by David Levy
Similar or even greater increases in processing speed during the next few decades are inevitable. Electronics experts are continually pushing forward the technologies that determine the speed at which silicon chips can process data, and entirely new computing technologies are being researched for this purpose, technologies that do not rely on silicon and that go under names such as molecular computing, optical computing, and quantum computing. While computer processing speeds continue to increase at dramatic rates, computer memory sizes are also increasing apace. The tiny memory devices you might be using to plug into your laptop, store your music downloads, or enable your camera to retain ever-escalating numbers of high resolution photographs are some of the more obvious signs of the increases in memory capacity that we have witnessed during recent years. The importance of increased memory for tasks in artificial intelligence has been succinctly expressed by Yorick Wilks, one of the world’s leading experts for the past three decades on computer translation and other aspects of processing linguistic skills. Wilks says that “Artificial Intelligence is a little software and a lot of data.” The reasoning behind this aphorism is that the more knowledge a computer program can access, the more intelligent it will appear to be.
I shall now go out on a limb to some extent and explain why my predicted time line for the development of sophisticated robot companions and lovers focuses on somewhere around the middle of this century as the likely time by when all the robotic goals described in this book will be achieved. My own background in AI is strongly rooted in the field of computer chess, so I’ll use that as the basis of my argument. Goethe described chess as “the touchstone of the intellect,” a definition that matches most people’s perception that to play chess at grandmaster level requires a high degree of intelligence. This is why, when the 1956 workshop on artificial intelligence inaugurated the discipline as an academic division in its own right, those scientists who had come together to define and discuss the goals of AI listed computer chess as one of them. Since then computer chess has often been described by eminent AI gurus as “the drosophila of AI,” a reference to the way in which the fruit fly has been seen as the classic test bed for genetics research.
When I first became aware of chess-playing programs that were capable of giving weak human players an interesting game, their standard of play was, in comparison with my own standard, abysmal.* Thirty years later I watched with awe as Garry Kasparov was ripped apart in just one hour, in the sixth and final game of his fateful match against IBM’s Deep Blue chess computer. So within those three decades, I had seen progress in this touchstone area of artificial intelligence, from the abysmal to world-shattering. Given that playing chess well is a task that requires much brainpower, I believe that another thirty years from now, give or take a few years, will see strides made in just about every other area of AI, including emotion, personality, and all the mental qualities required of a robot that can behave as you and I do, strides that raise the state of the art in those other branches of AI from where they are now, which in many cases is better than abysmal, to levels that will exceed those exhibited by the most intelligent, the most capable, the most sensitive, the most loving of humans. All this points to a time around the year 2035. When I asked the noted futurist Ray Kurzweil when he expected the first human-robot marriages to take place, his answer was 2029, but I am somewhat more conservative than Kurzweil in my predictions, and I prefer to add in some contingency to allow for unexpected dips in the enthusiasm for academic research in this area and/or for funding from the traditional sources(i.e., U.S. government agencies) in some of the crucial scientific disciplines. Hence my estimate of midcentury. But there is one major factor that could result in my being proved wrong, a factor that could bring marriageable robots onto the market in Kurzweil’s projected time frame or even earlier. That factor is the commercial avarice of big business, particularly in the “adult” sector, and I have witnessed its effects before. When, during the late 1970s and very early 1980s, I used to visit the twice-yearly Consumer Electronics Show in Las Vegas and Chicago, videocassette recorders were just another interesting product item. Within another year or two the market for VCRs, and even more for their bestselling videos (pornography), had expanded beyond all recognition. Sex machines such as the Sybian, the Thrillhammer, and the Stallion are not yet big business, but when their sales reach a certain threshold, watch out! Investment in new product developments might suddenly become available on a massive scale, with an eye to increasing the already astounding profits that the adult-entertainment industry reaps each year. If and when that happens, Kurzweil will be proved right.
Erotic Computation Research
I do not believe that it will take more than a decade for sexual applications of artificial intelligence and robotics to become mainstream research topics. When that happens, universities and research institutes will be able to tap into the creative talents of the huge number of students and other researchers who are eager to mix their natural desires with their academic goals. Take a look at this text, from the Web site of the Erotic Computation Group at MIT, for a taste of university life to come:
VISION
The broad goal of the MIT Media Lab is to explore the impact of modern computing on human society. Groups at the Media Lab study the implications of computational technology at all stages of the human life cycle, from the neurological development of infants and the behavioral learning patterns of children to the sophisticated interaction modalities of adults in digital communities.
Though we are at times reluctant to admit it, all humans are sexual beings. It is time that we overcame the antiquated societal taboos associated with the topic of human sexuality and began to explore it from a critical academic viewpoint. The Erotic Computation Group is devoted to this exciting field of inquiry.
