What is a monster? At least three concepts have been proposed: Aristotle thinks a monster to be a "mistake of purpose" in nature; Noël Carroll thinks a monster to be a scientifically impossible being that arouses disgust and fear; Cynthia Freeland thinks a monster to be an evil being. Thus a two-headed calf is an Aristotelian monster; a werewolf a monster on Carroll's definition; and Norman Bates of Hitchcock's Psycho a monster on Freeland's concept. These have no interesting overlaps. My project is to discuss Norman Bates and Mark Lewis (of Michael Powell's Peeping Tom). Bates and Lewis are monsters, but only on Aristotle's concept.
Monster, horror, Aristotle, Alfred Hitchcock, Noël Carroll, Cynthia Freeland, Michael Powell
1. Three concepts of monstrosity
Philosophers have recently become interested in exploring the genre of horror fiction and the concomitant concept of monstrosity. Books by Noël Carroll and Cynthia Freeland are the leading studies on these topics. Each has an opinion on Alfred Hitchcock's Psycho (1960) and Michael Powell's Peeping Tom (also 1960), and on the monstrosity of Psycho's Norman Bates (Anthony Perkins) and Peeping Tom's Mark Lewis (Carl Boehm). In brief: Carroll thinks that neither Psycho nor Peeping Tom are horror stories and that neither Bates nor Lewis are monsters; Freeland holds the opposite view. I will argue that both are wrong. Psycho and Peeping Tom are horror stories, though not for Carroll's reasons; and Bates and Lewis are monsters, but not for Freeland's reasons. I'll begin with an ancient account of monstrosity.
In Aristotle's Physics a monster is a mistake of nature, something that failed to attain its natural end. "Now surely as in intelligent action, so in nature; and as in nature, so it is in each action, if nothing interferes. Now intelligent action is for the sake of an end; therefore the nature of things also is so. ... Now mistakes come to pass even in the operations of art: the grammarian makes a mistake in writing and the doctor pours out the wrong dose. Hence clearly mistakes are possible in the operations of nature also. If then in art there are cases in which what is rightly produced serves a purpose, and if where mistakes occur there was a purpose in what was attempted, only it was not attained, so must it be also in natural products, and monstrosities will be failures in the purposive effort."
Aristotle's example of such a failure is borrowed from Empedocles, an ox with a man's face, though from his discussion it isn't clear whether Aristotle believes such a thing existed, or whether he simply presents this as a famous example of a possible monster. In any case, for Aristotle monsters now and then occur as missteps of purpose in nature just as grammatical mistakes now and then intrude on purpose in writing and wrong doses here and there ruin healing.
Noël Carroll takes a monster to be "a being in violation of the natural order, where the perimeter of the natural order is determined by contemporary science." Contemporary science, of course, is the state of scientific knowledge now, and scientific knowledge may enlarge over time. "Superman is not compossible with what is known of the natural order by science," Carroll writes. "He may at a later date become so, as knowledge of other planets and galaxies advances, but I wouldn't bet on it."
But Superman, while a monster, is not a horrific monster, the stories in which he appears not horror stories. A horror story must aim at arousing what Carroll calls "art-horror," a combination of fear and disgust caused by and directed towards a monster that is threatening and impure. A monster is impure if it is "categorically interstitial" (sometimes described as a "mixture of what is normally distinct") such as an animal that is part fly, part man; or if it is "categorically contradictory" such as a vampire that is both living and dead; or if it is "incomplete" such as a severed hand that acts on its own; or if it is "formless" such as a malevolent fog or a people-eating blob.
Another writer on horror, Cynthia Freeland, objects. "If monsters are really 'super-natural' as he [Carroll] thinks, then a real-life monster like Bob Rusk in Frenzy does not quite fit the paradigm." For Freeland, horror films are about evil, and "monsters are usually (though not always) evil in horror movies ...." Some of the characters Freeland takes to be monsters are obviously supernatural: the six-foot cockroaches that can assume human shapes in Guillermo del Toro's Mimic (1997), for example. Some of Freeland's monsters are empirically possible, including Catherine Deneuve's psychotically depressed character in Roman Polanski's Repulsion (1963) and the shark of Stephen Spielberg's Jaws (1975) - at least she assumes that film's great white is consistent with current shark science. For Freeland, what makes these monsters is their evilness.
