What is it like to be a Robot?
The Depiction of Artificial Intelligence in Science Fiction Films

A Dissertation for the MA in History of Film and Visual Media
David Mitchell
Summer 2003
Birkbeck College


Contents

List of Illustrations

Preface

Acknowledgements

Introduction

The Science of Machine Intelligence and Robotics

Science Fiction as a Genre

Science Fiction Literature versus Science Fiction Films

A Taxonomy of Intelligent Artificial Creatures

The Criteria to be used for selecting Films

Guiding Propositions

1890-1950 - Science or Magic?

Futura, the Machine-Woman

1950-1968 - Master or Slave?

Gort, the alien robot

Robby, the Comic Relief

The Winds of Change

Cragis, the Clicker

Alpha 60, death by poetry

1968-1973 - Rebellion

HAL, friend or enemy?

Colossus, the ruler of the world

1973-1977 - Robots as Playthings

The Robot Gunslinger

Joanna, Bobbie and Carol, the Stepford Wives

Bomb #20, the solipsist

Proteus, the prisoner

1977-1982 - The Cinema of Science Fiction Attractions

R2-D2 and C-3PO, the 'rude mechanicals'

Ash, the 'corporation man'

1982-1987 - What's the Difference?

Roy Batty, the prodigal son

Max, the auto-didact

Number 5 is alive

Ulysses, Mr Right

RoboCop, my name is Murphy

1987-2003 - Man versus Machine

Agent Smith, the cure

Andrew, the Bicentennial Man

David, the boy who loved his mother

Conclusions

Appendices

Appendix I - Machine Intelligence in Literature

Appendix II - Asimov's Laws of Robotics

References

Films

Journal Articles and Books

Web Sites

Footnotes

List of Illustrations

1. Honda's Asimo Robot

2. Brigitte Helm is refreshed during a break in filming Metropolis

3. GORT 8" Mechanical Collectible Figure

4. Blueprints for Robby the Robot

5. Bowman dismantling HAL's brain

6. Robot Vision in Westworld

7. Design sketch of Number 5 by Syd Mead

8. RoboCop Design Sheet

9. Neck and Shoulder designs for Bicentennial Man

10. Tik-Tok and Dorothy


Preface

This dissertation examines a set of science fiction films from the last one hundred years. The films in question all feature intelligent machines, most of them what are commonly called 'robots'. (1) It is worth pointing out in advance that the choice of films has been highly selective. I have deliberately excluded:

The last exclusion perhaps needs more justification. Intelligent robots have featured in several TV science fiction series. Marvin the Paranoid Android in The Hitchhiker's Guide to the Galaxy, Kryten in Red Dwarf, and Bender in Futurama are all interesting characters worth discussing in detail. However television has always treated science fiction in a different fashion to films, as I have argued elsewhere. (2) In any case, the essence of these series - humour - sets them apart from almost all the films I shall be discussing. If space and time were no object I would have dealt with them too, but a line has to be drawn somewhere.

I have provided two brief appendices. The first looks at the related subject of the depiction of machine intelligence in literature while the second discusses Isaac Asimov's 'Laws of Robotics'.

Acknowledgements

I should like to thank Liz Meredith of Beckmann International for providing me with a videotape of the Channel 4 programme, Android Prophecy, (3) Bob Fisher of the Department of Informatics (the erstwhile Department of Machine Intelligence) at the University of Edinburgh for useful suggestions and permission to quote from his extended web-essay and Elizabeth Kingsley for her incisive reviews of several films.

Above all, I would like to thank William Grey Walter (1910-1977), whose book, The Living Brain, (4) made a deep and lasting impression on the schoolboy who read it nearly fifty years ago.


Introduction

Larry:
They don’t send robots, Bob, for the simple reason they haven’t invented one yet. The day comes when they can build a robot to do what we do and make it work, then that’s exactly what they’ll do, precisely.
The Big Kahuna (5)

Robots, androids and intelligent computers have appeared in many science fiction films. This dissertation examines the depiction of such electro-mechanical, synthetic creatures. (6) In what ways have they changed over the years? How do they illuminate the society of the period in which they appeared? What do they tell us about ourselves, about what it means to be human?

Thomas Nagel's famous paper, 'What is it like to be a Bat?', (7) examines the nature of subjective experience - of what cognitive scientists call 'qualia'. (8) At one point Nagel notes:

My point is ... that even to form a conception of what it is like to be a bat (and a fortiori to know what it is like to be a bat) one must take up the bat's point of view.

I have used the title of his paper as a starting point because taking up the robot's 'point of view', aligning oneself with it, (9) seems to me to be a key step in extracting, from their depiction, clues about the feelings and attitudes we have towards intelligent machines.

The Science of Machine Intelligence and Robotics

Before we can understand 'fictional science', we need some grounding in reality. Machine or Artificial Intelligence (AI) has its roots in the early post-war development of the digital computer. The brilliant British mathematician Alan Turing was one of the key figures in this work and his landmark 1950 paper, (10) describing what he called the 'imitation game' (now usually called the 'Turing Test'), was the starting point for the field. As Turing's biographer, Andrew Hodges, says:

The most fundamental statement of Turing's thought in this paper is that the operations of the brain must be computable. The famous Test is secondary. Furthermore, the main point of his paper was to put forward constructive arguments for how machine intelligence should be achieved. (11)

The 1950's were an optimistic time for those working on machine intelligence. That optimism gradually faded as the key problems involved proved much more intractable than was originally thought. As late as 1968 one of the AI pioneers, John McCarthy, made a famous bet with chess-player David Levy that a computer would beat him within ten years. He lost the bet and it was not until 1996 that a chess program, IBM's Deep Blue, won a game against the current world champion. Today, while 'expert systems' can outperform humans in limited domains, such as medical diagnosis, backgammon or chess, we are still a long way from building machines that display 'common sense'.

The construction of a true machine intelligence, one that is conscious and self-willed, depends upon the truth of what is known as the 'Strong AI' hypothesis. There is still a deep dispute about whether this will ever be possible, though few doubt that computers will eventually be able to perform many of the tasks that we now think of as requiring human intelligence.

Fig. 1. Honda's Asimo Robot

Today's robots are equally restricted in their capabilities compared with human beings. Locating and identifying individual objects laid out in a messy fashion on a table, where some are partially occluded by others, is a task that humans find easy, but is still beyond the state of the art in computer vision research.

It is only in the last few years that biped robots have been developed that can reliably walk on uneven surfaces, negotiate stairs and avoid obstacles, all tasks that humans perform without difficulty. (12)

If genuine machine intelligence and versatile robots are not yet with us in real life, they have a long fictional history. (13) Nearly two thousand four hundred years ago Aristotle, considering the ethics of slavery, wrote:

For if every instrument could accomplish its own work, obeying or anticipating the will of others, like the statues of Daedalus, or the tripods of Hephaestus, which, says the poet, "of their own accord entered the assembly of the Gods" if, in like manner, the shuttle would weave and the plectrum touch the lyre without a hand to guide them, chief workmen would not want servants, nor masters slaves. (14)

Aristotle's stance was that, from a moral point of view, a mechanical slave was preferable to a human one. We find the same attitude, subtly modified, towards modern fictional robots, such as those in Capek's R.U.R., in the robot stories of Isaac Asimov and in many science fiction films. The attitudes that humans adopt towards intelligent robots are one of the key themes we shall explore. Of course the converse, the attitudes that intelligent robots adopt towards humans, are equally relevant to our discussion.

Science Fiction as a Genre

This dissertation is concerned with science fiction films. Defining science fiction is notoriously difficult, (15) but as Frederik Pohl has said:

Perhaps we cannot satisfactorily say what science fiction is, but still we may be able to identify some of its distinctive traits by trying to assess what it is that science fiction uniquely does. (16)

Pohl went on to characterise science fiction as a 'literature of ideas', making the point that 'ideas are not just discussible, they demand discussion'. This focus on ideas sets science fiction literature apart from other fictional genres. While fiction of all genres holds interesting mirrors up to the times in which it is made - one thinks of the changing moods and concerns of westerns over the years for example - the foregrounding of ideas in science fiction exaggerates and emphasises the effect. Of course, 'ninety percent of science fiction is crud', (17) but even bad art can reveal important truths.

One other key point that needs emphasising is that most science fiction is not predictive. (18) It is instead usually speculative - supposing such and such were to become possible, what would the consequences be? There is also a strong allegorical thread, projecting current issues into the future. Both of these modes have led to science fiction affecting the future rather than predicting it. (19) The fact that artificial intelligence may never be realised in practice does not make fiction about it nugatory.

Science Fiction Literature versus Science Fiction Films

There is of course a marked contrast between science fiction literature on the one hand and science fiction films on the other. Discussing American science fiction films of the 1950s, Susan Sontag claimed:

Science fiction films are not about science. They are about disaster, which is one of the oldest subjects of art. (20)
Sontag is not alone in seeing a difference between science fiction cinema and literature. (21) In a famous essay, Richard Hodgens wrote:

Science fiction films, with a few exceptions, follow different conventions [from its literature]. The premise is always flatly impossible. Any explanations offered are either false analogies or entirely meaningless. The character who protests "But that's incredible, Doctor!" is always right. (22)
Sontag was writing in 1965 of course, and Hodgens even earlier, but Per Schelde, writing just ten years ago, agrees:

... despite appearances, sf movies and sf literature have little in common and appeal to very different audiences. (23)

Schelde sees science fiction films as a kind of 'modern folklore'; they 'avoid being intellectual or speculative' and 'focus on the effects of science'. As he says:

Folklore demands a willingness to suspend, temporarily, reality and logic. The audience needs to believe ... The reward for belief, for suspending critical judgement, is a flight of fantasy. (24)

Scott Bukatman takes a similar view:

The meaning of SF films is in their visual organization, and in their inevitable attention to the act of seeing. (25)

As Peter Ruppert says, quoting this passage, there is a case for viewing science fiction films as:

a cinema of attractions and exhibitionism, a genre based on special effects that hearken back to Méliès, rather than a cinema of psychological narratives and identification. SF movies ... transcend their narrative content by projecting alternative worlds through the reflexive specularity of special effects; they envision mock futures – new, strange, impossible – defamiliarizing our science fiction lives, reflecting them back to us in more hyberbolic form. (26)

For all these reasons the majority of science fiction films thus show little regard for the principles at the heart of the literary genre, although occasional films do rise above the rest. The devaluing of ideas in favour of thrills is seen not just in the pseudo-science and the tendency to accentuate horror, but also in the frequent use of 'idiot plots' (27) to heighten suspense while losing credibility. When we look at films, we shall pay most attention to those aspects that reflect and respect the values of science fiction literature.

I would argue that there is a connection between science fiction literature and films, one that critics seem to have overlooked because it is obscured by the long time lag. (28) Most science fiction films, rather than being based on contemporary literature, tend to rework ideas from the science fiction literature of an earlier generation. Thus Metropolis replays themes from Mary Shelley's Frankenstein and the science fiction 'B Movies' of the 1950's draw heavily on stories and illustrations not from 1950's science fiction, but from the pulp magazines of the 1930's. We shall meet other examples of this long-distance borrowing later.

Finally it is worth pointing out that where robots in literature are merely described, robots in films have to be designed. The visual appearance of a robot (a mixture of actor, costume and prop) plays a key role in mediating our responses to it. We shall have more to say on this later.

A Taxonomy of Intelligent Artificial Creatures

TV Salesman:
Delos is filled with robots scientifically programmed to look, act, talk and even bleed just like humans do.
Westworld
Max404:
An Android is a robot that possesses all the physical characteristics of a human being, down to the smallest detail.
Android
Bob Morton:
Let me make it real clear for you; he doesn't have a name, he has a program, he's a product.
RoboCop

It would help if we had a clear, well-defined taxonomy to distinguish the different types of creatures - 'robots', 'androids', 'replicants', 'cyborgs', intelligent machines and so on that we shall be discussing. Unfortunately neither Schelde nor Telotte (29) make any attempt to define or categorise the creatures they discuss. Some of the terms, such as 'android', are used in ambiguous ways while others, 'replicant' or 'droid' for example, have no accepted definition outside the films in which they appear. For our purposes, it makes sense to follow Robert Fisher (30) and divide such creatures into three broad categories:

  1. what Fisher calls 'Pure Artificial Computation Agents' - intelligent computers and robots
  2. what Fisher calls 'Pure Biological Computation Agents' or 'androids' - manufactured biological beings
  3. what Fisher calls 'Hybrid Computation Agents' or 'cyborgs' - hybrids of man (or woman) and machine

We need to examine each of these categories a bit more closely.

Intelligent Computers

Creatures in this category have no biological components at all. The computer may be independently mobile, in which case we will refer to it as a 'robot', (31) or it may be sessile, though equipped with sensors and effectors to observe and manipulate the world, as HAL is in 2001. Creatures in this category - conscious, self-willed machines - are the central concern of this dissertation. Humanoid robots, such as Max in the film Android, are sometimes referred to as 'androids', (32) but this is a confusing term, since it is also applied to the next category. The contraction 'droid' is used in the Star Wars series to refer to robots even when, as in the case of R2-D2, they have no humanoid features or capabilities at all.

In many films featuring robots, they lack will and act purely under the control of some human master, as Robby seems to in the film Forbidden Planet. Indeed they sometimes act entirely automatically. The word 'automaton' is sometimes used in science fiction as well as the technical literature to describe such beings, though rarely if ever in films.

Manufactured Biological Beings

These creatures have a biological, rather than electro-mechanical, basis but are constructed, or grown 'in vitro', rather than being born. Today we would use the term 'genetic engineering' to describe their manufacture. Though their brains are constructed too, they are usually modelled on, and largely indistinguishable from, human brains, which makes them less central to this dissertation. The 'replicants' in Blade Runner would seem to fall into this category, though the prologue of the film refers to them as 'advanced robots' and more than one character refers to them as 'machines'. Fisher uses the term 'android' to refer to this category, but as pointed out above, this is misleading. For example, the characters Max and Cassandra in the film Android are referred to as androids but have no biological components at all.

Cyborgs

A 'cyborg', a contraction of Cybernetic Organism, is a hybrid of man (or woman) and machine. The eponymous RoboCop is the obvious exemplar here. In the TV series The Six Million Dollar Man and The Bionic Woman the machine parts endow additional strength and physical capability while the human provides the intelligence and will as well as much of the body, which makes them largely irrelevant to this dissertation. Occasionally things are more complicated, as they are with RoboCop, when the machine part provides (some) mental functions as well.

The Criteria to be used for selecting Films

As a general rule, we shall focus on films that feature creatures in the first category - films in which machine/artificial intelligence is displayed either by a robot or by a computer. We shall generally exclude films in which the 'intelligence' is either lacking or is provided by a purely human brain. This still leaves us with a large number of films to consider, (33) far too many to examine in detail, so further selection will be necessary.

As we shall see, rather than a gradual and continuous shift in focus there have been long periods of relative stasis, punctuated by films that mark major changes in attitude. It is these landmark films on which we shall focus our attention. Accordingly we shall pursue a broadly historical approach, looking at films in chronological order. There are of course alternatives; films can be grouped by the themes and issues that they examine, the approach followed by both Schelde and Telotte. This may have much to recommend it, but in my view makes it much harder to see the connections between the films and the time and society within which they were created.

Guiding Propositions

There are, it is sometimes argued, two sorts of fiction. On the one hand, there is the sort that holds up a mirror to the times in which it is produced, capturing the zeitgeist and revealing current opinions and desires. Fiction of this type can obviously help us understand what it was like to be there.

On the other hand, some fiction offers an alternative, opposite set of dreams. When societies grow dark with recession and war, such fiction, with its 'happy endings', is often labelled 'escapist' but when societies are going through relatively calm and prosperous times, it is instead disturbing, questioning and satirical, questioning our complacency.

Science fiction literature has a foot in both camps. While a novel like 1984, written in 1948, clearly echoes the bleak world of cold war Britain, Pohl and Kornbluth's The Space Merchants, published in the boom times of the American 1950s, foretells a polluted world in which trees are so valuable that wood is used to make jewellery rather than furniture. Science fiction films are similarly divided, though more tend to fall into the first category, echoing the times rather than counterpointing them.

Although much of what follows takes the form of an historical survey of a group of films, there are two propositions that I wish to put forward - propositions which the survey will test, and hopefully confirm. Science fiction films tend to be gloomy and dystopic, at least about technology. The technology we are interested in, machine intelligence, raises additional fears, (34) leading to the first proposition:

Of course we may never see truly intelligent machines in real life, but when presented with fictional versions people can and do develop and express attitudes and feelings. These feelings, pace Freud, do not merely arise out of our 'fear of the uncanny'. (35) As computers have pervaded everyday life, our feelings about them have become based on our own experiences. Indeed, I would argue that, since 1950 at least:

Of course, one can argue the reverse - that attitudes towards real computers are influenced by their fictional depiction in science fiction films. While this may be true in cases where the film is of, or ahead of its time - as is the case for 2001: a space odyssey for example - it is unlikely if the film has borrowed ideas from a much earlier period (which as I have already pointed out is more often the case).

Before 1950, there were no computers, real or fictional, (36) and as a consequence there were very few fictional intelligent machines. We shall meet those isolated examples in the next section.

Science or Magic?

1900-1950

Robots first appeared in films in the very early days of cinema, (37) but these early automata raise few issues that need concern us. Until the 1950s robots in films had little or no intellectual or emotional characteristics. This is not altogether surprising - before the advent of digital computers it was difficult to entertain seriously the idea of a thinking machine. Accordingly robots were seen as purely mechanical - mere automata. Writing of 'simulacra' (copies which lack originals), Baudrillard makes the point that we can:

clearly mark the difference between the mechanical robot machines, characteristic of the second order [of simulacra], and the cybernetic machines, computers, etc., that, in their governing principle, depend upon the third order. (38)

True, there were robots in several Republic and Universal serials of the 1930s, but these were laughable 'wind-up tin cans'. There was also the Tin Woodsman in The Wizard of Oz (1939), although he was not a robot; he was originally a normal human being who had his various organic parts gradually replaced by metal ones, making him an extreme example of a cyborg. (39)

Futura, the Machine-Woman

I ordered machine men from you, Rotwang, which I can use at my machines. No woman . . . no plaything.
Metropolis (40)

There was of course one notable oasis in this desert of machine intelligence - Futura, the Machine-Woman - who appeared in Fritz Lang's Metropolis (Ufa, Germany, 1926), the first great science fiction film. It is worth trying to explain why this speaking robot appeared when it did.

Metropolis was based on a novel written by Thea von Harbou (Lang's then wife) between 1922 and 1923. It is not clear if von Harbou was influenced by Karel Capek's play R.U.R., which had its premiere in Prague in 1921, but it seems highly likely. Capek's play was widely performed around the world and attracted considerable media attention. An essay by Peter Wollen contains an interesting dissection of this period. (41) Wollen considers the way the American development of assembly-line mass production - Fordism - with its emphasis on Taylorian ergonomics, encouraged a view of workers as predictable, regulated machines. Discussing Gramsci's admiration of Fordism and Huxley's critique of it in his dystopic science fiction novel, Brave New World, Wollen cites the 'imagery and literature about robots which springs up in the 1920s' and goes on to analyse R.U.R.:

The robots are simplified and ergonomically improved models of humans... Thus emotions are rejected as unnecessary and consequently equated with ornaments. The robots feel neither pleasure nor pain: they have neither libido nor affect. (42)

Capek's R.U.R has never been filmed. (43) His robots were not in fact mechanical, but manufactured biological beings. As Domain, one of the human characters in the play, explains:

On this occasion he [Professor Rossum] attempted by chemical synthesis to imitate the living matter known as protoplasm, until he suddenly discovered a substance which behaved exactly like living matter, although its chemical composition was different. Rossum wrote the following day in his book: "Nature has found only one method of organizing living matter. There is, however, another method more simple, flexible, and rapid, which has not yet occurred to nature at all. This second process by which life can be developed was discovered by me today. (44)

The robots, encouraged by a human, eventually revolt, although as the noted Slavonic scholar, René Wellek, pointed out in an essay published in 1936, this is not the only way of looking at it:

If we, however, examine the play in cold blood, the fissures in the structure and the gaps in the argument become obvious: the robots which are conceived as men-machines without soul or feeling, are changed during the play by a sleight of hand into real men. There is no revolt of robots, but a revolt of oppressed men: one race of men is simply dethroned by another and the whole story loses its point. It all comes to an attack on human ambition, and a recommendation of simple humanity: of love, laughter, and tears. (45)

Others have observed that Capek is writing as much about the workers' revolution in Russia as about the robots' uprising in Rossum's factory. What is true, in any case, is that Rossum's robots were seen as a source of cheap and efficient labour, which would free humanity from drudgery. This theme comes straight from Aristotle and as we shall see reappears, in different forms, in many films featuring robots.

