never know. This is the posthuman condition.
--------------------------------------------------------------------------------
Footnotes
* I am thankful for valuable comments from Hunter Washburne.
[1] Pearce, David. The Hedonistic Imperative
http://www.hedweb.com/
[2] “Famous American Trials: Illinois v. Nathan Leopold and Richard Loeb,”
http://www.law.umkc.edu/faculty/projects/ftrials/leoploeb/leopold.htm
[3] Bostrom, N. 2003. Are you living in a computer simulation? Philosophical Quarterly 53:243-55.
http://www.simulation-argument.com
[4] Hanson, Robert. “How to Live in a Simulation.” Journal of Evolution and Technology, vol. 7 (2001).
http://www.jetpress.org/volume7/simulation.html
[5] Bean, Matt. “‘Matrix’ makes its way into courtrooms as defense strategy,”
http://www.cnn.com/2003/LAW/05/21/ctv.matrix.insanity