So we have three concepts of a monster: a mistake in natural teleology, though one that occurs now and again in nature (Aristotle); an empirically impossible being that is "impure" and that arouses fear and disgust (Carroll); and an evil creature, sometimes empirically impossible, sometimes "real-life" (Freeland). These concepts concur with ordinary ways of speaking in which we might describe a two-headed calf, a werewolf, and Pol Pot, the architect of mass genocide against the Cambodians, as monsters.
Do these beings have anything interesting in common? That is, are the monsters described by Aristotle, Carroll, and Freeland species of an overarching genus of monstrosity? It would seem not. The two-headed calf is natural development gone wrong, the werewolf a scientifically impossible being, and Pol Pot quite natural but very evil. Perhaps some commonality might be found in our reaction to these beings. Even here, however, there seems no common ground. Aristotle's natural deformities arouse pity and in extreme cases repulsion; Carroll tells us (his) monsters arouse fear and disgust; Freeland's evildoers arouse moral indignation. This suggests that "monster" has three distinct meanings - or to put it another way, the term "monster" can denote any one of three distinct concepts. (This is not to say that monsters bear a "family resemblance" to one another. Where are the "threads of similarity" that Wittgenstein required?)
2. Carroll and Freeland on Psycho and Peeping Tom
My interest here are the "natural monsters," Norman Bates and Mark Lewis. Because he takes a horror story by definition to evoke fear and disgust, Carroll thinks Psycho and Peeping Tom are properly tales of terror rather than tales of horror: "though eerie and unnerving, [they] achieve their frightening effects by exploring psychological phenomena that are all too human." But Norman Bates worries Carroll. Bates is, for Carroll, "a schizophrenic, a type of being that science countenances." Thus he is not a monster "technically speaking." Yet Bates "resembles the impure beings at the core of the concept of art-horror. He is Nor-man: neither man nor woman but both. He is son and mother. He is of the living and the dead. He is both victim and victimizer. He is two persons in one. He is abnormal, that is, because he is interstitial. In Norman's case, this is a function of psychology rather than biology." It is indisputable that Norman Bates is possible within the laws of nature. Robert Bloch, author of the novel Psycho, based Norman Bates on Ed Gein, the horrific serial killer of Wisconsin - or I should say on what he knew about Ed Gein. Suffice it to say that Gein's atrocities were far more horrible than Norman Bates's. However, if the impurities Carroll purports to find in Bates's psychology are fictional truths of Psycho - two persons in one; both living and dead; son and mother - then Bates is not (just) a schizophrenic consistent with science. He really is a "mixture of what is normally distinct" or a "categorically contradictory" being, hence truly (and not just apparently) a monster under Carroll's concept.
Carroll takes literally the criminal psychiatrist Dr. Richmond (Simon Oakland) who at the end of Psycho offers his diagnosis of Norman Bates to the gathered reporters. He's heard the story from Norman Bates's mother, he tells us; Norman Bates no longer exists. This, taken literally (which may not be how Dr. Richmond meant it), would imply that Psycho is a back-from-the-dead or spirit-possession film. Somehow, against what scientific knowledge countenances, Mrs. Bates's consciousness survived bodily death and has returned to reside in her son, eventually winning a psychic battle by extinguishing his mind. On such a reading, Norman Bates's body has incorporated his mother's consciousness and ultimately lost his own, a grotesque iteration of Locke's puzzle case: "For should the soul of a prince, carrying with it the consciousness of the prince's past life, enter and inform the body of a cobbler, as soon as deserted by his own soul, every one sees he would be the same person with the prince, accountable only for the prince's action ...."
But either the psychiatrist has got part of it wrong or he was speaking hyperbolically. Norman Bates isn't literally possessed by his mother. Nothing in Psycho signals that the afterlife is somehow involved. The Bates mansion is creepy but it is not a haunted house. Compare Psycho with The Possession of Joel Delaney (1972), a film in which it is clearly a fictional truth that Joel Delaney is possessed by the spirit of a recently deceased Hispanic friend (Delaney suddenly begins to speak Spanish, for one thing). Norman Bates, in contrast, is delusional. It is a fictional truth of Psycho that he thinks himself to be his mother. This is different from being his mother (in Locke's prince-cobbler sense). It is not, then, a fictional truth of Psycho that Bates is "interstitial." This makes him on Carroll's theory not a monster at all, a conclusion with which Carroll would agree, though it also subverts Carroll's attempts to explain why others take Bates to be a monster.