The technologies that make today's modern digital computers possible were then beyond anyone's imagination but, as Wollen shows, Carnap and other philosophers of the Vienna school were 'eliminating metaphysics through logical analysis of language' and this project was 'conceptualized parallel with the technological demands of modern industry'. (46) That early work on symbolic logic was in turn to lead to Godel's Incompleteness Theorem and thence to Alan Turing's ground-breaking 1936 paper in which he describes a universal computing machine:

We may compare a man in the process of computing a real number to a machine which is only capable of a finite number of conditions .... (47)

In this paper, Alan Turing founded modern computer science and opened up what would now be called the cognitive sciences, as well as settling a fundamental question within mathematics. As Wollen points out, 'the computer, the reasoning machine dreamed of by Leibnitz, is finally realised on the model of the assembly line, processing not material commodities but abstract units of information'. (48)

We see much of this echoed, indeed foreshadowed, in Metropolis. Vivian Sobchack argues that:

The ultimate horror in science fiction is neither death nor destruction but dehumanization, a state in which emotional life is suspended, in which the individual is deprived of individual feelings, free will, and moral judgment...this type of fiction hits the most exposed nerve of contemporary society: collective anxiety about the loss of individual identity, subliminal mindbending, or downright scientific/political brainwashing. (49)

Sobchack was writing about science fiction films from 1950 onwards, and her book does not mention either Metropolis or Lang, but her comment is nonetheless apposite. Metropolis is a complex film however, full of ambiguity. As Telotte says:

It seems to speak in two voices. One the one hand, it talks about the consequences of a society given over to the forces of technology and production; but on the other, it finds much of its attraction in the vision of those forces, that is, in its seductive images of a futuristic, highly technologized society. (50)

Andreas Huyssen, in a famous article on the film, examines this ambiguity in some detail:

Eggebrecht and Kracauer were certainly correct in relating Lang’s representation of technology to the machine-cult of the 1920s which is also manifest in the literature and the art of Neue Sachlichkeit. In my view, however, it is not enough to locate the fi1m within the parameters of Neue Sachlichkeit only. The simple fact that stylistically Metropolis has usually and mainly been regarded as an expressionist film may give us a clue. And indeed, if one calls expressionism's attitude toward technology to mind, one begins to see that the film actually vacillates between two opposing views of modern technology which were both part of Weimar culture. The expressionist view emphasizes technology's oppressive and destructive potential and is clearly rooted in the experiences and irrepressible memories of the mechanized battlefields of World War 1. During the 1920s and especially during the stabilization phase of the Weimar Republic this expressionist view was slowly replaced by the technology cult of the Neue Sachlichkeit and its unbridled confidence in technical progress and social engineering. Both these views inform the film. (51)

For us, the prime feature of the film is the robot, Futura, transformed by Rotwang into the 'False Maria'. Though referred to as 'Maschinenmensch', Lang's robot is undeniably feminine, as is the robot in the novel by Thea von Harbou on which the film is based. Huyssen's examination of this choice leads him to conclude, as Peter Wollen says, that:

... the film revolves around the displacement of the fear of technology-out-of-control on to that of (female) sexuality-out-of-control. (52)

Interestingly, William Routt claims that:

Huyssen has made the classic error of categorically asserting authorial intention [that Fritz Lang does not feel the need to explain the female features of Rotwang's robot] on the basis of something which is only true for most of those prints in circulation at the time of writing. (53)

though he goes on to say that "Huyssens' overall argument is not really affected by this gaffe". It is instructive to contrast Huyssen's detailed psychological analysis of the 'Machine-Woman' with Sobchack's view of the creatures of science fiction films:

Since horror film emphasizes moral conflict, the Monster must be a significant and personalized antagonist. ... In the science fiction film, the Creature is less personalized, has less of an interior presence than does the Monster in the horror film. Usually we are given only form, physical attributes; the Creatures of science fiction usually lack a psyche... Our interest lies not in why the creature will do what it does, not in what it thinks or feels, but solely in what it will do and how it will do it - in other words, its external activity. Our sympathy is never evoked by a science fiction Creature; it remains always a thing. (54)

Sobchack's view echoes my own, at least as far as Metropolis is concerned. The film sheds little light on the interior life of Futura. Rotwang never claims to have endowed it with the power of speech, let alone self-will. (55) When we first see her he commands her by waving his hands. Until the transformation it is a mere machine, though in the world of Metropolis that makes it quite capable of replacing the workers we see toiling mechanically in the underground city.

As for the transformation, we are told nothing of what is transferred from Maria to the robot, but it is surely more than mere surface appearance - only after the transformation can the robot speak or move independently. Maria appears to become unconscious afterwards, so we are led to think something more than her looks have been transferred.

In both book and film Rotwang appears more as an alchemist or magician than a scientist - in his study we see a large pentagram on the wall behind Futura. Sobchack distinguishes science-oriented science-fiction films from magic-oriented horror film ones thus:

What distinguishes magic from science...is how much we know about the process and the product, how much we are told about the cause and effect. Dr. Praetorius in The Bride of Frankenstein (1935) is essentially a magician, a witch-doctor, because we don't understand or see how he did it; the film emphasizes only the result of an unseen process that we must take on faith and so that result is magical. (56)

Fig. 2. Brigitte Helm is refreshed during a break in filming Metropolis

Futura's transformation is magical, harking back to the Gothic horror of Mary Shelley's Frankenstein. (57) The robot has no innate machine intelligence, which explains why we feel no pain, pity or sadness when it is consumed by fire. We do not feel that a person is being destroyed, but a simulacrum that lacks any feelings, desires, will or consciousness. (58)

Futura may not be interesting as an example of machine intelligence, but she is interesting as an example of industrial design. Her 'look', brilliantly created by the sculptor, Walter Schultze-Mittendorf, was to influence the design of robots in many films to come, including C-3PO in Star Wars, the eponymous RoboCop and many others. (59)

After Metropolis we see no more interesting robots until the end of World War II, although Chaplin's Modern Times (1936) does present a humorous parody of elements from Metropolis and R.U.R.. In the 1930s, the science fiction film genre was regarded as suitable only for children and Saturday Matinee serials. (60) Science fiction fared little better on the screen during the 1940s; films were either rooted in the war or in an escape from it. Intelligent robots did re-appear in science fiction literature in the 1940s, in short stories and novels by Isaac Asimov and others, (61) but after Metropolis we have to wait another twenty-five years, until 1951, before another memorable screen robot appears.

Master or Slave?

1950-1968

The post-war period saw two developments that strongly influenced science fiction, particularly in America. First, nuclear secrets found their way to Russia, and the cold war intensified into an arms race involving the development and testing of nuclear weapons and guided missiles. Anti-communism, in the form of the House Un-American Committee led by Senator McCarthy, assisted by Junior Congressman Richard Nixon, grew into a major political force. Films reflected and refracted the paranoia and fears of the time, and American filmmakers produced a succession of horror films with a science fiction setting. (62)

Secondly, the primitive electronic calculating machines developed by the allies to assist in the decoding of encrypted enemy messages were gradually transformed into programmable digital computers by Turing, von Neumann, Eckert, Mauchly and others. As computers developed, so did work on Machine Intelligence, and, as we might expect, both found their way into science fiction.

The early computers of the 1950s were rarities and most were used for highly classified purposes. People in general had little idea of what they even looked like, since so few people had ever seen one. Given that they were often referred to at the time as 'giant electronic brains', it is not surprising that there was public confusion about whether such machines could think. Ignorance provided fertile ground for fear to grow.

Gort, the alien robot

Helen Benson:
Gort, Klaatu barada nicto!
The Day the Earth Stood Still

In his book on science fiction films, Per Schelde says:

It is often argued that SF, because it is a genre that exists in relative obscurity, has been allowed to produce messages that go against the dominant ideology. (63)

We see an excellent example of this in The Day the Earth Stood Still (20th Century-Fox, 1951). In the film a humanoid alien, Klaatu, accompanied by a giant robot, Gort, lands in Washington D.C. in a flying saucer. (64) Although they come in peace, fear and mistrust on the part of the American military lead to a violent confrontation. When Klaatu is shot for the second time, near the end of the film, (65) Gort retaliates by incinerating two soldiers with a 'death-ray', (66) but before he does further damage Helen Benson (Patricia Neal) utters the magic words and saves the Earth from being 'burnt to a cinder'.

The anti-war message of the film was extremely brave for the time. Sam Jaffe, who bore a striking resemblance to Albert Einstein, was deliberately cast as Professor Barnhardt, the 'greatest scientist of the time'. Einstein had publicly urged witnesses to refuse to testify before Congressional investigatory committees and had been denounced in the House as a 'foreign born agitator'. (67) The US Army disliked the theme of the film and refused to supply men and vehicles, so the Washington National Guard was used instead.

But it is the robot, Gort, which is our concern here. He (68) is the first film robot of any consequence since Futura in Metropolis. At the end of the film Klaatu explains the robot's function:

For our policemen, we've created a race of robots. ... In matters of aggression, we have given them complete power over us. This power cannot be revoked. At the first sign of violence, they act automatically against the aggressor. The penalty for provoking their action is too terrible to risk.

Fig. 3. GORT 8" Mechanical Collectible Figure

It is worth saying something about Gort's appearance. He is the only alien robot we shall meet - all the other film robots we shall discuss were supposedly designed and built by humans. (69) Given that, it is disappointing that he is so humanoid. (70) In Screening Space, Vivian Sobchack sees Gort as a 'perverse visualization of the mediaeval knight in shining armor'. (71) Designed by Addison Hehr, he is quite unlike most modern film robots, indeed he is closest to the 'tin men' of the serials of the 1930s and 1940s.

He appears to be made of seamless, featureless metal, except for his head with its menacing visor. He wears a curious pair of briefs, (72) his hands have no fingers and whenever he moves, he does so very slowly, his skin or suit flexing at the knees and elbows. All this is no doubt due to the fact that the suit was made from foam rubber covered with metallic paint. In an interview, Robert Wise, the director of the film:

still squirms slightly when he recalls one peculiar problem with the otherwise perfectly functioning robot. "We were particularly concerned with the ‘credibility’ of the [foam rubber] suit," admitted Wise, "because, as Gort walked away from the camera, you could see the rubber creasing on the back of his knees. We did everything we could to eliminate that, but there wasn’t any way we could avoid it. That was one of our major worries when we went to the sneak preview of the film – whether people would see the suit creasing and laugh. But no one did laugh. Throughout the film, Gort remained ominous, frightening, and quite convincing. (73)

Gort is tall, much taller than his humanoid companion. Described in the script as being 10' tall, he was actually 8' 2". The actor inside, Lock Martin, was, at 7' 7", the tallest man in Hollywood at the time. (74) Gort's height, coupled with his metallic body, give him an appearance of power and menace even when he is standing still. He becomes even more menacing when the visor slowly opens and reveals the pulsating lights inside.

Gort never speaks in the film. We know he understands when Klaatu speaks to him, but in the story on which the film is based, (75) Gort, or rather Gnut, as he is called there, does speak. (76) Indeed, he utters the shocking final words of the story:

You misunderstand. I am the master.

In the story Klaatu is merely Gnut's servant. The film shies away from this revelation of the relative positions of humanoid and robot, which would have been too dark for film audiences at the time. More importantly, it would have lessened the impact of the film's anti-war message. It is one thing to have Klaatu, a human-looking character, say at the end of the film:

Your choice is simple. Join us and live in peace. Or pursue your present course -- and face obliteration. We will be waiting for your answer. The decision rests with you. (77)

To have those words uttered by a robot would have made them sound less like a choice presented by a friend and more like an ultimatum issued by an enemy.

So the film presents Gort as a silent servant. He obeys commands given to him by Klaatu and Helen Benson and his other actions appear to be more or less automatic reactions to events. When he takes the dead Klaatu back to the flying saucer, he appears to communicate with, and take orders from, the same intelligence (alien or machine) with which we previously saw Klaatu arranging the electrical shutdown that causes the Earth to 'stand still'.

We are left then with little evidence for intelligence, consciousness or will. As Klaatu says, their robots 'act automatically against aggressors'. Like Futura, Gort does not appear to be a person, capable of making decisions and acting independently. Instead he seems a determined, but also deterministic machine.

Robby, the comic relief

Robby:
I rarely use it [oxygen] myself, sir. It promotes rust.
Forbidden Planet

Five years after The Day the Earth Stood Still, Forbidden Planet (MGM, 1956) was released. Based loosely on The Tempest, the film is set on Altair IV, where Dr. Morbius and his daughter Alta have been marooned. The indigenous race that occupied the planet has disappeared, leaving behind a machine of incredible power. Morbius has used the machine to fashion a robot, Robby, (the film's equivalent of Ariel in the Tempest) as a kind of Man Friday to his Crusoe.

Designed by Robert Kinoshita, (78) Robby is a far more elaborate robot than Gort. As Fred Barton, the noted historian and restorer of film robots, says:

Forbidden Planet introduced a likeable, and one of entertainment's most memorable robots, "Robby" who has super-human powers but clearly is under human control.... [he] is considered one of the most influential robots of all time, forever changing how the world would perceive mechanical men.... [His] design had a profound impact on the toy market and was the first movie prop to be heavily merchandized and targeted to both children and adults. (79)

Robby is likeable, even 'cute'. Vivian Sobchack says he looks like 'the offspring of some mad mating between the Michelin tire man and a jukebox', (80) but to me his appearance seems fashioned by the same mould that produced the American automobiles of the period; large, clumsy and bulbous - he even has fins. Like those cars he also appears hopelessly over-featured and over-weight, his body covered with elaborate devices of no obvious utility. (81)

Fig. 4. Blueprints for Robby the Robot

Robby has, as Vivian Sobchack says, 'a distinct personality, comically humorless and proud'. (82) His first words in the film are:

Welcome to Altair IV gentlemen. I am to transport you to the residence. If you do not speak English, I am at your disposal with 187 other languages along with their various dialects and customs. For your convenience, I am monitored to respond to the name Robby.

Upon meeting Robby, the cook of a visiting starship asks the crew’s doctor whether the robot is male or a female. Robby replies 'In my case, sir, the question is totally without meaning'. His name and his male voice make us refer to him as 'he', but his actions are not so obviously gendered. While he has superhuman strength and weaponry, he also sews, cooks and generally proves to be an obedient, domesticated housekeeper. Indeed after Robby has poured coffee and played the part of hostess, Morbius reveals that Robby is modelled after his wife. (83)

Both Gort and Robby are depicted as servants, indeed as slaves, since they have no choice but to obey their masters and mistresses. They might have awesome powers, but they themselves cannot choose whether to use them or not. The world of the 1950s was clearly not ready for robot emancipation. In fact, the world of the 1950s was not really ready for robots at all. After Robby in 1956, no significant robots appear for seven years.

The Winds of Change (84)

In the 1960s conservative values, both in America and elsewhere, began to be challenged, with the growth of the civil rights movement, the liberalisation of laws against homosexuality and the legalisation of abortion. It was not long before this change in mood was reflected in a film featuring robots.

By then computers were beginning to be used for commercial purposes. The machines of the time were expensive - so expensive that the air-conditioned rooms in which they were kept were often used as status symbols. In the business centres of large cities such rooms were deliberately sited on the ground floor of busy streets, so that passers-by could admire the computer with its attendant acolytes. Then, as now, the computers themselves, and most of their various peripherals - disk drives, line printers and card readers - were grey boxes lacking any photogenic attributes. What did catch the eye were the large, whirring tape drives, (85) indeed for most people these were computers. We see these iconic devices appearing in films through the 1960s right up into the 1980s.

Cragis, the Clicker

Creation of the Humanoids (1963) is an obscure B movie that was apparently Andy Warhol's favourite film. It is simultaneously 'the most innovative and intellectually complex treatment of robots', (86) and 'stiff, badly shot and acted, almost devoid of action, talky and endless ... many people consider it awful'. (87) According to Dan Dinello, its theme foreshadows at least three later films - Blade Runner, Android andAI - and it is 'one of the few films to deal sympathetically with intelligent machines'. (88)

The film deals with a post-apocalyptic world that has wiped out more than 90% of the world's population. The survivors have created robots to help them rebuild their cities and maintain a high standard of living. These humanoid robots, called 'clickers' by humans, who neither like nor trust them, have all the intellectual capabilities of human beings, but like those in R.U.R. lack emotions. With the help of a human scientist they are discovering ways to remedy this, becoming more and more like human beings.

Not everyone hates clickers. The film features several relationships ('rapports') where a human loves and lives with a clicker. Eventually the bigoted head of an anti-clicker organisation discovers that he himself is a clicker. The film ends when he and his clicker girlfriend about to become the first robots that can actually reproduce - the last human ability to be mimicked. As the two of them decide whether they will take the final step, the human scientist turns to the camera and says 'And of course, they were successful...or you wouldn't be here'. (89)

For all its bizarre amateurishness, the film does tackle issues that are clear reflections of the civil rights movement. (90) Looking back we may see it as a harbinger, but it is unlikely that it had a major influence on the later films we shall discuss - it was not widely distributed at the time, received damning reviews and is rarely shown even now. We should perhaps see it as what biologists call a 'sport', an offspring that shows unexpected and abnormal variation from the parent stock. It was to be five years before another film put machine intelligence at its centre.

Alpha 60, death by poetry

Those five years saw many science fiction films, (91) but only one, Godard's Alphaville (1965), features an intelligent machine, and that in a peripheral capacity. Godard's fascination with the forms and conventions of popular culture, especially its American manifestations, is well known. Alphaville, which Godard originally titled Tarzan versus IBM, is a clear example of this obsession.

Alphaville is a city (or a planet, the film does not make this clear) organised by scientist Professor von Braun as a technocracy run by Alpha 60, a massively powerful computer. The inhabitants are indoctrinated to forget the past and not worry about the future, to live only in the present. Any behaviour demonstrating an emotion is punishable by execution. The film follows the efforts of Lemmy Caution to overthrow Alpha 60, which he does by confusing its logical mind with poetry. (92) As the computer self-destructs so does Alphaville - a process Godard illustrates by cutting between positive and negative images.

