Paracelsus's homunculi and Asimov's robots: the laboratory ethics of personhood and artificial life in medieval and modern thought.
What is the most important thing about living beings? This question, abstract and philosophical as it sounds, has direct applicability to questions of bioethics and roboethics in the modern world. Genetic engineering, in a sometimes spectacular fashion, breaches the boundary between real and artificial creatures. The FDA, despite pleas from consumers, has proven completely unable to come up with a definition for "natural". (12) Even the question Do clones have souls?, meaningless as it may sound to those outside the traditional metaphysical idiom, troubles many when they consider questions of stem-cell research or cloning. Those who think further to the future ponder the logic of digital intelligence, the question of whether hypothetical advanced computer intelligences can and should be considered the same as human ones. Depending on the perspective, the potential products of science's advance can embody either humanist progress and the possibility of eliminating poverty and disease or an inevitable march toward any particular sort of posthuman dystopia. In either case, the ethical quagmire of identity and souls and fundamental rights cannot be avoided when considering the issues raised by the near-inevitable relentless creation and refinement of scientific methods which continue to transcend, as they have in every era, the comfortable categories of human experience as we know it.
The ethical and philosophical questions raised by the spectre of artificial life and artificial intelligence are more-or-less as follows:
1) By what criteria may we differentiate a living creature from a nonliving creature?
2) By what criteria may we define and recognise intelligence?
3) What inherent rights are to be afforded to a living being?
4) What inherent rights do we ascribe to an intelligent being?
5) How, if at all, are natural and artificial creatures qualitatively different?
These same questions have been equally pressing since people began seriously considering and debating the possibility of creating life through laboratory or engineering processes.
The question of whether it was either possible or ethically justifiable for human efforts to generate life from nonlife was just as immediate and vital a question to the philosophers of antiquity and the alchemists of the Middle Ages as it is to today's genetic engineers and neural-computation programmers. The prevalence of stories in the mythologies of various cultures in which a person is created by human agency alone - from Galatea to the Golems - bear evidence to the compelling nature of such suppositions. In the medieval discipline of alchemy, huge amounts of effort were devoted to elaborating and analysing the concept of the homunculus. This creature, in general a theoretical construct although some chymists claimed to have created it in a laboratory, was specifically envisioned as a laboratory-grown man who, although theology forbade its having an immortal soul, would have anomalous superhuman properties due to its circumstances of conception and growth. More generally, the theme and potential creatability of the homunculus served as both thought-experiment and motivator for those who wished to explore the same questions that surround the notion of artificially-created life in present-day scientific endeavour.
Forebears and counterparts of the homunculus in folklore and literature.
One familiar example of this motif of the artificial man can be seen in the common folk-tale narrative common to tales such as Tom Thumb (5) and Thumbelina in which a couple, unable to bear children, are apprized by some wise old person (in Tom Thumb, this role goes to Merlin) of a method to make for themselves a child by other means. The child is generated by the couple by the planting of a seed that grows a flower inside which the child is hiding, or by cracking open a rock given them by a witch, or some other similar magical mechanism; the child matures rapidly to a precocious intellect, and invariably proves to be of anomalously tiny size - indeed a mannikin, the word itself an English calque of "homunculus". Here we see the first hints of the theme of precocity in the artificial person.
Another prominent instance of a homunculus theme in legend may be seen in the cloud of mythology surrounding the mandrake, Mandragora officinarum, the classic "witches' herb"; its root, which can grow into remarkably anthropomorphic forms, was historically harvested for occult purposes and carved to accentuate its similarity to the human form. These creatures were credited with the ability to multiply money, fend off the evil eye, and a number of other magical uses. In addition, it was reputed to be used in witches' rites for its narcotic effects, and thus obtained an occult reputation and association with demonic forces. (10) Its association with humankind and with soulless sorcery continued into the present century: Eliphas Levi, most famous of the turn-of-the-century High Magicians, claimed that "man came out of the slime of the earth, and his first appearance must have been in the form of a rough sketchthe first men were, in this case, a family of gigantic, sensitive mandragores, animated by the sun, who rooted themselves up from the earth." (14) In German Alraune, these creatures were, in Northern European legend, supposed to grow up where hanged men's semen fell on the earth. In some cases, mandrake figures were cast as familiars, gifts from the Devil to the magician who might consult them or set them to deliver messages or any small task.