One of the more interesting projects undertaken by the group is that of Dan Maynes-Aminzade, the group’s leader, which he calls “Sex Toys of the Future” and which the site describes as follows:
From the simple beauty of the Ibrator to the delicate complexity of the Jackhammer Jesus, sexual appliances have a long history of sophisticated industrial design. Unfortunately, these tools are often ineffective, and their manner of usage is frustratingly crude, having changed little since the first century AD. Dan has a special interest in inventing sexual instruments that understand the artistic intentions of their user, allowing for the enhancement and extension of erotic expression. By designing haptic interfaces that scale gracefully from the sexual neophyte to the experienced professional, he hopes to make sex more accessible, intuitive, and fulfilling.
Any reader wishing to join the group will doubtless be intrigued by its admissions policy:
The Erotic Computation Group seeks creative, hard-working, team-oriented, and sexually competent graduate students. Successful applicants possess varied skills in computer programming, electrical engineering, and mechanical engineering. Special sexual abilities are also important assets. Oral and written communication skills are essential, as our work is regularly presented to visitors and submitted to major conferences and journals.
As I have mentioned earlier, I do not expect it to take more than a decade for such groups to form part of mainstream AI and robotics research. But at the time of writing (early 2007), there are no such groups. The “Erotic Computation Group” is a hoax, perpetrated by MIT students. But what appears on their Web site is all quite believable. In fact, when this first appeared on part of MIT’s main Web site, some members of the MIT faculty and staff believed that it was genuine.
8 The Mental Leap to Sex with Robots
The only unnatural sexual act is that which you cannot perform.
—Alfred Kinsey1
In the early years of the twenty-first century, the idea of sex with robots is regarded by many people as outlandish, outrageous, even perverted. But sexual ideas, attitudes, and mores evolve with time, making it interesting to speculate on just how much current thinking needs to change before
sex with robots is accepted as one of the normal expressions of human sexuality rather than one of its more bizarre offshoots and for us to ask what the processes will be that bring about such a change. In order to demonstrate the extent to which sexual thinking has altered, particularly during the past century or so, we shall examine changes in attitudes, principally in Britain and America, toward four different aspects of sexuality: homosexuality, oral sex, fornication, and masturbation.
Homosexuality
Since Victorian times, no aspect of human sexuality has been the subject of more dramatic changes of attitudes than homosexuality. Ancient Jewish law had prescribed the death penalty for sodomy, based on the biblical teachings of Leviticus 20:13: “If a man lies with a male as with a woman, both of them have committed an abomination; they shall be put to death, their blood is upon them.” And since biblical times, several countries and civilizations, including Britain and the United States, have similarly meted out the death penalty for sodomy, as documented by Richard Davenport-Hines, Reay Tannahill, and Gordon Taylor:
Aztec law included the death penalty for homosexuals, male and female; in Peru anyone guilty of sodomy was condemned to be dragged through the streets and hanged, and then burned with all his clothes; the Incas burned sodomites alive in the public square.2
In 1627, Pedro Simón, in his Conquistas de Tierra Firme, reported on five Italian soldiers, serving in Venezuela, who were “strangled and burnt, with general applause” at the orders of their Spanish commander.3
The first Russian state laws against buggery appeared in military statutes drawn up during the eighteenth century reign of Peter the Great. Initially this was punished by burning at the stake, later changed to corporal punishment.4
In England the appetite for punishing homosexual behavior with the death penalty appears to have been not completely consistent. An ecclesiastical law of 1290 ordered sodomites to be buried alive, but this sentence seems never to have been imposed, and the few sodomites who were convicted by Church courts were hanged. This was also the punishment prescribed by King Henry VIII in 1533, when a “Buggery Statute” was enacted in Britain, defining sodomy as sexual activity between two men or as bestiality involving an animal and either a man or woman. This brought homosexual behavior, and in particular anal sex, within the jurisdiction of the state courts rather than the ecclesiastical courts as it had been previously, but despite the continuing capital nature of the offense, there were cases in 1541 and 1594 of headmasters who were found to have sexually enjoyed male pupils but survived, not only with their lives but also with their reputations scarcely tarnished.5
Following the hanging of the Earl of Castlehaven in 1631, there appear to have been no more executions for sodomy in Britain until the eighteenth century. By the early part of the nineteenth century, executions of homosexuals were steadily increasing—in one English county alone, Middlesex, 28 men were hanged out of a total of 42 convicted sodomites during the period 1805–15. And sodomy was regarded as so base a crime in early-nineteenth-century Britain that in newspaper accounts of the trials and executions of those convicted, it was commonplace to write somewhat euphemistically about their offenses, in contrast to the reports of trials of murder, for which all the gory details would normally be published. In the Times of August 13, 1833, for example, the report on the execution of Henry Nicoll,* a retired captain from the Fourteenth Infantry Regiment, says of his crime only that he “was tried and found guilty of an unnatural offense.” It was a popular pastime for large crowds to watch executions in those days, and the Times reported that “amongst the spectators a large number of females also presented themselves, and, by their shouts manifested their abhorrence of the criminal.” The broadside† of Nicoll’s execution employs language of an even more venomous kind than was customary for hangings, reflecting the general view of the base level of depravity of his offense, but still without saying what he had done: “Heinous, horribly frightful, and disgusting was the crime for which the above poor Wretched Culprit suffered the severe penalty of the law this morning, Monday, August 12, 1833…. Thank heaven the public Gallows of Justice in England is very rarely disgraced by the Execution of such Wretches; but, every person must have observed, with dismay, how greatly the number of diabolical assaults of a similar nature, have lately multiplied in this country.”