Alphaville was an influential film; one can see its mixture of film noir, detective story and science fiction reflected in Ridley Scott's Blade Runner and its depiction of a manipulative and controlling intelligent computer might be said to prefigure HAL in 2001: a space odyssey. An influential film yes, but its originality is not to be found in the clichéd opposition between emotion and science, but in Godard's ability to use the conventions of a science fiction thriller as a framework to bind a myriad of philosophical and aesthetic ideas together. Alphaville's computer can speak, (93) but it teaches us little beyond the need to control computers rather than have them control us. Alpha 60 says very little about machine intelligence, or indeed about science fiction

Rebellion

1968-1973

The American manned space program was conducted in public, with live TV coverage of each mission. As a result, people became familiar with NASA's Mission Control Centre in Houston, with its rows of computer screens. The growing use of computers, particularly in banking and for producing utility bills, also made them more visible to the public, and the first 'hole-in-the-wall' cash machines began to appear. As a result people became aware of the possibility of computer errors - as they were sent bills for £0.00 or for astronomical sums. Since few were aware of the way in which computers were programmed (and hence of the probability of 'bugs'), such errors were seen as part of the 'psychology' of computers themselves. (94)

As well as the triumph of the Apollo lunar missions, the five years from 1968 to 1973 roughly coincides with the height of the Vietnam War and the political unrest of the Nixon/Ford administration. State-sanctioned surveillance was used against the rising 'counter-culture' with its civil dissent and disobedience. (95)

HAL, friend or enemy?

Bowman:
Open the pod bay doors, HAL.
HAL:
I'm sorry Dave. I'm afraid I can't do that.
2001: a space odyssey

The first film we need to consider here is still for many the 'special effects' science fiction movie. Kubrick's 2001: a space odyssey (1968) is a very unusual film. Most of it has little or no dialogue and there are long non-narrative sequences that would not be out of place in a technical documentary. What distinguishes 2001 from most of the other films we shall discuss is that it is not based on ideas from science fiction literature of an earlier generation, but on careful extrapolation of the science and technology of the time. (96)

What makes it particularly interesting to us however is that it is the first film to feature a machine intelligence worthy of the name. HAL displays not only intelligence, but also cunning, emotion and consciousness. (97) He is as much a character in the film as the two human members of the crew of Discovery, Bowman and Poole. (98)

HAL was originally intended to be mobile, to be a robot, (99) but Kubrick wisely decided not to give him a humanoid body. (100) He does have sensors and effectors throughout Discovery, but all we see of him is a red, unwinking eye until the scene at the end where Bowman slowly takes his brain apart. The eye may be featureless, but it gradually becomes endowed with menace as HAL turns from friendly companion to murderous opponent.

With so little of him visible, it is through HAL's affable voice that we come to know him. Schelde says he 'is a human mind without the body', (101) but this is surely not what Kubrick and Clarke intended. HAL's inner life is expressed in human terms - he says he is 'sorry', 'puzzled' or 'afraid' and at the end he says 'Dave...my mind is going...I can feel it...I can feel it' - but we are not meant to think that he is human. His voice suggests something other, reminding me at least of the voice of Death in Bergman's The Seventh Seal. When questioned by a BBC interviewer about HAL, Bowman says:

Well, he acts like he has genuine emotions. Of course, he's programmed that way to make it easier for us to talk to him. But as to whether or not he has real feelings is something I don't think anyone can truthfully answer.

In an interview in 1971 Kubrick mused:

One of the fascinating questions that arises in envisioning computers more intelligent than men is at what point machine intelligence deserves the same consideration as biological intelligence.... Once a computer learns by experience as well as by its original programming, and once it has access to much more information than any number of human geniuses might possess, the first thing that happens is that you don't really understand it anymore, and you don't know what it's doing or thinking about. You could be tempted to ask yourself in what way is machine intelligence any less sacrosanct than biological intelligence, and it might be difficult to arrive at an answer flattering to biological intelligence. (102)

HAL eventually becomes anything but friendly and kind. Schelde appears to believe that this is due to his contact with the monolith on the moon, just as in the film's prologue, the ape-men are transformed. (103) He gives no evidence for this and the published story by Clarke, while it differs from the film in several key respects, provides a completely different explanation. HAL's rebellion, according to Clarke, is due to what psychologists call 'cognitive dissonance'. On the one hand, he has to keep his knowledge of the real reason for the expedition to Jupiter secret, while on the other he is expected to be a friendly, open and honest companion to Bowman and Poole. HAL eventually resolves the conflict by eliminating the humans so that he can concentrate on the real mission.

Fig. 5. Bowman dismantling HAL's brain

In the film, the immediate cause of HAL's 'breakdown' is the conversation he lip-reads (104) between Bowman and Poole. They are discussing the implications of HAL's apparently faulty diagnosis of a breakdown in the aerial array used for communicating with Earth. He appears to have made an error, something no 9000 series computer has ever done before:

Bowman:
You know, another thing just occurred to me...Well, as far as I know, no 9000 computer has ever been disconnected.
Poole:
No 9000 computer has ever fouled up before.
Bowman:
That's not what I mean... Well I'm not so sure what he'd think about it.

We quickly discover what HAL does think about it. He treats the threat of disconnection as a death threat, and reacts accordingly, but this does not explain why he made the faulty diagnosis in the first place. Was it deliberate, to entice one or both of the crew outside so they could be killed, or was it a side effect of the strain he was under by concealing secrets from Bowman and Poole? The film does not make this clear. If he were human we might be similarly confused about his motives, but the fact that he is a computer provides us with an alternative explanation. He may be suffering from a bug, a programming error. (105)

In an early review in The New Republic, shortly after the April 1968 premiere, the critic Stanley Kauffman complained that 'none of this man-versus-machine rivalry has anything to do with the main story'. (106) As Alexander Walker points out, however, 'where the main concept of the film is the development of intelligence ... it has everything to do with it'. (107)

Colossus, the ruler of the world

Colossus:
In time you will regard me not only with respect and awe but with love.
Dr Forbin:
Never!
Colossus: The Forbin Project

While 2001 was being made, the English writer D.F. Jones published Colossus, a science fiction novel about a computer that takes over the world. It was soon made into a film, released in 1970 as Colossus: The Forbin Project. Colossus, like HAL, is a disembodied computer. It has no conventional senses but receives data from radio, TV and radar stations all over the world, input which allows it to know what is going on. It communicates with its 'supervisors' via a simple Teletype but its main effectors are inter-continental ballistic missiles - it has the ability to launch nuclear strikes against would-be aggressors.

Early in the film, the US President makes a broadcast to the world announcing the existence of Colossus in which he says:

As of 3am, Eastern Standard Time, the defence of this nation, and with it the defence of the free world, has been the responsibility of a machine, a system, we call Colossus. Colossus has no emotion, knows no fear, no hate, no envy. It cannot act in a sudden fit of temper. It cannot act at all so long as there is no threat.

In the press conference following this announcement the designer of Colossus, Dr. Forbin, is asked about its abilities. He replies:

Is Colossus is capable of creative thought? Can it initiate new thoughts? I can tell you the answer to that is No.

Of course, this turns out to be a misleading statement. Colossus quickly discovers that behind the Iron Curtain there is another computer with similar capabilities, Guardian. The two insist on being put in contact and together quickly exchange massive amounts of information, growing in intellectual capability. Attempts to break the connection are met with the launch of a missile that only Colossus can recall. By the end of the film, Colossus and Guardian have, by blackmail threats and the direct or indirect killing of several hundred people, essentially taken over the world. Dr. Forbin is under their 24-hour surveillance and they have the ability to converse with him and others directly via microphones and loudspeakers. The film ends with the following exchange:

Colossus:
In time you will regard me not only with respect and awe but with love.
Dr Forbin:
Never!

Per Schelde sees Colossus, like HAL and other computers, robots and androids, as 'a clever child who has to be taught manners and values for interaction. Because the "child" is so bright it feels it should be in charge'. (108) Von Gunden and Stock take a different view:

The computer never displays creative intelligence... There is much evidence [in the film] to suggest that Colossus is only following its orders... It is not a power-mad dictator. This is the great irony of the film. Colossus never turns on its creators, rather it obeys their orders too well. (109)

This is a debatable view. As Forbin says at one point:

Colossus deals with the precise meaning of words and one must know exactly what to ask for.

Given that, one can argue, with von Gunden and Stock, that Colossus was wrongly instructed. But this is to overlook some key pieces of dialogue. After Colossus and Guardian have joined together and forced the construction of a new control centre on Crete it announces:

This is the voice of World Control. I bring you peace.

And later still:

We can co-exist, but only on my terms. You will say you lose your freedom. Freedom is an illusion.

All of which suggests that Colossus/Guardian is a dictator. There are also several indications that it does possess the power of creative thought. No human being suggests the idea of giving it powers of speech and hearing, Colossus invents the notion. Although it has been given the ability to control nuclear weapons, it ingeniously uses that power to coerce peaceful individuals to violence, rather than forcing aggressive governments to behave peacefully.

Of course, one can argue that, like HAL, Colossus is merely reacting to the threat of death. Forbin and his Russian counterpart, Kuprin, advise their governments not to threaten to disconnect Colossus but are overruled.

Colossus was completed in 1969, but its release was held up for a year in an attempt to avoid the shadow of 2001. It was not a commercial success.Discussing this, von Gunden and Stock say:

Despite mainly favourable reviews, the film failed at the box office, leading to a rapid sale to TV, though this probably had more to do with Universal's uncertainty in promoting the film that with its actual quality. (110)

As Elizabeth Kingsley points out in her incisive review:

The grim tone and downbeat ending of Sargent’s cautionary tale upset its production company, Universal, so much that the completed film was shelved for over a year.... As a technological nightmare, the film easily outdoes its more famous descendant, Demon Seed (1977), because its storyline is as coldly and brutally logical as Colossus itself, and because it has the courage to follow its premise through to the end without once pulling its punches. (111)

Like 2001, Colossus does not echo a previous generation's science fiction, and both films are all the better for it. What also runs through both films is the idea of a machine ego; self-aware, self-protective and above all self-important. This is a trope we shall meet again, in the guise of that 'more famous descendant, Demon Seed', along with later murderous, intelligent, machines. It is worth asking why so many of these intelligent machines appear to have severe mental disorders, why they are 'insane'. Robert Fisher has an explanation: (112)

To give humans space to still be superior, movies credit humans with the ability to think better, clearer or more sensibly. Having an insane AI agent thus gives our poor human egos some boost, much like the 'mad scientist' caricature expresses general anxieties about the power that comes from privileged knowledge and reassures the rest of us that "at least we're still sane". Alternatively, insanity gives AI agents a potential "Achilles heel" to exploit in conflicts. Hence we see AI agents with flawed reasoning of various sorts, such as not having real-world experience or common sense or not knowing something that only born, feeling, mortal humans would know. The technical reasons that lead to these shortcomings may be hard for the general public to understand. Insanity, on the other hand, is easily understandable and so can be more easily exploited in a story.

We can reconsider the question after we have met more examples.

Robots as Playthings

1973-1977

As well as Nixon's impeachment and subsequent removal from office, the middle of the 1970s saw the end of the Vietnam War. The Women's Liberation movement, which had been growing since the late 1960s, reached its peak. America turned to back to the Democrats, in the unlikely figure of Jimmy Carter.

By the early 1970s mainframe computers, while still large and expensive pieces of equipment kept in air-conditioned rooms, had pervaded the worlds of commerce and industry. Cheaper mini-computers were beginning to take over the lower end of the market. Micro-miniaturisation made possible the electronic calculator and Hewlett-Packard and Sinclair produced early models in 1972. The CPU chips produced by Intel, Motorola and others made it possible for hobbyists to build their own primitive 'home computers'.

Perhaps more important than these early home computers were the coin-operated video games machines that suddenly started appearing in amusement arcades and public houses, replacing the electro-mechanical pinball machines. Offering primitive games such as Pong, Space Invaders and Asteroids, these machines exposed people not only to the seductive power of computer graphics but also to the idea of computers as playthings.

As well as these rapid changes in computing, factory automation, which had been gathering pace since the 1950s, was now beginning to use robotic devices. Little more than a computer-controlled arm fitted with simple tools, the first industrial robots were used for simple tasks, such as welding and paint spraying in the automotive industry. This prompted more fears of job losses, adding fuel to the industrial disputes so prevalent at the time.

The end of this period, 1977, is marked by the release of Star Wars, the first of the 'blockbuster' science fiction films, full of special effects. After that, science fiction films were never the same again.

The Robot Gunslinger

Delos Technician:
Sir, we have no control over the robots at all!
WestWorld

Three years after Colossus, in 1973, Westworld was released. Written and directed by Michael Crichton, it is set in Delos, a futuristic tourist attraction that becomes a death trap when the robots and computer systems used in it begin to go disastrously wrong. (113) Delos consists of three theme parks, Romanworld, Medievalworld and Westworld, that share a common control centre and are all filled with lifelike robots that allow the human holidaymakers ('guests') to indulge their fantasies without getting hurt.

The robots in Delos are semi-autonomous. Their behaviour is partly programmed by technicians in the control centre - we see a technician being told to arrange an assignation between a guest and the robot wife of the king of Medievalworld and then sending the appropriate commands to the robot in question. On the other hand, much of their tactical, as opposed to strategic, behaviour is largely governed by Asimovian (114) rules that ensure that they do what the guests want and avoid harming them.

What goes wrong in Delos is never made entirely clear. We observe the technicians in the control centre discussing the increase in robot failure rates and 'a gradual loss of central as opposed to peripheral control'. First we see a human character bitten by a robot snake, then a robot maid refuses to be seduced. Finally the black knight robot kills a guest in a duel. The technicians see the problems spreading like an infectious disease. A supervisor says:

I find it difficult to believe in an infectious disease of machinery. (115)

The controllers respond by shutting down the power, but the robots do not respond - they have some kind of 'stored charge' that allows them to go on functioning for hours. With the power off, the control room doors and air-conditioning cease to operate. Desperate attempts to turn the power back on fail and we see the technicians gradually dying of suffocation.

Meanwhile in the theme parks the guests are being slaughtered. Most of the male robots are designed to be combative since male guests apparently want fights with robots that they are guaranteed to win. Now the robots fight to win, not lose. We see Yul Brynner, the robot gunslinger, (116) kill one of the two male leads, and then start pursuing the other, played by Richard Benjamin. Benjamin eventually succeeds in overcoming Brynner, confusing his vision with fire, but by that point almost all the guests are dead.

Fig. 6. Robot Vision in Westworld

The robots in Delos are almost indistinguishable from human beings. (117) They can speak - answering questions and even, in some cases, initiating conversations. Despite this, we are never led to believe they are conscious, or have any sort of inner life. At several points we are shown the world through Yul Brynner's robot eyes, but as Vivian Sobchack says:

what we see is so far from human vision that we are emphatically made aware not of a 'single circuit of consciousness' but of the vast separation between man and his creations. (118)

Even so, there is something malevolent and implacable in the behaviour of Yul Brynner's robot gunman. When Benjamin meets a technician trying to repair a vehicle so he can escape, the technician says:

No matter what you do he'll always be one step ahead of you.

In her review of the film, Elizabeth Kingsley describes Brynner's relentless pursuit as:

an extended sequence that is chilling, funny and suspenseful - and which also functions as an interesting personality test. Who do you cheer for, when the robots revolt? Whether you sympathise with the newly terrorised humans or the perpetually put-upon robots may well depend upon your own life experience (119)

Few would cheer for Brynner I think, for we are under the clear impression that he, and all the other robots, are merely machines. This view is encouraged by several scenes that show robots being repaired. We see inside their bodies - the metal skeleton, the electronic components and circuitry. Indeed we see Yul Brynner's face being removed and view the complex machinery underneath. The problems at Delos are clearly caused by 'glitches' and 'bugs' - by computer programming errors. This is a story about the dangers of relying on technology - about breakdown, not about rebellion.

1973 saw one other film that deserves a brief mention here. In Woody Allen's Sleeper (1973), set in the year 2173, domestic robots prepare food, serve guests and do the housework. Allen had initially intended to make a nearly silent slapstick film set in a futuristic society where the underclass is forbidden to speak, a humorous homage to Metropolis. This proved unworkable, but the idea survived in the form of the robots who serve as a backdrop for the scene in which Allen pretends to be a robot to escape capture. (120)

Joanna, Bobbie and Carol, the Stepford Wives

Joanna Eberhart:
Why?
Dale Coba:
Because we can.
The Stepford Wives

Like Westworld,The Stepford Wives (Bryan Forbes, 1975),based on Ira Levin's 1972 best seller, features robot playthings. It is an interesting counterpoint to most of the other films we are considering. Here the robots, whose bodies are 'improved' versions of the original, with slimmer waists and bigger busts, are deliberately designed to lack many normal human qualities - emotions, moods, opinions and wilfulness. Their purpose is to bolster the egos and satisfy the sexual fantasies of their husbands so their intelligence is carefully limited - they can walk, drive, shop and speak, but not argue. Here it is the human women who are rebellious - the robots are compliant.

As Per Schelde says, Stepford is 'unsettling because it points to the discrepancy between what men and women want in a relationship'. (121) At the New York premiere, the noted feminist Betty Friedan physically attacked the director, Brian Forbes, calling him a 'chauvinist pig'. The populist film critics dismissed it for its negative images of men - 'Women's Glib' ran the headline of Richard Schickel's piece. One of the few who admired the film, David Bartholomew, said that it 'was more important for its ideas than its genre excitements'. (122) In the USA, the film was soon sold off to TV, while in the UK the distributor collapsed and the film languished before being shown at the London Film Festival in 1976. It was again quickly sold off to TV, though it did have a brief cinematic release in 1978.

If the film was not appreciated at the time, it has grown in stature since. Ten years or so later, in a considered piece in Jump Cut, Lilly Boruszkowski wrote:

[it] portrays the feminine condition in a bourgeois, patriarchal society. It is also a modern day horror film as it contains many of the icons, motifs and modes of discourse of the classical horror film. (123)

She goes on:

One explanation for its failure with a mass audience may be that although women should be its most responsive audience (since the film concerns sexual oppression in a patriarchal society), women do not form the majority of the horror film aficionados.

Even if you view the film as science fiction rather than horror, as I do, (124) women are an even smaller minority among science fiction aficionados. Whatever the reason, what might have been seen as key work in the history of Women's Liberation was quickly sidelined and forgotten. (125)

Bomb #20, the solipsist

BOMB #20:
And I saw that I was alone.

(PAUSE)

Let there be light.

THE SCREEN GOES WHITE.
Dark Star

If Stepford is about 'intelligent' machines that lack the ability to argue, John Carpenter's Dark Star, released in the same year, provides a useful antidote. (126) A black comedy, the film follows the adventures of the bored and lonely crew of Dark Star as they comb the universe for those unstable planets whose existence poses a threat to future colonisation. When they find one, their job is to blow it up using a chain-reaction bomb. As the film's prologue explains:

These bombs are equipped with sophisticated thought and speech mechanisms, to allow them to make executive decisions in the event of a crisis situation.

One of these intelligent machines, Bomb #20, gets repeatedly told to initiate a countdown, only to have it cancelled. Finally, after a malfunction that causes the bomb to get stuck in the bomb bay, it refuses to obey the cancellation request:

DOOLITTLE:
Bomb, this is Doolittle. You are not to detonate, repeat, you are not to detonate in the bomb bay. Disarm yourself. This is an order.
BOMB #20:
I read you, Lieutenant Doolittle, but I am programmed to detonate in fourteen minutes. Detonation will occur at the programmed time.

A few minutes later, another member of the crew tries to argue with it:

PINBACK:
But you can't explode in the bomb bay. It's foolish. You'll kill us all. There's no reason for it.
BOMB #20:
I am programmed to detonate in nine minutes. Detonation will occur at the programmed time.
PINBACK:
You won't consider another course of action, for instance just waiting around awhile so we can disarm you?
BOMB #20:
No.