One of the most famous of the Sufi fables, the Story of Salaman and Absal, was written in the 1400s by the Persian poet Jami but dates in its original form from around the year 0. It tells the story of a king who, on the advice of a wise man, abstains from sexual intercourse in order to improve his mind. When he wishes to produce an heir, he seals up his semen inside a mandrake-root vessel and gives it to the wise man, who then performs some unspecified ritual which results in the creation of a perfect and rational male child. This child then grows to adulthood, falls in love with a woman (much to his father's disgust), and eventually flees with her across the ocean, but eventually abandons her for Aphrodite and finally Aphrodite for a return to celibate philosophy and rulership of his father's kingdom. (2) In this story one can see rather clearly echoes of the mannikin or fairy-child stories mentioned above, as well as startlingly precognisant suggestions of later notions of the homunculus.
In the modern (post-Enlightenment) era, the new mechanistic philosophy of living things meant that when thinkers once again turned to the question of artificial life, the new idea was not one of a clay statue infused with a soul, but rather a complex feat of engineering from which a living thing appeared through some process which we would now describe as "emergent complexity". Frankenstein's electrically reanimated monster, probably the most famous example, can perhaps be viewed as an intermediate in this sequence, bridging the alchemical tradition (and the animistic and medievalist leanings of the Romantic movement) and the optimistic mechanism of the Industrial Revolution.
More recently, the writers who address the creation of intelligent life by humankind and the deeper questions thereby raised are largely science fiction authors. The classic work of this sort is Isaac Asimov's I, Robot, published in 1950, which was Asimov's attempt at a modern and rational approach to the question of artificial intelligence and the morality related thereto; more recently, such authors as Bruce Sterling have taken up the inquiry.
Alchemy and the creation of artificial life.
In order to understand the edifice of scientific thought in which the alchemical notion of the homunculus arose, it is necessary to realise that spontaneous generation was a fundamental assumption of Egyptian and Greek physical theory. Ovid, for example, specifies rather closely in the Metamorphoses the different sorts of animals that are spawned from the putrefaction of various sorts of matter - bees from dead cattle, scorpions from crab-shells. (3) By Aristotle's time, then, the idea that animals could be spawned directly from nonliving substance was an almost unchallenged assumption, as so many data seemed to confirm it. The theory being repeated and elaborated upon by Aristotle, it then was elaborated further by the Islamic scholars and became a mainstay of medieval natural philosophy.
Aristotle also appears to have originated the theory of sexual generation in which the male partner's semen serves as almost complete form, in contrast to the almost pure matter of the woman's menstrual blood. Consequently, the male "principle" is thereby defined as the hot, active life-giving principle, and the female as the cold, passive matter. (4) Since all his arguments later became dogma, this notion was taken to much greater lengths by the medieval scholars, who began to wonder why the particular sort of matter provided by the female parent was so important, and indeed whether the totipotent semen could generate life if placed in a different matrix.
The bizarre pseudo-Platonic alchemical text Liber vaccae, an almost-indecipherable text ascribed to such diverse sources as Galenus, Plato, and Albertus Magnus but most likely dating from the period of the Arabic scholars, is the first appearance of a recipe for what is unequivocally a homunculus in the later widely-accepted sense of the term. (8) The procedure given for the creation of this "animal form that is agreeable to many miracles", consists of impregnating a cow or ewe with a mixture of human semen and some sort of phosphorescent stone called lapis solis (possibly the Bologna stone, as cited by the later alchemists), keeping it in the dark and feeding it on blood until it gives birth, at which point the offspring is put into a concoction of powders to give it a human skin and then fed on the surrogate mother's birthing-blood and milk in the darkness for forty days. When this procedure is complete, the creature's creator may either immediately kill it and anoint his feet with its blood, gaining the ability to walk on water, or raise it for another year on milk and rainwater, at which point it will tell him all manner of secret things. (1) Bizarre as this may sound, it has a certain logic when considered in the light of the theory of spontaneous generation: rather than the gross matter of human menstrual blood, the semen's form is imposed upon pulverised lapis solis, making it a being of light rather than matter, and thus inhuman in every sense of the word - lacking humanity, it has no soul and thus can be killed with impunity, but conversely neither is it tied down to inert feminine matter, enabling its spirit to far outstrip that of ordinary humans.
Possibly the greatest of the Arabian alchemists, Jabir ibn Hayyan, called Geber, was entirely in disagreement with this work. Jabir also gives a system for the artificial production of intelligent beings, which also begins with the sealing of a substance - variously translated "essence, "matter", and "sperm" - into an incubating apparatus for putrefaction. However, there the similarity ends; rejecting the methods of the Liber vaccae as demonic, Jabir instead recommends the construction of a series of rotating concentric spheres centred around the incubating apparatus, in order to imitate as closely as possible the celestial structure and thereby induce life to arise in this simulacrum of the universe. In fact, rather than treating life as a product to be chemically produced through the simple action of semen on matter, Jabir seems to set up his system after the style of the supposedly inhabited statues in the antique temples of the gods, so as to invite a spirit in.