Freeland takes Hitchcock's Psycho and Michael Powell's Peeping Tom to be films that "permanently altered the face of the horror-film monster," not only by "turning him into the boy next door" but by inviting audience sympathy for him. Freeland is right about audience sympathy. Hitchcock, scriptwriter Joseph Stefano, and Anthony Perkins conspired to turn Bloch's Norman Bates, a fat, middle-aged, bespectacled, charmless, and alcoholic mess into the young, handsome, shy, philosophical motel keeper movie-goers have come to know. (The film also tones down the crimes of the novel's Bates who decapitates Marion Crane in the shower.) After the abrupt dispatch of Marion Crane (Janet Leigh), Psycho's audience turns to Bates as an object of sympathetic concern. We want to sympathize with him - because of his loneliness, his boyish good looks, the burden he has in caring for a destructive mother, the fact that he seems his mother's victim, and most of all because he seems to have been deprived of a real life.
Mark Lewis, also young, shy, intelligent, and handsome, is positioned as a victim of his father, a scientist who conducted cruel experiments on his little boy to record on film and in books the boy's reactions to fear. We see Mark's violence as a direct cause of his father's cold cruelty - his father's voice is still heard on film saying, "Don't be a silly boy. There's nothing to be afraid of." While we must wait until nearly the end of Psycho to discover that it was Norman not Mrs. Bates who murdered Marion Crane and the private detective Arbogast (Martin Balsam), we know right from the beginning of Peeping Tom that Mark Lewis has murdered at least one woman, a prostitute, while filming the deed. (We must wait until nearly the end to discover that he also uses a mirror to reflect the fear of his victims back to them.) Our sympathy for Mark Lewis is thus even more compromised than our sympathy for Norman Bates (which might in part account for the extremely hostile reception accorded Peeping Tom upon its initial release).
However, it is not clear whether Bates or Lewis should count as monsters under Freeland's concept. Are they evil? If this means, Are they intentionally evil, doing evil for evil's sake, then the answer is, arguably, No. Bates is, by the time of Marion Crane's arrival at the motel, very, very ill. He kills in the persona of his mother and when back to himself really seems not to realize what "he" has done. "Mother! Oh God, what-? Blood, blood! Mother!" Norman cries from the house after Marion's murder. He rushes to the motel, and nearly retches from the sight. Mark Lewis is more in control of himself - at least he doesn't go into the trance-like state Norman Bates does when killing - yet as he himself recognizes, he is in the grip of a compulsion which he calls "scoptophilia".
Both Norman Bates and Mark Lewis are acting under internal psychological compulsion. They thus resemble repetitive exhibitionists and kleptomaniacs, though they are, of course, far more dangerous. The compulsive can't fully control his actions, even though he realizes that what he does is irrational and possibly dangerous to himself and others. One writer on legal insanity, Joel Feinberg, gives an example of a kleptomaniac high school student who had in his possession stolen property including "14 silverine watches, 2 old brass watches, 2 old clocks, 24 razors, 21 pairs of cuff buttons, 15 watch chains, 6 pistols, 7 combs, 34 jack knives, 9 bicycle wrenches, 4 padlocks, 7 pairs of clippers ..." and so on. Norman Bates is too divorced from his mother persona to recognize his illness, let alone control his actions. Mark Lewis, however, is aware of his scoptophilia. He even briefly consults a psychiatrist, a consultant on the movie set where Mark works, about curing his condition, and is breezily told that it would take three years of thrice-weekly analysis. Mark has some control over his actions, but not enough. He is in the grip of impulses he barely understands and which overtake him now and again, which is part of how Feinberg demarcates compulsive from normal behavior: the sick or compulsive's motives appear unintelligible both to us and to the compulsive himself; his motives, further, are irrational in that the actions performed are often done at great risk with little gain for the actor; that is, they do not further any of the actor's other interests. In sum, neither Norman Bates nor Mark Lewis have enough intentionality at the root of their murders to clearly count as evil; and it is this lack of control that allows them some moral exculpation for their terrible deeds.