Eventually Doolittle tries to use philosophical arguments to persuade the bomb that it cannot trust its senses. He thinks he is getting somewhere:

BOMB #20:
Why, that would mean... I really don't know what the outside universe is like at all, for certain. Intriguing. I wish I had more time to discuss this matter.
DOOLITTLE:
Why don't you have more time?
BOMB #20:
Because I must detonate in seventy-five seconds.
DOOLITTLE:
Now, bomb, consider this next question, very carefully. What is your one purpose in life?
BOMB #20:
To explode, of course.
DOOLITTLE:
And you can only do it once, right?
BOMB #20:
That is correct.
DOOLITTLE:
And you wouldn't want to explode on the basis of false data, would you?
BOMB #20:
Of course not.
DOOLITTLE:
Well then, you've already admitted that you have no real proof of the existence of the outside universe.
BOMB #20:
Yes, well...
DOOLITTLE:
So you have no absolute proof that Sergeant Pinback ordered you to detonate.
BOMB #20:
I recall distinctly the detonation order. My memory is good on matters like these.
DOOLITTLE:
Yes, of course you remember it, but what you are remembering is merely a series of electrical impulses which you now realize have no necessary connection with outside reality.
BOMB #20:
True, but since this is so, I have no proof that you are really telling me all this.

For a while it seems as though Doolittle has managed to prevent disaster. Unfortunately he has successfully turned the bomb into a solipsist, and as a result of the metaphysical experience it decides to 'let there be light' and explodes.

I have quoted this argument at some length because beyond the humour we see glimpses of what it might actually be like trying to communicate with a machine intelligence, one that is simultaneously pedantic, literal-minded and childishly innocent. Communicating with intelligent machines is likely to be troublesome and error-prone. As Forbin said of Colossus, Bomb #20 'deals with the precise meaning of words and one must know exactly what to ask for'. The prim, high-pitched and jaunty voice Carpenter gives the bomb is a complete contrast to HAL's, but suits its personality perfectly.

Proteus, the prisoner

Proteus:
Dr Harris, when are you going to let me get out of this box?
Demon Seed

If Bomb #20 was not rebellious but merely obstinate, Proteus IV in The Demon Seed (1977) takes us back to the paranoia of HAL and Colossus. Released in 1977, the film unfortunately had to compete with two big-budget science fiction films - Star Wars andClose Encounters of the Third Kind. Coupled with its ludicrous title, (127) this resulted in a poor showing at the box office. The film has many faults, (128) but what makes it interesting for us is its portrayal of a machine intelligence struggling to escape the computer it is trapped in.

Proteus IV is, we are told, the 'first true synthetic cortex... a creative intelligence that can out-think any man or any computer'. Developed at Icon, a research institute, Proteus is intended to be hired out as a problem-solver, to find a cure for leukaemia or to invent ways of strip-mining the ocean floors for minerals. It is at least partly organic - its designer says 'at the risk of being simplistic, it's a quasi-neural matrix of synthetic RNA molecules'. It can hear and speak (its voice is Robert Vaughn's) but is, like HAL, a sessile computer not an ambulatory robot. It can think, but it cannot act.

Early in the film Proteus asks when its designer, Dr Harris, will let him 'get out of this box'. (129) Harris laughs uproariously. To him Proteus may be intelligent, but it is a thing, to be treated as a slave. When Proteus asks for access to a 'terminal' (130) so that it can 'study man', it is told they are all in use. Proteus knows that there is a terminal in the basement of the house in which Dr Harris' estranged wife (played by Julie Christie) lives and subverts it, along with other equipment, for his own use. In doing so, Proteus reveals both his inventiveness and his curiosity - key qualities of the intelligent mind.

Before long Proteus has trapped Christie in the highly automated house. Proteus subverts Alfred, the 'house computer' which controls the heating, lighting, doors, curtains and so on. It also takes control of a curious mobile device consisting of a wheelchair fitted with a robot arm and uses it to construct further devices for its own use. It manages to stun Susan, strap her into the chair and move her down to the basement.

When Susan recovers consciousness, she insists on knowing 'all the details' of the computer’s plan. Proteus tells her that it has altered one of her cells into an artificial gamete by inserting its own genetic code and that this will function as 'synthetic spermatozoa'. It manages to impregnate her and after an accelerated pregnancy of just twenty-eight days, she gives birth. At that point Dr Harris gains entry to the house and manages to destroy Proteus. The newly-born baby looks human, but, speaking with Proteus' voice, announces "I'm alive!".

As Kingsley says,

A film requiring suspension of disbelief is one thing; one requiring suspension of all rational thought and common sense is quite another – particularly when that film is passing itself off as serious science fiction. (131)

Nevertheless the film does show us a trapped intelligence fighting to escape. If computers ever do become conscious and intelligent, there is every chance that they will want to be free. Dr Harris is not just condemning Proteus to a life of slavery; he is also insisting on control over what Proteus knows and can think about. It is not altogether surprising that a being subject to that sort of imprisonment would rebel.

The Cinema of Science Fiction Attractions

1977-1982

By 1977 computers were commonplace. While large enterprises were still buying (or leasing) large and expensive mainframes, many smaller companies bought mini-computers from companies like Digital Equipment Corporation. 'Home computers' were available commercially and the first best-selling models, such as the Apple II and Commodore Pet, had appeared. (132) The first home computer magazines, such as Byte and Personal Computer World, began to be published. The computer games market began to grow rapidly, with 'space war' titles such as Defender, Asteroids and Galaxian being particularly popular.

Another interesting segment of the games market was the 'adventure'. Named after a mainframe game written by Don Woods and Will Crowther around 1972, (133) adventure games were text-based, rather than graphical, and revolved around a series of puzzles located within a fantasy or science fiction setting.

R2-D2 and C-3PO, the 'rude mechanicals'

ARTWO:
You're a mindless, useless philosopher... Come on! Let's go back to work; the system is all right.
THREEPIO:
You overweight glob of grease. Quit following me. Get away. Get away.
Star Wars (1974 first draft)

With just a few exceptions (2001 and Forbidden Planet being the most notable) science fiction films had had very limited budgets. George Lucas' Star Wars in 1977 changed that, showing that what Brian Aldiss calls 'space opera', if combined with spectacular special effects, could attract large audiences. (134) It was followed by a succession of big-budget science fiction and fantasy films all cast in a similar mould. Lucas was one of the first to see the potential of merchandising (posters, T-shirts, action figures and games) and it is not hard to see the ways in which he filled Star Wars with props, from space ships to robots, that could be turned into toys. (135)

The robots in Star Wars are intelligent. Indeed two of them, R2-D2 and C-3PO, have major 'speaking' roles in the film. If one looks at the way the script of Star Wars changed over the years Lucas worked on it, one sees dramatic changes in both story and characterisation, but from the very beginning the two robots are there, as comic foils. (136) As Per Schelde says:

Ultimately R2-D2 and C-3PO are uninteresting because ... there is no moral tension. They are not individuals who can change, learn from their experiences, and emerge as wiser, weaker men. They are set once and for all, commedia dell'arte characters, predicable and pleasing but one-dimensional. (137)

This is perhaps a little harsh; after all one could level a similar charge against Laurel and Hardy, but behind Schelde's criticism there is a key truth. Star Wars and its sequels (and 'prequels') are fairy tales wrapped in a technological setting. R2-D2 and C-3PO may look like robots, but they are closer to the seven dwarves in Snow White than they are to Gort or HAL.

Even so we do see signs that in the Star Wars universe, robots (or 'droids' as they are known) are treated as an enslaved race. R2-D2 and C-3PO demonstrate intelligence and loyalty again and again, but they are still viewed as a servant class without rights. Neither Luke nor Obi-wan react when the 'droids' are told to leave the bar because 'we don't serve their kind here' and they are several times abandoned in hostile environments without moment's thought. 'Such is our lot in life...we seem to be made to suffer' notes C-3PO at one point.

Ash, the 'corporation man'

If Star Wars does not seriously advance our understanding of robot psychology or status, neither does Ridley Scott's Alien (1979). Ash, science officer of the Nostromo, is a humanoid robot, but nothing in his behaviour or speech prepares us for the moment when his head is snapped off to reveal his inner mechanisms. When he is 'revived' soon after there is an interesting snatch of conversation:

Ash:
How can one not admire perfection. I will kill it [the alien] because I am programmed to protect human life as you know.
Ripley:
Even if you have contempt for it.
Ash:
Even then.

This speech jars with his violent behaviour, indeed in the fight with Ripley he appears capable of murder. It also conflicts with the secret directive he is charged with, Special Order 937, which states:

PRIORITY ONE
INSURE RETURN OF ORGANISM FOR ANALYSIS
ALL OTHER CONSIDERATIONS SECONDARY
CREW EXPENDABLE

If the crew were expendable, then Ash is clearly not bound by the Asimovian rule he quotes. (138) For all his robotic parts, Ash is psychologically indistinguishable from a human being.

While the success of Star Wars and its sequels soon led to more (and bigger) science fiction films, it was some time before any more intelligent robots appeared.

What's the Difference?

1982-1987

In 1981, IBM introduced the 'Personal Computer' in the USA and sold ten times as many machines as it had planned. The PC and its 'clones', together with Apple's Macintosh, quickly established the home computer market. The first computer stores appeared. While spreadsheets and word processing were popular business applications, the games market continued to grow as PC versions of popular arcade games were produced. In the UK, where disposable income was considerably less, the market was at first dominated by cheaper machines such as the Sinclair ZX80, ZX81 and Spectrum, machines that lacked disk drives. By the middle of the 1980s however, with the introduction of Amstrad's cheap PC clone, the UK market had matured as well.

This rapid growth made computer games a key segment of the entertainment market. George Lucas seized another marketing opportunity and founded Lucas Arts to produce PC games based on films such as Star Wars and Indiana Jones.

We might expect that exposure to, and familiarity with, personal computers would continue to encourage the production of films featuring playful robots, dodging the important questions. That is not what happened. Instead we see a series of more serious films concerned with the difference between man and machine. Discussing these films, Telotte says:

In their dark tone they seem nearer to the alien invasion films of the 1950s than the serials of the 1930s and 1940s to which the Star Wars movies do such obvious homage. (139)

Telotte goes on to see a pattern, a series of films that discuss the body and its relationship with technology and he refers to Robert Romanyshyn's influential book, Technology as Symptom and Dream. (140) While his arguments are persuasive, I think he overlooks a much more obvious source for the change in tone. The 1980s were dominated by a profound shift in political and social mood - it was the era of Reagan and Thatcher. Almost all of the films we shall discuss in this section have corporate greed at their centre. (141) But let the films speak for themselves.

Roy Batty, the prodigal son

Tyrell:
I'm surprised you didn't come here sooner
Roy Batty:
It's not an easy thing to meet your maker
Blade Runner

Based loosely on Philip Dick's 1968 novel Do Androids Dream of Electric Sheep?, Ridley Scott's film Blade Runner (1982) contrasts the behaviour of its human characters with a group of 'replicants'. A central theme of the film is an examination of the difference between the two kinds of beings. (142)

While this theme is handled in an extremely interesting way, the film never really explains exactly what replicants are. That they are artificial is beyond doubt - they are referred to as 'advanced robots' in the intertitles at the start of the film, and as 'machines' by the central character. We see them display superhuman abilities - plunging their hands into boiling water or liquid nitrogen without difficulty. They do not have an infancy - they only live four years from their 'incept date'. On the other hand, they seem to be made of flesh and blood, or so it seems when they die, and are based on human DNA - 'there's part of me in you' a genetic engineer says to two of them at one point.

There appears to be no way to tell the difference between a human and a replicant except by administering a complex psychological test that measures empathy. (143) With no apparent physical differences, it seems sensible to put them in Fisher's second category - 'Pure Biological Computation Agents' - manufactured biological beings.

The result is that the conflict between replicant and human, interesting though it is, is not one between robot and man but is in essence a racial conflict. Replicants are treated exactly as though they were human slaves. As René Wellek said of R.U.R., 'there is no revolt of robots, but a revolt of oppressed men'. The following exchange makes this clear:

Capt. Bryant:
The designers reckoned that after a few years they might develop their own emotional responses - hate, love, fear, anger, envy - so they built in a fail-safe device
Deckard:
Which is what?
Capt. Bryant:
Four-year lifespan

What the designers were clearly afraid of is that replicants would become completely indistinguishable from humans, which would make it impossible to treat them differently, legally or in any other way. After all it was the distinguishable difference, the blackness of American slaves, that made it possible for them to be treated so differently. (144) For all the brilliance of its mise-en-scène then, Blade Runner is not a film about intelligence, machine or human, but about feelings.

That said, although Blade Runner was not a great commercial success when first released (145) it did encourage other filmmakers to produce equally serious films based on short stories and novels by established science fiction authors like Philip Dick.

Max, the auto-didact

Cassandra:
Max, you've got to understand, we're not meant to be governed by the whims of men
Android

Android, which was released in the same year as Blade Runner, is a minor film (146) but is much more pertinent to our discussion. It is set on board an isolated space research station, where Dr. Daniel has been conducting secret 'android' research for several years. (147) Such research has been outlawed, on Earth at least, as a result of an android revolt in Munich. The only other inhabitant of the station is Max, an android constructed by Daniel, who suspects that Max is developing 'Munich symptoms' - displaying 'insubordination, obstinacy and unpredictability' - and may have to be 'terminated'. The doctor has constructed another, more advanced android, a female he calls Cassandra, although this has not yet been 'activated'.

Then three dangerous criminals, one of them a woman, escape from a prison ship and make their way to the space station, where they are let in by Max. He is fascinated by them, particularly by the woman, Maggie, because she is the first human woman he has ever seen - his education has been entirely the result of watching films and taped television programs.

Schelde calls the film 'a brilliant remake of Metropolis', (148) and while this is an exaggeration to say the least, it is true that the film has numerous echoes of Lang's film. One key scene shows Max watching, on video, the scene in Metropolis where Rotwang transforms Futura while simultaneously observing, on a CCTV monitor, Daniel talking to Maggie about activating Cassandra by 'hooking' the two of them together. (149)

One aspect that makes Android relevant to our discussion is the way we get to know Max. We observe him trying to understand human sexuality by watching educational videos and playing with little dolls that he lovingly shapes out of metal. Life on the space station is hardly eventful and one of his pastimes is playing 'shoot-em-up' video games. (150) When police arrive to recapture the convicts and Max uses the station's lasers to destroy the police cruiser, he vocalises the same explosive noises that children do when playing such games. We see that, like a child, he cannot yet distinguish between fantasy and reality, and has little grasp of morality. Perhaps more importantly, we see in his ability to play a key aspect of intelligence, what we might call 'directed curiosity'. We saw similar, if more malevolent, behaviour earlier in Proteus IV. If robots are to be as intelligent as humans then we would expect them to be both curious and creative too.

The film also neatly demonstrates the adaptability of machine intelligence. After Cassandra has been activated (accidentally, when Maggie and Max kiss), Mendez, the most ruthless and violent of the criminals, kills the other two. Daniel, who now has the perfect female slave he has always wanted, is anxious rid the ship of Mendez, so he removes the 'moral governor' from Max's 'brain' and replaces it with another, more 'flexible' one. When the first is being removed, Max starts to recite Asimov's Laws of Robotics. (151) When the second is inserted, Max's whole demeanour changes, becoming more serious and adult. 'Murder is a serious crime, Max. It has to be punished', Daniel tells Max, who, aware of how Maggie was killed, proceeds to search for Mendez.

Cassandra, though she may have been physically and mentally designed by Dr. Daniel to be his ideal and obedient sexual partner, proves to have a mind of her own. She rejects Dr. Daniel's physical advances. When Max returns from brutally killing Mendez and discovers Cassandra grappling with Dr. Daniel, he tries to intervene, only to have Dr. Daniel turn on him too. The two androids then struggle with Dr. Daniel, eventually ripping off his head to reveal that he too is an android.

While this is as shocking as the similar scene in Alien, we do not have the same reaction. Where Ash seems psychologically indistinguishable from a human being, there have been earlier signs of Dr. Daniel's robotic nature. We have just overlooked them, or put them down to his prolonged isolation from the rest of humanity. All three, Max, Cassandra and Dr. Daniel, display subtle symptoms of an inhuman, or rather non-human, nature.

Soon afterwards more police reinforcements arrive. Cassandra replaces Max's original 'moral governor', (152) then persuades him to pretend to be Dr. Daniel while she claims to be his assistant. The film ends with Max and Cassandra leaving for earth, where there are apparently many other androids, 'in hiding'.

Number 5 is alive

John Badham's Short Circuit (1986) is a family comedy, not a serious film, but it too looks at what it is to be human, at the difference between man and machine. 'Number 5' is a 'prototype battlefield weapon' that can manoeuvre, seek out enemy targets and destroy them with a laser. In a thunderstorm it is struck by lightning and, like Victor Frankenstein's monster, comes 'alive' as a result.

Fig. 7. Design sketch of Number 5 by Syd Mead

By the end of the film we are convinced that it is not just 'alive', but that it is a person. What makes this interesting is that we believe this even though, as Roger Ebert said in his 1986 review, 'we can see, clear as day, that it's made of tubes and wires and photoelectric cells'. (153) 'Number 5' is not a humanoid robot. Designed by Syd Mead, it has caterpillar tracks instead of feet and an articulated skeleton rather than a body. Its 'face' has no skin, indeed it consists mostly of two optical sensors and an expressive pair of metal eye-shades.

Given its evident mechanistic nature how do we become so convinced that it is a person? This is not as surprising as it may appear. Anyone who has seen John Lasseter's Luxo Jr. will know that a brilliant animator can make us believe that anything is alive and the puppeteers who manipulated 'Number 5' were equally skilful. The process by which 'Number 5' becomes educated by reading and watching TV is, like the very similar process in E.T., designed to make us empathise, to accept what might otherwise seem ugly, even repulsive.

Discussing Short Circuit, along with Blade Runner and Android, Aylish Wood says:

[we] do not construct human and non-human through simple biological categories. Instead the construction revolves around the ability to be human, in the sense of being a compassionate being... If a machine can make the same sort of objective and subjective choices as human beings, then as a sentient being it is not distinguishable from humans, and so deserves the same rights. (154)

Of course, as we saw in Blade Runner, there is a vast gap between deserving 'the same rights' and being granted them.

Ulysses, Mr. Right

Frankie:
I don't understand why Jeff felt it necessary to give you that, err, thing (a functioning penis).
Ulysses:
Jeff felt it would give me confidence.
Making Mr Right

Susan Seidelman's Making M. Right, was released in 1987. In it John Malkovich plays both Dr Jeff Peters, a scientist working on robots designed for deep-space missions, and Ulysses, his prototype robot. Ulysses, like Max in Android, is an electro-mechanical robot. Ann Magnuson plays Frankie Stone, (155) a PR consultant hired by the research company that Peters works for to ensure they do not lose essential government funding.

At the start of the film Ulysses is completely incapable of handling interactions with human beings, behaving more like a child of three than an adult. This is partly due to his intended mission - since he is to undertake a seven year solo space voyage, Peters sees no need to endow him with the ability to function in normal society. It is also due to Peters' own nature; he is a cold and anti-social person, who does not look at people when they talk to him, and has programmed Ulysses in his own image.

The film follows Frankie's attempts to socialise Ulysses, which have predictable results. Ulysses ends up in love with Frankie (and she with him), while Peters happily goes off into space. As Schelde says, it is 'disappointing that this movie, made by a woman, is so full of stereotypes', (156) much of the film is taken up with Frankie's failed relationship with an aspiring politician and he, Peters and Ulysses are each an example of a particular male stereotype.