The homunculus in European alchemy.
Several writers in the early years of the High Middle Ages European revival of alchemy make arguments regarding the theoretical aspects of the Arab scholars' "artificial man", but the only actual attempt at generation of such a creature is attributed to Arnald of Villanova, who kept a flask of semen according to the practices of the 9th-century scholar al-Razi until it began to quicken and form a human body inside the flask, at which point Arnald destroyed the creature so as not to "seem to tempt God". Arnald, the foremost physician and scholar of his day, was able to read Arabic, and so was one of the major conduits of Arabic philosophy and science into western Europe. Although he followed the dictates of al-Razi, a strict rationalist, in creating the homunculus, in destroying it he was following the dictates of such as al-Kindi and other orthodox philosophers of the caliphate.
Such writers as Alonso Tostado and William of Auvergne praise Arnold's piety and portray the practice of artificial generation as demonic, although a few writers such as the pseudo-Thomas Aquinas of De essentiis essentiarum side with the Arab physicians in claiming a naturalistic origin. In general, though, the High Middle Ages were not favourable toward unnatural generation, due to a widespread atmosphere of piety and dread of attracting the interest of demons or the wrath of God.
However, the situation changes somewhat in the 16th century. By the time Magiae naturalis was written, Giambattista della Porta was able to cite a long list of examples of spontaneous generation "of putrefaction", Ovid's along with many others', follow it with husbandry advice for causing the breeding of hypothetical hybrids between various disparate animals. (6) The first recorded use of the actual word "homunculus" was in the famous process originally described in 1572 by the rogue alchemist-physician Theophrastus Aureolus Philippus Bombastus von Hohenheim, better known by his assumed name Paracelsus. This now-famous recipe for the creation of "a true living infant, having all the members of a child...but much smaller", taken in part from al-Razi, consists of incubating a flask full of semen in a warm environment of composting horse-dung for "forty days, or until it begins at last to live, move, and be agitated". Once it quickens, the homunculus is kept forty days more in its incubator and fed every day on a distillate of human blood; then it is decanted and educated intensively.
Created as it was without the use of the woman's usual contribution to reproduction, Paracelsus believed such a creature would be small and transparent of body, but pure and incorrupt in all its parts, and thus possessed of preternatural intelligence and an insight into the workings of Nature that amounted almost to a second sight, "know[ing] all secret and hidden matters". In fact, when considered from a Platonic perspective, the homunculus embodied human Form, and as such would be born with the faculties of a full-grown man's spirit and none of the impurities of matter. On the other hand, if the identical procedure was undertaken using menstrual blood rather than semen, the product of the procedure would be instead a basilisk, an evil and mindless creature whose gaze was poison, since it would be comprised entirely of unformed matter imbued only with the inherently corrupt and impure essence of the woman's original Fall from grace. (7) The only characteristic that united these two creatures, then, being made from totally opposite isolates of the human spirit, would be the lack of a soul, which Paracelsus believes is a faculty gained only through the union of male and female seed; thus he claims that only men and women born the usual way could possess one.
Sexual concerns and transhumanism.
Even compared to his 16th-century peers, Paracelsus had exceedingly negative views regarding sexuality. In his later writings, he claimed that homunculi could further be accidentally generated in nearly any environment by non-procreative emission of semen, or even internally if one were celibate and experienced lust - for example, anal sodomy is blamed for intestinal worms - and he claims that the only way to avoid such accidents is either to have only procreative sex, or to self-castrate and thereby avoid the problem entirely. By this interpretation, the creatures sound less like artificial life and more like a sort of metaphysical biohazard. The only legitimate expenditures of semen, he seems to claim, are sexual reproduction and wholly-artificial laboratory creation of homunculi.