Freeland finds Peeping Tom "problematic" because of "how attractive it makes the serial killer at its center... Everything about him is explained so that we can understand his motives and see why he does what he does... So is he evil?... Reason tells us that Mark is an evil perverted killer, but our emotions might be guided by Helen's responses of sympathy and empathic interest." This, however, side-steps an important question, and disregards the information conveyed by our emotional responses. If Mark Lewis deserves some moral exculpation for his actions, and I don't see how one can avoid that conclusion, then he is in that sense not evil - or his evil is substantially lessened, probably below monster-level evil. Mark is certainly dangerous, pathetic, perhaps deserving of incarceration and treatment. But is he as evil as a monster needs to be? Compare Mark Lewis with other, later, cinematic serial killers who act with far more intent and control, such as Freddie Krueger of Wes Craven's Nightmare on Elm Street (1984) and Hannibal Lecter of Jonathan Demme's The Silence of the Lambs (1990). Lecter has an almost preternatural combination of reason in control of horrible desires. Even Frank Booth (Dennis Hopper) of David Lynch's Blue Velvet (1986), who is obviously crazy, still acts with control and intent. He would probably count as a monster on Freeland's view, but Booth is different from Norman Bates and Mark Lewis. Booth is evil, but Bates and Lewis are pathetic.
The problem Freeland runs into with Mark Lewis is a fuzziness in her concept of evil. How evil must something be before it counts as a monster? Hitler, Stalin, and Pol Pot had millions murdered. They were acting intentionally and certainly have achieved monster-level evil. Must someone be a murderer of millions before he is a monster? Or would only a handful count? Are the kindly old aunts of Frank Capra's Arsenic and Old Lace (1944) monsters? Suppose a person stops at one murder. Would this exonerate him from monster-hood, or would it depend on how he killed his victim? What about a man who bilks people out of their retirement savings? Is he a monster? Is a professor who plagiarizes from his graduate student? Is someone who jumps the turnstile at the subway a monster? Must a person be sane (in control of his actions) before he can count as a monster? If we don't require intentional evil, should rabid dogs and hurricanes count as monsters?
Ironically, Carroll who wants to avoid the conclusion that Norman Bates is a monster actually invites it, though for the wrong reasons; and Freeland who believes Mark Lewis to be a monster probably cannot make good her claim in light of the preceding objections.
3. Bates and Lewis as Aristotelian monsters
I think that Norman Bates and Mark Lewis are monsters. But there is no firm justification for this fact in either Carroll's or Freeland's views. One will find one in Aristotle's concept of monstrosity - or I should say, in Aristotle's view modified. Aristotle, of course, believed that purpose was pervasive throughout nature, an idea that today has little backing. The idea of an Aristotelian monster, however, needs only the concept of the normal development of members of a species and of individual deviations from this normal development. However, individuals can deviate a bit from the normal development of their species without turning into monsters - for example, a dog with one gray eye and one brown eye. An Aristotelian monster, then, is an individual that deviates strikingly from the normal development of a species - that is, who deviates sufficiently to elicit our pity and possibly revulsion.
Norman Bates and Mark Lewis are such individuals. They are psychological analogues of the two-headed calf. We expect that the normal development of a human being will produce a person with an autonomous will, someone capable of acting on desires and values that are freely chosen. We think that normal people should "be themselves," as we say. Yet Norman Bates suffers from an extreme form of maternal identification compounded by severe schizophrenia. He is driven to behave as he thinks his mother would, though when himself he recoils from what "she" has done. And his madness is manifested in such macabre ways (not only does he murder young women as his mother, he keeps his mother's preserved corpse as a housemate).
We also expect that in the course of normal development a human will come to be a person whose sexual drive when directed at another is aimed at mutual pleasure. But Mark Lewis is afflicted not just with voyeurism but with a perversion of voyeurism. Lewis doesn't want to watch naked women or people having sex like a normal voyeur; he wants to watch women watching themselves as they're murdered. After meeting Helen, he doesn't want to do these things any more and briefly seeks treatment, though his compulsion reasserts itself and suicide seems his only other option.
Besides Norman Bates and Mark Lewis, other fictional Aristotelian monsters include the cast of Tod Browning's Freaks (1932); John Merrick of The Elephant Man (1980) and the deformed baby in Eraserhead (1977), both by David Lynch; and the creepy twin gynecologists of David Cronenberg's Dead Ringers (1988). Certainly, these unfortunates are striking deviations from the human norm.