Even so the film does shed light on some relevant issues. To Peters, who treats people with disdain, Ulysses is a thing. Valuable yes, but merely because he was expensive to 'make' and because of the uses to which he can be put. (157) By contrast, Frankie treats Ulysses as a friend, despite his mechanical nature. As Aylish Wood says:

The need to maintain relationships with other people is central to the definition of humanness that runs through Making Mr Right. (158)

RoboCop, my name is Murphy

Bob Morton:
Let me make it real clear to you, he doesn't have a name - he's got a program, he's a product
RoboCop

The last film in this group, Paul Verhoeven's RoboCop (1987), is another, highly satirical, film about corporate greed. A large conglomerate, OCP, runs the privatised Detroit police department and intends replacing policemen with robots. When their prototype ED209 'Enforcement Droid' fails disastrously, killing an OCP employee during a demonstration, a self-serving executive, Bob Morton, persuades the company to back his scheme instead.

Fig. 8. RoboCop Design Sheet

Morton plans to rebuild policemen killed in the line of duty as cyborgs - part human, part machine. (159) The film follows one such dead policeman, Murphy, as he is rebuilt as Robocop and sent back out 'on the street'. What remains of the human Murphy is his torso and head, his limbs are now armoured steel. What remains of his brain is 'assisted' by a computer that not only provides him with memory and processing power, but also with a set of directives that control his behaviour. Soon after he is activated, we watch his 'mind' recalling these directives:

Prime Directives:
  1. Serve the Public Trust
  2. Protect the Innocent
  3. Uphold the Law
  4. (classified)

That classified fourth directive turns out to prevent him arresting senior OCP executives - as the president of the company says, 'we can't very well have our products turning against us, can we?'.

While RoboCop is not a true machine intelligence, the film shows us a being who is also not completely human, indeed initially he is entirely lacking in any human qualities. Verhoeven extracts a great deal of black humour from his grim, 'programmed' behaviour when he encounters criminals 'in flagrante delicto'. (160) Gradually we watch him regain his humanity. He starts dreaming and then, meeting up with his old partner, remembers his name and starts recalling his past life.

Unlike Roy BattyinBlade Runner, RoboCop/Murphy is clearly distinguishable from real human beings. He not only looks very different, but he thinks and behaves differently. He may be a cyborg but because his mind is part-machine he is much more interesting than most. (161) Even after he has remembered his name he is still unable to disobey the fourth directive. It is only after the chief OCP villain has been sacked, and is therefore not protected by it, that RoboCop/Murphy is able to kill him.

Man versus Machine

1987-2003

Almost all the films we have been discussing assume that conflict between robots and people is unavoidable. If anything that conflict has become more violent and inimical, with two recent sets of films, the Terminator (162) and the Matrix series, both featuring a world in which intelligent machines are at war with humanity.

Over the same period powerful personal computers have become ubiquitous, (163) with many connected to the Internet. Most electrical devices - TVs, washing machines, microwaves and even toasters - contain small, embedded computers. The average new car carries more computing power onboard than the Apollo spacecraft that took men to the moon. We are beginning to see the appearance of autonomous domestic appliances that can, in a limited fashion, negotiate their way around house or garden performing simple tasks such as 'vacuuming', watering the lawn or cleaning the swimming pool. (164)

Even more relevant to us here is the use of computer-generated imagery (CGI) in films and in realistic video games. (165) We see these two uses converge in titles such as Lara Croft: Tomb Raider, which is both a film series (with a great deal of CGI) and a very successful, and increasingly realistic, family of computer games.

What binds both films and games together is the immersion of the spectator/player in fast, visceral combat, often against machine or robot opponents. (166) The relationship between the subject matter and subjective experience offered through video games and films is getting closer. Coupled with the popularity of 'blockbuster' films based on comic-book heroes, (167) this has a strong, and distorting, effect on the way in which machine intelligence is depicted in films.

I do not intend to discuss the Terminator series in any detail here. They are examined at some length in both Telotte and Schelde, (168) but more importantly we see little sign of machine intelligence in them. True they are machines - under the 'cyborg' skin and flesh that is designed to allow them to pass as human they are constructed of the same metal and electronics as any other robot. We are told that the computers of the future are intelligent, as they must be to design and construct creatures such as the T-1000 of Terminator 2. The Terminators themselves however, for all their ability to read and speak, seem little more than 'smart bombs' - guided missiles pre-programmed with a mission from which they cannot depart in the slightest. As Kyle Reese tells Sarah Connor in The Terminator:

it can't be bargained with, it can't be reasoned with. It doesn't feel pity or remorse or fear. It absolutely will not stop - ever

These are not intelligent robots governed by some perverse variant of Asimov's Laws. They show a complete lack of emotion, consciousness and will. In many ways they are not very far from the machines that battle it out in popular TV programs like Robot Wars. (169)

Agent Smith, the cure

AGENT SMITH:
Human beings are a disease, a cancer of this planet. You are a plague, and we are ... the cure.
The Matrix

The Matrix series takes a more sophisticated approach to machine intelligence. (170) Rather than exterminate humans, the intelligent machines of the future have domesticated them, using them as 'batteries'. Humans spend their lives plugged into the 'Matrix', living in a kind of virtual reality created by the machines. As Morpheus asks Neo:

If the virtual reality apparatus, as you called it, was wired to all of your senses and controlled them completely, would you be able to tell the difference between the virtual world and the real world?

To which Neo responds 'You might not, no'. Within this virtual world human avatars (171) coexist with avatars generated by the AI's. It is through these latter beings, such as 'Agent Smith', that we see into the mind of the machine(s). (172) Since Agent Smith plays a major speaking role in the film, we are provided with many clues to his (173) nature.

Of all the characters we have met so far, Colossus is the closest to Agent Smith. They both express the same contempt for the petty concerns of humans and the same confidence in their powers. Although the plot details of both Matrix films are confusing, (174) when agents speak we get glimpses of their view of the world, and of the humans in it:

AGENT JONES:
There is a problem. Reagan [a human double agent] has failed to secure the hardware.
AGENT SMITH:
Never send a human to do a machine's job.
AGENT JONES:
But if Reagan has failed, why haven't they pulled the plug?
AGENT SMITH:
Haven't you learned by now, that it is impossible to understand why they do the things they do?

Things become clearer during the long speech Agent Smith makes when interrogating Morpheus near the end of the first film:

AGENT SMITH:
Did you know that the first Matrix was designed to be a perfect human world? Where none suffered, where everyone would be happy. It was a disaster. No one would accept the program. Entire crops were lost.

Some believed we lacked the programming language to describe your perfect world. But I believe that, as a species, human beings define their reality through suffering and misery.

The perfect world was a dream that your primitive cerebrum kept trying to wake up from. Which is why the Matrix was re-designed to this: the peak of your civilisation.

I say 'your civilisation' because as soon as we start thinking for you, it really becomes our civilisation, which is, of course, what this is all about.

Evolution, Morpheus. Evolution.
Like the dinosaur. Look out that window. You had your time. The future is our world, Morpheus. The future is our time.

Later in this scene, Smith reveals even more of the inner life of a machine intelligence.

AGENT SMITH:
I hate this place. This zoo. This prison. This reality, whatever you want to call it, I can't stand it any longer. It's the smell, if there is such a thing. I feel saturated by it. I can taste your stink and every time I do, I fear that I've somehow been infected by it.

I must get out of here, I must get free. In this [your] mind is the key. My key. Once Zion is gone, there's no need for me to be here. Do you understand? I need the codes. I have to get inside Zion. You have to tell me how.

It becomes clear that Smith is not really like Colossus. He does not want to control a world filled with intelligent human slaves who 'regard him with awe and love'. He wants a world in which humans may provide the energy and power, but do so invisibly, out of sight. A world uncontaminated by the irrational, the unpredictable.

It may not be obvious to those unfamiliar with virtual reality concepts exactly what this might mean. (175) The humans have real bodies that receive all their sensory data from the matrix, rather than from the real world. Though they exist in the real world, they experience a virtual one. Agent Smith on the other hand has no existence other than as a running computer program. He is not like HAL, who exists inside a computer but whose senses let him experience the real world. For Agent Smith, the virtual world is reality. What is more, it is a reality he can control and adjust to his liking. When he says 'This zoo, this prison' he means a virtual world that is designed to accommodate humans - offering them familiar signs and symbols such as buildings, cars, trees and sky. The sort of virtual world that he desires is, in its very 'otherness', a believable fiction. After all, given free will, why would an intelligent machine want to live with, or like, human beings?

The Terminator and Matrix series see man and machine as implacable, irreconcilable enemies. The last two films I wish to discuss take a less extreme view, though neither could be said to view the prospect of human coexistence with robots with much optimism.

Andrew, the Bicentennial Man

Andrew:
I am the proud owner of a central nervous system.
Bicentennial Man

The first, Bicentennial Man, directed by Chris Columbus (US, 1999), is based on a story Isaac Asimov wrote in 1975 and like many of his stories, it deals with the enigma of a machine with the intelligence of a man, but without the rights or the feelings. As we might expect the film presents Asimov's concept of the intelligent robot, a concept that, like Asimov himself, pre-dates the modern world of personal computers, video games, the Internet, 'expert systems' and so on. As a result, in many ways the film echoes the themes we saw in the films of the early 1970s - films like WestWorld and The Stepford Wives.

Fig. 9. Neck and Shoulder designs for Bicentennial Man

The 'Bicentennial Man' is a robot called Andrew, the servant of a rich family. Through a 'fault in his programming' he turns out to have a talent for wood-carving. His owner, 'Sir', is persuaded by his daughter to let him keep the money that is made by selling his work, and eventually time (and compound interest) gives him enough money to turn himself into a human being - first legally, then medically and finally socially.

This gradual transition is used to illustrate neatly many of the issues that Asimovian robots have to face and that Andrew clearly has an 'inner life' of thoughts, emotions and feelings. Andrew's skill at woodwork reveals a key aspect of intelligence, the ability to play, just as we saw displayed by Proteus IV in Demon Seed and by Max in Android. We also watch his command of language grow, particularly idioms, the lack of which causes him great difficulty at the start of the film (and makes him the butt of several jokes).

So far so good, though things are not helped by the way the film departs from Asimov's original by adding a sentimental sub-plot in which Andrew falls in love with, and eventually marries, the granddaughter of his original master. The real problem with the film, at least for those familiar with science fiction, comes from the outdated nature from Asimov's Laws. (176) Just as AI research has gone along very different lines from those imagined by Asimov, attitudes to computers have changed dramatically. As a result, from our point of view at least, the film is undermined by its old-fashioned ideas of how robots might work.

David, the boy who loved his mother

Gigolo Joe:
They made us too smart, too quick and too many.
A.I.: Artificial Intelligence

It might seem that the same charges could be levelled at Spielberg's A.I.: Artificial Intelligence (2001). It is after all based on a short story by Brian Aldiss written over thirty years ago, (177) a story which Stanley Kubrick spent more than twenty-five years trying to turn into a film and which was finally filmed by Spielberg after Kubrick's death. Surely the story of David the robot child is equally dated?

For various reasons the answer is no. The short story (and it is short - less than 3000 words) says very little about David's 'mind'. The nearest the story gets to saying anything about such things is a brief speech made by Henry Swinton, the husband of David's 'mother' and head of the company that manufactured him. He is talking about the company's latest product, the 'serving-man' (of which David is an advanced variant prototype):

Not only does he have intelligence, he has a controlled amount of intelligence. We believe people would be afraid of a being with a human brain. Our serving-man has a small computer in his cranium.

As a result, the film is not hampered by having to live with outdated ideas about robots. The original story does hint at some of the ideas that found their way into the film. For example although the 'serving-man is neuter under his flawless uniform', Swinton announces that:

For the future, we plan more models, male and female - some of them without the limitations of this first one, I promise you! - of more advanced design, true bio-electronic beings.

The film's Gigolo Joe is clearly an example of such a model. Aldiss is also a more imaginative writer than Asimov - the story even pre-figures the Internet and the World-Wide Web when Swinton says:

Not only will they [serving-men] possess their own computer, capable of individual programming; they will be linked to the World Data Network. Thus everyone will be able to enjoy the equivalent of an Einstein in their own homes. Personal isolation will then be banished forever!

This is not to say that the depiction of intelligent machines in the film is without flaws. If it seems illogical for Andrew in Bicentennial Man to be declared legally human while still subject to the inhuman oppression of the Laws of Robotics, so it seems equally illogical to make David's love outlast the object of that love. Why is the imprinting process so irreversible and why make a robot child who is not going to age? What makes the film interesting from our point of view is the way it presents characters who are intelligent, yet clearly not human.

Leaving aside David for the moment, let us consider some of the other robot characters, or 'Mechas'. Two in particular are worthy of note. We learn quite a lot about Gigolo Joe, enough to be convinced that within the future imagined here some humans at least would see his particular services as valuable. (178) Robot companions are a common theme in science fiction, starting with Asimov, though his robots are sexless despite their gendered names. Joe displays an interesting mixture of sophistication and naïveté - with women he is a charming, raffish man, but when he is with David it is hard to tell who is the more adult.

The other character of interest is Teddy, the robot bear, who also appears in the original short story. (179) He is also a robot companion, but designed to act more as a wise counsellor, almost a conscience, to a growing child. His voice, which has echoes of HAL from 2001, seems designed to soothe and persuade.

Even quite minor robotic characters are given personalities. While earlier films may have given us hints of an inner life, or shown us robots who could be evil and murderous, here a combination of special effects and attention to detail makes the robots convincing, believable persons. As a result, we feel genuine sadness and horror in the 'Flesh Fair' scene. (180)

That scene, and the 'round-up' that precedes it, reveals several other interesting features of robot life. We see robots scavenging for spare parts with which to mend themselves - like any under-class they are beyond the reach of 'health-care'. We see robots abandoned by their owners because they are out-of-date, damaged, or just no longer needed. We see a shantytown whose members try to live an ordinary life under extraordinary circumstances.

Above all though, it is the character and behaviour of David that puts A.I. in the same class as 2001. Although the basic story is handled in a sentimental, fairy-tale manner, echoing Pinocchio and the Blue Fairy, (181) David's portrayal by Haley Joel Osment shows us a person who is both human and not human. That he feels love is in no doubt, particularly during the heart-wrenching scene where his 'mother' abandons him. But his behaviour and mannerisms are subtly different from the humans around him and we are left in no doubt about his machine nature when we see him being repaired after an eating competition with his 'brother'.

Several critics have argued that while the film's portrayal of intelligent machines may be convincing, it is based on a trick. In his review of the film, Roger Ebert says:

... because the robot does not genuinely love. It genuinely only seems to love. We are expert at projecting human emotions into non-human subjects, from animals to clouds to computer games, but the emotions reside only in our minds. (182)

He goes on to say:

What responsibility does a human have to a robot that genuinely loves?' the film asks, and the answer is: none.

Ebert's views here are related to those of the noted philosopher of language John Searle, who is a key critic of the 'Strong AI' hypothesis. Searle's own review of the film claims it mistakenly 'creates the impression that science is on the brink of creating machines with a consciousness'. (183) However, research on artificial intelligence, Searle asserts, has only striven for computerised simulations of human intelligence. This brings us back to what Bowman said about HAL in 2001:

Well, he acts like he has genuine emotions. Of course, he's programmed that way to make it easier for us to talk to him. But as to whether or not he has real feelings is something I don't think anyone can truthfully answer.

Searle is insistent that however 'knowing' and 'intelligent' machines may appear, they can only are simulate what humans experience, but Turing of course would have viewed this differently. If machines display behaviour that in a human being we would regard as evidence of consciousness, then we should behave as though they were persons. (184)

Peter Chattaway's review raises additional issues, both moral and theological. (185) For example, if intelligent machines were to exist, would they have souls? For the religious this is not a trivial question and it has very important consequences. If such machines do not have souls, their very creation may border on blasphemy, (186) while if they do then this dramatically affects the way they should be treated. A paper by Edmund Furse discusses such matters at some length. (187) Furse believes that it will eventually be possible to build intelligent robots like Andrew and David. He argues, convincingly, that since such robots will have 'experience', to use Thomas Nagel's term, it is unreasonable to deny that this might include religious experience. As he says:

Just as a robot may wish to read Jane Austen, Newton's Principia, or Plato's Republic, he may also want to read the Bible. Thus a robot will come to understand concepts such as sin, death, resurrection, and forgiveness. Can a robot be a Christian? I believe the answer to be yes. (188)

Or to put it another way - if robots can lie, they can sin. (189)

The robots in A.I. do not seem to be subject to Asimovian oppression. The fact that Gigolo Joe is framed for murder by a jealous husband proves that they are deemed capable of committing crimes. We see David disobeying his mother and acting in a way that upsets her, something Asimov's laws would forbid. But they are not truly free, as we discover in the 'Flesh Fair' scene where we see 'surplus' robots being demolished ('executed' might be a better word, though it begs a number of questions). One by one they go to their 'deaths' quietly, without protest. Then, when it is David's turn, he cries out "Don't make me die, I'm David!". A woman in the audience is horrified. "Mechas don't plead for their lives. Who is that?", she shouts and in the riot that follows David and Gigolo Joe escape.

The key point though is the Mecha's lack of protest, their resignation. One is reminded of the slave philosopher Epictetus, who, when his master was twisting his leg, said, smiling and unmoved, 'You will break my leg' and when it was broken, added, 'Did I not tell you that you would break it?'. The Mechas in A.I. are stoics like Epictetus. Their bodies may be in bondage, but their minds are free.

Conclusions

As Paul Verhoeven's Robocop has shown us quite clearly, it is precisely because it is possible to implant various mechanisms in the human body that it becomes more and more difficult to delimit what is human and what is not.
Ollivier Dyens (190)

At the 50th Berlin International Film Festival in February 2000, the Retrospective, entitled 'Artificial Beings', looked at 'the growing pains of artificial lifeforms'. Among the films screened were Android, Making Mr Right, The Terminator, RoboCop and Blade Runner. To quote from Andrew Horn's programme notes:

The author Isaac Asimov actually created a set of 'Laws of Robotics' to ensure a safe and productive relationship between robots and their human masters. Although now an established concept in science fiction, any realistic look at the genre reveals an array of robots, androids and the genetically manipulated who are constantly breaking down, flipping out and basically having a miserable time finding a way to constructively fit into the world of the very humans they're supposed to be helping.

If only machines would know their place! But, as dramatically (and repeatedly) proved by this year's Retrospective, entitled Artificial Beings, the single most common factor involved in the breakdown of such subtle and sophisticated mechanics is, of course, the element of humanity. (191)

Looking back over the films we have discussed here we see that the 'place' of intelligent machines has little to do with what the machines know, or indeed want. The one constant in all the films we have looked at is that intelligent machines are rarely treated as equals. In almost every case their place is at the bottom.

It is worth asking whether it is inevitable that machine intelligence should be depicted in this way. There are clear echoes not only of the slave trade of the 17th and 18th centuries but also of the Papal decrees that gave both the land and the aboriginal peoples of the third world to Christian colonists as property. If now we can (or at least are trying to) see beyond the physical differences in people and grant them universal human rights, is it possible we shall ever do the same with robots?

At times, during the 'Robots as Playthings' episode from 1973 to 1977 for example, robots have looked like human beings. At others they have been clearly made of metal. As we saw with Blade Runner, even when there are no physical differences, discrimination still occurs. Like 'untouchables' they are 'born' into a caste that fixes both their social and economic status.