In the preceding century, the theologian Alonso Tostado had come down against the "spermists", pointing out that children do often resemble their mothers and thus draw some form from them, and further opining that if the strict Aristotelian spermism were true, women would be merely hollow flasks, unworthy of the name and honour of motherhood. In fact, this is framed in terms of an insult to the Virgin Mary, to whom honour (he says) is due for her role as holy Mother and not merely as a "vas clausum" for the incubation of a preformed homunculus, even one created by God. Moreover, he claims that Arnald's experiment implies that demons who collect the semen of sleeping men can then become "mothers" by sealing the stolen semen in warm places; though, he admits, the demons do in fact do this and thereby generate giants and monsters (and furthermore that the Huns are an example of such a degenerate breed of humanlike creatures), it is not worthy of the dignity of the office of motherhood. Rather, the mother must supply human menstrual matter and thereby take equal part in the generation of the offspring, as did the Virgin Mary.
Tostado's arguments, combined with the unsavoury implications of Paracelsus' severe misogyny and strange attitudes regarding sexuality, are startlingly reminiscent of modern criticisms of the notion of the "artificial womb". Any casual watcher of the world of biotechnology will notice that every five years or so, a researcher somewhere announces another step in the direction of a suitable environment for in vitro incubation of fetuses (frequently called "exogenesis"), prompting a flood of opinionated analyses ranging from nave animist alarmism to sophisticated analyses of possible harm to the social status of women. On the other hand, some humanist critiques of exogenesis come down in favour of the liberation of women from the constraints of mammalian biology; although the alchemists were by religious dogma constrained from truly celebrating such a notion, one can see a similar strain of enthusiasm in those authors who thought of male laboratory parthenogenesis as a means of escaping the intellectual constraints of material existence. The debate, of course, is still difficult to resolve.
Religion, philosophy, and ethics: the need to define the personhood paradigm.
The ethical debate regarding cloning, for many scientific moralists, tends to nucleate around the question of whether it is justifiable to create a living thing for medical use. This consideration, entirely beyond the question of whether the creature in question is living,
The pseudo-Thomas of De essentiis essentiarum explicitly advocates the use of the homunculus as a source of blood "useful against many infirmities, according to what he says" - to a modern reader not only a bizarre suggestion that the homunculus is advocating its own destruction and dismemberment for therapeutic purposes, but also a demonstration of complete disregard for a clearly sentient creature. (1) This idea, also present in the Liber vaccae, is theologically justified by the idea of the tripartite soul taken from Plato's Republic and filtered through centuries of Catholic exegesis, in which the parts are appetite, reason, and spirit. Pseudo-Thomas claims that the homunculus has no rational (divine) soul, this fraction being imparted only by the Creator and only upon the birth of a human being, and thus that it is as fit as any other animal for use as a research subject.
By contrast, William of Auvergne in the 13th century condemns as unnatural and possibly demonic the practice of generating creatures in order to slaughter them for magical purposes, though oddly making no challenge or reference to pseudo-Thomas's claims regarding ensoulment. His analysis seems strange in that it does not treat at all the question of whether the homunculus is a human, rather focusing entirely on the question of blasphemy and offense against nature.
Reactions to the publication and immediate popularization of De natura rerum were vociferous, both doctrinal and practical, and chiefly Catholic. Athanasius Kircher in Mundus subterraneus denounced Paracelsus' method as impious, Henry More in his Enthusiasmus triumphatus called "it one of the latest sanctuaries for the Atheist and the very prop of ancien Paganism", Joanes Bickerus in Hermes redivivus made incoherent claims about magnetic force attracting sorcery, and advocated prayer instead. (8) However, although the Jesuits predictably entangled themselves in a mesh of worries about original sin, theological error, and making demands on God, the real concern was that of "playing God". Admittedly Paracelsus did little to alleviate this fear, stating that "nothing is so secret that it cannot be made apparentGod can do everything through His wisdom and art. Likewise, we shall be able to do everything. Nothing shall resist us, neither magic nor spells; for these things are from God, and they are His artsI under Him as far as His realm goes, but He under me in my realm." (7) Clearly, though the technology involved has changed drastically, the ethical concerns that struck the medieval scholars in dealing with the question of laboratory-created life are startlingly similar to those which are brought up on a regular basis by those on both sides of the modern debate.
These thinkers spent great amounts of their effort on questions of defining the way in which the artificial man was to be considered, long before there was any reliable demonstration that such a creature could actually be created. This may seem a waste of effort when one considers that their scientific framework was undermined over the following few hundred years, the concept on which they had spent so much ink being proven a physical impossibility and their philosophical and theological paradigm far less fundamental than they supposed. However, upon reflection one realises that it is important to establish a legal and social framework for defining and preserving the rights of artificial intelligences, before it becomes a crucial issue and the great world powers decide to make do with some new version of the Three-Fifths Compromise.