Aristotelian monsters differ from Carroll's monsters in that Aristotelian monsters are possible in the real world while Carroll's monsters are impossible in the real world. The giant ants of Gordon Douglas's classic Them (1954), mutated by radiation from bomb tests, and the attacking birds of Hitchcock's The Birds (1963) are not simply striking deviations from the normal development of ants and birds. They are scientifically impossible deviations (and in any case, elicit fear not pity). As Hitchcock's amateur ornithologist points out in the seaside cafe: gulls and crows do not attack humans, different species of birds never flock together, birds are not intelligent enough to plan attacks, and so on.
Clearly, Aristotelian monsters cannot be entirely fictional, for there must be an existent species - a real norm outside the fiction - against which a fictional being exhibits deviant development. Man-eating blobs, brain-stealing pods, werewolves, and other entirely fictional monsters cannot be mistakes of nature for they have no counterparts in nature from which they deviate. The acid-drooling dinosaur-like beast of Ridley Scott's Alien (1979) is probably a perfectly developed example of its (fictional) species (certainly it behaves as the robot-scientist aboard the spaceship expected). Ridley's creation is a monster on Carroll's concept for it elicits fear and disgust, but it is not an Aristotelian monster.
Carroll and Freeland each have a plausible concept of monstrosity; they err only by insisting that theirs is the unique concept of monstrosity (if indeed they do so insist). No one concept of monstrosity fits all monsters, and I've argued that only Aristotle's can account for the monstrosity we sense in Norman Bates and Mark Lewis. It should be pointed out that Carroll's and Freeland's main target is the definition of the genre of horror. Their concepts of monstrosity are designed to fit that project. For each writer, a horror story is by definition one that includes a monster. My sense is that only the Aristotelian monsters that arouse pity and revulsion will make a story into one of horror, and this because revulsion is an emotion similar to horror. This is why The Elephant Man is not a horror film (the John Merrick character arouses pity but not revulsion), and why Psycho and Peeping Tom are horror films. The character of Norman Bates and Mark Lewis initially arouse sympathy and perhaps pity, but then at the end elicit revulsion.
 Aristotle, Physics II.8, trans. R. P. Hardie and R. K. Gaye, in Richard McKeon, ed., The Basic Works of Aristotle (New York: Random House, 1941).
 Noël Carroll, The Philosophy of Horror or Paradoxes of the Heart (New York: Routledge, 1990), p. 40.
 Ibid. p. 27. I have abbreviated Carroll's definition.
 Ibid., pp. 32-33. Carroll acknowledges his debt to Mary Doug-las, Purity and Danger (London: Routledge and Kegan Paul, 1966).
 Cynthia Freeland, The Naked and the Undead: Evil and the Appeal of Horror (Boulder, CO: Westview Press, 1999), pp. 10, 11.
 Carroll, Philosophy of Horror, p. 15.
 Ibid., pp. 38, 39.
 Stephen Rebello gives an account of Ed Gein's crimes. Robert Bloch worked from the 1957 newspaper accounts, which reported Gein's murders and alleged cannibalism but suppressed Gein's other even more horrific deeds. Alfred Hitchcock and the Making of Psycho (New York: St. Martin's Griffin, 1990), Ch. 1. The killers Jame Gumb ("Buffalo Bill") and Hannibal Lecter of The Silence of the Lambs (1990) are, jointly, closer to the real Ed Gein.
 John Locke, An Essay Concerning Human Understanding (1690), II.xxvii.15, "Of Ideas of Identity and Diversity." From the Everyman edition abridged and edited by John Yolton (London: J. M. Dent and Sons, 1977).
 Freeland, The Naked and the Undead, p. 161.
 Joel Feinberg, "What Is So Special About Mental Illness?" in Joel Feinberg and Jules Coleman, eds., Philosophy of Law, Sixth Edition (Belmont, CA: Wad-sworth/Thompson Learning, 2000), pp. 677-678. The kleptomaniac and his list of stolen property is drawn from an actual Iowa case, State v. McCullough (1901).
 Ibid., pp.678-681.
 Freeland, The Naked and the Undead, pp. 167-168.
 I wish to thank several people who commented on this paper: Herb Granger, Mark Huston, Sean Stidd (who, I believe, suggested the title of this paper), and three anonymous referees of this journal.
Robert Yanal, Wayne State University