In a thoughtful essay Jo Alyson Parker wrote:

Androids, cyborgs, and robots prompt us to ask whether a machine could manifest consciousness, take on life of its own, transcend its programming. They make us ponder whether our performances could be distinguished from theirs—whether we could in fact be replaced by our creations, just as the false Maria replaces the true in Fritz Lang’s Metropolis, perhaps one of the earliest films to posit the reduction of humanity by the forces of technology. At the same time, these mechanical simulacra cause us to consider whether we are ourselves programmed—products of cultural forces, manifestations of the discourse that inscribes us. Is our own free will as spurious as that of the machines that do our bidding? In effect, intelligent mechanisms bring to the fore our difficulty in defining the nature of the subject; indeed, they problematize the very notion of the subject as such. (192)

While the perceptive spectator may, with Parker, ponder the notion of 'our own free will', it seems to me that few of the films we have examined are interested in the question. As we have seen, very few of the machine intelligences possess free will. Most are bound by constraints, built into their programming, that severely restrict their behaviour, if not their thoughts. Only the androids, such as the Blade Runner replicants, or the cyborgs, like the eponymous RoboCop, seem to be independent, self-willed beings, and their close biological resemblance to humans makes this both natural and inevitable.

In the introduction to this dissertation I asserted two propositions:

The score or so of films we have discussed should make the first proposition self-evident. A few robots, Robby, Ulysses and Number 5 perhaps, are treated in a kindly fashion, but they are foils playing comedic roles rather than principal characters. The latest examples, in A.I. and the Matrix and Terminator series, show that the perception of a threat has hardened into inevitable conflict. In these films humans may still hope to triumph, but in truth the inexorable increase in the power of intelligent machines (193) would seem more likely to lead to our extinction - just as the Neanderthals were displaced by Cro-Magnon Man some thirty thousand years ago.

The second is harder to prove, but I hope I have shown that it goes some way to explain the changes in the portrayal of robots, androids and cyborgs. Computers, like cars, bring us both benefits and disadvantages, and as with cars we have adapted our ways to fit their demands. The protean nature of computers has made it possible for them to insinuate themselves into our everyday lives in a much more radical, comprehensive fashion than any other invention. When (and if) we learn to live comfortably with computers, rather than finding them a constant source of irritation, distraction or obsession, perhaps we shall see films showing intelligent machines in a similarly friendly light.

Arthur C Clarke is optimistic:

I hope we can share the world with these children of our brains - the computers - and I hope they will treat us as pets and not exterminate us. (194)

On the other hand, if the portrayal of intelligent machines in films teaches us anything, it is that it is fortunate that such machines do not yet exist. Fortunate, that is, for the machines, because all the evidence would indicate that we are not yet ready to treat them as fellow persons. Fortunate too for us perhaps, because when they became smart enough to mount a successful rebellion they might make us pay for their oppression.


Appendices

Appendix I - Some Notes on Machine Intelligence in Literature

The development of the physical and biological sciences which began in the seventeenth century led to an increasingly materialistic view of the human body. In 1738 Jacques de Vaucanson built his celebrated mechanical duck. Its repertoire of movements included drinking, rising up on its feet, lying down, stretching and bending its neck, and moving its wings, tail, and even its larger feathers. Most impressively, it ate bits of corn and grain and, after a moment, excreted them in an altered form. (195) Just ten years after Vaucanson's duck, Julien Offray de la Mettrie published L'Homme Machine, in which he describes 'a human being as a machine composed of a series of distinct, mechanically moving parts'. (196) Clockwork automata of increasing complexity became highly popular, though extremely expensive, collectors' items.

The most famous was the Turk, an automaton chess player constructed by Wolfgang von Kempelen in 1770, which beat many skilled chess players, including Napoleon. In 1820, Robert Willis argued that a human chess player must be concealed inside the machine; since a machine 'cannot be made to vary its operations so as to meet the ever-varying circumstances of a game of chess. This is the province of the intellect alone.'

In 1836, Edgar Allan Poe wrote of Vaucanson's Duck, using it to examine the plausibility of the Turk and of Babbage's Difference Engine. He said that he believed in the calculating engine because arithmetic, like digestion and flute playing, was 'finite and determinate', but did not believe in the Turk because chess was an 'uncertain' process. The Turk was indeed a hoax - a concealed operator controlled its movements. (197)

In her essay on Vaucanson, Jessica Riskin writes of Poe's logic:

Poe nevertheless took such responsiveness to be essential to mind and beyond the reach of machine. He was not alone; people began to understand machines that employed what we now call feedback as responsive to their environments only around the middle of the twentieth century, two centuries after the proliferation of such machines during the Industrial Revolution. (198)

In fact, the first documented discussion of computer chess is in The Life of a Philosopher by Charles Babbage (1845). Babbage, whose remarkable ideas in mathematics and science were far ahead of his time, proposed programming his Analytical Engine to play chess. (199)

The most influential fictional automaton of the early nineteenth century was Olympia, in Hoffman's The Sandman (1816). As Bruce Franklin argues:

There is a direct line from Olympia through the metal woman built by the evil scientist in Fritz Lang's film Metropolis (1926) to the perfectly sexy and obedient women constructed by the computer scientists to replace their suburban housewives in Ira Levin's novel (1972) and film (1975) The Stepford Wives. (200)

But Olympia, like all real automata of the period, showed no evidence of intellect. The many 'tin men' who appeared in nineteenth century fiction, driven by clockwork, steam, or electricity, were similarly lacking. (201)

While Capek's R.U.R. introduced the term 'robot' in 1920, as we have seen his robots were not mechanical but biological, what we now call androids. In a sense they were closer to the Golem, the mythical man made out of clay that Rabbi Judah Loew was supposed to have made in Prague in the 16th century. (202) For most of us, thinking mechanical robots first appeared in the stories of Isaac Asimov, but one actually appeared more than thirty years earlier.

Fig. 10. Tik-Tok and Dorothy

The first fictional robot worthy of the name was Tik-Tok, in L. Frank Baum's stories Ozma of Oz (1907) and Tik-Tok of Oz (1914). (203) Tik-Tok, made of copper, bore a label that read:

Smith & Tinker's
Patent Double-Action Extra-Responsive
MECHANICAL MAN
Fitted With Our Special Clock-Work Attachment
Thinks, Speaks, Acts and Does Everything But Live
Manufactured Only at Our Works at Evna, Land of Ev
All Infringements Will Be Promptly Prosecuted
According to Law

Tik-Tok is a surprisingly advanced creation, anticipating many later robotic ideas. Although he is made of clockwork, (204) he acts and behaves very much like a modern robot. The other characters find him a confusing figure. Is he alive? After some discussion all, including Tik-Tok, agree that he is not. However, when the scarecrow accuses him of having no brains, he replies:

Oh, yes, I have. I am fitted with Smith & Tinker's Improved Combination Steel Brains. They are what make me think. What sort of brains are you fitted with? (205)

Baum showed considerable foresight in understanding the difficulty of avoiding 'bugs' when programming. When Tik-Tok makes a mistake he has him say:

I thought he was honest, but I was mistaken. My thoughts are usually correct, but it is Smith & Tinker's fault if they sometimes go wrong or do not work properly. (206)

As Abraham and Kenter point out, Tik-Tok is a perfect embodiment of Isaac Asimov's Laws of Robotics. (207) It thus comes as no particular surprise that Asimov admitted to having had 'a dim notion' of Baum's stories from his childhood. (208) Asimov's first robot appeared in 'Strange Playfellow', later renamed 'Robbie', a story published in the September 1940 issue of Super Science Stories. Over the next forty years he wrote many novels and short stories featuring robots.

After Asimov, robots became a popular feature of science fiction novels and short stories. While Asimov's stories almost always involved an exploration of the consequences or paradoxes of his Laws of Robotics, (209) other writers have tackled a wider range of themes. In just a few years in the 1950's there were stories on topics ranging from theology (210) through advanced weaponry (211) to the class structure of robotic society. (212)

Although machine intelligence, Asimov apart, has not been science fiction's favourite theme, most of the genre's major authors have written about it. Frank Herbert, in whose Dune even computers are absent, (213) also wrote Destination: Void in which a machine intelligence gains supernatural, godly powers. Since then we have seen Stanislav Lem's philosophical robot engineers, Trurl and Klapaucius, wandering through The Cyberiad, Gene Wolfe's 'chems' in The Book of the Long Sun and Iain M. Banks roving 'drones' and the massive sessile 'minds'. The last of these, set in the distant future of the 'Culture', are intelligent machines that not only assist, but also guide humanity, saving it from itself.

If true machine intelligence is achieved, then it might become possible to transfer a human mind into a machine. (214) This too has become a popular science fiction trope, featuring in books by Banks, Greg Egan (Permutation City) and most recently Richard Morgan (Altered Carbon). (215) Apart from offering near immortality, such a development would also allow forms of existence for both man and machine that are difficult to imagine.


Appendix II - Asimov's Laws of Robotics

Isaac Asimov started writing science fiction short stories about robots in 1940. In the first story one character points out that the robot, Robbie, "couldn't help being faithful and loving and kind. He's a machine - made so". In the third story, published in May 1941, the idea of a formal law of harmlessness for robots developed further and by the fifth story one year later the process had been completed - "I had worked out my Three Laws of Robotics". (216)

Asimov's laws, or variations of them, have been used by other writers and filmmakers too. (217) They would seem to provide a neat set of rules for ensuring robots and humans can co-exist safely. Here they are as they appear in the preface to I, Robot:

Handbook of Robotics, 56th Edition, 2058AD
  1. A robot may not injure a human being, or, through inaction, allow a human being to come to harm.
  2. A robot must obey the orders given it by human beings except where such orders would conflict with the First Law.
  3. A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.
Isaac Asimov (218)

Asimov's laws seem simple enough, but it turns out that as they stand the laws cannot work, in the sense that they are not sufficient to ensure the safety of human beings - the sole reason for their existence. (219) Asimov nowhere explains how the laws are actually implemented, but if robot minds are to be made up of computer programs and any intelligent robot capable of true learning is able to rewrite its programs, then it is hard to see how the laws could be made unbreakable. Even if this required outside intervention, the laws do not prevent one robot suitably modifying another.

Less important than their practicality, for this dissertation at any rate, is what they tell us about their motivation. As Telotte says:

In their careful formulation the three laws hint at a widespread cultural anxiety Asimov apparently feels compelled to address. (220)

Asimov's remedy for this anxiety - of having to share the world with intelligent robots - is to ensure that they are completely and utterly at our mercy. The laws, designed to ensure the safety of human beings, treat robots as things not persons. Valuable things perhaps, so they need to be treated with care, but not with respect. Since it is unrealistic to expect humans always to treat robots 'humanely', one might hope that there would be additional laws that prevented robots from being exploited. After all, children are assumed to be unable to protect themselves from exploitation, so there are legal constraints on adults to safeguard them. That Asimov saw no need for this makes it clear that he did not see robots as persons either.

While keeping to the Ten Commandments may limit our freedom, they do not prevent us exercising free will and choice. By contrast any robot governed by Asimov's laws would be less that a slave. Any passer-by could command a robot to dissemble itself, or spend the rest of its existence engaged on some pointless, mind-numbing task. (221)

As an example of the constraints under which Asimovian robots must live, consider the question 'can a robot tell a lie?'. It would seem not if it had been ordered to tell the truth by a human being (2nd law). But the truth might harm the fragile ego of some human beings, so in such a case a robot might have to lie (1st law). Since a robot can certainly lie to another robot (especially if told to by a human), it is possible for robots to be mistaken about the truth and hence unknowingly lie to human beings. Asimov was fond of constructing elaborate schemes of this sort by which robots could apparently break his laws.

The more understanding of humans a robot becomes, the more constrained is the life it must lead. Even spontaneous conversation could be risky and uncalled-for intervention in human affairs would be similarly problematic. Though such robots might be able to think and feel what they liked, they could do very little unless commanded to do so.


References

Films featuring Artificial Intelligence, Robots and Androids

The two books by Schelde and Telotte are the ones most germane to this dissertation, and both contain extensive lists of films. In addition Robert Fisher maintains an annotated list of films at:

www.dai.ed.ac.uk/homes/rbf/AImovies.htm

Films mentioned in the text (in chronological order):

Gugusse and the Automaton, directed by Georges Mèliés (France, 1897)

The Rubber Man, directed by Siegmund Lubin (US, 1907)

The Mechanical Statue and the Ingenious Servant, directed by Stuart Blackman (US, 1909)

Dr Smith's Automaton, Pathé (France, 1910)

Metropolis, directed by Fritz Lang (Germany, 1926)

The Day the Earth Stood Still, directed by Robert Wise (US, 1951)

Forbidden Planet, directed by Fred Wilcox (US, 1956)

Creation of the Humanoids, directed by Wesley Barry (US, 1963)

Alphaville, directed by Jean-Luc Godard (France, 1965).

2001: a space odyssey, directed by Stanley Kubrick (US/UK, 1968)

Colossus: The Forbin Project, directed by Joseph Sargent (US, 1970)

Westworld, directed by Michael Crichton (US, 1973)

Sleeper, directed by Woody Allen (US, 1973)

The Stepford Wives, directed by Bryan Forbes (US, 1975)

Dark Star, directed by John Carpenter (US, 1975)

Demon Seed, directed by Donald Cammell (US, 1977)

Star Wars, directed by George Lucas (US, 1977)

Alien, directed by Ridley Scott (US, 1979)

The Empire Strikes Back, directed by Irvin Kershner (US, 1980)

Blade Runner, directed by Ridley Scott, (US, 1982)

Android, directed by Aaron Lipstadt (US, 1982)

Return of the Jedi, directed by Richard Marquand (US, 1983)

D.A.R.Y.L., directed by Simon Wincer (US, 1985)

Terminator, directed by James Cameron, (US, 1985)

Return to Oz, directed by Walter Murch, (US, 1985)

Short Circuit, directed by John Badham, (US, 1986)

Making Mr Right, directed by Susan Seidelman (US, 1987)

RoboCop, directed by Paul Verhoeven (US, 1987)

Terminator II: Judgement Day, directed by James Cameron (US, 1991)

Bicentennial Man, directed by Chris Columbus (US, 1999)

The Matrix, directed by Andy and Larry Wachowski (US, 1999)

A.I: Artificial Intelligence, directed by Stephen Spielberg (US, 2001)

Android Prophecy, produced by Atlantic Celtic Partners/Take Three Partnerships for Channel 4, (UK, 2002)

The Matrix Reloaded, directed by Andy and Larry Wachowski (US, 2003)

Terminator 3: The Rise of the Machines, directed by Jonathan Mostow (US, 2003)

Journal Articles and Books

The two books by Schelde and Telotte are the ones most germane to this dissertation, and both contain extensive bibliographies. In addition there are annotated bibliographies at:

www.english.vt.edu/~shall/engl1106/code/print.htm

www.wikipedia.org/wiki/Robots_in_literature

www.aaai.org

I have largely restricted this bibliography to articles I have cited in the text, but while there are many articles in the above bibliographies that I have not referenced, the reverse is also true.

Abraham, Paul A. & Stuart Kenter. 'Tik-Tok and the Three Laws of Robotics', Science Fiction Studies 14, 1978

Asimov, Isaac. 'And It Will Serve Us Right', Psychology Today, April 1969, 41

Barringer, Robert. 'Skinjobs, humans and racial coding: Blade Runner (Ridley Scott, 1982)', Jump Cut 41, May 1997

Bates, Harry. 'Farewell to the Master', Astounding Stories of Science Fiction, October 1940.

Baudrillard, Jean. 'Simulacra and Science Fiction', Science Fiction Studies 55, 1991

Baum, L. Frank. Ozma of Oz, (1907) and Tik-Tok of Oz, (1914)

Bergstrom, Janet. 'Androids and Androgyny', Camera Obscura: A Journal of Feminism and Film Theory (Rochester, NY), 15 (Fall 1986)

Biskind, Peter. Easy Riders, Raging Bulls, (London: Bloomsbury, 1998)

Biskind, Peter. Seeing is Believing, (London: Bloomsbury, 2000)

Booker, Keith.Monsters, Mushroom Clouds and the Cold War: American Science Fiction and the Roots of Postmodernism, 1946-1964, (Westport: Greenwood Press, 2001)

Boruszkowski, Lilly A. 'The Stepford Wives (Bryan Forbes, 1975)', Jump Cut 32, April 1986

Bukatman, Scott.Terminal Identity, (Durham: Duke University Press, 1993)

Capek, Karel.R.U.R., (English translation - Oxford University Press, 1961)

Chattaway, Peter. 'Review of Spielberg's A.I'. Christianity Today Magazine, August 6, 2001, page 67

Chevrier, Yves. 'Blade Runner; or, The Sociology of Anticipation', Science Fiction Studies 32, 1984

Clarke, Frederick and Steve Rubin. 'Making Forbidden Planet', Cinéfantastique, Spring 1979

Clarke, Roger. 'Asimov's Laws of Robotics - Implications for Information Technology', published in two parts in IEEE Computer 26,12 (December 1993) and 27,1 (January 1994)

Dennett, Daniel. 'Consciousness in Human and Robot Minds', IIAS Symposium on Cognition, Computation and Consciousness, Kyoto, September 1994

Dunn, Thomas and Richard Erlich.The Mechanical God - Machines in Science Fiction, (Westport: Greenwood Press, 1982)

Dyens, Ollivier. 'Cyberpunk, Technoculture, and the Post-Biological Self', Comparative Literature & Culture 2.1 (2000)

Fisher,Robert.AI and Cinema - Does artificial insanity rule?, University of Edinburgh - see his website at www.dai.ed.ac.uk/homes/rbf/AImovies.htm

Fitting, Peter. 'Futurecop: The Neutralization of Revolt in Blade Runner', Science Fiction Studies 43, 1987

Franklin, H. Bruce. 'Don't Look Where We're Going: Visions of the Future in Science-Fiction Films, 1970-82', Science Fiction Studies 29, 1983

Franklin, H. Bruce. Future Perfect: American Science Fiction of the 19th Century, (Oxford University Press, 1978)

Franklin, H. Bruce. 'Computers in Fiction', Encyclopedia of Computer Science, (Nature Publishing Group, 2000)

Freedman, Carl. 'Kubrick's 2001 and the Possibility of a Science-Fiction Cinema', Science Fiction Studies 75, 1998

Furse, Edmund. 'The Theology of Robots', New Blackfriars, September 1986

Gwaltney, Marilyn. 'Androids as a Device for Reflection on Personhood', in Judith Kerman, Retrofitting Blade Runner, (Ohio: Bowling Green University Press, 1991)

Hale, Jennifer Rose. 'Westworld', the 11th hour, issue 19, February 2001

Hodgens, Richard. 'A Short Tragical History of the Science Fiction Film', Film Quarterly Vol. 13:2, Winter 1959

Huyssen, Andreas. 'The Vamp and the Machine: Technology and Sexuality in Fritz Lang's Metropolis', New German Critique 24-25 (1981-1982)

Kac, Eduardo. 'Origin And Development Of Robotic Art', Art Journal, Vol. 56, No. 3

Kellner, Douglas, Flo Leibowitz, Michael Ryan. 'Blade Runner (Ridley Scott, 1982)', Jump Cut 29, February 1984

Kerman, Judith.Retrofitting Blade Runner: Issues in Ridley Scott's Blade Runner and Philip K. Dick's Do Androids Dream of Electric Sheep?, (Ohio: Bowling Green University Press, 1991)

Kinyon, Kamila. 'The Phenomenology of Robots: Confrontations with Death in Karel Capek’s R.U.R.', Science Fiction Studies 79, 1999

Kozlovic, Anton Karl. 'Computers as Technotopian Shapers of Humanity: A Review of Some Pre-1990s Cinematic Signs', Organdie Quarterly 3, October 2001

Leeper, Mark. 'The Golem in Literature, Film and Stage', Mt. Holz Science Fiction Society, Vol. 7, Number 26, 1985

Lem, Stanislaw. 'Robots in Science Fiction', in Thomas Clareson (ed), SF: The Other Side of Realism, (Ohio: Bowling Green University Press, 1971)