When the homunculus of Paracelsus became a major part of scientific discourse, thinkers were alarmed by the lack of restraints upon a creature with no soul; since by the logic of the Church fathers the reason to avoid evil actions was to avoid imperiling one's immortal soul, the homunculus would have no restraint upon its actions. This lent a somewhat frightening aspect to the creature; with nearly unbounded intelligence and no fear of imperiling any immortal soul, it became a powerful amoral agent which could reinforce the creator's efforts in any direction he wished. The converse of this, of course, was that a creature who had no concept of original sin and no corruption of female (material) nature would perhaps be more pure and virtuous than an ordinary sinful man - the conclusion one might draw from familiarity with Jami's "Salaman and Absal". Similarly, Bruce Sterling, at a "Symposium on Roboethics", suggests that our interest in creating robots is in creating that thinking robot whom Dr. Calvin of I, Robot describes: someone humanlike in appearance but with immutable and incorruptible ethics. (11) However, he points out, the current state of robotics much closer to Paracelsus' amoral intelligence than to Asimov's Three Laws, since an essential feature of a robot at this point - such as the Sony Qrio, "a human-shaped, self-propelled puppet that can walk, talk, pinch, and take pictures, and it has no more ethics than a tire iron" - is that it is unintelligent and freely programmable; whatever its physical and computational capabilities, the ability to guide its own actions is missing and it is thus merely a tool to its owner. And, of course, a tool with all the physical capabilities of a souped-up human being is a frightening prospect, especially when one considers such threats as autonomous weaponry; as Sterling observes, this is not one of Asimov's robots. "Qrioknows nothing, cares nothing, and reasons not one whit. Improperly programmed, it could shoot handguns, set fire to buildings, and even slit your throat as you sleep before capering into a crowded mall to detonate itself while screaming political slogans." This goes beyond the fear of soulless intelligence into the cautionary tales of the Golem, and here too we must be careful of unregulated capabilities.
Intelligence being the closest thing to ensoulment, in the realm of A-life and computer science, the obvious goal - but how will we do that, and how will we know when we have? As Turing famously observed, we won't. (13) The only way to tell whether something is alive, then, is through observation; life as a fundamental characteristic is terribly ill-defined. Those present-day thinkers who argue and pontificate in editorials about such quasi-theological matters as clones' ensoulment or lack thereof are falling into a pattern of thinking which will, unfortunately, probably prove unhelpful due to its necessary lack of universality; as we see, the only universal, and therefore the only useful, approach is a legal and sociological definition of life. In this period of scientific history, we are closer than ever before to the exogenetic creation of something empirically indistinguishable in its mental properties from a living human. The way in which we approach this question, then, will prove vital to the entire future development of the treatment of intelligent beings.
1. Newman, William R. Promethean Ambitions: Alchemy and the Quest to Perfect Nature. U. of Chicago Press, 2004.
2. Jami, Moulana Nuruddin Abdorrahman. Salaman and Absal. Trans. Edward Fitzgerald. Octagon Press, c. 1450.
3. Smith, Justin. The Problem of Animal Generation in Early Modern Philosophy. Trans. Horace Gregory. Cambridge Studies in Philosophy and Biology. Cambridge University Press, 1958.
4. Aristotle. De Generatione Animalium. Trans. A. von Frantzius. Ed. 4: Leipzig: W. Engelmann, 1853.
5. Altemus, Henry. The History of Tom Thumb. 1999. Project Gutenberg.
6. Della Porta, Giambattista. Magiae Naturalis. Ed. 6, 1558.
8. Pinto-Correia, Clara. The Ovary of Eve: Egg and Sperm and Preformation. U. of Chicago Press, 1997.
9. Moran, Bruce T. Distilling Knowledge: Alchemy, Chemistry, and the Scientific Revolution. Harvard University Press, 2005.
10. WALA.de. Mandrake. 2006. Website. Available: http://www.wala.de/english/pflanze/archiv/mandrake.htm.
11. Sterling, Bruce. Robots and the Rest of Us. Wired 2004. Available: http://www.wired.com/wired/archive/12.05/view.html?pg=4
12. just-food.com. Us: Consumers Back Calls for Fda Definition of 'Natural'. 2006. Online news service article. Available: http://www.just-food.com/article.aspx?art=63415&type=1.
13. Oppy, Graham. The Turing Test. Stanford Encyclopedia of Philosophy, 2005.
Paracelsus (von Hohenheim), (Theophrastus Philippus Bombastus). De Natura Rerum. Ed. 7, 1572.
14. Levi, Eliphas. Transcendental Magic. Trans. A. E. Waite. 1896.