THE EXECUTION OF CAPTAIN HENRY NICOLL, IN AUGUST 1833.
Equally euphemistic was the Times’s wording in reporting on the September 1835 trial at the Old Bailey of John Smith and James Pratt, the last men to be executed in Britain for sodomy. Their crime was described simply as “an abominable offense in a house kept by William Bonnell, who was charged as an accessory. The jury returned a verdict of Guilty against all the prisoners. Sentence—Death.”*
When in 1828 new legislation retained buggery and oral sex as capital offenses in Britain, the number of capital convictions rose to such an extent that during 1842–49 only murder exceeded sodomy as a cause of death sentences, and during one year (1846) there were actually more death sentences passed on those guilty of sodomy than on murderers. However, after 1835 all such sentences were commuted, reflecting the beginnings of a marginally more tolerant attitude in Britain toward homosexuality, and in 1861 the statutory punishment of homosexual behavior was changed from hanging to penal servitude of between ten years and life.
Prior to the American Revolution, British laws were in force in the colonies, and so offenses committed in Virginia, for example, would be subject to those laws and their punishments. Bruce Robinson has chronicled the history of sodomy legislation in the United States, a history paralleling that of Britain. Thus we find that in 1624 a Captain Richard Cornish was charged in Virginia under the British buggery statute with having raped his male servant. He was found guilty, and both he and his servant were hanged. Sodomy became a capital crime in Massachusetts in 1641 (but only between males); the following year Connecticut included sodomy among its twelve capital crimes, and other states soon followed suit. In 1682 the Quaker colony of Pennsylvania became the first jurisdiction in America to make sodomy a non-capital offense. Initially the prescribed punishment was a whipping, a fine equal to one-third of the offender’s estate, and six months’ hard labor, but in 1700 the punishment was made much harsher—imprisonment for life or castration.
During the years following America’s independence from Britain, the death penalty for sodomy was gradually removed from the former colonial laws, though in several states this took quite some time. South Carolina, for example, abolished the death penalty as its punishment only in 1869. And over the next one hundred years, in both the United States and Britain, the prison sentences passed for homosexual behavior gradually grew shorter and shorter, being abolished completely in Britain in 1967.
By the second half of the twentieth century, the punishment of homosexual behavior had also become a rarity in the United States. In 1950 all American states had antisodomy laws on their books, and even in 1974, when the American Psychiatric Association crossed homosexuality off its list of diagnoses, the laws of some states still regarded homosexual behaviors as criminal offenses deserving of harsh punishments. Such laws were not dealt a final blow until, on June 26, 2003, after months of public media debate, the U.S. Supreme Court struck down a Texas law banning sexual relationships between gay couples on the grounds that the law was unconstitutional. The basis for that judgment was simply that gay men “are entitled to respect for their private lives.” This ruling apparently invalidated laws in four states—Texas, Kansas, Oklahoma, and Missouri—that still prohibited oral and anal sex within same-sex couples, as well as laws in nine other states—Alabama, Florida, Idaho, Louisiana, Mississippi, North Carolina, South Carolina, Utah, and Virginia—that prohibited consensual sodomy (defined for the purposes of those laws to include oral sex) for everyone, homosexual, heterosexual, married or not. Until then the punishment prescribed in Idaho was the harshest remaining in the United States—imprisonment from five years to life.
Oral Sex
Many American states had laws that in theory treated oral sex as an act similar to sodomy, an attitude that appeared to have gained currency during the last twenty years of the nineteenth century. In 1900 there were thirteen states with sodomy laws that also encompassed fellatio, and of these there were ten states that had specifically changed their sodomy laws so that they did provide for punishing fellatio. By 1920 the number of such states had almost doubled, from thirteen to twenty-four. But despite this legislation, prosecutions were exceedingly rare and the practice was exceedingly popular—by the late 1940s, Kinsey’s research indicated that 49 percent of married men in American had performed cunnilingus, while 46 percent of married women had performed fellatio.
Today oral sex is viewed very differently from the way it was eighty years ago. Nowadays it is close to being de rigueur in most heterosexual relationships, with 35 percent of men in the United States and 40 percent of those in France reporting in one study that they had engaged in oral sex (fellatio or cunnilingus or both) during their most recent sexual encounter. For women the percentages were 34 in France and 26 in the United States.
Fornication
Another sexual “sin” that we consider in terms of changing attitudes and mores is that of fornication—sexual intercourse other than with one’s spouse. The Old Testament prescribes death by stoning for fornication, a punishment that could be meted out, for example, to a woman who was found not to be a virgin at the time of her marriage: “Then they shall bring out the damsel to the door of her father’s house, and the men of her city shall stone her with stones that she die: because she hath wrought folly in Israel, to play the whore in her father’s house: so shalt thou put evil away from among you.” (Deuteronomy 22:21)