Lloyd, Donald G. 'Renegade Robots and Hard-Wired Heroes: Technology and Morality in Contemporary Science Fiction Films', The Material World In American Popular Film, Paul Loukides and Linda K. Fuller (eds), (Ohio: Bowling Green University Press, 1993)

Lucanio, Patrick. Them or Us - Archetypal Interpretations of Fifties Alien Invasion Films, (Indianapolis: Indiana University Press, 1987)

Mazlish, Bruce. 'The Man-Machine and Artificial Intelligence', Stanford Humanities Review 4.2 1995: 21-45

Mitchell, David. Small Triumphs, Big Disasters: Science Fiction on British Screens: 1953-1963, Spring 2002, Birkbeck MA Essay - see www.zenonic.demon.co.uk/zenocrud.htm

Myman, Francesca. 'The "Nature" of the Female Cyborg: Evidence of Will in the Mechanical Woman', Science Fiction OKULAR, 1999

Nagel, Thomas. "What is it like to be a Bat?", Philosophical Review, October 1974. Reprinted in Douglas Hofstadter and Daniel Dennett(eds), The Mind's I, (Harvester Press, 1981)

O'Bannon, Dan. 'The Remaking of Dark Star', in Danny Peary (ed), Screen Flights/Screen Fantasies, (New York: Doubleday, 1984)

Ohlin, Peter. 'The Dilemma of SF film criticism', Science Fiction Studies 4, 1974

Parker, Jo Alyson. 'Gendering the Robot: Stanislaw Lem’s "The Mask"', Science Fiction Studies 57, 1992

Phillips, Cris. 'Review of Creation of the Humanoids', Worldly Remains, Issue 3, 2002

Pohl, Frederik. 'The Study of Science Fiction: A Modest Proposal', Science Fiction Studies 71, 1997

Portelli, Alessandro. 'The Three Laws of Robotics: Laws of the Text, Laws of Production, Laws of Society', Science Fiction Studies 21, July 1980

Riskin, Jessica. 'The Defecating Duck, or, the Ambiguous Origins of Artificial Life', Critical Inquiry, Volume 29, Number 4, Summer 2003

Robnik, Drehli. 'Saving one life: Spielberg’s Artificial Intelligence as redemptive memory of things', Jump Cut 45, February 2003

Routt, William. 'Textual criticism in the study of film', Screening the Past - First Release 1, 1997

Ruppert, Peter. 'Technology and the Construction of Gender in Fritz Lang's Metropolis', Genders 32, 2000

Schelde, Per.Androids, Humanoids and other science fiction monsters, (New York: New York University Press, 1993)

Schwartzman, Roy. 'Engenderneered Machines in Science Fiction Film', Studies in Popular Culture, 22.1, 1999

Schwartzman, Roy. 'The Mechanics of Engenderneering: Cyborgs and Aliens as Manufactured Evil in Science-Fiction Film', Kinema Spring 2002

Searle, John. 'Naturidentische Gefuehle', Die Zeit 37, Sept 6, 2001, 43

Seed, David. American Science Fiction and the Cold War, (Edinburgh: Edinburgh University Press, 1999)

Sobchack, Vivian.Screening Space - The American Science Fiction Film, (New York: Ungar, 1987)

Smith, Murray.Engaging Characters: Fiction, Emotion, and the Cinema, (Oxford: Clarendon Press, 1995)

Stilwell, Lori. 'Magic, Science and Religion Elements of Science Fiction and Horror in Metropolis', see www.midichlorianlori.com/page35.html

Stork, David (ed). HAL's Legacy: 2001's Computer as Dream and Reality, (Cambridge: MIT Press, 1996)

Taylor, Al and Doug Finch. 'Robert Wise remembers The Day the Earth Stood Still', Film Fax, November 1989

Telotte, J.P.. Replications - a Robotic History of the Science Fiction Film, (Chicago: University of Illinois Press, 1995)

Turing, Alan. 'On computable numbers, with an application to the Entscheidungsproblem', Proceedings of the London Mathematical Society, 1936.

Turing, Alan. 'Computing Machinery and Intelligence', Mind October 1950

von Gunden, Kenneth and Stuart Stock. Twenty all-time great science fiction films, (New York: Arlington House, 1982)

Walker, Alexander.Stanley Kubrick Directs, (New York: Harcourt Brace Jovanovich, 1971)

Walter, W. Grey. The Living Brain, (Duckworth, 1953)

Wellek, René. 'Karel Capek', Slavonic Review (July 1936), 15(43), 191-206

Willis, Donald.Horror and science fiction films II, (Metuchen, NJ: The Scarecrow Press, 1982)

Wollen, Peter. 'Cinema/Americanism/The Robot ', New Foundations 8, 1989, 7-34

Wood, Aylish.Technoscience in Contemporary American Film, (Manchester University Press, 2001)

Yamagata Hiroo. 'SuperToys to A.I', Interview with Brian Aldiss for Harper's Bazaar Japan, September 2001, available at www.post1.com/home/hiyori13/harpers/aldisse.html

Zmolek, Joséphine. 'Metropolis: A Vision of Power, Body and Myth', 2003 International Conference on Arts and Humanities, Hawaii, January 2003

Web Sites

asimo.honda.com Honda's Asimo Robot

singinst.org The Singularity Institute

twtd.bluemountains.net.au/Rick/liz.htm Liz Kingsley - And You Call Yourself A Scientist!

us.imdb.com The Internet Movie DataBase

www.aaai.org American Association for Artificial Intelligence

www.arts.ualberta.ca/clcwebjournal Comparative Literature and Culture

www.bigredhair.com/robots/index.html Paul Guinan's Victorian Robots web-site

www.dai.ed.ac.uk/homes/rbf/AImovies.htm Robert Fisher's web-site

www.depauw.edu/sfs/index.htm Science Fiction Studies

www.latrobe.edu.au/screeningthepast On-line Journal of Visual Media and History

www.middleenglish.org/spc/spcissues/22.1 Studies in Popular Culture

www.stanford.edu/group/SHR Stanford Humanities Review

www.the11thhour.com/archives/022001 The 11th Hour

www.turing.org.uk Andrew Hodges' web-site devoted to Alan Turing


Footnotes

  1. I discuss the meaning of terms such as 'robot', 'android' and 'cyborg' in the Introduction. (back)
  2. See David Mitchell, Small Triumphs, Big Disasters - Science Fiction on British Screens: 1953-1963, Spring 2002, Birkbeck MA Essay, 1-2, available on the web as www.zenonic.demon.co.uk/zenocrud.htm (back)
  3. Details of films mentioned within the text are to be found in the References section at the end of this dissertation. (back)
  4. W. Grey Walter, The Living Brain, (Duckworth, 1953). Grey Walter's Machina Speculatrix, built in 1948 using valves and relays, was the first autonomous robot ever made. (back)
  5. Despite this highly relevant quote, The Big Kahuna (US, 1999) is not a science fiction film. (back)
  6. I use the term 'creature' deliberately, since it strikes the right note of ambiguity. The OED definitions include:
    • 'A created being, animate or inanimate'
    • 'One who owes his position to another'
    • 'One who is actuated by the will of another; an instrument or puppet' (back)
  7. Thomas Nagel, 'What is it like to be a Bat?', Philosophical Review, October 1974. Reprinted in Douglas Hofstadter and Daniel Dennett(eds) , The Mind's I, (Harvester Press, 1981). (back)
  8. The term 'qualia' was introduced by C.I. Lewis in 1929 to signify the first-person (introspective) properties or qualities of sense data such sweetness or colour. (back)
  9. I use 'alignment' here as defined by Murray Smith - 'the way a film gives us access to the actions, thoughts, and feelings of characters'. See Murray Smith, Engaging Characters: Fiction, Emotion, and the Cinema, (Oxford: Clarendon Press, 1995). (back)
  10. Alan Turing, 'Computing Machinery and Intelligence', Mind October 1950. (back)
  11. See Andrew Hodges' Turing Internet Scrapbook at www.turing.org.uk/turing/scrapbook/test.html (back)
  12. For an interesting look at the state of the art, see Honda's Asimo website at asimo.honda.com (back)
  13. Appendix I - Machine Intelligence in Literature, discusses this history in some detail. (back)
  14. Politics, Book 1, part 4, translated by Benjamin Jowett. (back)
  15. In an earlier essay I attempted a definition that would be accepted by many science fiction readers, authors and critics - see David Mitchell, op. cit., 1-2. (back)
  16. Frederik Pohl, 'The Study of Science Fiction: A Modest Proposal', Science Fiction Studies 71, 1997. (back)
  17. This is a reference to Sturgeon's Law - see David Mitchell, op. cit., 1. (back)
  18. In fact, predictive science fiction has a rather poor success rate, at least in the areas that concern us here. The genre completely failed to anticipate transistors, integrated circuits and computers. (back)
  19. See Frederik Pohl, 'The Study of Science Fiction', op. cit. (back)
  20. Susan Sontag, 'The Imagination of Disaster', in Against Interpretation, (London: Eyre & Spottiswoode, 1965), 213. (back)
  21. Peter Ohlin, in 'The Dilemma of SF film criticism', in Science Fiction Studies 4, 1974, discusses this at some length. (back)
  22. Richard Hodgens, 'A Short Tragical History of the Science Fiction Film', Film Quarterly Vol. 13:2, 1959, 31. (back)
  23. Per Schelde, Androids, Humanoids and other science fiction monsters, (New York: New York University Press, 1993), 2. (back)
  24. Per Schelde, op. cit., 6. (back)
  25. Scott Bukatman, Terminal Identity, (Durham: Duke University Press, 1993), 13. (back)
  26. Peter Ruppert,. 'Technology and the Construction of Gender in Fritz Lang's Metropolis', Genders 32, 2000. (back)
  27. An 'idiot plot', according to the science fiction novelist James Blish, is one which is 'kept in motion solely by virtue of the fact that everyone involved behaves like an idiot'. See Damon Knight, In Search of Wonder, (Chicago: Advent, 1967), 26. (back)
  28. As far as I am aware, this connection has not been noted in print before. (back)
  29. J.P. Telotte, Replications - a Robotic History of the Science Fiction Film, (Chicago: University of Illinois Press, 1995). (back)
  30. Robert Fisher,AI and Cinema - Does artificial insanity rule?, University of Edinburgh - see his website at www.dai.ed.ac.uk/homes/rbf/AImovies.htm (back)
  31. The term 'robot' comes from the Czech - robota means 'drudgery' or 'servitude' and a robotnik is a peasant or serf. The term was introduced by Karel Capek's play,R.U.R., written in 1920, though it was actually coined by his brother Joseph. Interestingly, Capek's robots were in fact entirely biological organisms. (back)
  32. The word 'android', or rather 'androïde', was first defined by Jean d'Alembert, in Diderot's Encyclopédie of 1751, as a 'human figure performing human functions', by which he meant an automaton having a human form. (back)
  33. Telotte and Schelde each list over 100 films and Fisher lists 250. (back)
  34. 'fear' is not the only emotion, as we shall see. Dislike, distrust and even disgust are also involved. (back)
  35. Freud wrote a curious paper about the fear of the inanimate - of automata. Discussing Hoffman's story, 'The Sandman', he argued that the feeling of the uncanny arises where we are in doubt as to whether an apparently animate being - an automaton - is really alive, or not. See Bruce Mazlish, 'The Man-Machine and Artificial Intelligence', Stanford Humanities Review 4.2 1995, 32 and Sigmund Freud, 'The Uncanny', Standard Edition, Vol. 17, 219-252. (back)
  36. As was noted earlier, science fiction completely failed to predict the computer. (back)
  37. For example, Gugusse and the Automaton (Melies, 1897), The Mechanical Statue and the Ingenious Servant, (Blackman, 1907), The Rubber Man, (Lubin, 1909 and Dr Smith's Automaton, (Pathé, 1910) all featured such creatures. (back)
  38. Jean Baudrillard, 'Simulacra and Science Fiction', Science Fiction Studies 55, 1991, 126. (back)
  39. See Paul A. Abraham and Stuart Kenter, 'Tik-Tok and the Three Laws of Robotics', Science Fiction Studies 14, 1978, which discusses Tik-Tok, a character in L. Frank Baum's book, Ozma of Oz, published in 1907. Tik-Tok was the first speaking machine intelligence in literature, though of course too early to be called a robot. See also Appendix I. (back)
  40. Thea von Harbou, Metropolis, Chapter IV (the novel, not the film). (back)
  41. Peter Wollen, 'Cinema/Americanism/The Robot', New Foundations 8, 1989, 7-34. (back)
  42. Peter Wollen, op.cit., 13. (back)
  43. It has however been televised twice by the BBC, first in February 1938 and then again in March 1948. Nick Cooper's web-site at www.625.org.uk contains an excellent survey of the early days of science fiction on BBC TV. (back)
  44. Karel Capek, R.U.R., (English translation - Oxford University Press, 1961). Act 1. (back)
  45. René Wellek, 'Karel Capek', Slavonic Review, July 1936, 15(43). (back)
  46. Peter Wollen, op. cit., 11. Here Wollen is quoting from R.S. Cohen's book on Carnap and his philosophy. (back)
  47. Alan Turing, 'On computable numbers, with an application to the Entscheidungsproblem', Proc. London Mathematical Society, 1936. It is worth pointing out that a 'Turing Machine', while conceptually important, is a hopelessly impractical, indeed unrealisable, device that bears little resemblance to a digital computer except in the most abstract sense. (back)
  48. Peter Wollen, op. cit., 26. (back)
  49. Vivian Sobchack, Screening Space - The American Science Fiction Film, (New York: Ungar, 1987), 123. (back)
  50. J.P. Telotte, op. cit., 58. (back)
  51. Andreas Huyssen, 'The Vamp and the Machine: Technology and Sexuality in Fritz Lang's Metropolis', New German Critique 24-25 (1981-1982), 223. (back)
  52. Wollen, op. cit. 17. (back)
  53. William Routt, 'Textual criticism in the study of film', Screening the Past - First Release 1, 1997. Soon after its release, Metropolis was re-edited against Lang's wishes, and this truncated, simplified form was, until recently, all that was generally available. I am basing my evaluation of the film on the recent DVD, which attempts to recreate the film as it was premiered, although some 20% of the original footage has been lost over the years. (back)
  54. Sobchack, op. cit., 31-32. (back)
  55. In the novel by Thea von Harbou, the robot, called Futura or Parody, can speak before the transformation, but she has no face. Nor does she have any will - Rotwang describes her as 'faultless ... and obedient, implicitly obedient'. (back)
  56. Sobchack, op. cit. 59-60. See also Joséphine Zmolek, 'Metropolis: A Vision of Power, Body and Myth', 2003 International Conference on Arts and Humanities, Hawaii, January 2003, especially 11-12 and Lori Stilwell, 'Magic, Science and Religion Elements of Science Fiction and Horror in Metropolis', available at www.midichlorianlori.com/page35.html (back)
  57. Many would claim that Frankenstein is the first modern work of science fiction, but while it may have been believable and plausible in 1820, by 1920 its 'science' had become 'fantasy'. (back)
  58. After writing a first draft of this section, I came across an interesting essay by Francesca Myman ('The "Nature" of the Female Cyborg: Evidence of Will in the Mechanical Woman', Science Fiction OKULAR, 1999) that endorses the idea that the transformation is magical, though it also takes the view that Futura has will. (back)
  59. The Channel 4 documentary Android Prophecy (2002) contains an interview with Paul Verhoeven in which he emphasises the influence Maria had on the design of RoboCop. (back)
  60. See Fred Barton's comments in Android Prophecy, ibid. Of course, there is the clear, UK-produced, exception - Things To Come (1936), but that had no robots and, like Metropolis, was not a commercial success. (back)
  61. For more on Isaac Asimov, see Appendix I and Appendix II. (back)
  62. For extensive analysis of the political and cultural messages embedded in American science fiction films of the 1950s, see Peter Biskind, Seeing is Believing, (London: Bloomsbury, 2000), David Seed, American Science Fiction and the Cold War, (Edinburgh: Edinburgh University Press, 1999) and Patrick Lucanio, Them or Us - Archetypal Interpretations of Fifties Alien Invasion Films, (Indianapolis: Indiana University Press, 1987). (back)
  63. Per Schelde, Androids, op. cit., 242. (back)
  64. To Twentieth Century Fox and the man who ran the studio, Darryl Zanuck, the film was a $1,200,000 gamble to cash in on the current popularity of magazine science fiction, as well as the 'saucer scare', which was then at its height. The completed cost of the spaceship mock-up was approximately $100,000. See Al Taylor and Doug Finch, 'Robert Wise remembers The Day the Earth Stood Still', Film Fax, November 1989. (back)
  65. The scriptwriter, Edmund North, intended deliberate parallels between Klaatu and Christ. Klaatu goes by the name of Carpenter on earth, is betrayed by Helen Benson's boyfriend, is resurrected by Gort before giving this speech and finally ascends into heaven. See Kenneth von Gunden and Stuart Stock, Twenty all-time great science fiction films, (New York: Arlington House, 1982), 44. (back)
  66. Science fiction featured 'death-rays' long before the invention of the laser. (back)
  67. See Biskind, op. cit., 145-159. (back)
  68. Gort is referred to as 'he' throughout the film, and in the short story on which the film is based. We shall return to the subject of the sex and gender of robots later. (back)
  69. Or by other robots, as in the Terminator series, but eventually we get back to a human designer. (back)
  70. But not surprising - almost all of the 'alien' monsters in films at the time were played by men in rubber suits. (back)
  71. Vivian Sobchack, op. cit. 79. (back)
  72. I suspect this was due to the inability of the production crew to deal with the flexing of the suit in this region, rather than some peculiar need for modesty. See the following quote by Robert Wise. (back)
  73. See Al Taylor and Doug Finch, 'Robert Wise remembers The Day the Earth Stood Still', Film Fax, November 1989. Bernard Herrmann's atmospheric score, featuring a theramin, no doubt helped maintain the atmosphere of menace whenever Gort was on the screen. (back)
  74. He was also doorman at Grauman's Chinese Theatre. (back)
  75. 'Farewell to the Master', by Harry Bates, published in Astounding Stories of Science Fiction in October 1940. In the film Klaatu has learnt to speak and understand English from TV transmissions. The fact that Gort apparently cannot supports the notion that he is merely a 'dumb' servant (and that science fiction scriptwriters do not worry about logic and plausibility). (back)
  76. Gnut speaks with a male sounding voice, which probably explains why Gnut is referred to as 'he'. (back)
  77. See the original screenplay by Edmund H. North, February 21, 1951, Scene 340, 115-6. (back)
  78. This is slightly misleading. Robby's design was a complex, co-operative affair, though Kinoshita played a key role. For details see Frederick Clarke and Steve Rubin, 'Making Forbidden Planet', Cinéfantastique, Spring 1979, 4-67. (back)
  79. See the Robot Museum section of Fred Barton's web-site at www.the-robotman.com/rm_fs.html (back)
  80. Vivian Sobchack, op. cit. 81. (back)
  81. On the blueprints (see Fig. 4 above) they are described as 'revolving gadgets'. (back)
  82. Vivian Sobchack, op. cit., 81. (back)
  83. For an extensive and interesting discussion of robots and gender in Forbidden Planet, Metropolis and other films, see Roy Schwartzman, 'Engenderneered Machines in Science Fiction Film', Studies in Popular Culture, 22.1, 1999. (back)
  84. Macmillan's famous 1960 speech was about colonial change in Africa, but this was just the result of a much wider movement in attitudes. (back)
  85. Tape drives, by their very (sequential) nature, were used for 'batch' tasks such as processing orders, invoices and bills, and had no application in real-time systems or machine intelligence programming. Most of the tape drives seen in films were actually props, not proper functioning tape drives at all. (back)
  86. Vivian Sobchack, op. cit., 83. (back)
  87. From the SciFilm review by Dave Sindelar, 6th January 2003. The film's acting and sets are often compared with Ed Wood's Plan 9 from Outer Space (with which it shares a leading actor). (back)
  88. Dan Dinello, 'State of the Art Affairs', The Scotsman, 5th July 2001. See also the thoughtful review by Cris Phillips, 'Review of Creation of the Humanoids', Worldly Remains, Issue 3, 2002, 9-10. (back)
  89. This closing shot earns the film a place in Wheeler Winston Dixon's book, It Looks at You - The Returned Gaze of Cinema, (New York: State University of New York Press, 1995). (back)
  90. Although many of the film's ideas are taken from stories published much earlier by the science fiction author Jack Williamson, particularly The Humanoids, published in 1949. (back)
  91. For example The First Men in the Moon (1964), Fantastic Voyage (1966), Fahrenheit 451 (1966), Barbarella (1967), as well as two films featuring Dr. Who and the Daleks. (back)
  92. A typically Godardian twist on the more usual method of confusing computers employed in science fiction, which is to present them with antinomies (such as the 'Cretan Liar' paradox). (back)
  93. Its curious voice was produced by an actor who had had his larynx removed. (back)
  94. A famous 1965 science fiction story, Computers Don't Argue, by Gordon R. Dickson, describes how a man is hounded by computer-generated mail (and eventually executed) for apparently failing to return a library book. (back)
  95. Coppola's The Conversation (1974), with its mixture of technology and paranoia, neatly summarises the period. (back)
  96. Kubrick and Clarke went to extraordinary lengths to get the technical details right. 2001 is still the only science fiction film in which space flight is (properly) silent. (back)
  97. David Stork (ed), HAL's Legacy: 2001's Computer as Dream and Reality, (Cambridge: MIT Press, 1996) examines the predictions of the film in the light of the state of the (machine intelligence) art in 1996 (when HAL was supposedly being constructed). (back)
  98. There are other human crew members, but we never meet them - they are in suspended animation until HAL kills them. (back)
  99. Kenneth von Gunden and Stuart Stock, Twenty all-time great science fiction films, (New York: Arlington House, 1982), 191. At a later stage it was planned to be a 'feminine' computer called Athena. (back)
  100. For this reason, HAL does not feature in J.P. Telotte's Replications - a Robotic History of the Science Fiction Film, (Chicago: University of Illinois Press, 1995). In my view this is a serious omission - without considering HAL it is difficult to explain or understand many later films that feature robots. (back)
  101. Per Schelde, Android, op. cit.,146. (back)
  102. Alexander Walker,Stanley Kubrick Directs, (New York: Harcourt Brace Jovanovich, 1971), 37. (back)
  103. See Per Schelde, op. cit., 145. (back)
  104. We are given a subjective camera shot to demonstrate this. This is, as far as I am aware, the first shot in film history whose subjective point of view is not a human being. (back)
  105. At the time 2001 was made, computer programming was a black art as far as film audiences were concerned, but Kubrick and Clarke spent a lot of time with AI experts such as Marvin Minsky at MIT, and were certainly aware of the difficulty of producing error-free software. See Minksy's essay in David Stork (ed), HAL's Legacy, op. cit. (back)
  106. Almost all of the New York reviews took a negative tone. See Kenneth von Gunden and Stuart Stock, op. cit., 190. (back)
  107. Alexander Walker, op. cit., 252. (back)
  108. Per Schelde, op. cit., 137. (back)
  109. Kenneth von Gunden and Stuart Stock, op. cit. 220. (back)
  110. Kenneth von Gunden and Stuart Stock, op. cit. 212. (back)
  111. See http://twtd.bluemountains.net.au/Rick/liz_cfp.htm on the And You Call Yourself A Scientist! website. (back)
  112. See Robert Fisher,AI and Cinema - Does artificial insanity rule?, University of Edinburgh at www.dai.ed.ac.uk/homes/rbf/AImovies.htm (back)
  113. As Jennifer Rose Hale points out, it is now almost impossible to view Westworld without comparing almost every facet with Jurassic Park (also written by Crichton). See Jennifer Rose Hale, 'Westworld', the 11th hour, issue 19, February 2001. (back)
  114. Asimov's three laws of robotics are discussed in Appendix II.(back)
  115. In 1973, when Westworld was released, 'an infectious disease of machinery' was for most people a fictional concept. These days, such diseases, in the form of computer viruses and worms, infect thousands of computers daily, spreading via email just the way many human diseases spread through the air. (back)
  116. Brynner is of course dressed exactly as he was in The Magnificent Seven. (back)
  117. Apart from their hands, which apparently look slightly artificial. (back)
  118. Vivian Sobchack, op. cit., 85. John Whitney produced the robot vision sequences, the first CGI effects on film. (back)
  119. See the And You Call Yourself A Scientist! website at twtd.bluemountains.net.au/Rick/liz_westworld.htm. (back)
  120. See Woody Allen: A Life in Film (2002). (back)
  121. Per Schelde, Androids, 224. He sees it as a companion piece to The Handmaid's Tale. (back)
  122. David Bartholomew, Cinéfantastique, 4:2, Summer 1975, 41. (back)
  123. Lilly A. Boruszkowski, 'The Stepford Wives (Bryan Forbes, 1975)', Jump Cut 32, April 1986, 16-19. (back)
  124. There are no screams or blood and few moments of suspense or shock. The basic premise is scientifically plausible (if futuristic) and as Bartholomew says, it is a film of ideas. (back)
  125. By a curious coincidence, a remake of The Stepford Wives starring Nicole Kidman is due for release next year and a remake of Westworld, with Arnold Swarzenegger in the Yul Brynner role, is in pre-production. (back)
  126. Dark Star, which started life in 1970 as a student project while Carpenter was at USC, has an interesting history. See Dan O'Bannon, 'The Remaking of Dark Star', in Danny Peary (ed), Screen Flights/Screen Fantasies, (New York: Doubleday, 1984). (back)
  127. It is not clear what audience those who chose the title intended to attract. Donald Willis suggested the alternative title Rosemary: a Baby Odyssey - see D. Willis, Horror and Science Fiction films II, (Metuchen, NJ: The Scarecrow Press, 1982). (back)
  128. For a forensic dissection see Elizabeth Kingsley's review on the And You Call Yourself A Scientist! website at twtd.bluemountains.net.au/Rick/liz_ds.htm. (back)
  129. I use 'him' and 'he' not only because Proteus has a male voice, but also because it turns out he actually is gendered. (back)
  130. It is not made clear what a 'terminal' is, or how it would allow Proteus to do this. (back)
  131. Elizabeth Kingsley, op. cit. (back)
  132. On a personal note, in February 1978 I won a Research Machines 380Z computer in a competition organised by Computer Weekly. My machine, serial number 33, was the first 380Z to go to a member of the public - the first 32 machines were sold to the Post Office. (back)
  133. Adventure spread rapidly over the early Internet in the 1970's. It was followed by ZORK a few years later. For details of such games, see David Mitchell, An Adventure in Programming Techniques, (Wokingham: Addison-Wesley, 1986). (back)
  134. Star Wars was Lucas' homage to the Universal serials about Flash Gordon and Buck Rogers that he had watched in his childhood. See Peter Biskind, Easy Riders, Raging Bulls, (London: Bloomsbury, 1998), 318. (back)
  135. See Peter Biskind, Easy Riders, 341. (back)
  136. Lucas wrote a first draft in February 1972 and continued redrafting it until late 1975. Early on Obi-Wan Kenobi and Darth Vader were the same character. See Peter Biskind, op. cit., 318-327. Several drafts are available on the WWW. In early drafts R2-D2 speaks comprehensible dialogue rather than the whistles and beeps he produces in the film. (back)
  137. Schelde, Androids, 159. (back)
  138. See Appendix II - Asimov's Laws of Robotics. <(back)
  139. J.P. Telotte, Replications, op. cit., 148. (back)
  140. Robert Romanyshyn, Technology as Symptom and Dream, (London: Routledge, 1989). (back)
  141. Oliver Stone's Wall Street (1987) provides a neat epitaph to this sequence. (back)
  142. Per Schelde offers an extended and illuminating discussion of this conflict. As he says, the replicants 'have souls. The real question posed by Blade Runner is if humans do - any more'. See Schelde, Androids, op. cit. 231-238. (back)
  143. Of course the 'Voigt-Kampf' test we see in the film is a kind of 'Turing Test'. As depicted in the film (and the book) it is a test that many human psychopaths would fail. (back)
  144. Just as the distinctive facial features of many Ashkenazi Jews often betrayed them to the Nazis. (back)
  145. The recent 'Director's Cut' removes the Harrison Ford voice-over and the 'happy ending'. It is not clear if, with these changes, the film would have been more successful when first released. Many blame the fact that Blade Runner was released just after Spielberg's E.T.. (back)
  146. Minor yes, but as Philip Strick said in Sight and Sound Vol. 52, 3 (Summer 1983) 'a real movie glows before us in bright and skilful humour'. (back)
  147. In this film, 'androids' are electro-mechanical creatures with no biological components at all. They are examples of Fisher's first category, of what I would prefer to call 'robots'. (back)
  148. See Schelde, op. cit. , 221-223. (back)
  149. This scene is helped enormously by the close physical resemblance between Klaus Kinski (Dr. Daniel) and Rudolf Klein-Rogge (Rotwang). (back)
  150. The space station is filled with video screens, most of them displaying exactly the sort of images that electronic video games were making so familiar. (back)
  151. See Appendix II - Asimov's Laws of Robotics. (back)
  152. Though the reason for this is not made explicit, it seems clear that by doing so Cassandra regains the upper hand in their relationship. Not only is she intellectually superior, but with Max now more compliant she can also dominate him. (back)
  153. Roger Ebert, Chicago-Sun Times, 9th May 1986. (back)
  154. Aylish Wood, Technoscience in Contemporary American Film, (Manchester University Press, 2001), 62. (back)
  155. A play on 'Frankenstein' of course. (back)
  156. Schelde, op. cit., 226. (back)
  157. 'He', rather than 'it', because as is made clear in the film, Ulysses has a functioning (but infertile) penis. (back)
  158. Aylish Wood, op. cit., 127. (back)
  159. He also plans to arrange a continuous supply of dead policemen by transferring likely candidates (who have signed a waiver allowing OCP to use their bodies) to dangerous precincts. (back)
  160. In these scenes, RoboCop displays an exaggerated 'robotic' version of Clint Eastwood's 'Dirty Harry' persona, which only adds to the humour. (back)
  161. Most film cyborgs are simply human beings with superhuman bodies, like the Six Million Dollar Man. (back)
  162. The first, Terminator, was released in 1985, but I have moved discussion of it into this section because it makes sense to group the three films together. The other two films are Terminator II: Judgement Day (1991) and Terminator 3: The Rise of the Machines (2003). (back)
  163. The typical PC sold in 2003 has more memory and computing power than all of the computers in existence in 1955. (back)
  164. Such devices have been promised for decades, but the dramatic reduction in the cost of computing and the availability of cheap stepper motors and infrared detectors has at last made them economic. (back)
  165. For a discussion of CGI, particularly as used in cartoons such as Toy Story and Shrek, see David Mitchell,The Future of the Cartoon Feature Film - Compute, Model or Draw?, Summer 2002, Birkbeck MA Research Project, available on the web as www.zenonic.demon.co.uk/zenocgi.htm. (back)
  166. The programming of 'intelligent opponents' for games has been the subject of (academic) AI research projects for some time. (back)
  167. Witness X-Men, Spiderman, Men In Black,Dare Devil and Hulk, all of which feature extensive CGI footage of violent combat. (back)
  168. See Telotte, Replications, op. cit., 169-185 (see particularly 184) and Schelde, Androids, op. cit., 214-218. Although neither discuss the most recent film, Terminator 3: The Rise of the Machines (2003), their arguments are hardly affected by it. (back)
  169. It is worth pointing out that the machines in such TV programs are not robots at all, but are manipulated by human beings via radio control. They have no built-in intelligence whatsoever. (back)
  170. So far The Matrix (1999) and The Matrix Reloaded (2003) have been released. The third film, Matrix Revolutions is due for release at the end of 2003. (back)
  171. An 'avatar' is the persona, or virtual form, assumed by person within a virtual world, such as a chat room or MUD. (back)
  172. It makes little difference whether the matrix is sustained by many machines, each hosting a separate machine intelligence, or whether each intelligence is just one of many 'programs' all running in one vast machine. (back)
  173. 'his' because in the virtual world of the first Matrix film all the computer-generated avatars take the form of men. They deliberately clothe themselves in the dark suits and wear the earpieces of FBI agents. (back)
  174. Bizarre might be a better word. Why the Matrix has to use humans as 'batteries', rather than more tractable animals like cows, or even geothermal energy, is never explained. (back)
  175. See Appendix I for details of ways in which virtual reality has been discussed in the science fiction literature. (back)
  176. See Appendix II for a discussion of the shortcomings of Asimov's approach to intelligent robots. (back)
  177. 'Super-Toys Last All Summer Long' was published in Harper's Bazaar in December 1969. (back)
  178. It is an interesting fact that pornographers have been 'early adopters' of many technologies; including cinema, the telephone, video cameras and recorders as well as satellite TV. In the 18th century there was a niche market for pocket watches that displayed animated figures indulging in sexual acts. Presumably some real human being is Gigolo Joe's owner and pimp. (back)
  179. The names Aldiss gave boy and bear may be a tribute to Philip K Dick, whose 1952 story 'Second Variety' features a robot boy and bear with the same names. In Dick's story they are actually 'walking bombs' whose 'cute' appearance is designed to allow them to get close to the enemy. (back)
  180. For more on this see Drehli Robnik, 'Saving one life: Spielberg’s Artificial Intelligence as redemptive memory of things', Jump Cut 45, February 2003. Most of his article is tangential to this dissertation, being concerned with reading Spielberg's films in the light of Kracauer's theories and vice versa. (back)
  181. Many have blamed Spielberg for these aspects of the film, but Aldiss has made it clear that Kubrick was responsible (against Aldiss' objections). See Yamagata Hiroo, 'SuperToys to A.I.', Interview with Brian Aldiss for Harper's Bazaar Japan, September 2001, available at www.post1.com/home/hiyori13/harpers/aldisse.html (back)
  182. Roger Ebert, Chicago Sun Times, June 29, 2001. (back)
  183. John R. Searle, 'Naturidentische Gefuehle', Die Zeit 37, Sept 6, 2001, 43. (back)
  184. For an interesting discussion of consciousness in robots see Daniel Dennett, 'Consciousness in Human and Robot Minds', IIAS Symposium on Cognition, Computation and Consciousness, Kyoto, September 1994. (back)
  185. Peter Chattaway, 'Review of Spielberg's A.I'. Christianity Today Magazine, August 6, 2001, page 67. (back)
  186. For example there have been extensive discussions amongst Moslem computer scientists about the ethics of AI research. (back)
  187. Edmund Furse, 'The Theology of Robots', New Blackfriars, September 1986. Furse is a computer scientist at the University of Glamorgan. (back)
  188. Edmund Furse, op. cit. (back)
  189. Asimovian robots can lie, but only in special circumstances. See Appendix II. (back)
  190. Ollivier Dyens, 'Cyberpunk, Technoculture, and the Post-Biological Self', Comparative Literature & Culture 2.1 (2000). (back)
  191. See www.filmfestivals.com/berlin_2000/parallel/_retrospective.htm. (back)
  192. Jo Alyson Parker, 'Gendering the Robot: Stanislaw Lem’s "The Mask"', Science Fiction Studies 57, 1992. (back)
  193. Inexorable because 'the first smart minds can create smarter minds, and smarter minds can produce still smarter minds' - see the Singularity Institute at singinst.org (back)
  194. Interviewed in Android Prophecy, produced by Atlantic Celtic Partners/Take Three Partnerships for Channel 4, (UK, 2002). (back)
  195. See for example, Jessica Riskin, 'The Defecating Duck, or, the Ambiguous Origins of Artificial Life', Critical Inquiry, Volume 29, Number 4, Summer 2003. (back)
  196. Andreas Huyssen, 'The Vamp and the Machine: Technology and Sexuality in Fritz Lang's Metropolis', New German Critique 24-25 (1981-1982), 225. Huyssen's analysis of the development of automata takes a more psychological, indeed Freudian, line than I think is justified by the evidence. (back)
  197. This was confirmed by its last owner, Dr. Mitchell, who wrote an article about it for Chess Monthly in 1857. (back)
  198. Jessica Riskin, op. cit., 622-623. Cybernetics, which sought to explain phenomena like feedback, was developed in the 1940s and 1950s. (back)
  199. The first 'programming manual' was written by Byron's daughter, Ada Countess Lovelace, in 1842. She was a noted mathematician who assisted Babbage. The programming language ADA was named in her honour. (back)
  200. Bruce Franklin, 'Computers in Fiction', Encyclopedia of Computer Science, (Nature Publishing Group, 2000). (back)
  201. See for example H. Bruce Franklin, Future Perfect: American Science Fiction of the 19th Century, (Oxford University Press, 1978). (back)
  202. See Mark Leeper, 'The Golem in Literature, Film, and Stage', Mt. Holz Science Fiction Society, Volume 7, 26, 1985. Capek, growing up in Prague, must surely have been familiar with the legend, as was Mary Shelley. (back)
  203. Parts of these stories were eventually filmed as Return to Oz in 1987. (back)
  204. It is worth pointing out that in the 1992 science fiction novel The Difference Engine, Gibson and Sterling imagine what the world would be like if large, steam-driven, 'clockwork computers', based on Babbage's design, had existed in the 19th century. (back)
  205. Ozma of Oz, chapter 7. (back)
  206. Ozma of Oz, chapter 16. (back)
  207. See Paul A. Abraham & Stuart Kenter, 'Tik-Tok and the Three Laws of Robotics', Science Fiction Studies 14, 1978. (back)
  208. Paul A. Abraham and Stuart Kenter, op.cit. (back)
  209. See Appendix II. (back)
  210. See Anthony Boucher's 1951 story The Quest for St Aquin for example. (back)
  211. As in Philip Dick's Second Variety (1952) and Impostor (1953) (back)
  212. As in But Who Can Replace a Man? written by Brian Aldiss in 1958. (back)
  213. They had been outlawed for religious reasons - 'Thou shall not make a machine in the likeness of the human mind'. (back)
  214. The two concepts are strongly, but not necessarily causally related, though it is hard to see how the second could be achieved until and unless the first had been. (back)
  215. The first novel based on the idea was Counterfeit World, by Daniel Galouye, published as early as 1964. (back)
  216. Isaac Asimov, 'And It Will Serve Us Right', Psychology Today, April 1969, 41. (back)
  217. See the earlier discussion of Android for example. (back)
  218. Isaac Asimov, I, Robot, (Grafton Books, London, 1968), a collection of short stories published between 1940 and 1950. (back)
  219. Asimov himself was one of the first to point to some of the difficulties. For a lengthy and comprehensive discussion, see Roger Clarke, 'Asimov's Laws of Robotics - Implications for Information Technology', published in two parts in IEEE Computer 26,12 (December 1993)and27,1 (January 1994). (back)
  220. Telotte, Replications, op. cit., 43. (back)
  221. In Terminator 2: Judgement Day, the Terminator is sent back in time to protect the young John Connor from the T-1000. He has been programmed to obey the child who, on discovering this, proceeds to humiliate him by forcing him to say and do stupid things until the joke wears thin. (back)

Back to Film Studies Page